Expression of P2X5 in lymphoid malignancies results in LRH-1-specific cytotoxic T-cell-mediated lysis

Authors


Dr Harry Dolstra, Central Haematology Laboratory, Radboud University Nijmegen Medical Centre, Geert Grooteplein 8, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: h.dolstra@chl.umcn.nl

Summary

Minor histocompatibility antigens (MiHA) selectively expressed by haematopoietic cells are attractive targets for specific immunotherapy after allogeneic stem cell transplantation (SCT). Previously, we described LRH-1 as a haematopoietic-lineage restricted MiHA that is capable of eliciting an allogeneic cytotoxic T-lymphocyte (CTL) response after SCT and donor lymphocyte infusion. Importantly, the gene encoding LRH-1, P2X5, is not expressed in prominent graft-versus-host-disease target tissues such as skin, liver and gut. Here, we investigate whether LRH-1-specific immunotherapy may be exploited for the treatment of lymphoid malignancies. We examined P2X5 mRNA expression in a large panel of patient samples and cell lines from different types of lymphoid malignancies by real-time quantitative reverse transcription polymerase chain reaction. P2X5 mRNA was highly expressed in malignant cells from all stages of lymphoid development. Furthermore, all LRH-1-positive lymphoid tumour cell lines were susceptible to LRH-1 CTL-mediated lysis in flow cytometry-based cytotoxicity assays. However, interferon-γ production was low or absent after stimulation with some cell lines, possibly due to differences in activation thresholds for CTL effector functions. Importantly, primary cells from patients with lymphoid malignancies were effectively lysed by LRH-1-specific CTL. These findings indicate that MiHA LRH-1 is a potential therapeutic target for cellular immunotherapy of lymphoid malignancies.

Allogeneic haematopoietic stem cell transplantation (SCT) in combination with donor lymphocyte infusion (DLI) is an effective treatment for patients with haematological malignancies (Horowitz et al, 1990; Appelbaum, 2001). The therapeutic efficacy is attributed to the graft-versus-tumour (GVT) response, an immune reaction during which donor-derived cytotoxic T lymphocytes (CTL) eliminate malignant cells of the recipient (Goulmy, 1997; Appelbaum, 2001; Riddell et al, 2003; Bleakley & Riddell, 2004). Unfortunately, the beneficial GVT reactivity is often accompanied by graft-versus-host-disease (GVHD), which results in CTL-mediated damage of normal tissues. Minor histocompatibility antigens (MiHA) are the major target antigens of these immune responses and expansion of MiHA-specific CTL has been shown to coincide with tumour remission (Goulmy, 1997; Warren et al, 1998; Marijt et al, 2003; Spierings et al, 2004; Hambach & Goulmy, 2005; de Rijke et al, 2005; Akatsuka et al, 2007). Selective GVT responses in the absence of GVHD may result from MiHA with expression limited to haematopoietic cells or haematopoietic cell lineages, including their malignant counterparts (Goulmy, 1997; Appelbaum, 2001; Bleakley & Riddell, 2004). Therefore, haematopoietic-restricted MiHA are attractive targets for specific immunotherapy after allogeneic SCT.

We reported on the molecular characterization of HLA-B7-restricted MiHA LRH-1, which displays haematopoietic-restricted tissue distribution (de Rijke et al, 2005). Expression of the gene that encodes LRH-1, P2X5, has been demonstrated in peripheral blood T cells, B cells and natural killer (NK) cells as well as in myeloid progenitor cells and in lymphoid organs. Importantly, P2X5 mRNA is not expressed in prominent GVHD target tissues such as skin, liver and gut (de Rijke et al, 2005). Functional analysis with LRH-1-specific CTL showed efficient lysis of target cells of lymphoid origin, whereas no CTL recognition was observed when mature myeloid cells or non-haematopoietic cells were used as targets cells (de Rijke et al, 2005). These findings suggest that LRH-1 may be a target for specific immunotherapy after allogeneic SCT for lymphoid malignancies.

Here, we studied the expression of LRH-1 in a large panel of lymphoid leukaemias, different subtypes of non-Hodgkin lymphoma (NHL) and multiple myeloma (MM) as potential targets for LRH-1-specific immunotherapy. We found that LRH-1 is expressed in malignancies derived from all stages of lymphoid development. Furthermore, we demonstrated that lymphoid tumour cell lines as well as primary lymphoid tumour cells from patients are significantly recognized and lysed by LRH-1-specific CTL. These findings illustrate that LRH-1-specific immunotherapy may be applicable for a broad range of lymphoid malignancies.

Materials and methods

Patient and donor material

Primary tumour samples were used for real-time quantitative reverse transcription polymerase chain reaction (RT-PCR) analysis: B-cell acute lymphoblastic leukaemia (B-ALL; n = 14), T-cell ALL (T-ALL; n = 11), B-cell chronic lymphocytic leukaemia (B-CLL, n = 20), follicular lymphoma (FL, n = 12), diffuse large B-cell lymphoma (DLBCL, n = 11), Burkitt lymphoma (BL, n = 5), mantle cell lymphoma (MCL, n = 6), splenic marginal zone lymphoma (SMZL, n = 5) and MM (n = 6). Primary B-ALL, T-ALL and B-CLL cells were obtained from bone marrow or peripheral blood mononuclear cells (PBMC) of patients at diagnosis. Frozen cells from CLL patients were kindly provided by Dr J.W. Gratama (Department of Internal Oncology, Erasmus Medical Centre, Rotterdam, The Netherlands). The amount of tumour cells in the ALL and CLL samples was >90% as determined by flow cytometric analysis. The different NHL samples were obtained from cryopreserved tumour tissue. Diagnosis and percentage of tumour cells of the NHL samples was determined by histological examination and supplementary immunohistochemical staining (data not shown). The FL samples contained >80% tumour cells. The BL and MCL samples contained >90% tumour cells. The DLBCL and SMZL samples contained 50–60% tumour cells. CD38 brightly positive plasma cells with a purity of >95% were sorted from bone marrow cells of MM patients using the Coulter Epics Elite flow cytometer (Beckman Coulter, Fullerton, CA, USA). Fibroblasts were isolated from bone marrow of healthy donors and cultured as previously described (de Rijke et al, 2005). All patient and donor material was obtained with informed consent obtained within clinical protocols in our institute.

Cell culture

LRH-1-specific CD8+ CTL clone RP1 was isolated from a CML patient who was successfully treated with DLI after allogeneic SCT (de Rijke et al, 2005). CTL RP1 recognizes the 9-mer epitope TPNQRQNVC in the context of HLA-B*0702. The HLA-B7-alloreactive CD8+ CTL clone KOR18 was kindly provided by Prof. Dr E. Goulmy (Dept. of Immunohematology, LUMC, The Netherlands). Both CTL RP1 and KOR18 (0·5 × 106) were cultured in Iscove’s modified Dulbecco’s medium (IMDM) (Invitrogen, Carlsbad, CA, USA) supplemented with 10% human serum (HS) (PAA laboratories, Pasching, Austria) containing irradiated (80 Gy) HLA-B7+ LRH-1+ Epstein–Barr virus (EBV)-transformed lymphoblastoid cell line (LCL) (0·5 × 106), irradiated (60 Gy) allogeneic PBMC (0·5 × 106) from two donors, 100 IU/ml interleukin (IL)-2 (Chiron, Emeryville, CA, USA) and 1 μg/ml phytohaemagglutinin (PHA)-M (Boehringer Ingelheim, Ingelheim, Germany) in a total volume of 2 ml. CTL cultures were used in stimulation assays 5–7 d after stimulation. Cell lines were cultured in IMDM/10% fetal calf serum (FCS). T-cell blasts were generated by stimulating CD3+ T cells in IMDM/10% HS with 20 μg/ml PHA-M for 3 d. Thereafter, PHA-activated T cells were washed and further cultured with 100 IU/ml IL-2 for two additional days.

Retroviral transduction of HLA-B*0702 in cell lines and CTL stimulation assay

HLA-B*0702 cDNA (kindly provided by Prof. Dr E. Goulmy) was subcloned in the LZRS-IRES-EGFP vector. The resulting LZRS-HLA-B*0702-IRES-EGFP vector was used to generate a stable producer cell line. Retroviral transduction was performed using non-tissue culture-treated 35-mm2 dishes (Becton Dickinson, Franklin Lakes, NJ, USA) coated with 10 mg/ml RetroNectin (Takara Biomedicals, Otsu, Japan). Target cells (106) were resuspended in 2 ml virus supernatant and transferred to RetroNectin-coated dishes. After 24 h of incubation, cells were collected and incubated with fresh virus supernatant. Finally, transduced cells were cultured for at least five additional days before use in CTL stimulation assays. Briefly, CTL (5 × 103) were cultured in 96-well flat-bottom plates in the presence of target cells (3 × 104) and 25 U/ml IL-2 in a total volume of 200 μl. Cells were incubated for 18 h at 37°C. Release of interferon-γ (IFN-γ) was determined by enzyme-linked immunosorbent assay (ELISA; Pierce Endogen, Rockford, IL, USA).

Real-time quantitative RT-PCR analysis of P2X5 gene

Total RNA was isolated from cell samples using Trizol (Invitrogen) or the Zymo RNA isolation kit II (Zymo Research Corporation, Orange, CA, USA). First strand cDNA was prepared from 2 μg total RNA using oligo-dT, random hexamers and Mo-MuLV reverse transciptase (Invitrogen). The housekeeping gene hydroxymethylbilane synthase (HMBS) was used to normalize P2X5 expression. One microlitre of cDNA was amplified in a 50 μl reaction mixture containing 1·25 U AmpliTaq Gold (Applied Biosystems, Foster City, CA, USA), 300 nmol/l gene-specific forward and reverse primer, 150–300 nmol/l gene-specific Taqman probe (150 nmol/l for P2X5 and 300 nmol/l for HMBS), 250 μmol/l of each dNTP, 5 mmol/l MgCl2 and 1x Taqman PCR buffer (Applied Biosystems). The following gene-specific primers and Taqman probes were used: P2X5; P2X5-F 5′-TCCTGGCGTACCTGGTCGT-3′, P2X5-R 5′-CTTCATTCTCAGCACAGACGTTC-3′ and P2X5-probe 5′-(TET)-TGGGTGTTCCTGATAAAGAAGGGTTACCA-(TAMRA)-3′, and HMBS; HMBS-F 5′-GGCAATGCGGCTGCAA-3′, HMBS-R 5′-GGGTACCCACGCGAATCAC-3′ and HMBS-probe 5′-(VIC)-CTCATCTTTGGGCTGTTTTCTTCCGCC-(TAMRA)-3′. PCR amplification was performed using an ABI Prism 7700 (Applied Biosystems) with the following PCR conditions: enzyme activation for 10 min at 95°C, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min. P2X5 mRNA expression was quantified by determining calibration functions using JVM-2 as reference cell line. The level of P2X5 expression of test samples was calculated relative to the P2X5 expression in the JVM-2 cell line.

P2X5 genotyping

Genomic DNA was isolated from cell lines using the QIAamp DNA Blood mini kit (Qiagen, Hilden, Germany). Nine nanograms of DNA was amplified in a 50 μl reaction mixture containing 1·25 U AmpliTaq Gold (Applied Biosystems), 300 nmol/l P2X5-specific forward and reverse primer (Biolegio, Nijmegen, The Netherlands), 200 nmol/l of each allele-specific Taqman probe, 250 μmol/l of each dNTP, 5 mmol/l MgCl2 and 1x Taqman PCR buffer (Applied Biosystems). The following gene-specific primers and Taqman probes were used: P2X5; P2X5-EXON3-F 5′-CCAAATCAAACCTCAGCACAGAC-3′, P2X5-EXON3-R 5′-CTCAGTGCCTCTCTGGTTCCTTA-3′, P2X5 5C allele-specific probe 5′-(FAM)-ATTGTGACCCCCAACCA-(MGB)-3′ and P2X5 4C allele-specific probe 5′-(VIC)-TGTGACCCCAACCAG-(MGB)-3′. PCR amplification was performed using an ABI Prism 7700 (Applied Biosystems) with the following PCR conditions: enzyme activation for 10 min at 95°C, followed by 45 cycles of 92°C for 15 s and 62°C for 1 min. After amplification, an end-plate read was performed using ABI Prism 7700 allelic discrimination software.

Flow cytometry-based cytotoxicity studies

Flow cytometry-based cytotoxicity assays were performed as described by Jedema et al (2004) with some adaptations. Briefly, target cells were washed with phosphate-buffered saline and resuspended at 20 × 106 cells/ml. Endogenous HLA-B7+ cell lines, HLA-B7+ EBV-LCL and HLA-B7+ primary B-ALL cells were labelled with 5, 1 and 0·125 μmol/l carboxyfluorescein diacetate succimidyl ester (CFSE; Molecular Probes Europe, Leiden, The Netherlands) respectively. The HLA-B7+ primary B-ALL cell samples from patients 1 and 2 contained 90% CD34+ tumour cells. Cells were labeled with CFSE for 10 min at 37°C. The reaction was terminated by adding an equal volume of FCS, followed by incubation at room temperature for 2 min and washing. Cell lines retrovirally transduced with LZRS-HLA-B*0702-IRES-EGFP were used as targets without CFSE labelling. Target cells (1 × 104) were cultured with effector cells (5 × 103) at an effector:target (E:T) ratio of 0·5:1 in a total volume of 200 μl IMDM/10% FCS containing 25 U/ml IL-2 in 96-well round-bottom plates. Alternatively, HLA-B7+ primary MM cells were used unlabelled at an E:T ratio of 3:1 calculated for the plasma cell content of the sample, which was 7 and 17% CD138+ cells for patients 3 and 4 respectively. To determine absolute cell numbers prior to co-culture with CTL a fixed amount of Flow-Count Fluorospheres (1 × 104) (Beckman Coulter) was added to 100 μl of target cells and 5 × 103 beads were acquired by flowcytometry. After 1–3 d of co-culture, cells were harvested and Flow-Count Fluorospheres and 7-amino-actinomycin D (7AAD; Sigma, St. Louis, MO, USA) were added to quantify the number of viable target cells. In addition, primary B-ALL cells were labelled with anti-CD34 (clone 581-ECD; Immunotech Beckman Coulter) and primary MM cells were labeled with anti-CD138 (clone B-B4-PE; IQ Products, Groningen, The Netherlands) and anti-CD38 (clone HB7-FITC; Becton Dickinson) to analyse tumour cell populations. For each sample 5 × 103 beads were acquired and the absolute numbers of viable target cells were determined. The percentage of cell survival in the presence of CTL compared to medium was calculated using the following formula: % survival = {[absolute no. viable CFSE+ or GFP+ target cells co-cultured with CTL(t = x)]/[absolute no. viable CFSE+ or GFP+ target cells cultured in medium (t = x)]} × 100%. The percentage of CTL specific lysis was calculated as follows: inline image

Results

P2X5 mRNA expression in primary tumour cells of lymphoid malignancies

To define which lymphoid malignancies could be targets for LRH-1-specific CTL, P2X5 mRNA expression was determined by real-time quantitative RT-PCR in a large panel of primary lymphoid tumour samples including B-ALL, T-ALL, different subtypes of B-NHL and MM. Previously, we demonstrated that EBV-LCL, which had a mean expression level of 2·7 compared with the reference B-cell line JVM-2, were significantly recognized by LRH-1-specific CTL. In contrast, monocytes and fibroblasts with a mean P2X5 mRNA expression level of 0·28 and 0·10, respectively, are not susceptible to LRH-1 CTL-mediated lysis. Based on these observations, we used a cut-off P2X5 mRNA level of 0·4 to distinguish LRH-1-positive from -negative cell types (de Rijke et al, 2005) Interestingly, all the lymphoid malignancies analysed showed significant P2X5 mRNA expression at higher levels than in EBV-LCL and PHA-activated T cells (Fig 1). The highest P2X5 expression level was observed in B-CLL samples (mean = 23·0). Furthermore, high mRNA transcription levels were detected in B-ALL (mean = 11·6), whereas the lowest P2X5 mRNA levels were found in T-ALL samples (mean = 3·1). In the different types of NHL, mean expression levels varied between 8·1 for FL, 8·0 for DLBCL, 4·6 for BL, 11·0 for MCL and 15·2 for SMZL. The mean P2X5 mRNA expression in CD38++ MM cells was 7·4. Collectively, these results demonstrate that LRH-1 encoding P2X5 mRNA is highly expressed in tumour cells originating from all stages of lymphoid development.

Figure 1.

P2X5 expression in primary tumour samples of lymphoid malignancies. P2X5 mRNA expression was determined by real-time quantitative RT-PCR in lymphoid tumour cells. The following tumour types were analysed: B-ALL (n = 14), T-ALL (n = 11), B-CLL (n = 20), FL (n = 12), DLBCL (n = 11), BL (n = 5), MCL (n = 6), SMZL (n = 5) and MM (n = 6). Expression is shown relative to the P2X5 expression measured in the reference cell line JVM-2, which is susceptible to lysis by the LRH-1-specific CTL (de Rijke et al, 2005). The housekeeping gene HMBS was used for normalization. Tumour cells with P2X5 expression higher than 0·4 were considered positive and may be targets for LRH-1-specific CTL (de Rijke et al, 2005). This arbitrary threshold is indicated by the horizontal dashed line. The mean expression level for each cell population is shown by a thick horizontal line.

P2X5 expressing malignant lymphoid cell lines are recognized by LRH-1-specific CTL

To investigate potential CTL recognition of lymphoid malignancies, P2X5 gene expression levels were measured by real-time quantitative RT-PCR in a panel of 18 malignant lymphoid cell lines. All cell lines tested were P2X5 positive (i.e. P2X5 mRNA level >0·4) (Table I). Since target cell recognition by LRH-1-specific CTL is controlled by a single cytosine deletion polymorphism in exon 3 of P2X5, we performed P2X5 genotyping analysis on the panel of tumour cell lines. Eleven of the 18 cell lines (61%) were either homozygous (C/C genotype) or heterozygous (C/− genotype) positive for the LRH-1-encoding allele and, therefore, could be potentially targeted by LRH-1-specific CTL. Since only four of the 18 cell lines endogenously expressed HLA-B7, the other 14 cell lines were retrovirally transduced with HLA-B*0702. Next, lymphoid tumour cell lines with endogenous or ectopic expression of HLA-B7 were tested for stimulation of IFN-γ release by LRH-1-specific CTL (Table I). For comparison, EBV-LCL generated from the CML patient (UPN389) from whom the LRH-1-specific CTL was originally isolated (de Rijke et al, 2005), and the EBV-LCL generated from her transplant donor were used as positive and negative controls respectively. All HLA-B7-expressing lymphoid tumour cell lines were significantly recognized by the allo-HLA-B7-specific CTL, indicating susceptibility of the cell lines for appropriate T-cell stimulation (range 243–1584 pg/ml IFN-γ; Table I). The majority of the homozygous and heterozygous LRH-1+ tumour cell lines were significantly recognized by LRH-1-specific CTL (>100 pg/ml IFN-γ). However, the cell lines HPB-ALL, HS-SULTAN and RPMI8226 stimulated IFN-γ production by LRH-1-specific CTL just slightly above background level (123, 141 and 107 pg/ml IFN-γ respectively). No IFN-γ production could be induced by the cell lines KM3 and RPMI1788 despite the LRH-1-positive genotype and functional recognition by the allo-HLA-B7-specific CTL (Table I). These data demonstrate that most LRH-1-postitive lymphoid tumour cell lines can be recognized by LRH-1-specific CTL, although some tumour cell lines are relatively defective in stimulating significant IFN-γ production.

Table I. P2X5 mRNA expression, genotype and CTL recognition of lymphoid tumour cell lines.
Tumour typeCell lineP2X5 expression†P2X5 genotype‡IFN-γ production (pg/ml)§
LRH-1 CTLHLA-B7 CTL
mock+ B7mock+ B7
  1. *Cell lines with endogenous expression of HLA-B7.

  2. †Normalized P2X5 mRNA expression was determined by real-time quantitative RT-PCR and expressed relative to the expression level in the B-cell prolymphocytic leukaemia (B-PLL) JVM-2 cell line.

  3. P2X5 genotyping was performed by PCR amplification of genomic DNA using allele-specific probes as described in Materials and methods.

  4. §Cell lines were retrovirally transduced with HLA-B*0702 and tested for recognition by LRH-1-specific CTL and HLA-B7-specific CTL. IFN-γ production was measured by ELISA.

  5. <, no significant IFN-γ release by CTL (<100 pg/ml); n.d., not determined, EBV, Epstein–Barr virus; LCL, lymphoblastoid cell line; B-ALL, B-cell acute lymphoblastic leukaemia; B-NHL, B-cell non-Hodgkin lymphoma; MM, multiple myeloma.

EBV-Recipient*2·8C/−1296n.d.1612n.d.
LCLDonor*2·5−/−<n.d.1063n.d.
B-ALLKM32·2C/−<<<1264
BV1731·5−/−<<< 833
T-ALLHSB28·6C/−< 568<1531
HPB-ALL1·7C/−< 123<1096
MOLT41·7−/−<<< 243
CEM1·1−/−<<< 574
B-PLLJVM-2*1·0C/−1273125316121584
B-NHLRAJI8·0C/C< 892<1557
HS-SULTAN6·8C/C< 141< 754
BJAB3·9 C/C< 291<1557
RPMI 17882·0C/−<<< 989
U-698-M*2·5−/−<< 7711030
RAMOS1·6−/−<<< 304
SU-DHL-61·3−/−<<< 311
MMRPMI 82261·5C/C< 107< 833
UM9*1·3C/− 109 25916981455
UM10·7C/−< 281<1455
U266*0·6−/−<<11811455

P2X5 expressing malignant lymphoid cell lines are efficiently lysed by LRH-1-specific CTL

To investigate whether malignant lymphoid cell lines are susceptible for lysis by LRH-1-specific CTL, we performed flow cytometry-based cytotoxicity assays. This assay facilitates determination of target cell proliferation and death by both rapid and more slowly T-cell effector mechanisms (Jedema et al, 2004). Moreover, the addition of low-dose IL-2 to the co-cultures prolongs CTL survival, which allows continuous exposure and serial killing of tumour cells by the CTL. Using this in vitro assay we observed that LRH-1-specific CTL efficiently lysed and inhibited growth of the LRH-1+ cell lines RAJI and UM9 at a very low E:T ratio of 0·5:1 (Fig 2A and C). No cytotoxicity by LRH-1-specific CTL was observed against the LRH-1 cell lines RAMOS and U266 (Fig 2B and D). By contrast, allo-HLA-B7-specific CTL lysed both LRH-1+ and LRH-1 cell lines (Fig 2A–D). An overview of the percentage specific lysis by LRH-1-specific CTL compared to medium of all 18 malignant lymphoid cell lines tested is shown in Fig 2E. All HLA-B7-expressing tumour cell lines were significantly lysed by the allo-HLA-B7-specific CTL, indicating that HLA-B7 molecules were properly expressed (data not shown). Interestingly, LRH-1-specific CTL effectively lysed all tumour cell lines that were either homozygous (C/C) or heterozygous (C/−) positive for the LRH-1-encoding allele at a very low E:T ratio of 0·5:1 after 2 d of co-culture (Fig 2). No or low background cytotoxicity was observed against cell lines that were homozygous (−/−) negative for the LRH-1 encoding allele, confirming the specificity of LRH-1-specific CTL. These data clearly demonstrate that all tested LRH1+ tumour cell lines, including those that are inefficient in stimulating IFN-γ production, are highly susceptible to direct lysis by LRH-1-specific CTL.

Figure 2.

 Specific cytotoxicity of LRH-1-specific CTL against lymphoid tumour cell lines. Cytotoxicity against B-NHL and MM cell lines was determined after incubation with LRH-1-specific CTL (bsl00066), HLA-B7-specific CTL (bsl00072; i.e. positive control) or medium (♦) in flow cytometry-based cytotoxicity assays (A–D). Survival of cells is shown in the absence or presence of CTL at an E:T ratio of 0·5:1. Data are depicted as mean ± SD of triplicate wells. Specific lysis of lymphoid cell lines (n = 18) of several subtypes (B-ALL, T-ALL, B-PLL, B-NHL and MM) was determined after 2 d of co-culture in a flow cytometry-based cytotoxicity assay at an E:T ratio of 0·5:1 (E). Black bars represent cell lines positive for the LRH-1-encoding allele (C/C or C/− genotype). Grey bars represent cell lines negative for the LRH-1-encoding allele (−/−). * indicates cell lines with endogenous expression of HLA-B7.

Primary B-ALL and MM cells are efficiently lysed by LRH-1-specific CTL

To investigate whether primary lymphoid tumour cells are susceptible to LRH-1 CTL-mediated lysis, we performed flow cytometry-based cytotoxicity assays with tumour cells from patients with lymphoid malignancies. For these assays we selected HLA-B7+ patients to circumvent the need for retroviral transduction of primary patient cells. P2X5 genotyping analysis on a cohort of 15 patients showed that nine of 15 patients samples (60%) were either homozygous (C/C genotype) or heterozygous (C/− genotype) positive for the LRH-1-encoding allele and, therefore, potentially susceptible to LRH-1 CTL-mediated lysis. Primary tumour cells from two HLA-B7+ B-ALL and two HLA-B7+ MM patients were tested in flow cytometry-based cytotoxicity assays. We observed that CD34+ B-ALL cells from LRH-1+ patient 1 were efficiently lysed by LRH-1-specific CTL at a very low E:T ratio of 0·5:1 (Fig 3A). No cytotoxicity by LRH-1-specific CTL was observed against CD34+ B-ALL cells from LRH-1 patient 2 (Fig 3B). By contrast, CD34+ B-ALL cells from both patients were efficiently lysed by allo-HLA-B7-specific CTL (Fig 3A and B), which confirmed expression of the HLA-B7 molecule. In addition, primary cells from two LRH-1+ MM patients were tested for susceptibility to LRH-1 CTL-mediated lysis. Both LRH-1-specific CTL and allo-HLA-B7-specific CTL efficiently killed the CD138+ LRH-1+ plasma cells (Fig 3C and D) at an E:T ratio of 3:1. The LRH-1+ EBV-LCL generated from CML patient UPN389 (de Rijke et al, 2005) and the LRH-1 EBV-LCL generated from her transplant donor were tested as controls (Fig 3E and F) at an E:T ratio of 3:1. These data demonstrate that LRH-1-specific CTL effectively recognized and killed primary tumour cells from both immature and mature lymphoid malignancies.

Figure 3.

 Primary B-ALL and MM cells are efficiently lysed by LRH-1-specific CTL. Cytotoxicity against B-ALL and MM cells from four HLA-B7+ patients was determined after incubation with LRH-1-specific CTL (bsl00066), HLA-B7-specific CTL (bsl00072; i.e. positive control) or medium (♦) in flow cytometry-based cytotoxicity assays. Survival of viable CFSE-labelled CD34+ B-ALL cells is shown from a LRH-1+ (A) and a LRH-1 (B) B-ALL patient in the absence or presence of CTL at an E:T ratio of 0·5:1. Survival of viable CD138+ MM cells is shown from two LRH-1+ MM patients (C, D) in the absence or presence of CTL at an E:T ratio of 3:1. Survival of control EBV-LCL generated from CML patient UPN389 (LRH-1+) and her transplant donor (LRH-1) was tested in the absence or presence of CTL at an E:T ratio of 3:1 (E, F). Data are depicted as mean ± SD of triplicate wells.

Discussion

Human haematopoietic-restricted MiHA represent suitable targets for effective cellular immunotherapy to prevent or to treat relapse of haematological malignancies after allogeneic SCT. Previously, we identified the molecular nature of MiHA LRH-1 using genetic linkage analysis (de Rijke et al, 2005). Gene expression analysis using real-time quantitative RT-PCR revealed significant P2X5 mRNA levels in peripheral blood T cells, B cells and NK cells, lymphoid tissues and in normal and malignant CD34+ progenitor subpopulations (de Rijke et al, 2005). Regarding normal tissues, only limited P2X5 expression was found in normal adult brain and skeletal muscle (de Rijke et al, 2005). Importantly, no P2X5 expression was detected in prominent GVHD target organs, such as skin, liver and gut, indicating that LRH-1-specific CTL could play an important role in GVT-specific immunity without causing severe GVHD.

This study investigated which types of lymphoid malignancies are potential targets for LRH-1-specific immunotherapy. A large number of patient samples from different lymphoid malignancies were analysed for P2X5 expression using real-time quantitative RT-PCR. Analysis revealed that the P2X5 transcript encoding the LRH-1 epitope is significantly expressed in a broad range of primary lymphoid malignancies originating from all stages of lymphoid development, including B-ALL, T-ALL, B-CLL, MM and various B-NHL subtypes. The highest mean P2X5 mRNA expression level was detected in B-CLL, whereas the lowest mean level was detected in T-ALL. Despite the fact that in B-ALL a P2X5 mRNA positive group (expression levels between 0·7 and 2·5), and a group with higher expression levels (range 9·7–38·7) seem to be present, no correlation between B-ALL subtype (pro-/pre-/common-B-ALL) and P2X5 expression could be detected (data not shown). Nevertheless, both groups had significant P2X5 mRNA levels which are expected to be sufficient for LRH-1 CTL-mediated lysis (de Rijke et al, 2005). Of the B-NHL subtypes tested, the FL, BL and MCL samples contained high percentages of tumour cells and significant P2X5 mRNA levels of 8·1, 4·6 and 11·0 were detected respectively. The DLBCL and SMZL samples contained lower percentages of tumour cells and had higher numbers of T cells. However, the P2X5 mRNA expression of 8·0 in DLBCL and 15·2 in SMZL is much higher than the expression level detected in T cells activated with PHA and IL-2 (mean 2·0). This indicates that the P2X5 mRNA expression in these samples is not only attributed to the presence of T cells, but is most probably due to high expression in B-lymphoid tumour cells. Furthermore, this finding is supported by the P2X5 expression observed in B-NHL cell lines and the susceptibility of these cell lines to LRH-1-specific CTL.

Flow cytometry-based cytotoxicity assays with lymphoid tumour cell lines clearly demonstrated that LRH-1-specific CTL were able to significantly lyse LRH-1+ lymphoid tumour cell lines, which had P2X5 mRNA expression levels ranging from 0·7 to 8·6. However, some lymphoid tumour cell lines appeared to be defective in stimulating IFN-γ production by LRH-1-specific CTL. Similarly, we demonstrated that both immature CD34+ LRH-1+ B-ALL cells and mature CD138+ LRH-1+ MM cells are effectively lysed by LRH-1-specific CTL in flow cytometry-based cytotoxicity assays, whereas LRH-1-specific CTL were not stimulated sufficiently for production of IFN-γ (data not shown). This discrepancy may be explained by the presence of dual activation thresholds for cytotoxicity and cytokine production in CTL upon antigen-specific interactions with its target (Valitutti et al, 1996; Faroudi et al, 2003). For instance, at low antigen concentrations a lytic immunological synapse (IS) can be formed, which is sufficient to elicit cytotoxicity towards multiple targets via polarized secretion of lytic granules (Faroudi et al, 2003; Wiedemann et al, 2006). However, higher antigen concentrations and the formation of a mature IS are needed for the induction of IFN-γ production (Valitutti et al, 1996; Faroudi et al, 2003). Besides a sufficient antigen concentration, the presence of adhesion molecules on the target cells, such as intercellular adhesion molecule 1, is needed for the formation of a mature IS (Cemerski & Shaw, 2006). Both the antigen concentration and the expression levels of adhesion molecules probably differ between the various lymphoid malignancies resulting in different capacity to stimulate IFN-γ production by LRH-1-specific CTL. Nevertheless, our data clearly show that LRH-1 epitope density on primary lymphoid tumour cells is high enough for efficient lysis by CTL. Furthermore, our findings indicate that the flow cytometry-based cytotoxicity assay is a more valuable and sensitive method to determine CTL-mediated recognition of malignant target cells compared to CTL stimulation assays in which IFN-γ production is measured.

In conclusion, we demonstrated that P2X5 is highly expressed in a broad range of lymphoid malignancies and that primary lymphoid tumour cells are susceptible to LRH-1-specific lysis. Therefore, LRH-1 represents an attractive GVT-specific target to exploit by cellular immunotherapeutic strategies to treat patients with relapsed lymphoid malignancies or residual disease after allogeneic SCT. In the studied patient cohort, we observed that 60% of the patients was positive for the LRH-1-encoding allele with genotype frequencies of 7% C/C, 53% C/− and 40%−/− respectively. Using the allele frequencies (P = 0·33 and q = 0·67) and formula pxq3 + 0·75xp2xq2 (Martin, 1997), we calculated that the recipient disparity between siblings for the LRH-1+ allele was 13·6%. The overall applicability of the HLA-B7-restricted LRH-1 epitope for immunotherapy after allogeneic SCT is estimated around 2·3%, based on LRH-1 recipient disparity (c. 13·6%) and the HLA-B7 phenotype frequency (c. 17% in the Caucasian population) (Marsh et al, 2000). This underscores the potential of LRH-1 as T-cell target in MiHA-specific immunotherapy.

Acknowledgements

We thank Frans Maas, Arie Pennings (deceased), Jeroen van Velzen and Rob Woestenenk for technical assistance. We thank Prof. Dr Frederik Falkenburg and Dr Inge Jedema (Department of Hematology, Leiden University Medical Center, Leiden, The Netherlands) for their advice on the flow cytometry-based cytotoxicity assays. We thank Dr Andries Bloem (Department of Immunology, University Medical Center Utrecht, Utrecht, The Netherlands) for providing the UM multiple myeloma cell lines. This work was supported by grants from the Dutch Cancer Society (KUN 2000–2294), Radboud University Nijmegen (UMCN 2002-29), the Bekales Foundation and the Ger Janssen Foundation.

Ancillary