chronic myeloid leukemia cells express tumor-associated antigens eliciting specific cd8+ t-cell...

Experimental Hematology 34 (2006) 1709–1719

Chronic myeloid leukemia cells expresstumor-associated antigens eliciting specific CD8þ

T-cell responses and are lacking costimulatory molecules

Michael Schmitta, Li Lia, Krzysztof Giannopoulosa,Jinfei Chena, Christian Brunnera, Thomas Barthb, Anita Schmittc,

Markus Wiesnethd, Konstanze Dohnera, Hartmut Dohnera, and Jochen Greinera

a3rd Department of Internal Medicine; bDepartment of Pathology;cDepartment of Otorhinolaryngology, University of Ulm; dInstitute of Clinical Transfusion Medicine, Ulm, Germany

(Received 28 November 2005; revised 11 July 2006; accepted 17 July 2006)

Specific immunotherapies for patients with chronic myeloid leukemia (CML) might eliminateresidual CML cells after therapy with imatinib or chemotherapy and might enhance a specificgraft-versus-leukemia effect after allogeneic stem cell transplantation. Here, we investigatedthe mRNA expression and T-cell recognition of tumor-associated antigens or leukemia-associ-ated antigens (LAAs) in 34 patients with CML. Several LAAs are expressed in CML andtherefore are candidate structures for specific immunotherapies: bcr-abl (100%), G250(24%), hTERT (53%), MPP11 (91%), NEWREN60 (94%), PRAME (62%), Proteinase3(71%), RHAMM/CD168 (83%), and WT1 (53%), but not BAGE, MAGE-A1, SSX2, or NY-ESO-1. The frequency of mRNA expression of RHAMM/CD168, Proteinase3, and PRAMEwas higher in acceleration phase and blast crisis. In flow cytometry, CD34+ progenitor cellstyped positive for HLA molecules but were deficient for CD40, CD80, CD83, and CD86. How-ever, RHAMM/CD168 R3-peptide (ILSLELMKL)-specific T-cell responses in CML patientswere demonstrated by ELISPOT analysis and specific lysis of RHAMM/CD168 R3-pulsedT2 cells and CD34+ CML cells in chromium-51 release assays. RHAMM-R3-specific T cellscould be phenotyped as CD8+R3*tetramer+CD45RA+CCR7LCD27L early effector T cellsby tetramer staining. Therefore, vaccination strategies inducing such RHAMM-R3-directedeffector T cells might be a promising approach to enhance specific immune responses againstCML cells. � 2006 International Society for Experimental Hematology. Published byElsevier Inc.

The success of imatinib has dramatically revised the way oftreating patients with chronic myeloid leukemia (CML).The prolonged cytogenetic remissions obtained by the treat-ment with imatinib made this particular therapy a first-lineapproach, even in patients who hitherto might have beencandidates for transplantation strategies [1–3]. Neverthe-less, allogeneic stem cell transplantation remains the onlycurative approach in CML to date [4]. In patients in chronicphase, a survival rate ranging from 70 to 80% at 5 years canbe achieved by the treatment with allogeneic hematopoeticstem cell transplantation [4–6]. In case of relapse, donorlymphocyte infusions (DLI) constitute an effective thera-

Offprint requests to: Jochen Greiner, M.D., 3rd Department of Internal

Medicine, University of Ulm, Robert-Koch-Str. 8, 89081 Ulm, Germany;

E-mail: [email protected]

0301-472X/06 $–see front matter. Copyright � 2006 International Society for

doi: 10.1016/j.exphem.2006.07.009

peutic approach [7–10]. The graft-versus-leukemia (GVL)effect through DLI is more pronounced in CML than inother leukemias, characterizing CML as a highly immuno-genic hematologic disease [10] and suggesting the exis-tence of target structures recognized by specific T cells.Molldrem et al. [11] recently defined T-cell subtypes char-acterized by high avidity to a specific antigen epitope ascells involved in the rejection of CML cells. For the designof further specific cell therapies in CML, we addressed thefollowing questions in this work:

1. Which tumor-/leukemia-associated antigens (TAAs/LAAs) are expressed in peripheral blood mononuclearcells from CML patients?

2. Does the expression of TAAs/LAAs vanish at differentstages of the clinical course (i.e. chronic phase, accel-eration, blast crisis)?

Experimental Hematology. Published by Elsevier Inc.

mailto:[email protected]

1710 M. Schmitt et al./ Experimental Hematology 34 (2006) 1709–1719

3. Do CD34 and bcr-abl double-positive cells express hu-man leukocyte antigens (HLA-ABC, HLA-DR) andcostimulatory molecules (CD40, CD80, CD83, andCD86) for an efficient presentation of TAAs/LAAs to-wards the priming of T cells?

4. Are specific CD8þ T cell–mediated immune responsesagainst tumor-specific expressed antigens likeRHAMM/CD168 detectable and are these CD8þ Tcells effector cells?

In this study, the mRNA expression of several TAAs/LAAs was assessed by conventional and quantitative re-verse-transcriptase polymerase chain reaction (RT-PCR) as-says from CML patients at different stages of the disease.Furthermore, CD34 and bcr-abl double-positive cells wereimmunophenotyped for their expression of HLA-ABC,HLA–DR, CD40, CD80, CD83, and CD86. T-cell assaysand cytotoxicity were performed to assess the efficiencyof specific immune responses in CML patients against theTAAs/LAAs RHAMM/CD168 showing a tumor-restricted

expression pattern and a high expression frequency inCML patients.

Materials and methods

Cell samplesPeripheral blood mononuclear cell samples were collected from34 patients with CML; the samples contained 6 to 100% bcr-abl-positive cells (Table 1). Eleven CML patients were women,23 men. The median age was 51 years (range: 25–79 years). Weexamined 24 CML patients in chronic phase and 10 patients in ac-celeration or blast crisis. Total leukocytes were prepared andstored for RNA preparation at �80�C. Furthermore, peripheralblood was obtained from healthy volunteers.

Informed consentSamples were taken from CML patients treated at our institutionwithin CML treatment protocols approved by the local ethics com-mittee. Informed consent was obtained from all patients and fromhealthy volunteer blood donors concerning the use of their bloodfor scientific purposes.

Table 1. Clinical characteristics of the CML patients tested for the mRNA expression of tumor-associated antigens

Patient no. bcr-abl (RT-PCR) bcr-abl/abl (%) Sex Age (y) Therapy status Stage

1 þ 100 F 28 After allogeneic transplantation Chronic phase

2 þ 55 M 46 After allogeneic transplantation Chronic phase

3 þ 30 F 51 No treatment Chronic phase

4 þ 57 M 25 No treatment Chronic phase

5 þ 16 M 59 After autologous transplantation Chronic phase


7 þ 6 M 59 After autologous transplantation Chronic phase



10 þ 100 M 61 Alpha interferon Chronic phase

11 þ 100 M 65 Alpha interferon Acceleration

12 þ 100 F 56 Alpha interferon Chronic phase

13 þ 27 F 66 Alpha interferon/hydroxyurea Chronic phase


15 þ 30 F 64 Alpha interferon Blast crisis (lymphatic)

16 þ 81 F 61 Alpha interferon Acceleration

17 þ 100 F 69 Hydroyxurea Blast crisis


19 þ 62 M 46 Alpha interferon/hydroxyurea Blast crisis

20 þ 67 M 40 Alpha interferon/hydroxyurea Chronic phase

21 þ 91 F 61 Hydroyxurea Chronic phase

22 þ 46 M 53 Hydroyxurea Blast crisis (lymphatic)

23 þ 24 F 48 Hydroyxurea Chronic phase


25 þ 81 M 69 Hydroyxurea Acceleration

26 þ 56 F 44 No treatment Chronic phase

27 þ 64 M 36 Hydroyxurea Blast crisis


29 þ 100 M 29 Alpha interferon/hydroxyurea Acceleration


31 þ 28 M 45 Hydroyxurea Chronic phase



34 þ 60 F 36 Hydroyxurea Blast crisis

1711M. Schmitt et al. / Experimental Hematology 34 (2006) 1709–1719

Cytogenetic analysisCytogenetic analysis was performed employing standard proce-dures as described elsewhere [12] from bone marrow and/or pe-ripheral blood, at diagnosis and/or in follow-up. All CMLpatients showed the chromosomal translocation t(9;22) (n 5 21).

RT-PCR for LAAs/TAAsLAAs have been examined in CML with known expression inother myeloid leukemias (AML) like BAGE, G250, hTERT,MPP11, NEW-REN-60, PRAME, Proteinase3, RHAMM, WT1[13] or TAAs that are under current clinical investigation in solidtumors like MAGE-A1, NY-ESO-1, and SSX2. Total RNA wasisolated using the RNAeasy Midi-Kit (Amersham Pharmacia Bio-tech, Little Chalfont, England, UK). mRNA was prepared by usingthe mRNA QuickPrep Micro purification kit (Amersham Pharma-cia Biotech); 2 mg of each sample was subjected to cDNA synthe-sis (Superscript II Gibco BRL, Frederick, MD, USA). Thesequence of the primers for RT-PCR for the antigens BAGE,G250, hTERT, MAGE-A1, MPP11, NY-ESO-1, NEW-REN-60,PRAME, Proteinase3, RHAMM, SSX2, WT1, temperatures of de-naturation, annealing and elongation, MgCl2 concentration, andthe cycle numbers were employed as described earlier [13]. ThemRNA expression of ß-actin and TATA-box binding proteinwere used in addition to the gene abl as housekeeping genes.The expression of the genes of interest was assessed by RT-PCRas described [13].

Real-time RT-PCR for bcr-abl and ablFor the quantitative measurement of the mRNA expression of bcr-abl and abl, real-time RT-PCR was performed using the light cy-cler SYBR Green I technology according to the manufacturer’sprotocol, with the primers 50 TTC AGA AGC TTC TCC CTGACA T 30 and 50 CGG CTC TCG GAG GAG ACG TAG A 30

for the bcr-abl transcript p210, and 50 CAG ATC TGG CCCAAC GAT GG 30 and 50 CCC AAC CTT TTC GTT GCA CTGT 30 for the bcr-abl transcript p190. An initial denaturation stepat 95�C for 1 minute was followed by 45 cycles of 1 second at95�C, 10 seconds at 64�C, 26 seconds at 72�C for the bcr-abl tran-script p210. For the bcr-abl transcript p190, an initial denaturationstep at 95�C for 1 minute was followed by 45 cycles of 1 second at95�C, 10 seconds at 64�C, 26 seconds at 72�C; 0.1 mg of mRNAwas used per RT-PCR. To quantify the mRNA expression of bcr-abl, the expression of the gene bcr-abl of the samples was corre-lated to the mRNA expression of the abl gene abl.

Nested RT-PCR for bcr-ablNested RT-PCR for bcr-abl p210 was performed using the follow-ing primers: 50 GAG CGT GCA GAG TGG AGG GAG AAC A 30

and 50 GGT ACC AGG AGT GTT TCT CCA GAC TG 30 for thefirst round, and the primer 50 TTC AGA AGC TTC TCC CTGACA T 30 and 50 TGT TGA CTG GCG TGA TGT AGT TGCTTG G 30 for the nested RT-PCR. An initial denaturation step at95�C for 1 minute was followed by 30 cycles of 60 seconds at95�C, 50 seconds at 64�C, 60 seconds at 72�C (‘‘PCR 1’’). Forthe second round of the nested RT-PCR (‘‘PCR 2’’), an initial de-naturation step at 95�C for 1 minute was followed by 35 cycles of60 seconds at 95�C, 50 seconds at 64�C, 60 seconds at 72�C; 0.1mg of mRNA per reaction was used per reaction.

Real-time RT-PCR for theTAAs/LAAs RHAMM, PRAME, and WT1For the quantitative measurement of the mRNA expression ofRHAMM, PRAME, and WT1 in CML samples, real-time RT-PCR using the light cycler SYBR Green technology was used ac-cording to the manufacturer’s protocol and the primers as de-scribed earlier [13].

Characterization of theimmunophenotype of CD 34þ CML cellsImmunophenotyping was performed for CD34þ cells that tested100% for the ratio bcr-abl/abl in the quantitative RT-PCR. Thesamples from 10 CML patients were incubated with a fluoroiso-thiocyanate (FITC)-labeled monoclonal antibody (mAB) againstCD34*FITC and second, with one of the following phycoerythrin(PE)-conjugated mABs: HLA-ABC, HLA-DR, CD33, CD40,CD80, CD83, or CD86. CD34þ selected cell samples (n 5 10)were examined. An HLA-A2-directed antibody (kindly providedfrom Dr. Alois Wolpl, Institute for Medical Genetics, Munich,Germany) was not conjugated, therefore staining was followedby incubation with a PE-conjugated secondary mAB (Dako Diag-nostics, Hamburg, Germany). After incubation the CD34þ andbcr-ablþ cells were washed twice in phosphate-buffered saline(PBS) and centrifuged for 5 minutes. Pellets were resuspendedin PBS and submitted to fluorescein-activated cell sorting(FACS) analysis using a FACScalibur flow cytometer. Data wereanalyzed by the Cellquest software (BD Biosciences, Heidelberg,Germany).

ImmunocytologyImmunocytology was performed as described elsewhere [14].Briefly, cytospins of CML cells (K562) and CD34-selected periph-eral and bone-marrow stem cells (PBSC/BMSC) were incubatedwith a monoclonal anti-CD168 antibody (supernatant of anIgG2a clone kindly provided by Len Cohen and Gerd Ritter, Lud-wig Institute for Cancer Research, New York, NY, USA) at a 1:25dilution. A polyclonal biotinylated Fab antibody to mouse IgG(Dako Diagnostics) and a tertiary rabbit anti-mouse antibody cou-pled with alkaline phosphatase were used in consecutive steps.The Fast Red Substrate System (Dako Diagnostics) was used assubstrate for the enzyme. By this method we were able to circum-vent the problem of intrinsic peroxidase activity of the leukemicblasts or CD34þ stem cells. The cytospin preparations were faintlycounterstained with Harris’ hematoxylin. Negative controls wereperformed without primary antibody.

Western blotWestern Blot analysis was performed to detect protein expressionof RHAMM in CML cells, CD34þ-separated hematopoietic pro-genitor cells, and cells of peripheral blood of healthy volunteers.Cells were lysed and 10 mg/lane were separated for dodecyl sul-fate-polyacrylamide gel electrophoresis (SDS-PAGE) as described[15]. After blotting of proteins onto nitrocellulose membranes,blots were incubated with RHAMM-antibody (RHAMM H-90,Sta. Cruz Biotech, Santa Cruz, CA, USA) specific for the human85- to 90-kDa RHAMM protein. All splice variants were detectedby this antibody. Blots were incubated with peroxidase-conjugatedgoat-anti-rabbit IgG (1:5,000 dilution) and visualized usingchemoluminescence (ECL, Amersham, Quebec, Canada).


Mixed lymphocyte peptide cultureperipheral blood mononuclear cells (PBMC) from CML patients(n 5 5) were separated by Ficoll and subsequently selected bymagnetic beads through a magnetic cell separation column.CD8-negative antigen presenting cells (APCs) were irradiatedwith 30 Gy and pulsed with a RHAMM/CD168 R3 and an irrele-vant MAGE-3 control peptide [14]. Mixed lymphocyte peptideculture (MLPC) was performed as described elsewhere [14].

Interferon-g and Granzyme B ELISPOT assaysELISPOT assays for the secretion of interferon-g and Granzyme Bwere performed as previously described to determine the specificlysis of RHAMM/CD168-positive target cells according to themanufacturer’s instructions (BD, San Diego, CA, USA) [14].

Chromium-51 release cytotoxicity assayIn CML patients with positive reactions in ELISPOT analysis,RHAMM-pulsed T2 cells were labeled with 51CrO4 as described[14]. Radioactivity was measured by a gamma counter (Perki-nElmer, Boston, MA, USA) and the percentage of specific lysiswas calculated.

Tetramer stainingCD8þ T cells (1 � 106), stimulated with irradiated CD8-negativeAPCs in the presence of R3 peptide, were stained with the tetra-mer HLA-A2/R3*PE (Ludwig Institute for Cancer Research, Lau-sanne, Switzerland) at a concentration of 2 mg/mL and anti-CCR7pure rat antibody (BD, Heidelberg, Germany) in the dark and in-cubated for 40 minutes at room temperature. Thereafter, anti-CD8*PerCP (BD, Heidelberg, Germany), anti-CD45RA*FITC,or anti-CD27*FITC (BD, Heidelberg, Germany) and secondaryantibody goat anti-rat IgG*APC (Caltag Laboratories, Burlin-game, CA, USA) were added at 4�C for 20 minutes in the dark.After washing twice, stained cells were analyzed by flow cytom-

etry; 100,000 events were collected. As a negative control cellswere stimulated with an irrelevant MAGE-3 peptide as describedelsewhere [14].

Results

Differential mRNA expression analysis of TAAs/LAAsThe mRNA expression of different immunogenic antigenswas evaluated in PBMC samples from 34 CML patients(Fig. 1). Several antigens with expression in other leuke-mias (AML or leukemia cell lines) showed no mRNA ex-pression in CML samples: No mRNA expression inprimary CML cells was found for NY-ESO-1, whereasthe TAAs SSX2, MAGE-A1, and BAGE showed only pos-itive mRNA expression in the leukemia cell line K562, butnot in PBMC samples from CML patients. The mRNA ex-pression of the other TAAs tested was in detail as follows(Fig. 1): The antigen G250 was detected in 8/34 (24%)and hTERT in 18/34 (53%) of CML patients. mRNA ex-pression of MPP11 was found in the cell line K562, servingas a positive control and in all other leukemia cell lines weexamined. Thirty-one of 34 (91%) of CML patients showedstrong mRNA expression of the antigen MPP11. mRNA ex-pression of the antigen NEWREN60 was detected in 32/34(94%) of CML patients and in all leukemia cell lines. Theantigen PRAME showed positive mRNA expression in 21/34 (62%) of CML patients. Twenty-four of 34 (71%) ofCML patients tested positive for mRNA of proteinase3.mRNA expression of RHAMM was found in 28/34

Frequency (%)

CP: Chronic phase

AP: Acceleration phase

BC: Blast crisis

0

20

40

60

80

100

ß-actin

TB

P

BA

GE

G250

hTE

RT

MA

GE

A1

MP

P11

NR

60

NY

-E

SO

-1

PR

AM

E

Prot. 3

RH

AM

M

SS

X2

WT1

BC

R-A

BL

CP

AP / BC

Figure 1. mRNA expression pattern of immunogenic antigens in samples from 34 CML patients in chronic phase and acceleration/blast crisis by conven-

tional RT-PCR. The LAAs bcr-abl, hTERT, MPP11, NEWREN60, PRAME, Proteinase3, RHAMM/CD168, and WT1 were expressed in CML at a high and

G250 at a lower frequency. The frequency of mRNA expression of RHAMM/CD168, Proteinase3, and PRAME was higher in acceleration phase and blast

crisis (AP/BC; white bars) than in chronic phase (black bars), whereas G250 was expressed at a lower percentage in AP/BC. CP, chronic phase; AP, accel-

eration phase; BC, blast crisis.


(83%). The gene WT1 has been shown to be expressed ata high level in patients with AML. In this study, 18/34(53%) of CML patients showed mRNA expression of WT1.

mRNA expression of TAAs/LAAs in chronicphase versus acceleration phase or blast crisisAll antigens with positive expression in CML patients wereexpressed in chronic phase (Fig. 1, black bars) but also inacceleration phase/blast crisis of the examined CML pa-tients (Fig. 1). Differences were found for the antigensRHAMM, G250, PRAME, and Proteinase3: In accelerationphase and blast crisis, RHAMM, Proteinase3, and PRAMEshowed a higher frequency of mRNA expression.RHAMM, PRAME, and proteinase3 were expressed in100%, 90%, and 70%, respectively, in acceleration phase/blast crisis (AP/BC), whereas these antigens were detectedin 75%, 63%, and 58% of CML patients in chronic phase.G250 was expressed in AP/BC, at a lower frequency (10%vs 29%). mRNA expression of the antigen MPP11 was ex-pressed at high frequency in chronic phase and in AP/BC(91% vs 92%).

Correlation of the expression ofTAAs/LAAs and bcr-abl during the clinical courseThe expression of TAAs/LAAs was correlated to the ex-pression of bcr-abl during the clinical course of threeCML patients in chronic phase. All samples showedmRNA expression of bcr-abl. The ratio bcr-abl/abl is listedin Table 1. Data of one patient are presented in Figure 2.The expression of TAAs/LAAs and bcr-abl was measuredby real-time RT-PCR using the light cycler technology forthe antigens PRAME, RHAMM/CD168, and WT1(Fig. 2) and using conventional RT-PCR for the antigensMPP11, NEWREN60, and Proteinase3. The expression ofthe antigens Proteinase3 and NEWREN60 showed no cor-relation to the expression of bcr-abl (data not shown). Theseantigens were also expressed at the same level in sampleswith negative bcr-abl expression tested by real-time RT-PCR and also by nested RT-PCR (sensitivity: one bcr-ablpositive cell in 1 � 105 to 1 � 106 bcr-abl negative cellscould be detected). In contrast, the expression of RHAMMand PRAME was reduced according to the quantitative bcr-abl expression (Fig. 2). The LAA RHAMM/CD168 wasnegative in all samples showing a complete molecular re-mission. WT-1 was expressed in CML cells with a high ra-tio of bcr-abl/abl. The level of expression was lower inpatients with a more controlled disease, but still positivein some samples of CML patients who tested negative forbcr-abl in nested RT-PCR.

Immunophenotypic staining of CD34þ CML cellsFlow cytometry of CD34þ cells that tested positive for bcr-abl (quotient bcr-abl/abl 5 100%) in the RT-PCR assay re-vealed a high expression frequency (O95%) for HLA-ABCand HLA-DR in all CML samples. This was observed in allCML patients (n 5 10); 79.3% (12.8–98%) of the CD34þ

cells expressed the cell surface marker CD33. In contrast,no expression (!1%) of the costimulatory moleculesCD40, CD80, CD83, and CD86 was found on the CD34þ

cell population. Figure 3 displays typical results obtainedby flow cytometry.

Immunocytochemistry and Western blot analysisHigh protein expression of the LAA RHAMM was detectedin CML cells (K562) by immunocytochemistry and West-ern blot analysis (Fig. 4A and B). In contrast, in CD34þ-se-lected peripheral stem cells from healthy donors mobilizedthrough granulocyte colony stimulating factor (G-CSF),only a very low protein expression of this antigen was de-tectable. Normal CD34þ-selected stem cells isolated fromthe bone marrow of healthy donors showed no expressionof RHAMM analyzed with both methods (Fig. 4A andB). Negative results of RHAMM protein expression in

Copy number

Copy number

Copy number

10

0%

(ra

tio

b

cr-a

bl/a

bl)

62%

ne

ga

tiv

e

ne

ga

tiv

e

0.0

1-0

.0

01

%

0.1

4%

1.2%

11

.6

%

Dis

tille

d w

ate

r

0%

(P

BM

N o

f H

V)

1

100

10.000

1.000.000

1 2 3 4 5 6 7 8 9 10

1 2 3 4 5 6 7 8 9 10

1 2 3 4 5 6 7 8 9 10

PRAME

1

100

10.000

1.000.000

1

10

100

1000

PRAME

RHAMM

WT1

Figure 2. Quantitative mRNA expression of TAAs/LAAs in correlation

with the mRNA expression of bcr-abl. Quantitative mRNA expression of

the LAAs PRAME, RHAMM/CD168, and WT-1 was assessed in a CML

patient in chronic phase at time of diagnosis (lane 1), during the further

clinical course (lane 2), in complete molecular remission (lanes 3 and

4), and at relapse (lanes 5–8). K562 was used as a positive control, DW

and PBMN of healthy volunteers as negative controls. The LAA

RHAMM/CD168 tested negative in complete molecular remission (lanes

3 and 4) and showed the best correlation to the expression of bcr-abl of

these antigens. The percentages indicate the ratio bcr-abl/abl, the columns

in the figure are the copy number of the respective LAA.


CD

34

HLA-ABC

HLA-DR

CD

34

CD

33

CD34

CD

40

CD

80

CD

86

CD

83

CD34 CD34

CD34 CD34

0% 35%

1% 64%

0% 30% 62% 8%

53% 41% 5% 1%

0% 0%

64% 36%

0% 0%

67% 33%

5% 0%

60% 35%

5% 0%

61% 34%

Figure 3. Flow cytometry analysis of the expression of HLA-ABC, HLA-DR, CD33, CD34, CD40, CD80, CD83, and CD86 on CD34þ CML cells. Cells

were CD34-positive selected from CML patients (n 5 10) in chronic phase (the cells testing 100% positive for bcr-abl/abl in quantitative RT-PCR). The

expression of HLA-ABC, HLA-DR was preserved on these CD34þ bcr-ablþ CML cells (O 95%) as a prerequisite to elicit efficient T-cell responses. In

contrast, the costimulatory molecules CD40, CD80, CD83, and CD86 were absent (!5%) on these CD34þ cells. This might constitute a mechanism for

the tumor escape of CML progenitor cells.


No

rm

al C

D34+

B

MS

C

No

rm

al C

D34+

P

BS

C

CM

L b

lasts (K

562)

No

rm

al P

erip

heral b

lo

od

Marker

K562 PBSC BMSC

RHAMM

- 80 kDa

ß-actin

- 30 kDa

B

A

Figure 4. RHAMM/CD168 protein expression in CML blasts, G-CSF-mobilized CD34þ selected peripheral stem cells (PBSC), and normal bone-marrow

stem cells (BMSC) of healthy donors by immunocytochemistry (A) and Western blot analysis (B). A high protein expression of RHAMM/CD168 could be

demonstrated in CML cells (K562). In contrast, only a very low expression was found in CD34þ PBSC mobilized from healthy donors with G-CSF, but not in

normal CD34þ BMSC obtained from healthy donors without G-CSF stimulation.

normal peripheral blood cells were obtained in accordancewith results we described elsewhere [15].

Immunophenotypic staining of CD34þ

hematopoietic stem cells from healthy volunteersFlow cytometry of CD34þ stem/progenitor cells fromhealthy volunteers (n 5 10) revealed a complete deficiencyfor the costimulatory molecules CD40, CD80, CD83, andCD86, although HLA-ABC and HLA-DR molecules wereexpressed in O95% of the cells (data not shown).

Specific immune responses in CMLpatients against the RHAMM/CD168-R3 peptideSpecific T-cell reactions to the RHAMM/CD168-derivedpeptide R3 have been already demonstrated in AML pa-tients [14]. For the characterization of specific immune re-sponses against RHAMM/CD168 in CML (n 5 5), fivepatients in chronic phase were evaluated. In four of fiveCML patients, RHAMM/CD168-R3-specific immuneresponses were detected. Figure 5 shows a Granzyme Brelease by CD8þ T cells from an CML patient in chronicphase against the RHAMM/CD168-derived peptide R3.Approximately 30 spots/10,000 CD8þ T cells were de-tected for the Influenza-Matrix-Protein (IMP) peptide asa positive control. In this patient, the frequency of CD8þ

T cells recognizing the RHAMM/CD168-derived R3 pep-tide was even higher as for IMP. Cr-51 release assaysshowed a strong specific lysis of R3-pulsed T2 cells byR3-peptide-presensitized CD8þ T lymphocytes from

Sp

ots p

er 1 x 10E

4 C

D8+

cells

0

10

20

30

40

50

60

70

80

Negative control IMP RHAMM/CD168-R3

*

**

* p< 0.05 ** p<0.01

IMP: Influenza matrix protein

RHAMM/CD168-R3: R3 peptide derived from RHAMM/CD168

Figure 5. Specific immune response (assessed by ELISPOT assay) against

RHAMM/CD168 R3-pulsed T2 cells in CML. This figure shows specific

CD8þ-mediated T-cell responses against the RHAMM/CD168-R3 peptide

in an ELISPOT assay for Granzyme B. The error bars indicate the standard

deviation of the assay performed in triplicate. *p ! 0.05; **p ! 0.01.


CML patients. In contrast, T2 cells pulsed with an irrele-vant HLA-A2 matching MAGE-3-derived peptide werenot recognized/lysed by such presensitized CD8þ T cells,thus excluding cross-reactivity. Moreover, K562 cells,which express RHAMM/CD168 but not HLA-ABC mole-cules, were not lysed by R3-primed T cells, indicatingthat CD8þ T cells and not contaminating NK cells werethe effector cells (Fig. 6A).

In Figure 6B, specific lysis of RHAMMþ blasts froma CML patient by R3-peptide-presensitized CD8þ T cellswas shown. In contrast, neither T2 cells pulsed withMAGE-3 peptide could be recognized by R3-peptide-pre-sensitized T cells, nor CML blasts by MAGE-3 peptide-pre-sensitized CD8þ T cells (Fig. 6B).

T2-R3

T2-MAGE3

K562

20:1 10:1 5:1 2.5:1

Sp

ecific lysis %

Effector / Target (E/T) cell ratio

20:1 10:1 5:1 2.5:1

Effector / Target (E/T) cell ratio

100

80

60

40

20

0

Sp

ecific lysis %

100

80

60

40

20

0

CML blasts/R3-primed

CD8+ T cells

T2 cells+MAGE-3/R3-

primed CD8+ T cells

CML blasts/MAGE-3

primed CD8+ T cells

A

B

Figure 6. High specific lysis of RHAMM/CD168-R3-pulsed T2 cells by

CD8þ T cells from the peripheral blood of CML patients. R3-peptide-pre-

sensitized CD8þ T lymphocytes isolated from the peripheral blood of

a CML patient were able to lyse T2 cells when pulsed with the R3 peptide

but not when pulsed with an irrelevant HLA-A2 matching MAGE3 pep-

tide. K562 cells (RHAMM/CD168 positive, but deficient for HLA-mole-

cules) were also used as a negative control to demonstrate that T cells

but not NK cells are the lytic effector cells (A). Specific lysis of CML

blasts expressing RHAMM protein by R3-presensitized CD8þ T cells

could be demonstrated in Cr-51 release assays (B).

Tetramer staining of CD8þ Tlymphocytes recognizing the RHAMM/CD168-derived peptide R3 in the context of HLA-A2Four-color FACS analysis of lymphocytes (Fig. 7) revealed1.1% of CD8þ T cells (gate R1 of Fig. 7A) specifically rec-ognizing the R3 peptide (as stained in R3-tetramers inFig. 7B) after one round of MLPC presensitization withthe R3 peptide. CD8þHLA-A2/R3-tetramerþ T lympho-cytes elicited by stimulation through the R3 peptide weregated (R2 gate) and were further characterized as predom-inantly (72%) CCR7�CD45RAþ (Fig. 7C) and CD27�

CCR7� (57%) (Fig. 7D) early effector T cells.Figure 8 demonstrates the specificity of this T-cell reac-

tion: CD8þ T cells stimulated with the peptide R3 but not Tcells stimulated with an irrelevant peptide MAGE-3 couldbe stained by R3-tetramers (Fig. 8A–C).

DiscussionSpecific CTL responses against CML progenitor cellsmight eliminate effectively minimal residual disease(MRD) after treatment with imatinib or chemotherapy.Moreover, immunotherapies targeting TAAs/LAAs mightconstitute an option for the specific enhancement ofthe GVL effect without aggravation of the graft-versus-host disease (GVHD) after hematopoetic stem celltransplantation.

A molecular monitoring of CML is available through thequantitative RT-PCR for bcr-abl expression [16,17]. Thefusion protein bcr-abl is also a potential target structurefor immunotherapeutic approaches [18,19]. However, re-ports on bcr-abl epitopes have been contradictory, and spe-cific immune responses against bcr-abl have been restrictedto rather infrequent HLA molecules [18–21]. Peptides de-rived from Proteinase3 and WT-1 are targets of specificT-cell responses [22–25] and have already been used inclinical vaccination trials for different myeloid disorders[24,26–28]. However, WT1 is also expressed in CD34þ

cells of the normal hematopoiesis [28–30] and Proteinase3in several normal tissues [13,31].

Therefore, we investigated in this work further potentialtarget antigens in CML. Several LAAs described in the lit-erature for other myeloid leukemias (AML) [13,14] are ex-pressed in bcr-abl-positive CML cells: RHAMM/CD168,hTERT, MPP11, PRAME, Proteinase3, NEWREN60, andWT1. These antigens were expressed in more than 50%of tested CML samples, G250 at a lower frequency. Thecancer germ line antigens NY-ESO-1, SSX2, MAGE A1,and BAGE were not expressed. For the antigensRHAMM/CD168, WT1, PRAME, Proteinase3, hTERT,and G250 expressed in CML, the induction of specific T-cell responses against peptides has already been reportedfor patients with different tumor entities [11,14,32–40].The antigens NEWREN60 and Proteinase3 bear the poten-tial danger of eliciting autoimmune reactions as they are


R2:1.1%

0.5%

23%

1.5%

57%

CD8*PerCP

SS

C -

FSC-Height

A2-

R3tetram

er*P

E

CC

R7*A

PC

CD45RA*FITC CD27*FITC

R1

A B

C D

CC

R7*A

PC

256

192

128

64

0

0 64 128 192 256

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

2.5%

39%72%

4.5%

Figure 7. (A): Four-color staining of R3-specific T lymphocytes by tetramer staining. Four-color staining of CD8-positive lymphocytes stimulated with R3

peptide in an MLPC was performed (patients in chronic phase). The lymphocytes gated in R1 (A) stained double positive for CD8þ and HLA-A2/R3 tet-

ramer(þ). The majority of these T cells revealed to be CD8þHLA-A2/R3-tetramerþCCR7�CD45RAþ and CD8þHLA-A2/R3-tetramerþCCR7�CD27� early

effect or T cells (B–D).

expressed in different normal tissue [13,22,23,31]. In con-trast, for the antigens G250, hTERT, PRAME, andRHAMM/CD168, no expression has been detected in nor-mal hematopoiesis and in most normal tissues that alsohave been tested [13,15].

Antigens with positive expression in CML were found inall stages of the disease. However, in acceleration phase andblast crisis RHAMM/CD168, PRAME, and Proteinase3showed a higher expression frequency whereas G250 wasexpressed less frequently. This is in accordance to theexpression levels detected in solid tumors showing higherexpression of RHAMM/CD168 in metastasis and in highertumor stages [41–44]. In contrast to WT1, the antigenRHAMM/CD168 could not be detected in CML samplesof complete molecular remission by real-time RT-PCR.Moreover, RHAMM/CD168 protein could not be detectedin CD34þ bone marrow stem cells. The very low expressionof RHAMM/CD168 protein by G-CSF-mobilized CD34þ

peripheral blood stem cells indicates an influence of G-CSF on the LAA expression.

CD34þ CML samples expressed HLA-ABC and HLA-DR molecules, but all CD34þ samples showed a lack of ex-pression of the costimulatory molecules CD40, CD80,CD83, and CD86, as typical for CD34þ early hematopoi-etic precursor cells (data not shown). These findings con-firm earlier data in leukemic cells [45,46]. In our study,we specified this observation for CML cells by gating onCD34þ cells. The lack of costimulatory molecules mightbe an essential step in the process of tumor escape ofCML cells, especially of progenitor cells. However, wewere able to detect specific CD8þ T-cell responses againstthe RHAMM/CD168-derived R3 peptide in CML patients.Such peptide primed CD8þ T cells were able to lyse R3-peptide pulsed T2 cells. The presence of R3 peptide-spe-cific early effector T cells in the peripheral blood ofCML patients could be confirmed by tetramer staining asCD8þHLA-A2/R3-peptide tetramer þCCR7�CD27�CD45RAþ T cells.

Taking into consideration the expression pattern ofLAAs in normal tissue (including hematopoetic cells), the


Figure 8. Two-color staining of R3-specific CD8þ T lymphocytes (A) with additional controls using an irrelevant MAGE-3-peptide. CD8þ lymphocytes

from a CML patient were subjected to one round of stimulation with autologous CD8� APCs in the presence of R3 (B) or an irrelevant MAGE3-derived

peptide (C). A difference in frequency of R3-specific CD8þ T cells could be noted in both cultures underlining the specificity of the tetramer staining.

SSC, sideward scatter; FSC, forward scatter.

efficient induction of specific immune responses and thehigh frequency of expression of the antigen in CML sam-ples, RHAMM/CD168 seems to be an ideal candidate forvaccination strategies in CML. With respect to these crite-ria, G250, hTERT, PRAME, and WT-1 might also be fur-ther target structures in CML for peptide vaccinationstrategies for CML patients. Therefore, at our institutionan R3-peptide vaccination trial has already been startedfor patients with hematologic malignancies.

AcknowledgmentsWe thank Mrs. Anita Szmaragowska, Mrs. Marlies Gotz, and Mrs.Michaela Buck for their excellent technical work in this project.This project has been supported by a generous grant of theDeutsche Jose Carreras Leukaemie-Stiftung e.V., project no.DJCL-R03/13.

References1. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific

inhibitor of the Bcr-Abl tyrosine kinase in chronic myeloid leukemia.

N Engl J Med. 2001;344:1031–1037.

2. Deininger M, Buchdunger E, Druker BJ. The development of imatinib

as a therapeutic agent for chronic myeloid leukemia. Blood. 2005;105:

2640–2653.

3. Hochhaus A, La Rosee P. Imatinib therapy in chronic myelogenous

leukemia: strategies to avoid and overcome resistance. Leukemia.

2004;18:1321–1331.

4. Druker BJ, O’Brien SG, Cortes J, Radich J. Chronic myelogenous leu-

kaemia. Hematology (Am Soc Hematol Educ Program). 2002;111–

135.

5. Horowitz MM, Rowlings PA, Passweg JR. Allogeneic bone marrow

transplantation for CML: a report from the International Bone Marrow

Transplant Registry. Bone Marrow Transplant. 1996;17(Suppl 3):5–6.

6. Clift RA, Radich J, Appelbaum FR, et al. Long-term follow-up of

a randomized study comparing cyclophosphamide and total body irra-

diation with busulfan and cyclophosphamide for patients receiving al-

logeneic marrow transplants during chronic phase of chronic myeloid

leukemia. Blood. 1999;94:3960–3962.

7. Falkenburg JH, Wafelman AR, Joosten P, et al. Complete remission of

accelerated phase chronic myeloid leukemia by treatment with leuke-

mia-reactive cytotoxic T lymphocytes. Blood. 1999;94:1201–1208.

8. Kolb HJ, Schattenberg A, Goldman JM, et al. Graft-versus-leukemia

effect of donor lymphocyte transfusions in marrow grafted patients.

Blood. 1995;86:2041–2050.

9. Collins RH Jr, Shpilberg O, Drobyski WR, et al. Donor leukocyte in-

fusions in 140 patients with relapsed malignancy after allogeneic bone

marrow transplantation. J Clin Oncol. 1997;15:416–417.

10. Kolb HJ, Schmid C, Barrett AJ, Schendel DJ. Graft-versus-leukemia

reactions in allogeneic chimeras. Blood. 2004;103:767–776.


11. Molldrem JJ, Lee PP, Kant S, et al. Chronic myelogenous leukemia

shapes host immunity by selective deletion of high-avidity leuke-

mia-specific T cells. J Clin Invest. 2003;111:639–647.

12. Bentz M, Cabot G, Moos M, et al. Detection of chimeric BCR-ABL

genes on bone marrow samples and blood smears in chronic myeloid

and acute lymphoblastic leukemia by in situ hybridization. Blood.

1994;83:1922–1928.

13. Greiner J, Ringhoffer M, Taniguchi M, et al. mRNA expression for

leukemia-associated antigens in patients with acute myeloid leukemia

for the development of specific immunotherapies. Int J Cancer. 2004;

108:704–711.

14. Greiner J, Li L, Ringhoffer M, et al. Identification and characterization

of epitopes of the receptor for hyaluronic acid mediated motility

(RHAMM/CD168) recognized by CD8 positive T cells of HLA-A2

positive patients with acute myeloid leukemia. Blood. 2005;106:

938–945.

15. Greiner J, Ringhoffer M, Taniguchi M, et al. Receptor for hyaluronan

acid-mediated motility (RHAMM) is a new immunogenic leukemia-

associated antigen in acute and chronic myeloid leukemia. Exp Hem-

atol. 2002;30:1029–1035.

16. Muller MC, Gattermann N, Lahaye T, et al. Dynamics of BCR-ABL

mRNA expression in first-line therapy of chronic myelogenous leuke-

mia patients with imatinib or interferon alpha/ara-C. Leukemia. 2003;

17:2392–2400.

17. Hughes TP, Kaeda J, Branford S, et al. International Randomised

Study of Interferon versus STI571 (IRIS) Study Group. Frequency

of major molecular responses to imatinib or interferon alfa plus cytar-

abine in newly diagnosed chronic myeloid leukemia. N Engl J Med.

2003;349:1423–1432.

18. Yotnda P, Firat H, Garcia-Pons F, et al. Cytotoxic T cell response

against the chimeric p210 BCR-ABL protein in patients with chronic

myelogenous leukemia. J Clin Invest. 1998;101:2290–2296.

19. Clark RE, Dodi IA, Hill SC, et al. Direct evidence that leukemic cells

present HLA-associated immunogenic peptides derived from the

BCR-ABL b3a2 fusion protein. Blood. 2001;98:2887–2893.

20. Posthuma EF, van Bergen CA, Kester MG, et al. Proteosomal degra-

dation of BCR/ABL protein can generate an HLA-A*0301-restricted

peptide, but high-avidity T cells recognizing this leukemia-specific an-

tigen were not demonstrated. Haematologica. 2004;89:1062–1071.

21. Chen W, Qin H, Reese VA, Cheever MA. CTLs specific for bcr-abl

joining region segment peptides fail to lyse leukemia cells expressing

p210 bcr-abl protein. J Immunother. 1998;21:257–268.

22. Molldrem J, Dermime S, Parker K, et al. Targeted T-cell therapy for

human leukemia: cytotoxic T lymphocytes specific for a peptide de-

rived from proteinase 3 preferentially lyse human myeloid leukemia

cells. Blood. 1996;88:2450–2457.

23. Molldrem J, Lee PP, Wang C, Champlin RE, Davis MM. A PR1-hu-

man leukocyte antigen-A2 tetramer can be used to isolate low-fre-

quency cytotoxic T lymphocytes from healthy donors that

selectively lyse chronic myeloid leukemia. Cancer Res. 1999;59:

2675–2681.

24. Mailander V, Scheibenbogen C, Thiel E, Letsch A, Blau IW, Keilholz

U. Complete remission in a patient with recurrent acute myeloid

leukemia induced by vaccination with WT1 peptide in the absence

of hematological or renal toxicity. Leukemia. 2004;18:165–166.

25. Kurbegov D, Molldrem JJ. Immunity to chronic myelogenous leuke-

mia. Hematol Oncol Clin North Am. 2004;18:733–752.

26. Oka Y, Tsuboi A, Taguchi T, et al. Induction of WT1 (Wilms’ tumor

gene)-specific cytotoxic T lymphocytes by WT1 peptide vaccine and

the resultant cancer regression. Proc Natl Acad Sci U S A. 2004;101:

13885–13890.

27. Qazilbash MH, Wieder E, Rios R, et al. Vaccination with the PR1 leu-

kemia-associated antigen can induce complete remission in patients

with myeloid leukemia. Blood. 2004;104:259 (Abstract).

28. Bergmann L, Miething C, Maurer U, et al. High levels of Wilms’ tu-

mor gene (wt1) mRNA in acute myeloid leukemias are associated with

a worse long-term outcome. Blood. 1997;90:1217–1225.

29. Maurer U, Weidmann E, Karakas T, Hoelzer D, Bergmann L. Wilms

tumor gene (wt1) mRNA is equally expressed in blast cells from acute

myeloid leukemia and normal CD34þ progenitors. Blood. 1997;90:

4230–4232.

30. Greiner J, Ringhoffer M, Simikopinko O, et al. Simultaneous expres-

sion of different immunogenic antigens in acute myeloid leukemia

(AML). Exp J Hematol. 2000;28:1413–1422.

31. Dengler R, Munstermann U, al-Batran S, et al. Immunocytochemi-

cal and flow cytometric detection of proteinase 3 (myeloblastin) in

normal and leukaemic myeloid cells. Br J Haematol. 1995;89:250–

257.

32. Boel P, Wildmann C, Sensi ML, et al. BAGE: a new gene encoding an

antigen recognized on human melanomas by cytolytic T lymphocytes.

Immunity. 1995;2:167–175.

33. Van der Bruggen P, Traversari C, Chomez P, et al. A gene encoding an

antigen recognized by cytolytic T lymphocytes on a human melanoma.

Science. 1991;254:1643–1647.

34. Jager E, Chen YT, Drijfhout JW, et al. Simultaneous humoral and

cellular immune response against cancer-testis antigen NY-ESO-1:

definition of human histocompatibility leukocyte antigen (HLA)-A2-

binding peptide epitopes. J Exp Med. 1998;187:265–270.

35. Chen YT, Scanlan MJ, Sahin U, et al. A testicular antigen aberrantly

expressed in human cancers detected by autologous antibody screen-

ing. Proc Natl Acad Sci U S A. 1997;94:1914–1918.

36. Tureci O, Chen YT, Sahin U, et al. Expression of SSX genes in human

tumors. Int J Cancer. 1998;77:19–23.

37. Ikeda H, Lethe B, Lehmann F, et al. Characterization of an antigen

that is recognized on a melanoma showing partial HLA loss by

CTL expressing an NK inhibitory receptor. Immunity. 1997;6:199–

208.

38. Ohminami H, Yasukawa M, Fujita S. HLA class I-restricted lysis of

leukemia cells by a CD8þ cytotoxic T-lymphocyte clone specific

for WT1 peptide. Blood. 2000;95:286–293.

39. Scheibenbogen C, Letsch A, Thiel E, et al. CD8 T-cell responses to

Wilms tumor gene product WT1 and proteinase 3 in patients with

acute myeloid leukemia. Blood. 2002;100:2132–2137.

40. Kessler JH, Beekman NJ, Bres-Vloemans SA, et al. Efficient

identification of novel HLA-A(*)0201-presented cytotoxic T lym-

phocyte epitopes in the widely expressed tumor antigen PRAME

by proteasome-mediated digestion analysis. J Exp Med. 2001;193:

73–88.

41. Hall CL, Yang B, Yang X, et al. Overexpression of the hyaluronan

receptor RHAMM is transforming and is also required for H-ras

transformation. Cell. 1995;82:19–26.

42. Wang C, Thor AD, Moore DH II, et al. The overexpression of

RHAMM, a hyaluronan-binding protein that regulates ras signaling,

correlates with overexpression of mitogen-activated protein kinase

and is a significant parameter in breast cancer progression. Clin Can-

cer Res. 1998;4:567–576.

43. Rein DT, Roehrig K, Schondorf T, et al. Expression of the hyaluronan

receptor RHAMM in endometrial carcinomas suggests a role in tu-

mour progression and metastasis. J Cancer Res Clin Oncol. 2003;

129:161–164.

44. Maxwell CA, Rasmussen E, Zhan F, et al. RHAMM expression and

isoform balance predict aggressive disease and poor survival in multi-

ple myeloma. Blood. 2004;104:1151–1158.

45. Hirano N, Takahashi T, Takahashi T, et al. Expression of costimulatory

molecules in human leukemias. Leukemia. 1996;10:1168–1176.

46. Zheng Z, Takahashi M, Aoki S, et al. Expression patterns of costimu-

latory molecules on cells derived from human hematological malig-

nancies. J Exp Clin Cancer Res. 1998;17:251–258.

chronic myeloid leukemia cells express tumor-associated antigens eliciting specific cd8+ t-cell...

Documents