isolation of human cd4/cd8 double-positive, graft-versus- host

12
of February 18, 2018. This information is current as Clone Restricted HY-Specific CD4 - HLA-DR7 Specific Regulatory T Cells and of a Novel - Protective, Minor Histocompatibility Antigen - Double-Positive, Graft-Versus-Host Disease Isolation of Human CD4/CD8 Rigal and Diane Scott Thomas, Pierre Tiberghien, Elizabeth Simpson, Dominique Farre, Caroline Addey, Marie-Laure Tartelin, Xavier Assia Eljaafari, Ozel Yuruker, Christophe Ferrand, Annie http://www.jimmunol.org/content/190/1/184 doi: 10.4049/jimmunol.1201163 December 2012; 2013; 190:184-194; Prepublished online 7 J Immunol Material Supplementary 3.DC1 http://www.jimmunol.org/content/suppl/2012/12/07/jimmunol.120116 References http://www.jimmunol.org/content/190/1/184.full#ref-list-1 , 19 of which you can access for free at: cites 52 articles This article average * 4 weeks from acceptance to publication Fast Publication! Every submission reviewed by practicing scientists No Triage! from submission to initial decision Rapid Reviews! 30 days* Submit online. ? The JI Why Subscription http://jimmunol.org/subscription is online at: The Journal of Immunology Information about subscribing to Permissions http://www.aai.org/About/Publications/JI/copyright.html Submit copyright permission requests at: Email Alerts http://jimmunol.org/alerts Receive free email-alerts when new articles cite this article. Sign up at: Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved. Copyright © 2012 by The American Association of 1451 Rockville Pike, Suite 650, Rockville, MD 20852 The American Association of Immunologists, Inc., is published twice each month by The Journal of Immunology by guest on February 18, 2018 http://www.jimmunol.org/ Downloaded from by guest on February 18, 2018 http://www.jimmunol.org/ Downloaded from

Upload: phamkhanh

Post on 05-Jan-2017

219 views

Category:

Documents


0 download

TRANSCRIPT

of February 18, 2018.This information is current as

CloneRestricted HY-Specific CD4−HLA-DR7

Specific Regulatory T Cells and of a Novel−Protective, Minor Histocompatibility Antigen

−Double-Positive, Graft-Versus-Host Disease Isolation of Human CD4/CD8

Rigal and Diane ScottThomas, Pierre Tiberghien, Elizabeth Simpson, DominiqueFarre, Caroline Addey, Marie-Laure Tartelin, Xavier Assia Eljaafari, Ozel Yuruker, Christophe Ferrand, Annie

http://www.jimmunol.org/content/190/1/184doi: 10.4049/jimmunol.1201163December 2012;

2013; 190:184-194; Prepublished online 7J Immunol 

MaterialSupplementary

3.DC1http://www.jimmunol.org/content/suppl/2012/12/07/jimmunol.120116

Referenceshttp://www.jimmunol.org/content/190/1/184.full#ref-list-1

, 19 of which you can access for free at: cites 52 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. Immunologists, Inc. All rights reserved.Copyright © 2012 by The American Association of1451 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 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 18, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Isolation of Human CD4/CD8 Double-Positive, Graft-Versus-Host Disease–Protective, Minor HistocompatibilityAntigen–Specific Regulatory T Cells and of a NovelHLA-DR7–Restricted HY-Specific CD4 Clone

Assia Eljaafari,*,†,‡ Ozel Yuruker,x Christophe Ferrand,{,‖ Annie Farre,*

Caroline Addey,x Marie-Laure Tartelin,†,‡ Xavier Thomas,# Pierre Tiberghien,{

Elizabeth Simpson,x Dominique Rigal,* and Diane Scottx,**,1

Minor histocompatibility (H) Ags are classically described as self-peptides derived from intracellular proteins that are expressed

at the cell surface by MHC class I and class II molecules and that induce T cell alloresponses. We have isolated three different

T cell populations from a skin biopsy of a patient suffering from acute graft-versus-host disease following sex-mismatched HLA-

identical bone marrow transplantation. The first population was: 1) CD4+/CD8+ double-positive; 2) specific for an HLA class I–

restricted autosomal Ag; 3) expressed a Tr1 profile with high levels of IL-10, but low IL-2 and IFN-g; and 4) exerted regulatory

function in the presence of recipient APCs. The second was CD8 positive, specific for an HLA class I–restricted autosomally

encoded minor H Ag, but was only weakly cytotoxic. The third was CD4 single positive, specific for an HLA-DR7–restricted HY

epitope and exerted both proliferative and cytotoxic functions. Identification of the peptide recognized by these latter cells

revealed a new human HY epitope, TGKIINFIKFDTGNL, encoded by RPS4Y and restricted by HLA-DR7. In this paper, we

show human CD4/CD8 double-positive, acute graft-versus-host disease–protective, minor H Ag–specific regulatory T cells and

identify a novel HLA-DR7/ HY T cell epitope, encoded by RPS4Y, a potential new therapeutic target. The Journal of Immu-

nology, 2013, 190: 184–194.

Minor histocompatibility (H) Ags are HLA-restrictedpeptides encoded by autosomal or sex-chromosomegenes (1). Their disparity between HLA-matched donor/

recipient pairs carries increased risk of graft failure and/or graft-versus-host disease (GVHD) following bone marrow transplan-tation (BMT) (2–4). One such is HY, targeted by T cells fromfemale donors following sex-mismatched BMT between HLA-identical siblings (5–7). However, as well as causing GVHD,

minor H Ag mismatches can also induce therapeutic graft-versus-leukemia (GVL), or graft-versus-tumor effects, depending on their

tissue distribution: expression of minor H Ags exclusively on

malignant cells or hematopoietic cells can facilitate GVL or graft-

versus-tumor, whereas ubiquitously expressed minor H Ags allow

GVHD (7, 8). Thus, following BMT and in the absence of graft

failure, re-emergent recipient hematopoietic cells, likely to be

malignant, can be targeted by donor T cells specific for recipient

minor H Ags. For example, MHC class I–restricted minor H Ags

HA-1 and HA-2 with expression limited to hematopoietic cells

have been shown to induce GVL by stimulating donor-derived

cytotoxic CD8+ T cells specific for recipient malignant cells (4,

9, 10). Another way to target recipient malignant cells is via se-

lective expression of the HLA restriction molecule of the minor

H epitope, whereas HLA-class I molecules are ubiquitously ex-

pressed, and HLA class II molecules are expressed primarily on

APCs, B cells, endothelial cells, and, in humans, activated T cells.

Leukemia, myeloma cells, and other tumor cells expressing HLA

class II can thus be targeted by HLA class II–restricted minor

H Ag–specific T cells, without leading to detrimental effects on

other tissues (11, 12). Thus, characterization of HLA class II–

restricted minor H Ags can identify new efficacious antitumor

T cell therapy targets.Several HLA class I–restricted minor H Ags have been identi-

fied in humans by screening of plasmid cDNA libraries, elution

of HLA-bound peptides, and genetic linkage analysis (13–16).

Recently, a novel strategy using whole-genome association scan-

ning reported characterization of 10 new HLA class I–restricted

minor H Ags (17). However, identification of HLA class II–

restricted minor H Ags has been more difficult. In humans, among

the several candidate genes that might encode HY epitopes, only

*Etablissement Francais du Sang Rhone Alpes, HLA Department, Lyon 69007,France; †INSERM U1060, Ouillins Cedex 69921, France; ‡Hospices Civils de Lyon,Lyon 69002, France; xSection of Immunobiology, Division of Immunology and In-flammation, Imperial College London, London SW7 2AZ, United Kingdom; {Eta-blissement Francais du Sang Bourgogne Franche-Comte, Besancon Cedex 25020,France; ‖INSERM Unite Mixte de Recherche 1098-SFR FED 4234, Besancon 25020,France; #Service d9Hematologie, Centre Hospitalier Lyon Sud, Pierre BeniteCedex 69495, France; and **Centre for Complement and Inflammation Re-search, Imperial College London, London SW7 2AZ, United Kingdom

1Current address: Centre for Complement and Inflammation Research, ImperialCollege London, London, U.K.

Received for publication April 23, 2012. Accepted for publication November 1, 2012.

This work was supported by the Blood Bank Center of Rhone Alpes, the LigueNationale contre le Cancer Rhone-Alpes, and the Medical Research Council (UnitedKingdom).

Address correspondence and reprint requests to Dr. Assia Eljaafari, INSERM U1060,Hospices Civils de Lyon Faculte de Medecine Lyon Sud, 165 Rue du Grand Revoyet,Ouillins Cedex 69921, France. E-mail address: [email protected]

The online version of this article contains supplemental material.

Abbreviations used in this article: BMT, bone marrow transplantation; DP, double-positive; GVHD, graft-versus-host disease; GVL, graft-versus-leukemia; H, histo-compatibility; HTLP, helper T lymphocyte precursor frequency; IMGT, ImMunoGe-neTics; LCL, lymphoblastoid cell line; rh, recombinant human; SP, single-positive;Treg, regulatory T cell.

Copyright� 2012 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201163

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

six—RPS4Y, UPS9Y, DDX3Y, UTY, TMSB4Y, and SMCY—encodeepitopes that are clinically relevant. Most of these are HLA classI–restricted (18), and only two, DDX3Y and RPS4Y, encode HLAclass II–restricted male-specific minor H Ags (6, 7, 19). To date,very few peptide epitopes have been characterized (20–23).To identify T cells able to recognize re-emergent recipient

leukemic cells, one line of investigation is to isolate donor T cellsspecific for mismatched minor H Ags expressed by recipient cells.Such T cells can be found within GVHD lesions (24). We initiallyisolated both CD8 single-positive (SP) and CD4/CD8 double-positive (DP) T cells from skin cultures of a male patient withmild GVHD following BMT from his HLA-identical sister. Afterseveral rounds of stimulation with recipient APCs, CD4 SP T cellsexhibiting both helper and cytotoxic functions against HLA-DR7/HY–expressing cells emerged from this culture. Using retroviralgene expression followed by synthetic peptides designed usingpeptide–MHC binding databases, we identified a new HLA-DR7/HY peptide encoded by the RPS4Y gene.In this study, we also functionally characterized the CD4/CD8

DP T cells, showing that they exerted regulatory function fol-lowing recognition of their target autosomal HLA class I–restrictedminor H Ag. This demonstrates the existence of minor H Ag–specific CD4/CD8 DP T cells with regulatory function with po-tentially controlling clinical acute GVHD. Minor H–specific reg-ulatory CD8+ T cells with low avidity and the ability to diminishconcurrent CD8+ T cells responses have been described in mouseand human models (25, 26). Moreover, adoptive transfer of suchregulatory T cells (Treg) appeared to contribute to male grafttolerance in a murine sex-mismatched skin transplant model (25)and to natural tolerance to familial minor H Ags acquired duringhuman pregnancy (26). Thus, the use of Treg able to counteractpotentially devastating acute GVHD appears to be a promisingapproach for trials of therapeutic tolerance induction in BMT.

Materials and MethodsMedium and reagents

RPMI 1640 (Life Technologies, Eggenstein, Germany) was supplementedwith L-glutamine (2 mM), penicillin (100 IU/ml), streptomycin (100 mg/ml), NaHCO3 (1.5 mg/ml), and 10% pooled, heat-inactivated human ABserum. Recombinant human (rh)GM-CSF, rhIL-2, rhIL-4, and rhTNF-awere purchased from R&D Systems (Abingdon, U.K.). HLA typing wasassessed by serology followed by oligonucleotide typing. Anti–HLA classI (W632), anti–HLA-DR (L243), and anti–HLA-DP (B7-21) mAbs wereprovided by J. Chopin (Hospital Cochin, Cochin Institute for MolecularGenetics, Paris, France), and anti–HLA-DQ (SPVL3) mAb was fromImmunotech (Marseille, France). Anti-CD4 and anti-CD8 mAbs werepurchased from Immunotools (Friesoythe, Germany).

Generation of minor H Ag–specific T cell lines and clones

The patient was grafted with the HLA-identical bone marrow from hissister: A*02:05/*68:01; B*14:01/*44:03; C*08:02:*16:01; DRB1*07:01;DQB1*02:02; DPB1*04:01/*11:01. The conditioning regimen consisted ofcyclophosphamide 60 mg/kg/d, for 2 d, and total-body irradiation (12 Gy,fractionated). GVHD was controlled with methotrexate and cyclosporin A.A mild and acute GVHD occurred 1 mo after BMT and was histologicallyclassified as grade 2. This GVHD resolved rapidly. A skin biopsy wasperformed during the GVH episode. The skin extract was cultured for 2 wkwith 20 IU/ml rIL-2 in RPMI 1640 complete medium. A chimerism studydemonstrated that 100% of skin T cells were of donor origin (not shown).Skin T cells were expanded by several rounds of stimulation with 30 Gy–irradiated pretransplant recipient APCs, and then cultured for 2 wk with20 UI/ml IL-2 and after the third round with 50 Gy–irradiated EBV-Blymphoblastoid cell lines (LCLs) as APCs, generated from the recipientpre-BMT. This protocol gave rise to T cell lines.

These lines were cloned as previously described (27) by limiting dilutionat cell concentrations of 4, 1, or 0.4 cells/well in 96-well round-bottommicrotiter plates (BD Biosciences) in the presence of 2.5 3 105/ml PBMCplus 0.5 3 105/ml irradiated recipient EBV-B LCLs and 10 U/ml rIL-2.The T cell clones were expanded in the presence of 0.5 3 105 recipient

EBV-B LCLs and rIL-2. Their phenotype was analyzed by FACScan usingPE/FITC-labeled CD4 and CD8 Abs (BD Biosciences). Twelve and 13clones were isolated, respectively, from the CD4/CD8 DP and CD4 SPT cell lines. Results of representative clones (three or four tested per ex-periment) and/or cell lines are shown in the figures and tables as stated.

Cytofluorometry

The Abs used for flow cytometry were FITC- or PE-conjugated mouseantihuman CD3, CD4, CD8, CD56, CD69, and CD62L. All of the aboveAbs were purchased from BD Pharmingen (San Diego, CA). The culturedcells were collected, washed twice, and then resuspended in 200 ml PBScontaining 0.1% BSA. These cells were stained with specific labeled Absor appropriate isotopic controls for 15 min at 4˚C. The cells were incu-bated on ice for 30 min and washed with PBS containing 0.1% BSA andthen fixed with a 1% paraformaldehyde solution. Analyses were performedusing FACScan and CellQuest Software (BD Medical Systems). At least1 3 104 cells were analyzed in live gate with FACScan (BD MedicalSystems).

MLR assays

A total of 1 3 104 T cell lines or clones were incubated in triplicate with33 104 irradiated EBV-B LCLs in 96-well round-bottom microtiter plates.After 3 d, 1 mCi/well [3H]methylthymidine was added for 16–18 h. Cellswere collected using a Filtermate 196 multiple harvester (Packard Instru-ments). [3H]Thymidine incorporation was measured in a TopCount liquidscintillation counter (Packard Instruments).

Helper T lymphocyte precursor frequency modified assay

Graded numbers of third-party responder PBMC were stimulated with 3 3104 recipient EBV-B LCLs in round-bottom 96-well microtiter plates.Sixteen replicates and four dilutions (104, 5 3 103, 2.5 3 103, and 1.5 3103) were made. CD4/CD8 DP T cells were added at two different Treg/effector ratios, 1:4 or 1:1. The 24-h supernatants were harvested andassayed using the IL-2–dependent CTLL2 cell line. [3H]Thymidine in-corporation at 24 h was measured as above. Wells were defined as positiveif above mean cpm + 3 SD of wells with recipient cells alone. Results areexpressed as number of negative wells.

RT-PCR amplification and sequencing protocols

Total RNA from clones or lines was isolated from 1 3 106 cells using theReagent Kit (Promega, Madison, WI). Total RNAwas converted into first-strand cDNA using an oligo(dT) primer (Amersham Pharmacia Biotech,Orsay, France) and avian myeloblastosis virus reverse transcriptase,according to the manufacturer’s specifications (Promega).

PCR amplification (30 cycles) was carried out using 25 V regionsequence-specific 59 sense primers for TCR-Vb families and a 39 antisenseCb primer. As a positive internal control, 59 sense and the 39 antisense Cregion primers were included. Cycles consisted of 95˚C denaturation, 57˚Cprimer annealing, and 72˚C extension steps, 1 min each. PCR was carriedout in a Biomed Thermocycler 60 (Biomed Instruments) using 2.5 U TaqDNA polymerase (Cetus) in a solution containing 4 pmol/ml primers, 0.5mM each 2’-deoxynucleoside 5’-triphosphate, 50 mM KCl, 10 mM Tris-HCl (pH 8.4), 4 mM MgCl, and 5 mg sample. PCR products were se-quenced by Genoscreen (Lille, France). The TCR sequences were analyzedusing Internet ImMunoGeneTics (IMGT) database (http://www.imgt.org/IMGT_vquest/vquest?livret=0&Option=humanTcR).

ELISA

A total of 1.5 3 105/ml T cell lines or clones were stimulated with 3 3105/ml EBV-B LCLs in X-VIVO-20 medium (Cambrex) without serum.Supernatant was removed after 24 h. Cytokine concentrations were eval-uated by standard commercial ELISA following the manufacturers’instructions: IL-4, IL-10, IL-2, IFN-g (BioSource International), and TGF-b (R&D Systems).

T cell epitope identification

Two Y-chromosome genes, DDX3Y and RPS4Y, were cloned into the ret-roviral expression vector pMIGR1 and transfected into the Phoenix-Ampho helper-free amphotropic cell line that had the internal ribosomeentry site–CD8 surface marker introduced downstream of the gag-polconstruct (28) using Lipofectamine 2000 (Invitrogen, Paisley, U.K.).Virus-containing supernatant was harvested from cells expressing highlevels of CD8, filtered, concentrated by high-speed centrifugation, andused to transduce female HLA-DR7 EBV-B LCLs by spinfection. Briefly,8 ml concentrated virus-containing supernatant was mixed with 8 ml

The Journal of Immunology 185

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

polybrene (4 mg/ml; Sigma-Aldrich) and added to ∼15 3 106 HLA-DR7EBV-B LCLs. The cells were plated (1 ml/well) into a 24-well tissue-culture plate (BD Biosciences, Oxford, U.K.) and centrifuged for 90 minat 760 3 g. A total of 1 ml RPMI 1640 (Life Technologies) supplementedas described above but with 10% FCS instead of human AB serum wasthen added for 3-d culture. Cells were sorted for expression of high CD8,and the levels of DDX3Y and RPS4Y expression were confirmed byquantitative real-time PCR.

HY-specific CD4+ T-cells were stimulated with the DDX3Y and RPS4Ygene-transduced female EBV-B LCLs. Proliferation and IFN-g productionwere, respectively, measured by [3H]thymidine incorporation at day 3 andELISA at day 1. For mapping the MHC class II–restricted T cell epitope,several long peptides with potential DRB1*0701 binding were synthesizedfrom the regions of RPS4Y differing from RPS4X. They were identifiedfrom MHC–peptide binding databases SYFPEITHI, Net-MHC, Propred,and HotSpot Hunter. The peptides were tested at concentrations from 100mM to 10 nM using female EBV-B-LCL expressing HLA-DRB1*07:01 asAPCs. The short peptide was identified by testing a series of 15–17 aasynthetic peptides from within in the long peptide eliciting the positiveresponse.

ResultsIsolation of CD4/CD8 DP, CD8, and CD4 SP T cells from anacute GVHD skin biopsy

One month after BMT between the sex-mismatched HLA-identicalsiblings, the male recipient was diagnosed with acute skin GVHD.T cells were isolated from his skin biopsy following culture with20 U/ml IL-2. After 14 d, phenotypic analysis revealed two pop-ulations of cells, one CD4/CD8 DP and the other CD8 SP. TheCD4/CD8 DP population represented ∼60% of the culture. Overall,T cells isolated from the GVHD skin biopsy after 14 d culturewere CD3+CD562, and 95.9% expressed TCRa/b (Fig. 1A). CD8SP and CD4/CD8 DP populations were separated using anti-CD4

magnetic beads. The sorted CD4/CD8 DP cells were stimulatedwith recipient APCs and cultured without IL-2 for another week,when they expressed CD25 (84%) and CD69 (58%) (Fig. 1B).Following further restimulations, a new population expressingonly CD4 emerged. This new population was then isolated usinganti-CD8 magnetic beads (Fig. 1C).Thus, three different populations were obtained from the GVHD

skin biopsy: CD8 SP cells, CD4 SP cells, and CD4/CD8 DP cells.

Vb TCR spectratyping of the three populations

T cells were analyzed using immunoscope (Fig. 2). The GVHDskin-derived T cells after 14 d in culture (Fig. 2A) showed askewed, oligoclonal Vb TCR repertoire. Following furtherstimulation with recipient APCs, the Vb TCR repertoire of thethree subpopulations was more skewed. Six fewer Vb TCR werepresent in the CD8 SP and CD4 SP populations compared withthe CD4/CD8 DP. Furthermore, only one or two previously un-detected Vb TCR appeared in each respective population (Fig.2B). Encircled panels in Fig. 2B indicate differences in the VbTCR profiles of CD4 or CD8 SP compared with the CD4/CD8 DPcells. Comparison of the long-term cultured and sorted CD4/CD8

FIGURE 1. Isolation of three distinct subpopulation of T cells from

a GVHD skin lesion biopsy. (A) T cells isolated from a GVHD skin lesion

biopsy were phenotyped. CD4/CD8 DP and CD8 SP T cells were found.

Their expression of CD3, CD56, and TCRa/b is shown. This figure is

representative of two experiments. (B) Following sorting with anti-CD4–

coated magnetic beads, the DP cells were stimulated with recipient APCs.

They were maintained for 1 wk in culture without IL-2. The level of CD25,

CD69, and CD62L expression is shown. (C) Following culture of the DP

cells, anti-CD8–coated magnetic beads were used to isolate a third pop-

ulation, which was CD4 SP.

FIGURE 2. Vb TCR spectratyping of the three subpopulations. The Vb

TCR spectratyping of the initial D14 population (A), which contained

a mixture of CD8 SP cells and CD4/CD8 DP T cells, was compared with

those of the sorted CD8 SP, CD4/CD8 DP, or CD4 SP subpopulations (B).

Circles show the difference in the Vb TCR profiles of the CD4 or CD8 SP

T cells compared with those of the CD4/CD8 DP cells. Arrowheads show

loss or appearance of a Vb TCR comparing long-term cultures of CD4/

CD8 DP cells with unsorted day 14 skin T cells. (C) Vb TCR spectratyping

of clones 2, 7, and 10 is shown. In addition, the CDR3 region of these three

clones has been sequenced and characterized with the help of the IMGT

database.

186 NEW HUMAN MINOR H REGULATORY AND HY-SPECIFIC T CELLS

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

DP cells with the unsorted T cells after 14 d culture showed thesorted CD4/CD8 DP population losing just one Vb TCR, probablyrelated to the CD8 profile, because it was found in the CD8 SP, butnot the CD4 SP population. A previously undetected Vb TCR wasalso observed in the sorted CD4/CD8 DP cells and in the CD4 SPcell immunoscope profile. The finding of so few Vb TCR profilemodifications at the time of the initial CD4/CD8 DP cell screeningsuggested the GVHD skin T cells isolated after 14 d culture werealready oligoclonal, perhaps specific for one or a small number ofminor H Ags.

CD4/CD8 DP cells proliferate in response to but have nocytotoxic activity against recipient PBMC and recognize anautosomal minor H Ag presented by the HLA-B*14:01molecule

To evaluate the function of the CD4/CD8 DP cells, we stimulatedthem with donor, recipient, or HLA-mismatched PBMC. Highlevels of proliferation against recipient, but not donor or mis-matched PBMC, were observed. Cytotoxic function was tested inparallel, showing only weak CTL activity against recipient but notdonor or HLA class I–mismatched targets (Fig. 3). We determinedwhich HLA molecule was the restriction element of cloned CD4/CD8 DP T cells. Twelve recipient-specific clones were isolated, ofwhich five, along with the cell line, were stimulated with recipientAPCs in the presence of mAb against HLA class I, -DR,- DQ, or-DP molecules. Results showed that anti–HLA class I, but notanti–HLA class II mAb, anti-CD4, nor anti-CD8 inhibited theproliferation of CD4/CD8 DP cloned T cells and the line (Table I).Using anti-Bw6 and anti-Bw4 sera, HLA-B*14:01 was identifiedas the restriction molecule (Table I). PBMC from male and femaledonors sharing HLA-B*14:01 with recipient APCs were usedto investigate possible HY specificity: APCs from three HLA-B*14:01 males were tested. Only one was recognized by the CD4/CD8 DP T cell line, indicating that the specificity was for an

FIGURE 3. CD4/CD8 DP T cells are able to proliferate but not to exert

CTL activity following recognition of a minor H Ag on recipient cells.

Sorted CD4/CD8 DP T cells were stimulated with donor, recipient, or cells

from two to three HLA-mismatched controls (A). T cell proliferation or

cytolytic activity was measured by either [3HT] incorporation or [51Cr]

release (B). Results of proliferation assays are the mean 6 SD of three

experiments. Results of cytolytic assays are representative of two experi-

ments. Table

I.TheCD4/CD8DPTcellsrecognizean

HLA-B*14:01–restricted

autosomal

Ag

Responder

↓APC

↓No.Ab

Anti–HLA

Class

IAnti–HLA

DR

Anti–HLA

DP

Anti–HLA

DQ

Anti-CD4

Anti-CD8

Cellline

Recipient

11,0776

237

3,6246

52

15,5296

791

ND

16,2956

1021

ND

ND

Clone8

Recipient

14,5646

2264

4,0586

51

38,3676

1633

39,2636

1324

37,4916

1096

ND

ND

Clone11

Recipient

3,568

6228

7546

55

5,4196

419

3,308

6412

4,888

6416

ND

ND

Clone2

Recipient

4,054

6225

1,9486

71

ND

ND

ND

5,748

6168

5,152

6129

Clone10

Recipient

11,3556

460

9166

39

ND

ND

ND

11,6896

218

10,6776

255

Clone7

Recipient

9,211

6125

9026

49

ND

ND

ND

8,332

6369

7,904

6206

Responder

↓APC

↓No.Ab

Anti–HLA

Class

IAnti–HLA

DR

Anti–HLA

DP

Anti–HLA

DQ

Anti-BW4

Anti-BW6

Cellline

Recipient

17,9746

583

ND

ND

ND

ND

ND

2,167

6236

Clone8

Recipient

3,540

6218

1,6626

80

5,0116

258

5,3746

27

3,197

6184

4,233

6210

1,429

619

APC

→Donor↓

Recipient↓

HLA

B*14:01

HLA

B*14:01

HLA

B*14:01

HLA

B*14:01

HLA

B*14:01

HLA

B*14:01

Sex

Fem

ale

Male

Fem

ale

Fem

ale

Fem

ale

Male

Male

Male

Cellline

14756

14

15,0966

1043

5446

35

2136

11

4096

45

5726

22

7,423

6829

4066

23

Anti–HLA

classI,classII,-CD4,and-CD8Absandspecificanti-BW4,orBW6sera

wereusedto

inhibitproliferationoftheCD4/CD8DPTcellclones

orlinein

response

torecipientAPCs.Fem

aleandmalecellsfrom

differentdonors

each

expressingHLAB*1401wereusedas

APC.Resultsareexpressed

asmean6

SEM

ofcpm

of[3H]Tincorporation.

Resultsin

boldface

representstatisticallysignificantresultswithtwo-tailedunpairedttest

(p,

0.05).

The Journal of Immunology 187

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

autosomal Ag, not HY (Table I). Vb spectratyping of three of theclones, 2, 7, and 10 (Fig. 2), indicated a high degree of similaritywith the DP cell line, perhaps not unsurprisingly because the threeclones were also CD4/CD8 DP (not shown).At the clonal level, clones 2 and 7 appeared to share the same

TCRVB, TCRVB 7, whereas clone 10 expressed TCRBV 9.

The CD4/CD8 DP cells exhibit specific regulatory effects in thepresence of recipient APCs

We analyzed the cytokine profile of the CD4/CD8 DP T cell linefollowing stimulation with recipient APCs. The cells secreted lowlevels of IFN-g and TGF-b, moderate levels of IL-4, and highlevels of IL-10, even at 24 h poststimulation, but no IL-2 (TableII). These CD4/CD8 DP T cells were therefore tested for regu-

latory function. A population of CD3+ T cells, T90, obtained froma fully HLA-mismatched blood donor and therefore able respondto both donor and recipient APCs, was used in a primary MLR.Their proliferative response was measured in the presence or ab-sence of different ratios of CD4/CD8 DP T cells. The T90 cellsshowed a strong alloresponse against donor, recipient, and third-party APCs, as expected. However, a clear dose-dependent inhi-bition was observed when CD4/CD8 DP T cell line (Fig. 4A) orclones (not shown) were added to the MLR in the presence ofrecipient but not donor or HLA-mismatched APCs (Fig. 4A isrepresentative of the cell line and clones). This demonstrated thatonce activated by recognition of the specific minor H Ag, theCD4/CD8 DP T cells exerted regulatory function. However, anti–IL-10 mAb did not reverse inhibition (data not shown). Because

FIGURE 4. CD4/CD8 DP T cells exhibit regulatory activity. (A) Proliferation of sorted T90 CD3+ T cells was measured in response to APCs from donor,

recipient, or HLA class I– or class II–mismatched control and in the presence or absence of different ratios of CD4/CD8 DP T cells. The IL-2 respon-

siveness of the CD4/CD8 DP cells was compared with that of T90 cells (B) and the CD8 SP T cells (D). (C) HTLP assays were performed with responder

T90 cells in the presence or absence of the CD4/CD8 DP cells. The CTLL-2 cell line was used to measure IL-2 production. (E) Proliferation of T90 and

CD4/CD8 DP T cells was measured in the presence of different types of PHA, with and without 20 UI/ml rIL-2. T cell proliferation was measured by [3HT]

incorporation. (A, B, D, and E) Results are the mean 6 SD of three experiments. (C) Values correspond to 16-plicates.

Table II. Cytokine expression pattern of the CD4/CD8 DP T cells

IL-2 (IU/ml) IL-4 (pg/ml) IL-10 (pg/ml) IFN-g (IU/ml) TGF-b(pg/ml)

Cytokines Donor Recipient Donor Recipient Donor Recipient Donor Recipient Donor Recipient

APC24 h ,0.2 ,0.2 7 299 7 1370 ,0.2 10.8 60 26748 h ,0.2 ,0.2 ,5 242 ,5 1276 ,0.2 32.4 64 29572 h ,0.2 ,0.2 ,5 206 ,5 816 ,0.2 43.3 63 314

IL-2 (IU/ml) IL-4 (pg/ml) IL-10 (pg/ml) IFN-g (IU/ml) TGF-b (pg/ml)

24 h 5.3 6 2 768 6 212 1230 6 52 105 6 46 188 6 73

In the top portion of the table, the CD4/CD8 DP T cell line was stimulated with recipient or donor EBV-BLCL for 24, 48, or72 h. In the bottom portion of the table, results represent mean 6 SEM of four different experiments in which the CD4/CD8 DPcell line was stimulated through their TCR with recipient EBV-BLCL or anti-CD3+CD28 mAbs for 24 h. Cytokine secretionwas measured by ELISA and expressed in IU/ml or pg/ml.

188 NEW HUMAN MINOR H REGULATORY AND HY-SPECIFIC T CELLS

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

the DP cells expressed quite high levels of CD25 (Fig. 1B), wemeasured the IL-2 responsiveness of both CD4/CD8 DP T andT90 cells and found that of the CD4/CD8 DP T cell line muchhigher than that of the T90 cells (Fig. 4B). Moreover, using thehelper T lymphocyte precursor frequency (HTLP) assay, IL-2levels were significantly reduced in the presence of the CD4/CD8 DP cells. Even with low numbers of CD4/CD8 DP cells,almost all of the 16 wells were negative for IL-2 (Fig. 4C).Comparison of the CD4/CD8 DP cells and CD8 SP cells isolatedfrom the skin biopsy also showed that the IL-2 responsiveness wasmuch higher in the CD4/CD8 DP cell line (Fig. 4D). Finally,stimulation of the CD4/CD8 DP cells with PHA showed thesecells to be dependent on rIL-2. Irrespective of PHA source, and incontrast to the T90, CD4/CD8 DP cells did not proliferate in thepresence of PHA alone. When 20 UI/ml rIL-2 was added, pro-liferation of the CD4/CD8 DP T cell line matched that of T90 cells(Fig. 4E). These data indicate that the CD4/CD8 DP cells are ableto regulate T cell proliferation, probably due to their high re-sponsiveness to, and therefore consumption of, IL-2.

CD4+ T cells proliferate and exhibit cytotoxic activity onrecognition of an HLA-DR7–restricted HY epitope

The CD4 SP T cell line isolated from the skin biopsy was analyzedfollowing stimulation with recipient APCs in the presence of anti–HLA class II mAbs. Anti–HLA-DR, but not anti–HLA-DQ orHLA-DP mAb, significantly inhibited proliferative responses(Table III). The recipient was homozygous for HLA-DRB1*07:01:following stimulation of the CD4 SP cell line or clones with APCsfrom HLA-DRB1*07:01 males or females, we observed a clearDR7-restricted anti-HY response because the T cells responded toeach male but not to female APCs (Table III). CTL activityagainst recipient target cells by the CD4 SP T cells was found andthe restriction element confirmed as HLA-DRB1, as cytotoxicitywas significantly blocked by anti–HLA-DR but not anti–HLAclass I mAb (p = 0.0143). Cells from an unrelated HLA-DRB1*07:01 male donor were also lysed by the CD4 SP T cells(Table III).

Different patterns of minor H Ag recognition and cytokinemRNA expression by CD4/CD8 DP T cells and CD4 SP T cells

The different pattern of minor H Ag recognition by CD4/CD8 DPand CD4 SP T cell lines was confirmed in proliferative assays witheither HLA-B*14:01 or HLA-DRB1*07:01 PBMC as stimulators.As shown in Fig. 5A, HLA-B*14:01 APCs specifically activatedthe CD4/CD8 DP T cells, whereas HLA-DRB1*07:01 male APCsactivated the CD4 SP T cells, confirming that distinct minor H Agswere recognized by the two subpopulations of T cells. The samedifferential patterns were obtained with T cell clones (data notshown). Furthermore, following activation by their relevant APC,the mRNA profile of these two subpopulations of T cells was quitedistinct. The CD4 SP T cells expressed higher levels of IFN-g,IL-2 and FOXP3, but lower levels of IL-10 mRNA, whereas theCD4/CD8 DP cells expressed higher levels of IL10, but lowerlevels of FOXP3, IL-2, and IFN-g mRNA (Fig. 5B). Although, atlower levels, IL-4 and TGF-b mRNA were not differentiallyexpressed by CD4/CD8 DP T cells.

The HY epitope recognized by HLA-DRB1*07:01–restrictedCD4+ T cells is encoded by RPS4Y

To define which HY gene encoded the HY epitope recognized bythe CD4 SP T cells, we transduced female EBV cell lines withthe RPS4Y or DDX3Y genes. IFN-g production and proliferationmeasured T cell responses of the clone X2. The results indicatedthat RPS4Y, but not DDX3Y, was able to stimulate the proliferation Table

III.

Functional

characterizationoftheCD4Tcellline

Responder

↓APC

↓No.Ab

Anti–HLA

Class

IAnti–HLA-D

RAnti–HLA-D

QAnti–HLA-D

P

Celllinea

Donor

6156

49

ND

ND

ND

ND

Recipient

11,1886

973

12,5626

320

8866

55

14,6626

1,280

9,722

6517

HLA-m

ismatch

1,5596

68

ND

ND

ND

ND

APC

→Recipient

Male

Fem

ale

Responderb

↓HLA-D

RB1*07:01

HLA-D

RB1*07:01

#1

#2

#3

#1

#2

#3

Cellline

12,8756

512

4,552

6128

9,254

6210

9,478

6200

1,1706

58

1,274

680

8386

30

CloneR1

10,7966

410

7,466

6126

6,667

6834

ND

1,1446

34

1,340

671

ND

CloneJO

.419,0806

568

12,6206

269

8,700

6482

ND

11,1636

84

1,173

610

ND

Responder

cNo.Ab

Anti–HLA

classI

Anti–HLA

classII

E:T

cellratio→

10/1,3/1,1/1,0.3/1

10/1

10/1

Target

↓Donor

0,0,0,0

ND

ND

Recipient

126

0.8,126

1.3,106

0.9,76

0.55

146

1.6

66

0.9

HLA

classIsharingAPC

0,0,0,0

ND

ND

HLA

classIIsharingAPC

286

1.7,246

0.6,176

0.4,106

1.2

ND

ND

aCD4SPTcellswerestim

ulatedwithdonororrecipientEBV-BLCL.Anti–HLA

classIorclassIImAbswereusedto

inhibitTcellproliferation.

bHY

specificity

was

assessed

usingmaleorfemaleHLA-D

Rb1*07:01APCs.

cIn

a[51Cr]releaseassay,APCsfrom

donor,recipient,or

unrelatedmales

sharingHLAclassI(-B*14:01)

orclassII(-DRb1*07:01)moleculeswereused

astargetsofCD4SPeffector

cells.Anti–H

LAclassIorclassIImAbs

wereaddedas

indicated.

Resultsin

boldface

representstatisticallysignificantresultswithtwo-tailedunpairedttest

(p,

0.05).Resultsareexpressed

asmean6

SEM

ofcpm,[3HT]incorporation(1,2),orpercent[51Cr]

release(3).

The Journal of Immunology 189

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

of CD4 SP T cells and production of IFN-g (Fig. 6A, 6B). Frompeptide–MHC binding databases, we designed peptides of 25–40aa in length covering regions spanning one or more candidateHLA-DRB1*07:01 peptide-binding motifs (Supplemental TableI), using them to pulse female HLA-DR7 EBVAPCs. One peptide,QR-40, stimulated IFN-g production by the CD4 SP T cells (Fig.6C). Several potential 15-mer HLA-DRB1*07:01 binding peptideswithin the QR-40 sequence were tested. One, TL15, specificallyinduced IFN-g and weak IL-2 mRNA expression by the CD4 SPcells (Table IV). We thus identified TGKIINFIKFDTGNL as theHY peptide epitope recognized. Although FOXP3 was nonspe-cifically upregulated in the presence of every peptide, TL15 in-duced the highest FOXP3 mRNA response. TL15 also inducedhigher levels of IFN-g than QR40 (Table IV).

RPS4Y-specific CD4+ T cells express TCRVB19*01

To identify the TCR Vb expressed by the HY-specific CD4 SPT cells, the Vb TCR of the T cell line and clones was sequenced.There was a single CDR3 rearrangement in each (Fig. 6D), sug-gesting the initial CD4 SP T cell line was already clonal. Se-quencing the TCRVB/JB/DB regions of the CD4 SP T cellline and clone showed them to be identical, homologous toTRBV19*01 and TRBJ2-3*01 of the IMGT database (Supple-mental Table II).

DiscussionIn contrast to immature thymic T cells, expression of the CD4 andCD8 coreceptors on mature T cells is generally mutually exclusive.However, peripheral CD4/CD8 DP T cells have been describedin certain pathological as well as normal conditions (29–31). Forexample, a subset of peripheral blood CD4/CD8 DP T cells hasbeen reported in autoimmune and chronic inflammatory disorders,like thyroiditis, multiple sclerosis, and systemic sclerosis (32–34).

They have also been found in patients with Kawasaki syndromeand Hodgkin lymphoma (35, 36). Following transplantation, onlya single study has described them in organ biopsies, but not for-mally demonstrated coexpression of CD4 and CD8 (37). In animaltransplantation models, only one study has found such cells in theperiphery associated with cyclosporin A treatment (38).It is not known whether peripheral CD4/CD8 DP T cells are

a result of a failure of thymic selection or if the second coreceptor isexpressed in response to an immunogenic stimulus that enhancesintracellular signaling by recruiting p56lck (29). The relative levelsof CD4 and CD8 on these cells might provide insight into this:CD4high/CD8low DP T cells are found in small numbers in normalindividuals and in somewhat higher numbers of those with auto-immunity (29, 33). However, with the exception of a single studythat demonstrated such T cells infiltrating cutaneous lymphomaand exerting tumor-specific MHC class I–restricted lysis, nofunction could be attributed to them (39). With the exception ofthat study, CD4high/CD8low DP T cells were found resistant toapoptosis, to proliferate poorly upon CD3/TCR stimulation, andunable to produce IL-2 (29). In contrast, CD4high/CD8high T cellsare more likely to be Ag-specific effector cells, rather than im-mature cells released from the thymus. Indeed, such cells havebeen described to exert antiviral and antitumor immunity fol-lowing contact with the cognate Ag (40–44).The CD4high/CD8high DP T cell population described in this

study was isolated from skin of a patient with grade 2 acuteGVHD that resolved quickly. The donor/recipient pair was HLAidentical, sex-mismatched in the direction favoring the activationof HY-specific female donor T cells by recipient male cells. ThisCD4/CD8 DP population was coisolated together with a CD8 SPT cell population (Fig. 1). We were readily able to separate thesetwo populations, but not able to culture the CD8 SP T cells be-cause after sorting, CD4/CD8 DP T cells reappeared in the cul-

FIGURE 5. Different patterns of minor H Ag recognition and cytokine mRNA expression by the CD4/CD8 DP or CD4 SP T cells. (A) The CD4/CD8 DP

or the CD4 SP T cell clones were stimulated in the same experiment with donor, recipient, HLA-B*14:01, or-DRB1*07:01 EBV-B LCLs. Proliferation was

measured at day 3 by [3HT] incorporation. Results are the mean 6 SD of three experiments. (B) Following 24 h of stimulation, mRNA cytokine expression

profiles of T cell clones or lines were measured relative to CD3ε or cyclophilin B mRNA expression. Results are expressed as mRNA concentration ratio.

They are representative of two experiments.

190 NEW HUMAN MINOR H REGULATORY AND HY-SPECIFIC T CELLS

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

ture. This could be due either to acquisition of the CD4 coreceptorupon stimulation or competitive proliferation of small numbers ofCD4/CD8 DP T cells that remained after sorting. The results ofproliferation assays (Fig. 4D) suggested the latter was the case.This was supported by blocking experiments showing that neitheranti-CD4 or anti-CD8 mAbs inhibited the CD4/CD8 DP specificalloresponse (Table I), Similar results have been reported usingtumor-specific DP T cells (39). The CD4/CD8 DP T cell subsetwas only found in the patient’s skin, not peripheral blood (data notshown). These cells specifically proliferated in response to anautosomally encoded Ag expressed by recipient cells, but pro-duced negligible amounts of IL-2 and rather high levels of IL-10,together with lower levels of IFN-g and IL-4 (Table II). Thesecells exerted regulation in vitro in the presence of donor APCs(Fig. 4A). A similar finding has been reported for intraintestinalCD4/CD8 DP T cells: secretion IL-10 and regulatory function(45). However, anti–IL-10 did not reverse our inhibitory effect(data not shown), suggesting IL-10 played no role in the regula-tory activity. IFN-g has been shown by us and others to reduceT cell responses by inhibiting APC maturation (46) or by inducinginducible NO synthase production (47, 48). However, in this study,this was excluded because IFN-g was secreted at low levels, andanti–IFN-g Ab did not block inhibition. Furthermore, inducible

NO synthase was not increased upon stimulation of these cells(data not shown), and they expressed low levels of FOXP3 (Fig.5), suggesting another inhibitory mechanism was involved. Be-cause the CD4/CD8 DP T cells expressed high levels of IL-2R,even 7 d after activation (Fig. 1), and did not produce IL-2 (TableII), we hypothesized that suppression was due to consumption ofIL-2. Indeed, such a suppressor mechanism has been previouslyshown to be used by Treg (49, 50). This mechanism being likely inthis study was shown by: 1) their high IL-2 responsiveness, greaterthan the CD8 SP T cells (Fig. 4D) and an unrelated population ofCD3+ primary T90 T cells, (Fig. 4B); 2) their consumption of theIL-2 produced by third-party cells in an HTLP assay (Fig. 4C);and 3) their unresponsiveness to PHA in the absence of rIL-2 (Fig.4E). These observations might further explain why the acuteGVHD episode was brief and weak in this patient, suggestingcontrol in vivo by these Treg. Further studies are needed to ana-lyze the tissue distribution of this minor H Ag, as, because of itslow frequency of expression, only one of six individuals withHLA-B*14:01 expressed this Ag, we were unable to evaluate thisin our present study.After repeated rounds of stimulation with recipient APCs, we

isolated a subset of CD4 SP T cells (Fig. 1C). These produced IL-2and IFN-g and were cytotoxic (Fig. 5B, Table III). Further char-

FIGURE 6. The HY-specific T cell clone X2 recognizes an RPS4-Y–encoded epitope. Female EBV-B LCLs transduced with RPS4Y or DDX3Y genes

were used to stimulate the CD4 SP T cell clone X2. T cell proliferation (A) or IFN-g secretion (B) was measured. (C) IFN-g secretion was used to identify

the long peptide recognized by the CD4 T cell clone. (A) Results are the mean6 SD of three experiments. (B) Values are representative of two experiments.

(C) Results are from a single experiment. (D) Vb TCR spectratyping of the CD4/CD8 DP T cell clone or line was processed by PCR amplification, using

Vb TCR primers. Electrophoretic migration of the PCR products is shown and compared with that of polyclonal clonal T cells.

The Journal of Immunology 191

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

acterization demonstrated that the minor H Ag that they recog-nized was different from that recognized by the CD4/CD8 DPT cells. Indeed, although we formally excluded specificity for HYAg by the CD4/CD8 DP T cells (Table I), the CD4SP T cells wereclearly specific for HY (Table III). Moreover, the CD4/CD8 DPT cells were HLA-B*14:01 restricted, and the CD4SP T cellswere HLA-DRB1*07:01 restricted. These findings make it un-likely that the CD4 SP T cells were derived from the CD4/CD8DP T cells, rather that they were amplified during repeatedin vitro restimulation. Their cytokine profile was also different; theCD4/CD8 DP T cells preponderantly expressed transcripts forIL-10 and TGF-b, whereas the CD4 SP T cells preferentiallyexpressed IFN-g and FOXP3 mRNA following activation (Fig.5B). Finally, we found that the Vb TCR profile of these twopopulations was distinct, because some Vb TCR were lost in theCD4 SP population compared with the CD4/CD8 DP T cells(Fig. 2).The CD4 SP T cells were HY specific and HLA-DRB1*07:01

restricted; we were able to identify the gene and peptide encodingthis HY epitope, with a candidate gene approach similar to thatpreviously used to identify murine MHC class II–restricted HYepitopes (51). We retrovirally transduced female DR7-EBV cellswith the candidate genes DDX3Y and RPS4Y. Using our CD4 SPT cell clone or the line from which clones were derived, weidentified RPS4Y as the gene encoding the DR7-restricted HYepitope. The peptide epitope itself was identified first using pep-tides of 25–40 aa, designed to incorporate candidate DRB1*0701-binding peptides selected from several databases. This allowedus to pinpoint one peptide, QR40 (Fig. 6), and subsequently thespecific 15-mer peptide, TGKIINFIKFDTGNL, as the HY epitoperecognized by the CD4 SP T cells (Table IV). Comparing theresponse of the CD4 SP T cell clone to QR40 with TL15, theshorter peptide, presumably with no need for further processing,induced higher responses (Table IV). This reflects recent findingsthat long peptides have a lower affinity for MHC than shortpeptides (52). The TL15 peptide represents a novel HY epitope,which may indeed represent a new target for GVL. AlthoughRPS4Y has already been shown to encode an HLA class II–re-stricted HY epitope able to induce helper and cytolytic activity, inthat study, the epitope was restricted by HLA-DRB3. Not unsur-prisingly, the amino-acid sequence was different from the oneidentified in this study, as was the Vb TCR of the clone thatrecognized it (7).In conclusion, we describe in this study evidence for the exis-

tence of CD4/CD8 DP T cells in the skin of an acute GVHD patientand show that these cells exert regulatory function, probably asa result of IL-2 consumption. We suggest these cells may have hada role in the reduction of the pathogenic response causing acuteGVHD in this patient. We also identify a novel HLA class II–restricted HY minor H Ag, its HLA-restriction molecule, and itsamino acid sequence and describe the CDR3 region and Vb TCRof the clone that recognizes this peptide. Because HLA class II–restricted minor H Ags are important potential targets expressedby leukemic cells in vivo, due to their expression by hematopoieticcells, our findings could contribute to specifically target HLA-DR7 male leukemic cells in cell therapy programs.

AcknowledgmentsWe thank Jeanine Bernaud (Etablissement Francais du Sang Rhone Alpes)

for excellent technical assistance in flow cytometry.

DisclosuresThe authors have no financial conflicts of interest.T

able

IV.

DeterminationoftheTcellepitoperecognized

bytheCD4+Tcellclone

mRNAa↓

Concentration↓

Peptide→

Sequence

No.

TL15TGKIINFIK

FDTGNL

GV17GKIINFIK

FDTGNLCMV

TG17.1

TGNLCMVIG

GANLGRVG

TG17.2

TGNVCMVIA

GANLGRVG

IL-2/CD3ε

10mM

0.004

0.23

0.02

0.02

0.01

1mM

0.004

0.06

0.02

0.006

0.0047

IFN-g/CD3ε

10mM

0.15

12.63

0.82

0.11

0.5

1mM

0.15

9.91

0.12

0.11

0.009

FOXP3/CD3ε

10mM

0.009

0.44

0.27

0.32

0.3

1mM

0.009

0.42

0.12

0.12

0.23

Effectorb

HLA-D

RB1*07:01APC

Peptide

Concentration

IFN-g

(ng/m

l)

CloneX2

Fem

ale

No

0.002

Male

No

0.370

Fem

ale

QR40

100mM

0.676

Fem

ale

TL15

100mM

0.934

Fem

ale

TL15

1mM

0.515

Fem

ale

TL15

100nM

0.117

Fem

ale

TL15

10nM

0.041

Resultsin

boldface

representthehighestlevelsthat

wereobtained

inthesameexperim

ent.This

experim

entis

representativeoftwodifferentexperim

ents.

aSeveral

shortpeptides

from

within

theQR-40sequence

weresynthesized

andtested

fortheirabilityto

stim

ulate

theCD4+TcellcloneX2.Theresponse

was

evaluated

bycytokinemRNAexpressionpattern

ofIL-2,IFN-g,andFOXP3,which

weremeasuredrelatively

toCD3εexpression.Resultsareexpressed

asmRNA

concentrationratio.

bA

dose-response

curveofIFN-g

secretionwas

then

measuredin

thepresence

ofdifferentconcentrationsofTL15,thespecificepitope,

orQR-40.

192 NEW HUMAN MINOR H REGULATORY AND HY-SPECIFIC T CELLS

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

References1. Simpson, E., and D. Roopenian. 1997. Minor histocompatibility antigens. Curr.

Opin. Immunol. 9: 655–661.2. Hambach, L., and E. Goulmy. 2005. Immunotherapy of cancer through tar-

geting of minor histocompatibility antigens. Curr. Opin. Immunol. 17: 202–210.

3. Falkenburg, J. H., L. van de Corput, E. W. Marijt, and R. Willemze. 2003. Minorhistocompatibility antigens in human stem cell transplantation. Exp. Hematol.31: 743–751.

4. Marijt, W. A., M. H. Heemskerk, F. M. Kloosterboer, E. Goulmy, M. G. Kester,M. A. van der Hoorn, S. A. van Luxemburg-Heys, M. Hoogeboom, T. Mutis,J. W. Drijfhout, et al. 2003. Hematopoiesis-restricted minor histocompatibilityantigens HA-1- or HA-2-specific T cells can induce complete remissions ofrelapsed leukemia. Proc. Natl. Acad. Sci. USA 100: 2742–2747.

5. Advisory Committee of the International Bone Marrow Transplant Registry.1989. Report from the International Bone Marrow Transplant Registry. BoneMarrow Transplant. 4: 221–228.

6. Vogt, M. H., J. W. van den Muijsenberg, E. Goulmy, E. Spierings, P. Kluck,M. G. Kester, R. A. van Soest, J. W. Drijfhout, R. Willemze, andJ. H. Falkenburg. 2002. The DBY gene codes for an HLA-DQ5-restricted humanmale-specific minor histocompatibility antigen involved in graft-versus-hostdisease. Blood 99: 3027–3032.

7. Spierings, E., C. J. Vermeulen, M. H. Vogt, L. E. Doerner, J. H. Falkenburg,T. Mutis, and E. Goulmy. 2003. Identification of HLA class II-restricted H-Y-specific T-helper epitope evoking CD4+ T-helper cells in H-Y-mismatchedtransplantation. Lancet 362: 610–615.

8. Ferrara, J. L., J. E. Levine, P. Reddy, and E. Holler. 2009. Graft-versus-hostdisease. Lancet 373: 1550–1561.

9. Hambach, L., B. A. Nijmeijer, Z. Aghai, M. L. Schie, M. H. Wauben,J. H. Falkenburg, and E. Goulmy. 2006. Human cytotoxic T lymphocytesspecific for a single minor histocompatibility antigen HA-1 are effectiveagainst human lymphoblastic leukaemia in NOD/scid mice. Leukemia 20: 371–374.

10. Hambach, L., M. Vermeij, A. Buser, Z. Aghai, T. van der Kwast, and E. Goulmy.2008. Targeting a single mismatched minor histocompatibility antigen withtumor-restricted expression eradicates human solid tumors. Blood 112: 1844–1852.

11. Rutten, C. E., S. A. van Luxemburg-Heijs, M. Griffioen, E. W. Marijt, I. Jedema,M. H. Heemskerk, E. F. Posthuma, R. Willemze, and J. H. Falkenburg. 2008.HLA-DP as specific target for cellular immunotherapy in HLA class II-expressing B-cell leukemia. Leukemia 22: 1387–1394.

12. Meyer, R. G., C. M. Britten, D. Wehler, K. Bender, G. Hess, A. Konur,U. F. Hartwig, T. C. Wehler, A. J. Ullmann, C. Gentilini, et al. 2007. Prophylactictransfer of CD8-depleted donor lymphocytes after T-cell-depleted reduced-intensity transplantation. Blood 109: 374–382.

13. Kawase, T., Y. Nannya, H. Torikai, G. Yamamoto, M. Onizuka,S. Morishima, K. Tsujimura, K. Miyamura, Y. Kodera, Y. Morishima, et al.2008. Identification of human minor histocompatibility antigens based ongenetic association with highly parallel genotyping of pooled DNA. Blood111: 3286–3294.

14. Kamei, M., Y. Nannya, H. Torikai, T. Kawase, K. Taura, Y. Inamoto,T. Takahashi, M. Yazaki, S. Morishima, K. Tsujimura, et al. 2009. HapMapscanning of novel human minor histocompatibility antigens. Blood 113: 5041–5048.

15. Warren, E. H., N. J. Vigneron, M. A. Gavin, P. G. Coulie, V. Stroobant, A. Dalet,S. S. Tykodi, S. M. Xuereb, J. K. Mito, S. R. Riddell, and B. J. Van den Eynde.2006. An antigen produced by splicing of noncontiguous peptides in the reverseorder. Science 313: 1444–1447.

16. Murata, M., E. H. Warren, and S. R. Riddell. 2003. A human minor histocom-patibility antigen resulting from differential expression due to a gene deletion. J.Exp. Med. 197: 1279–1289.

17. Van Bergen, C. A., C. E. Rutten, E. D. Van Der Meijden, S. A. Van Luxemburg-Heijs, E. G. Lurvink, J. J. Houwing-Duistermaat, M. G. Kester, A. Mulder,R. Willemze, J. H. Falkenburg, and M. Griffioen. 2010. High-throughput char-acterization of 10 new minor histocompatibility antigens by whole genome as-sociation scanning. Cancer Res. 70: 9073–9083.

18. Simpson, E., D. Scott, E. James, G. Lombardi, K. Cwynarski, F. Dazzi,M. Millrain, and P. J. Dyson. 2002. Minor H antigens: genes and peptides.Transpl. Immunol. 10: 115–123.

19. Laurin, D., E. Spierings, L. T. van der Veken, A. Hamrouni, J. H. Falkenburg,G. Souillet, C. Vermeulen, A. Farre, C. Galambrun, D. Rigal, et al. 2006. Minorhistocompatibility antigen DDX3Y induces HLA-DQ5-restricted T cellresponses with limited TCR-Vbeta usage both in vivo and in vitro. Biol. BloodMarrow Transplant. 12: 1114–1124.

20. Stumpf, A. N., E. D. van der Meijden, C. A. van Bergen, R. Willemze,J. H. Falkenburg, and M. Griffioen. 2009. Identification of 4 new HLA-DR-restricted minor histocompatibility antigens as hematopoietic targets in antitu-mor immunity. Blood 114: 3684–3692.

21. van de Corput, L., P. Chaux, E. D. van der Meijden, E. De Plaen, J. H. FrederikFalkenburg, and P. van der Bruggen. 2005. A novel approach to identify antigensrecognized by CD4 T cells using complement-opsonized bacteria expressinga cDNA library. Leukemia 19: 279–285.

22. Griffioen, M., E. D. van der Meijden, E. H. Slager, M. W. Honders, C. E. Rutten,S. A. van Luxemburg-Heijs, P. A. von dem Borne, J. J. van Rood, R. Willemze,and J. H. Falkenburg. 2008. Identification of phosphatidylinositol 4-kinase typeII beta as HLA class II-restricted target in graft versus leukemia reactivity. Proc.Natl. Acad. Sci. USA 105: 3837–3842.

23. Spaapen, R. M., H. M. Lokhorst, K. van den Oudenalder, B. E. Otterud,H. Dolstra, M. F. Leppert, M. C. Minnema, A. C. Bloem, and T. Mutis. 2008.Toward targeting B cell cancers with CD4+ CTLs: identification of a CD19-encoded minor histocompatibility antigen using a novel genome-wide analysis.J. Exp. Med. 205: 2863–2872.

24. Reinsmoen, N. L., J. H. Kersey, and F. H. Bach. 1984. Detection of HLA re-stricted anti-minor histocompatibility antigen(s) reactive cells from skin GVHDlesions. Hum. Immunol. 11: 249–257.

25. Maile, R., S. M. Pop, R. Tisch, E. J. Collins, B. A. Cairns, and J. A. Frelinger.2006. Low-avidity CD8lo T cells induced by incomplete antigen stimulationin vivo regulate naive higher avidity CD8hi T cell responses to the same antigen.Eur. J. Immunol. 36: 397–410.

26. Burlingham, W. J., and E. Goulmy. 2008. Human CD8+ T-regulatory cells withlow-avidity T-cell receptor specific for minor histocompatibility antigens. Hum.Immunol. 69: 728–731.

27. Eljaafari, A., A. Farre, K. Duperrier, J. Even, H. Vie, M. Michallet, G. Souillet,A. Catherine Freidel, L. Gebuhrer, and D. Rigal. 2001. Generation of helper andcytotoxic CD4+T cell clones specific for the minor histocompatibility antigen H-Y, after in vitro priming of human T cells by HLA-identical monocyte-deriveddendritic cells. Transplantation 71: 1449–1455.

28. Kinsella, T. M., and G. P. Nolan. 1996. Episomal vectors rapidly and stablyproduce high-titer recombinant retrovirus. Hum. Gene Ther. 7: 1405–1413.

29. Parel, Y., and C. Chizzolini. 2004. CD4+ CD8+ double positive (DP) T cells inhealth and disease. Autoimmun. Rev. 3: 215–220.

30. Calado, R. T., A. B. Garcia, and R. P. Falcao. 1999. Age-related changes ofimmunophenotypically immature lymphocytes in normal human peripheralblood. Cytometry 38: 133–137.

31. Ortolani, C., E. Forti, E. Radin, R. Cibin, and A. Cossarizza. 1993. Cyto-fluorimetric identification of two populations of double positive (CD4+,CD8+)T lymphocytes in human peripheral blood. Biochem. Biophys. Res. Commun.191: 601–609.

32. Munschauer, F. E., C. Stewart, L. Jacobs, S. Kaba, Z. Ghorishi, S. J. Greenberg,and D. Cookfair. 1993. Circulating CD3+ CD4+ CD8+ T lymphocytes in mul-tiple sclerosis. J. Clin. Immunol. 13: 113–118.

33. Parel, Y., M. Aurrand-Lions, A. Scheja, J. M. Dayer, E. Roosnek, andC. Chizzolini. 2007. Presence of CD4+CD8+ double-positive T cells with veryhigh interleukin-4 production potential in lesional skin of patients with systemicsclerosis. Arthritis Rheum. 56: 3459–3467.

34. Iwatani, Y., Y. Hidaka, F. Matsuzuka, K. Kuma, and N. Amino. 1993.Intrathyroidal lymphocyte subsets, including unusual CD4+ CD8+ cells andCD3loTCR alpha beta lo/-CD4-CD8- cells, in autoimmune thyroid disease. Clin.Exp. Immunol. 93: 430–436.

35. Hirao, J., and K. Sugita. 1998. Circulating CD4+CD8+ T lymphocytes inpatients with Kawasaki disease. Clin. Exp. Immunol. 111: 397–401.

36. Rahemtullah, A., K. K. Reichard, F. I. Preffer, N. L. Harris, and R. P. Hasserjian.2006. A double-positive CD4+CD8+ T-cell population is commonly found innodular lymphocyte predominant Hodgkin lymphoma. Am. J. Clin. Pathol. 126:805–814.

37. Burdick, J. F., W. E. Beschorner, W. J. Smith, D. McGraw, W. L. Bender,G. M. Williams, and K. Solez. 1984. Characteristics of early routine renal al-lograft biopsies. Transplantation 38: 679–684.

38. Godden, U., J. Herbert, R. D. Stewart, and B. Roser. 1985. A novel cell typecarrying both Th and Tc/s markers in the blood of cyclosporine-treated, allog-rafted rats. Transplantation 39: 624–628.

39. Bagot, M., H. Echchakir, F. Mami-Chouaib, M. H. Delfau-Larue, D. Charue,A. Bernheim, S. Chouaib, L. Boumsell, and A. Bensussan. 1998. Isolation oftumor-specific cytotoxic CD4+ and CD4+CD8dim+ T-cell clones infiltratinga cutaneous T-cell lymphoma. Blood 91: 4331–4341.

40. Nascimbeni, M., E. C. Shin, L. Chiriboga, D. E. Kleiner, and B. Rehermann.2004. Peripheral CD4(+)CD8(+) T cells are differentiated effector memory cellswith antiviral functions. Blood 104: 478–486.

41. Zloza, A., Y. B. Sullivan, E. Connick, A. L. Landay, and L. Al-Harthi. 2003.CD8+ T cells that express CD4 on their surface (CD4dimCD8bright T cells)recognize an antigen-specific target, are detected in vivo, and can be produc-tively infected by T-tropic HIV. Blood 102: 2156–2164.

42. Weiss, L., A. Roux, S. Garcia, C. Demouchy, N. Haeffner-Cavaillon,M. D. Kazatchkine, and M. L. Gougeon. 1998. Persistent expansion, in a humanimmunodeficiency virus-infected person, of V beta-restricted CD4+CD8+T lymphocytes that express cytotoxicity-associated molecules and are committedto produce interferon-gamma and tumor necrosis factor-alpha. J. Infect. Dis. 178:1158–1162.

43. Desfrancois, J., L. Derre, M. Corvaisier, B. Le Mevel, V. Catros, F. Jotereau, andN. Gervois. 2009. Increased frequency of nonconventional double positiveCD4CD8 alphabeta T cells in human breast pleural effusions. Int. J. Cancer 125:374–380.

44. Desfrancois, J., A. Moreau-Aubry, V. Vignard, Y. Godet, A. Khammari,B. Dreno, F. Jotereau, and N. Gervois. 2010. Double positive CD4CD8 alphabetaT cells: a new tumor-reactive population in human melanomas. PLoS ONE 5:e8437.

45. Das, G., M. M. Augustine, J. Das, K. Bottomly, P. Ray, and A. Ray. 2003. Animportant regulatory role for CD4+CD8 alpha alpha T cells in the intestinalepithelial layer in the prevention of inflammatory bowel disease. Proc. Natl.Acad. Sci. USA 100: 5324–5329.

46. Eljaafari, A., Y. P. Li, and P. Miossec. 2009. IFN-gamma, as secreted during analloresponse, induces differentiation of monocytes into tolerogenic dendriticcells, resulting in FoxP3+ regulatory T cell promotion. J. Immunol. 183: 2932–2945.

The Journal of Immunology 193

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

47. Sawitzki, B., C. I. Kingsley, V. Oliveira, M. Karim, M. Herber, and K. J. Wood.

2005. IFN-gamma production by alloantigen-reactive regulatory T cells is im-

portant for their regulatory function in vivo. J. Exp. Med. 201: 1925–1935.48. Wood, K. J., G. Feng, B. Wei, B. Sawitzki, and A. R. Bushell. 2007. Interferon

gamma: friend or foe? Transplantation 84(Suppl): S4–S5.49. Vercoulen, Y., E. J. Wehrens, N. H. van Teijlingen, W. de Jager, J. M. Beekman,

and B. J. Prakken. 2009. Human regulatory T cell suppressive function is in-

dependent of apoptosis induction in activated effector T cells. PLoS ONE 4:

e7183.

50. Shevach, E. M. 2009. Mechanisms of foxp3+ T regulatory cell-mediated sup-pression. Immunity 30: 636–645.

51. Scott, D., C. Addey, P. Ellis, E. James, M. J. Mitchell, N. Saut, S. Jurcevic, andE. Simpson. 2000. Dendritic cells permit identification of genes encodingMHC class II-restricted epitopes of transplantation antigens. Immunity 12:711–720.

52. Hambach, L., Z. Aghai, J. Pool, N. Kroger, and E. Goulmy. 2010. Peptidelength extension skews the minor HA-1 antigen presentation toward activateddendritic cells but reduces its presentation efficiency. J. Immunol. 185: 4582–4589.

194 NEW HUMAN MINOR H REGULATORY AND HY-SPECIFIC T CELLS

by guest on February 18, 2018http://w

ww

.jimm

unol.org/D

ownloaded from