Novel strategies for identification and therapeutic application of minor histocompatibility antigens

Transcription

1 Novel strategies for identification and therapeutic application of minor histocompatibility antigens Robbert Spaapen

2 ISBN: Cover photo by Chris Timmers, cover layout by Robbert Spaapen Genetic differences between individuals are the driving force towards immunotherapy after allogeneic stem cell transplantation Copyright 2009 by Robbert Spaapen No part of this thesis may be reproduced or transmitted in any form or by any means without written permission of the copyright owner. Printed by GVO drukkers & vormgevers B.V. Ponsen & Looijen Printing of this thesis was financially supported by Ortho Biotech Inc., J.E. Jurriaanse Stichting, BD Biosciences, Sanquin Reagents, Molecular Devices and Infection & Immunity Center Utrecht. Support was also granted by Van Dobben, Katja Fassin B.V., Duo Penotti and Verkade B.V.

3 Novel strategies for identification and therapeutic application of minor histocompatibility antigens Nieuwe strategieën voor identificatie en therapeutische applicatie van minor histocompatibility antigenen (met een samenvatting in het Nederlands) Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. J.C. Stoof, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op vrijdag 26 februari 2010 des middags te 2.30 uur door Robbert Michiel Spaapen geboren op 29 februari 1980 te Utrecht

4 Promotor: Copromotor: Prof. dr. H.M. Lokhorst Dr. T. Mutis

5 Table of contents 1. General introduction: Targeting hematopoietic-specific minor histocompatibility antigens to distinguish graft-versus-tumor effects from graft-versus-host disease. Adapted from: Best Pract Res Clin Haematol Sep;21: Toward targeting B cell cancers with CD4 + CTLs: Identification of a CD19-encoded minor histocompatibility antigen using a novel genomewide analysis. J Exp Med Nov;205: Rapid identification of clinically relevant minor histocompatibility antigens via genome-wide zygosity-genotype correlation analysis. Clin Cancer Res Dec; 15: Genome-wide predictions of minor histocompatibility antigens based on differential HLA-binding of polymorphic peptides. Submitted Rebuilding human leukocyte antigen class II-restricted minor histocompatibility antigen specificity in recall antigen-specific T cells by adoptive T cell receptor transfer: Implications for adoptive immunotherapy. Clin Cancer Res Jul;13: Induction of potent in vivo anti-tumor effects by native and TCR-redirected CD4 + cytotoxic human T lymphocytes directed at minor histocompatibility antigens. Submitted General discussion 111 Nederlandse samenvatting Dankwoord Curriculum vitae & Publications

6

7 1 General introduction: Targeting hematopoietic-specific minor histocompatibility antigens to distinguish graftversus-tumor effects from graft-versus-host disease Robbert M. Spaapen 1, Tuna Mutis 1 1 Department of Clinical Chemistry and Ha ematology, UMCU, NL Adapted from: Best Pract Res Clin Haematol Sep;21:543-57

8 8 Chapter 1 Allogeneic stem cell transplantation (allo-sct) and donor lymphocyte infusions can induce durable remission in patients with hematological malignancies through a graft-versus-tumor (GvT) effect. In human leukocyte antigen (HLA)-matched settings, this powerful immunotherapeutic effect is predominantly mediated by donor T cells directed at the recipient s minor histocompatibility antigens (mhags) presented on malignant cells. The mhags are short peptides excised from polymorphic regions of intracellular proteins, and are presented by HLA molecules to donor T cells. Several ubiquitously expressed mhags are involved not only in GvT but also in graft-versus-host disease (GvHD). However, a specific set of mhags is expressed exclusively by hematopoietic cells and their malignant counterparts. Targeting these hematopoietic mhags is an attractive strategy to induce specific GvT effects without increasing the risk of GvHD. This chapter will summarize the current efforts to identify therapeutically relevant hematopoietic mhags, and outline the strategies to apply mhag-based cellular immunotherapy to treat recurrent malignancies after allo-sct. Graft-versus-tumor effect of allogeneic stem cell transplantation Allogeneic stem cell transplantation (allo-sct) represents a powerful immunotherapeutic approach in the battle against several hematological malignancies 1,2. Early transplantation studies demonstrated that allo-sct mediated a significant therapeutic effect, which was not observed after autologous or syngeneic transplantation 1,2. The price paid for this potentially curative graft-versustumor (GvT) effect was the development of life-threatening graft-versus-host disease (GvHD), which remains a major source of morbidity and mortality following allo-sct. Removal of donor T cells from a hematopoietic graft has been shown to prevent GvHD, but also abrogates the therapeutic GvT effect 3, confirming the important role for alloreactive T cells in GvHD as well as in GvT. Finally, treatment with donor lymphocytes induces sustained remission in a subset of patients who relapse following allogeneic transplantation 4. Although the most powerful GvT effects are observed in patients with chronic myeloid leukemia, clear GvT effects have also been demonstrated in several other types of acute and chronic leukemia, lymphoma and in multiple myeloma Minor histocompatibility antigens: major targets of the GvT effect Starting from the earliest transplantation studies, it was obvious that elimination of GvHD, while preserving the GvT effect, would be the most important task towards safe and effective allo-sct. Although the risk of GvHD is reduced significantly by matching the recipient and the donor for human leukocyte antigens (HLA) 2, even after genotypically HLA-matched SCT, 20-50% of the recipients develop GvHD,

9 General introduction 9 which is also the most predictive factor for GvT 2. Of note, patients with mild GvHD demonstrate the best outcomes due to a lower risk of relapse while minimizing the toxicity associated with more advanced GvHD. Thus, even in fully HLA-matched transplant settings, the alloreactivity towards recipient cells is the key factor for the development of GvHD and GvT effects. In an HLA-identical setting, alloreactive donor T cells are directed at non-hla-encoded polymorphic antigens expressed by the recipient cells. These polymorphic antigens were originally discovered by Barth et al in mice and were designated as weak (minor) histocompatibility antigens (mhags) 14. However, current knowledge on the immunobiology of mhags indicates that the minor designation is misplaced since these antigens can induce very potent CD4 + and CD8 + T cell responses after HLA-matched, mhag-mismatched transplantation 15. Chapter 1 Biochemical nature of mhags As mhags are considered to be the key targets of GvHD and GvT, unravelling their biochemical nature has been a major focus of research. Early studies showed that human mhags were inherited as a Mendelian trait 16. Thus, it seemed that mhags were immunological reflections of genetic variation between HLA-identical individuals. Indeed, the identification of the first series of human mhags demonstrated that they are HLA-bound polymorphic peptides derived from intracellular proteins In most cases, the difference between these polymorphic peptides is encoded by a single nucleotide polymorphisms (SNPs) in diallelic genes The mhag difference between the donor and recipient can also be generated by a gene deletion in the donor 28, by differential protein synthesis caused by a frame-shift polymorphism 29, or by alternative RNA splicing induced by an intronic SNP 30,31. Since the human genome contains a huge number of SNPs, it seems likely that there are thousands of mhags. Consequently, each HLA-identical donor-recipient pair is expected to be mismatched for several mhags. Indeed, ex vivo analyses of recipient peripheral blood mononuclear cells after HLA-identical SCT frequently revealed the presence of T cells directed at various mhags However, there is also experimental evidence suggesting the immunodominance of some mhags, such as the mhag HA-1 and the SMCY-derived H-Y antigen 37. Contribution of individual mhags to the GvT effect Over the past decade, more than 30 mhags have been identified at molecular level. The contribution of these mhags in GvT has been evaluated mainly by testing their capacity to lyse malignant cells. In numerous studies, CD8 + cytotoxic T cells (CTLs) specific for several HLA class I-restricted mhags have been shown to recognize and lyse malignant cells and/or their clonogenic precursors derived from patients with different types of leukemia or myeloma 20,23,26,29,34, CTLs specific for the mhags HA-1, HMSD, DDX3Y and SP110 could also inhibit the engraftment of lymphoblastoid leukemia cells in immunodeficient NOD/SCID mice 31,43,51,53,54.

10 10 Chapter 1 Other studies have demonstrated that several mhag-specific CD4 + T cells are also capable of lysing malignant hematological cells 41,55,56,60. Thus most, if not all, mhags described to date can contribute to the GvT effect. Nonetheless, demonstrating an inverse correlation between a single mhag and relapse rates is difficult in clinical studies, probably because the clinical GvT effect is established by the collective contribution of several mhags. Hematopoietic system-specific mhags: tools for separating GvT from GvHD Since mhags are involved in both GvHD and GvT, it was difficult to establish specific GvT effects without GvHD by targeting mhags. However, this general opinion has undergone a revolutionary revision after the discovery that two mhags, HA-1 and HA-2, are expressed exclusively on hematopoietic cells 57, including hematological malignant cells and their clonogenic precursors 39,40. Subsequently, the mhags HA-1 and HA-2 were characterized 17,19. and their hematopoietic restricted expression was confirmed at molecular level 58,59. The discovery of hematopoietic mhags has led to the development of a novel concept, which proposes to separate GvT from GvHD by targeting donor T cells towards the hematopoietic mhags expressed on malignant cells. As illustrated in Figure 1, this concept is actually based on three assumptions: (1) the expression of the immunogenic mhag on the malignant cells ensures effective targeting of the tumor by donor T cells; (2) the specific expression of the mhag on hematopoietic cells will spare the non-hematopoietic tissues and prevent GvHD; and (3) the specific expression of the mhag on the recipient cells alone will spare the hematopoietic cells of the donor, which are necessary to rebuild the recipient s hematopoietic system. Therapeutically relevant hematopoietic mhags The idea of separating GvT from GvHD by targeting hematopoietic mhags provided a great drive to identify additional hematopoietic mhags (Table 1). Soon after HA-1 and HA-2, HB-1 was identified as an acute B-lymphoblastoid-leukemia-related mhag 20,24. Another mhag, encoded by the alternative transcription of the PANE- 1 gene, is expressed specifically in B cell chronic lymphocytic leukemia cells 30. Also, the mhag LRH-1, encoded by the P2X5 gene, is hematopoietic-specific and expressed in leukemic cells and their CD34 + progenitors. Recently, the mhag ACC6 was shown to be derived from the HMSD gene, which is expressed abundantly in acute myeloid leukemia and multiple myeloma cells, activated dendritic cells (DCs), T cells and B cells, but is virtually absent in normal tissue 31. Finally, the first hematopoietic mhag presented by an HLA class II molecule has been identified. This antigen is encoded by the B cell lineage-specific molecule CD In addition, the mhags ACC1, ACC2 and SP110 were shown to be derived from genes which are only expressed in hematopoietic cells under non-inflammatory conditions 26,51. A number of mhags encoded by broadly expressed genes appear to be preferentially expressed in activated hematopoietic cells and malignant cells (Table 1). T cells

11 General introduction 11 hematopoietic mhags of the recipient Tumor cell allelic peptides on donor hematopoietic cells ubiquitous mhags of the recipient Lysis Donor CD4 + T cells Licensing Chapter 1 (1) Stimulation DC Lysis Donor CD8 + T cells (3) (2) Priming Non-hematopoietic cell Donor-derived hematopoietic cell Figure 1. The principles and optimal effector mechanisms for separation of graft-versus-tumor (GvT) effects from graft-versus-host disease (GvHD) by targeting hematopoietic minor histocompatibility antigens (mhags) after HLA-identical, mhag-mismatched allogeneic stem cell transplantation. Donorderived CD4 + and CD8 + T cells can mediate GvT effects by recognition of the recipient s hematopoietic mhags on malignant hematopoietic cells (1). The mhag non-hematopoietic cells (2) and donor-derived hematopoietic cells (3) will be spared. While GvT effects are established through direct lysis of malignant cells by CD4 + as well as CD8 + T cells, the CD4 + mhag-specific T cells can optimize the GvT effects by providing help for the in vivo survival and expansion of CD8 + T cells. In a dendritic cell (DC) vaccination setting, the GvT effects will probably start by presentation of hematopoietic mhags by DCs to donor T cells. In this setting, targeting with CD4 + as well as CD8 + T cells is perhaps important, since upon activation, CD4 + mhag-specific T cells can licence DCs to prime CD8 + mhag-specific T cells. directed against these mhags did not recognize resting hematopoietic cells and non-hematopoietic cells in in vitro assays. The ability of these mhags to functionally discriminate between malignant hematopoietic cells and normal tissue awaits further confirmation. Of note, in a previous trial, adoptive transfer of mhag-specific CTLs selected on the basis of recognition of recipient hematopoietic cells but not skin fibroblasts was still associated with GvHD in a number of patients 61. Despite the growing number of molecularly identified hematopoietic mhags, significant advances are required to make mhag-specific therapy available for a broad range of patients. From the immunotherapeutic point of view, some of the identified mhags are not very relevant as they are presented by relatively infrequent HLA-alleles, such as the HLA-B60-restricted HA-1 antigen. Another important issue is the phenotype frequency of mhags. Since mhag-specific therapy can only be applied after mhag-

12 12 Chapter 1 Table 1. Hematopoietic minor histocompatibility antigens (mhags) Gene expression profile c Expression profile according to T cell reactivity Expression in hematopoietic malignancies Prevalence d mhag a Peptide sequence b HLA molecule Gene Source Presenting HA-1 VLHDDLLEA A2 HMHA1 hematopoietic hematopoietic all leukemia, MM 58,8% ref 19 HA-1 KECVLHDDL(L) B60 HMHA1 hematopoietic hematopoietic all leukemia, MM 58,8% ref 25 HA-2 YIGEVLVSV A2 MYO1G hematopoietic hematopoietic all leukemia, MM 94,5% ref 17 HB-1H EEKRGSLHVW B44 HB-1 B-specific B blasts,b-all B-ALL 94,8% ref 20 HB-1Y EEKRGSLYVW B44 HB-1 B-specific B blasts, B-ALL B-ALL 46,3% ref 24 LRH-1 TPNQRQNVC B7 P2X5 hematopoietic hematopoietic leukemia, MM 54.0% ref 29 PANE1 RVWDLPGVLK A3 PANE1 B-specific B-specific B-CLL, MM 92,6% ref 30 ACC6 MEIFIEVFSHF B44 HMSD hematopoietic hematopoietic CML AML MM 35.6% ref 31 CD19-L PEIWEGEPPCLPPRD DQ2 CD19 B-specific hematopoietic B-CLL ALL NHL 53.4% ref 60 ACC1-Y DYLQYVLQI A24 BCL2A1 hematopoietic* hematopoietic AML ALL CLL, MM 46,5% ref 26 ACC2 KEFEDDIINW B44 BCL2A1 hematopoietic* hematopoietic AML ALL CLL 44,5% ref 26 ACC1-C DYLQCVLQI A24 BCL2A1 hematopoietic* hematopoietic AML ALL CLL, MM 93,0% ref 49 SP110 SLPRGTSTPK A3 SP110 hematopoietic* hematopoietic AML CLL ALL 85,3% ref 51 LB-ADIR-1F SVAPALALFPA A2 ADIR broad hematopoietic MM AML ALL 57.0% ref 47 ECGF-1 RPHAIRRPLAL B7 ECGF-1 broad hematopoietic MM CML AML ALL 11.4% ref 50 ACC4 WATLPLLCAR A33 CTSH broad hematopoietic AML 11.0% ref 52 PI4K2B-1S SRSSSAELDRSR DQ6 PI4K2B broad hematopoietic CML AML 50,0% ref 64 HY (UTY) LPHNHTDL B8 UTY male-specific broad hematopoietic all leukemia 50,0% ref 22 HY(RPS4Y) TIRYPDPVI B52 RPS4Y male-specific broad hematopoietic all leukemia 50,0% ref 63 MM, multiple myeloma; B-ALL, B cell acute lymphocytic leukemia; B-CLL, B cell chronic lymphocytic leukemia; CML, chronic myeloid leukemia; AML, acute myeloid leukemia; NHL, non-hodgkin s lymphoma. a Underlined mhags are therapeutically relevant; they display a balanced phenotype frequency and are presented by a prevalent human leukocyte antigen (HLA)-allele. b Amino acid(s) that differ from the allelic peptide (if they exist) are underlined. c Genes marked with an asterisk are hematopoietic-restricted under non-inflammatory conditions but can be induced by interferon-γ in other tissues. d mhag frequencies in the Caucasian population are either derived from the original reference or from Reference 90.

13 General introduction 13 mismatched transplantation, an unbalanced phenotype frequency of the mhag will limit the number of eligible recipient-donor pairs significantly. For instance, it is hardly possible to find HA-2- or HB-1H-mismatched donor-recipient pairs because these mhags are expressed in more than 94.5% of the Caucasian population. Therefore, an important future task is to focus on the identification of therapeutically relevant mhags which can be defined as: (1) hematopoietic-cell-specific; (2) highly expressed on leukemic cells; (3) presented by a prevalent HLA-allele; and (4) represented with a balanced phenotype frequency in the population (see Box 1). Chapter 1 Box 1. Properties of therapeutically relevant mhags Hematopoietic-specific expression Sufficient expression on (hematopoietic) malignant cells Presentation by a frequent HLA-allele (preferably HLA-A1, -A2, -A3, -A11 or -A24) Balanced phenotype frequency (preferably between 26% and 78%) It is proposed that, in the Caucasian population, specific focus should be on the identification of mhags presented by HLA-A1, -A2, -A3, -A11 and -A24 alleles since >95% of the population express at least one of these five HLA molecules 62. This list may be slightly different for other ethnic groups. However, combination of HLA-A and HLA-B alleles in such lists does not seem to be advantageous due to common HLA-A/HLA-B haplotypes of the most prevalent alleles. The phenotype frequency of mhags should preferably be between 26% and 78%, because frequencies beyond these limits will decrease the chance of a mhag-mismatch to less than 10% in an HLA-identical sibling transplantation setting. Strategies for the identification of therapeutically relevant mhags To date, the vast majority of mhags have been discovered by so-called forward approaches. These strategies aim to identify a peptide recognized by a mhagspecific T cell clone isolated from a transplant recipient. The peptide elution method is based on biochemical fractionation and mass spectrometric sequencing of the mhag-peptide eluted from cell surface major histocompatibility complex (MHC) molecules. This sophisticated method has been very successful, but only appears to be suitable for the identification of mhags presented by HLA class I molecules 17-19,27,30,47. Screening of mhag + cdna libraries with mhag-specific T cell clones is another popular approach 20,22,24,28,31,50,54,63. Recently, a modification of this technique was applied to identify the mhag PI4K2B-1S, which is presented by HLA class II molecules 64. Next to these classical T cell epitope identification methods, there are a number of genetic approaches to identify polymorphic genes. Genetic linkage analysis is used to identify the genomic locus of the mhags by

14 14 Chapter 1 pairwise correlation of the mhag phenotype of large pedigrees (CEPH families) with thousands of genetic markers identified in their genomes 65. Once the genetic locus is located, the identification of the mhag gene requires further tailor-made analyses and has been successful for three HLA class I-restricted mhags 26,29. In an alternative genetic approach, the mhag ACC1-C was identified by comparing the SNP arrays from mhag + and mhag CEPH individuals 49. The present authors recently developed a less laborious and more powerful genetic approach to identify a novel CD19-encoded mhag, which is presented by HLA class II molecules. In this method, correlation of the mhag phenotypes of only 23 CEPH individuals with the SNP genotypes derived from the international HapMap project ( was sufficient to fine-map the exact genomic locus of the mhag CD19 L 60. As illustrated by aforementioned studies, the T cell-to-antigen -based strategies have been very successful to date. Nonetheless, identification of therapeutically relevant mhags with these strategies is completely dependent on the availability of T cells that recognize relevant mhags. Since there is no consistently reliable approach to isolate T cells specific for hematopoietic mhags with balanced allele frequencies, it is necessary to execute careful analyses to select the relevant T cell clones prior to these forward identification strategies. Theoretically, a more predictable way to identify therapeutically relevant mhags is using reverse immunology. This approach is based on computational prediction of mhags on hematopoietic proteins. Amino acid sequences of these proteins are screened for the presence of polymorphic peptides that can bind to prevalent HLA molecules Since the tissue expression profiles of almost all genes as well as the allele frequencies of millions of SNPs are readily available from public databases, this strategy can be executed on selected sets of hematopoietic proteins. Current efforts are focused on increasing the specificity and sensitivity of predictions by integrating the prediction algorithms for MHC class I binding, transport efficiency of the transporter associated with antigen processing (TAP) and proteasomal cleavage 69. Nevertheless, even the integrated algorithms still produce a significant number of false-positive and false-negative predictions. Perhaps due to this reason, only a few mhags have been identified by reverse immunology 24,25. It is also possible that generation of immunogenic mhags from polymorphic peptides is dictated by additional parameters. An important additional requirement may be the differential surface expression of the allelic peptides, since not all mhag-specific T cells can discriminate the immunogenic mhag-peptide from the allelic peptide 70. mhag-based immunotherapy strategies: where do we stand? The molecular identification of hematopoietic mhags provides the necessary tools, i.e. mhag-peptides, mhag genes and mhag-specific T cell receptors (TCRs), for the development of specific immunotherapy strategies for relapsed patients after

15 General introduction 15 HLA-matched, mhag-mismatched transplantation. Possible scenarios of adoptive immunotherapy and mhag-based vaccination strategies are depicted in Box 2 and outlined below. Box 2. mhag-specific immunotherapy strategies Chapter 1 Adoptive immunotherapy with: - donor-derived CTLs generated by ex vivo stimulation of mhag-loaded DCs - recipient-derived unmodified or virus- or tetanus toxoid-specific T cells transduced with mhag-tcr - donor-derived virus- or tetanus toxoid-specific T cells transduced with mhag-tcr - combination of mhag-specific CD4 + and CD8 + T cells Vaccination of the: - recipient with mhag-loaded donor DCs - recipient with mhag DNA - donor with mhag-loaded DCs prior to DLI mhag-specific adoptive immunotherapy Conceptually, specific treatment of recurrent leukemia should be possible using adoptive immunotherapy directed against mhags. This strategy requires the ex vivo generation of sufficient numbers of mhag-specific CTLs from the stem cell donor. Furthermore, the ex vivo generated CTLs should be able to migrate into tumor sites to exert their effector functions. An additional requirement will be the in vivo persistence of the adoptively transferred T cells. In previous studies, the present authors and others have succeeded in ex vivo generation of HA-1- and HA-2-specific CTLs from mhag stem cell donors by repeated stimulation of donor T cells with peptide-pulsed or gene-transduced DCs 44,71,72. Nevertheless, clinical implementation of these laborious ex vivo culture protocols was cumbersome. The most important bottleneck was the generation of sufficient numbers of CTLs in a reasonable time period 70. Long term in vitro culturing of T cells also seemed detrimental for their in vivo survival. Fortunately, these difficulties appear to be overcome effectively by the so-called TCR-transfer approach. In a number of studies, mhag-specific T cells could be generated by retroviral transfer of mhagspecific TCR genes into unselected 73-76, cytomegalovirus (CMV)-specific 77 or tetanus toxoid-specific 56 T cells. The insertion of mhag-specific TCRs into virusor tetanus toxoid-specific T cells is particularly interesting, since these strategies may enable the in vivo boosting of TCR-transferred T cells via their endogenous

16 16 Chapter 1 TCRs. Furthermore, it is important to note that if only donor-derived T cells are available for TCR-transfer, it is not recommended to introduce the mhag-tcrs into unselected donor T cells, because unselected donor T cells can induce GvHD via the recognition of ubiquitous mhags via their endogenous TCRs. Enthusiasm for pursuing the TCR-transfer approach has been bolstered by the successful treatment of metastatic melanoma patients with T cells transferred with MART-1-specific TCRs 78. This study has also shown that a lymphopenic environment improves the engraftment of adoptively transferred T cells significantly. A more commonly encountered limitation of adoptive immunotherapy is the lack of in vivo persistence of adoptively transferred T cells. In an attempt to address this issue, a recent study injected CD8 + CMV-specific T cells, either derived from CD62L + central memory cells or from CD62L effector memory cells, into non-human primates. Only the T cells derived from CD62L + central memory cells persisted for long periods of time 79. Thus, a possible future scenario is the transfer of mhag-specific TCRs into virus or recall antigen-specific central memory T cells. Alternatively, an effective GvT effect may be achieved by repeated administration of TCR-transferred T cells into the recipient. Finally, the question remains whether only targeting mhags presented to CD8 + CTLs will be sufficient. Experimental studies have revealed that mhagspecific CD4 + T cells not only provide proper help for the induction and expansion of CD8 + T cells, but also kill malignant cells directly 60. Furthermore, it appears that host MHC class II-positive antigen-presenting cells (APCs) and CD4 + T cells are required for an effective GvT effect after donor lymphocyte infusions (DLI) 80. Thus, infusion of both CD4 + and CD8 + T cells directed at hematopoietic mhags may increase the efficacy of mhag-specific adoptive immunotherapy significantly. mhag-based vaccination strategies Over the past years, numerous studies have demonstrated the in vivo generation of anti-tumor T cell responses by vaccination of patients with autologous DCs loaded with tumor-associated antigens 81. Tumor-specific humoral and cellular responses could also be induced by vaccination of patients with naked DNA encoding for tumor or viral antigens 82. Frequently, these vaccination strategies are more successful if the vaccines are designed specifically to generate both CD4 + and CD8 + T cell responses 81,82. Analogous to these approaches, plausible mhag vaccination strategies involve the vaccination of relapsed recipients with donor-derived DCs loaded with peptides or genes of hematopoietic mhags. DNA vaccination of the recipients is also a possible scenario. Since high tumor burden is frequently associated with poor vaccine response, pre-emptive vaccinations, for instance at a stage of minimal residual disease, may also be considered. The safety of vaccination of recipients after allo-sct with tumor antigen-loaded donor DCs has been demonstrated by a recent study 83. Vaccination of myeloma patients with idiotype-pulsed allo-dcs plus soluble protein Id conjugated with KLH (Id-KLH) resulted in no serious side effects, induced KLH antibodies and triggered the ex vivo secretion of several cytokines

17 General introduction 17 by T cells. Nonetheless, current DC vaccination strategies frequently fail to induce effective clinical responses, indicating that improvement of the vaccine composition and/or administration schemes or increasing the immunogenicity of tumors is still necessary. A final hypothetical approach is the vaccination of SCT donors with hematopoietic mhags before harvesting the hematopoietic stem cells for DLI. This strategy may enrich the DLI products with in vivo induced immunotherapeutic mhag-specific T cells, hereby improving the efficacy of DLI. Nevertheless, since the potential hazards of vaccinating donors with mhags is not easy to estimate, this approach may remain hypothetical for a long time. Chapter 1 How safe is immunotherapy with hematopoietic mhags? While the development of immunotherapy strategies based on targeting hematopoietic mhags is still ongoing, a frequently asked question is whether such a therapy can also induce GvHD. This question is mainly originated by the puzzling clinical studies on the mhag HA-1. While all in vitro, genetic and in-situ analyses clearly indicate that HA-1 is a hematopoietic system-specific mhag and is not a primary target for GvHD 57-59,84, two studies reported that a mismatch for HA-1 showed a significant association with GvHD 85,86. Moreover, the development and peripheral expansion of HA-1-specific CTLs showed temporal correlation with GvHD after HA-1-mismatched SCT 87. Thus, it seems that hematopoietic mhags could also induce GvHD. Novel insights, especially those provided by murine studies, are shedding light on this puzzling paradox. A number of studies have clearly demonstrated that alloantigen expression on the host APCs is necessary and sufficient to induce GvHD 88,89. Thus, the association between HA-1 and GvHD could be due to the presentation of HA-1 by host APCs. In fact, early after SCT, non-hematopoietic tissues still contain hostderived APCs that can present hematopoietic mhags such as HA-1. Inflammatory/ necrotizing cytokines such as interferon-γ or tumor necrosis factor-α secreted by activated HA-1-specific CTLs may initiate GvHD, either via cytokine-mediated damage of tissue cells or by the recruitment of other CTLs directed at mhags expressed on GvHD target tissues. Hence, a critical parameter of mhag-specific immunotherapy appears to be the timing of T cell administration: T cells directed towards hematopoietic mhags may induce GvHD if administered soon after SCT. However, the therapy can be applied safely at later time points, after replacement of the majority of host APCs by donor APCs. SUMMARY OF THIS INTRODUCTION mhags that are expressed exclusively on normal and malignant hematopoietic cells are attractive tumor targets after allo-sct. Conceptually, targeting hematopoietic mhags can induce powerful GvT effects but the therapy will be safe because the mhag non-hematopoietic tissues and the donor-derived hematopoietic cells will not be attacked. To make this novel therapy available for many patients,

18 18 Chapter 1 tremendous efforts are being devoted to identify hematopoietic mhags. With the recent discovery of the first HLA class II-restricted hematopoietic mhag on CD19, the number of immunotherapeutic mhags has reached 13. Nevertheless, several of the identified mhags can only target a small fraction of hematological malignancies. Hence, focused efforts are still necessary to identify mhags that display balanced allele frequencies and are presented by prevalent HLA-alleles. Also important is the identification of more HLA class II-restricted hematopoietic mhags because CD4 + T cells recognizing hematopoietic mhags not only provide help for the induction and expansion of mhag-specific CD8 + effector T cells but also eliminate the malignant cells directly. While the initial attempts to clinically implement the mhag-specific therapy were not very successful, the development of novel techniques, such as the TCR-transfer approach, and the emergence of novel insights on the in vivo survival of adoptively transferred T cells increase hopes to realize the mhag-specific adoptive immunotherapy in the near future. Also, vaccination of recipients with mhag-loaded donor DCs or with DNA encoding for mhags are possible future strategies to treat recurrent leukemia or myeloma after HLA-identical, mhag-mismatched allo-sct. scope OF THIS THESIS T cells recognizing the highly immunogenic minor histocompatibility antigens (mhags) expressed by hematopoietic and tumor cells may be powerful tools in the battle against hematological malignancies. However, mhag-specific immunotherapy is still in its infancy, since currently only few clinically useful mhags are identified, and since the exploration of T cell based immunotherapy for the treatment of cancer was initiated only a decade ago. The scope of thesis is to tackle these issues by 1) identifying and facilitating the identification of novel clinically relevant mhags and 2) investigating the immunotherapeutic potential of mhag-specific CD4 + T cells in vitro as well as in a preclinical murine model. Chapter 2 describes the identification of the first HLA class II-restricted mhag with a truly hematopoietic tissue expression. The mhag is encoded by the CD19 gene, which is not only highly expressed on B cells, but also on their malignant derivatives. The identification of the CD19-derived mhag is performed using a novel genetic fine-mapping strategy, called zygosity-genotype correlation analysis. The genomewide application of this strategy is easy and rapid as we demonstrate in Chapter 3 with the molecular characterization of the novel SLC19A1-derived mhag. Here we also outline a directed strategy to make significant progress toward wide scale application of mhag-specific immunotherapy. In Chapter 4, we utilize the rationale behind this directed strategy towards the identification of new mhags using a reverse identification approach. We describe the advantages of reverse identification and

19 General introduction 19 provide an extensive list of putative mhags that are predicted using reliable antigenprocessing and HLA-binding algorithms combined with databases containing all currently known genes and genotyped SNPs. Chapter 5 shows the feasibility of transferring a TCR derived from mhag-specific CD4 + human T cells into recall antigen (tetanus toxoid; TT)-specific CD4 + T cells. The TCR-transgenic cells acquire mhag specificity, and expand efficiently in vitro via stimulation of their intrinsic TT-specific TCR. In Chapter 6, similar dual-specific T cells as well as the parent mhag-specific T cells are used as sole treatment for established multiple myeloma tumors in immunodeficient mice. This adoptive immunotherapy with sole mhag-specific CD4 + T cells is very effective against tumor in the bone marrow. However, the mice suffer from extramedullary tumor relapses, which are in vitro but not in vivo susceptible to further T cell therapy. These results seem also relevant for the clinic, since the relapse of multiple myeloma at extramedullary sites is a known complication in patients treated with DLI. Chapter 1 Finally, the general discussion describes how the current findings may contribute to the improvement of immunotherapeutic strategies based on mhag-specific (CD4 + ) T cells.

20 20 Chapter 1 REFERENCES 1. Weiden PL, Flournoy N, Thomas ED et al. Antileukemic effect of graft-versus-host disease in human recipients of allogeneic-marrow grafts. N Engl J Med. 1979;300: Horowitz MM, Gale RP, Sondel PM et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood. 1990;75: Maraninchi D, Gluckman E, Blaise D et al. Impact of T cell depletion on outcome of allogeneic bone-marrow transplantation for standard-risk leukaemias. Lancet. 1987;2: Kolb HJ, Schattenberg A, Goldman JM et al. Graft-versus-leukemia effect of donor lymphocyte transfusions in marrow grafted patients. European Group for Blood and Marrow Transplantation Working Party Chronic Leukemia. Blood. 1995;86: Ferster A, Bujan W, Mouraux T et al. Complete remission following donor leukocyte infusion in ALL relapsing after haploidentical bone marrow transplantation. Bone Marrow Transplant. 1994;14: Atra A, Millar B, Shepherd V et al. Donor lymphocyte infusion for childhood acute lymphoblastic leukaemia relapsing after bone marrow transplantation. Br J Haematol. 1997;97: Collins RH, Jr., Shpilberg O, Drobyski WR et al. Donor leukocyte infusions in 140 patients with relapsed malignancy after allogeneic bone marrow transplantation. J Clin Oncol. 1997;15: Lokhorst HM, Schattenberg A, Cornelissen JJ, Thomas LL, Verdonck LF. Donor leukocyte infusions are effective in relapsed multiple myeloma after allogeneic bone marrow transplantation. Blood. 1997;90: Castagna L, El WA, Bourhis JH et al. Successful donor lymphocyte infusion (DLI) in a patient with myelodysplastic syndrome (MDS) after failure of T cell-depleted bone marrow transplantation (TD-BMT). Br J Haematol. 1998;103: Lokhorst HM, Schattenberg A, Cornelissen JJ et al. Donor lymphocyte infusions for relapsed multiple myeloma after allogeneic stem-cell transplantation: predictive factors for response and long term outcome. J Clin Oncol. 2000;18: Bethge WA, Hegenbart U, Stuart MJ et al. Adoptive immunotherapy with donor lymphocyte infusions after allogeneic hematopoietic cell transplantation following nonmyeloablative conditioning. Blood. 2004;103: Schmid C, Labopin M, Nagler A et al. Donor lymphocyte infusion in the treatment of first hematological relapse after allogeneic stem-cell transplantation in adults with acute myeloid leukemia: a retrospective risk factors analysis and comparison with other strategies by the EBMT Acute Leukemia Working Party. J Clin Oncol. 2007;25: Bloor AJ, Thomson K, Chowdhry N et al. High response rate to donor lymphocyte infusion after allogeneic stem cell transplantation for indolent non-hodgkin lymphoma. Biol Blood Marrow Transplant. 2008;14: Barth R, Counce S, Smith P, Snell GD. Strong and weak histocompatibility gene differences in mice and their role in the rejection of homografts of tumors and skin. Ann Surg. 1956;144: Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: Schreuder GM, Pool J, Blokland E et al. A genetic analysis of human minor histocompatibility antigens demonstrates Mendelian segregation independent of HLA. Immunogenetics. 1993;38: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: Wang W, Meadows LR, den Haan JM et al. Human H-Y: a male-specific histocompatibility antigen derived from the SMCY protein. Science. 1995;269: den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Pierce RA, Field ED, den Haan JM et al. Cutting edge: the HLA-A*0101-restricted HY minor histocompatibility antigen originates from DFFRY and contains a cysteinylated cysteine residue as identified by a novel mass spectrometric technique. J Immunol. 1999;163: Warren EH, Gavin MA, Simpson E et al. The Human UTY Gene Encodes a Novel HLA-B8-Restricted H-Y Antigen. J Immunol. 2000;164: Brickner AG, Warren EH, Caldwell JA et al. The immunogenicity of a new human minor histocompatibility antigen results from differential antigen processing. J Exp Med. 2001;193: Dolstra H, de Rijke B, Fredrix H et al. Bi-directional allelic recognition of the human minor histocompatibility antigen HB-1 by cytotoxic T lymphocytes. Eur J Immunol. 2002;32:

21 General introduction Mommaas B, Kamp J, Drijfhout JW et al. Identification of a novel HLA-B60-restricted T cell epitope of the minor histocompatibility antigen HA-1 locus. J Immunol. 2002;169: Akatsuka Y, Nishida T, Kondo E et al. Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med. 2003;197: Spierings E, Brickner AG, Caldwell JA et al. The minor histocompatibility antigen HA-3 arises from differential proteasome-mediated cleavage of the lymphoid blast crisis (Lbc) oncoprotein. Blood. 2003;102: Murata M, Warren EH, Riddell SR. A human minor histocompatibility antigen resulting from differential expression due to a gene deletion. J Exp Med. 2003;197: de Rijke B., van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: Brickner AG, Evans AM, Mito JK et al. The PANE1 gene encodes a novel human minor histocompatibility antigen that is selectively expressed in B-lymphoid cells and B-CLL. Blood. 2006;107: Kawase T, Akatsuka Y, Torikai H et al. Alternative splicing due to an intronic SNP in HMSD generates a novel minor histocompatibility antigen. Blood. 2007;110: Goulmy E, Blokland E, van RJ et al. Production, expansion, and clonal analysis of T cells with specific HLA-restricted male lysis. J Exp Med. 1980;152:182s-190s. 33. van Els CA, Zantvoort E, Jacobs N et al. Graft-versus-host disease associated T helper cell responses specific for minor histocompatibility antigens are mainly restricted by HLA-DR molecules. Bone Marrow Transplant. 1990;5: Marijt WA, Veenhof WF, Brand A et al. Minor histocompatibility antigen-specific cytotoxic T cell lines, capable of lysing human hematopoietic progenitor cells, can be generated in vitro by stimulation with HLA-identical bone marrow cells. J Exp Med. 1991;173: Nishimura M, Akaza T, Mitomi Y et al. Establishment of human minor histocompatibility antigen-specific cytotoxic T cell clones restricted by HLA-DR9. Transplantation. 1993;55: Marijt WA, Kernan NA, az-barrientos T et al. Multiple minor histocompatibility antigen-specific cytotoxic T lymphocyte clones can be generated during graft rejection after HLA-identical bone marrow transplantation. Bone Marrow Transplant. 1995;16: Rufer N, Wolpert E, Helg C et al. HA-1 and the SMCY-derived peptide FIDSYICQV (H-Y) are immunodominant minor histocompatibility antigens after bone marrow transplantation. Transplantation. 1998;66: Warren EH, Greenberg PD, Riddell SR. Cytotoxic T-lymphocyte-defined human minor histocompatibility antigens with a restricted tissue distribution. Blood. 1998;91: Marijt WA, Veenhof WF, Goulmy E et al. Minor histocompatibility antigens HA-1-, -2-, and -4-, and HY-specific cytotoxic T cell clones inhibit human hematopoietic progenitor cell growth by a mechanism that is dependent on direct cell-cell contact. Blood. 1993;82: van der Harst D, Goulmy E, Falkenburg JH et al. Recognition of minor histocompatibility antigens on lymphocytic and myeloid leukemic cells by cytotoxic T cell clones. Blood. 1994;83: Faber LM, van der Hoeven J, Goulmy E et al. Recognition of clonogenic leukemic cells, remission bone marrow and HLA-identical donor bone marrow by CD8 + or CD4 + minor histocompatibility antigen-specific cytotoxic T lymphocytes. J Clin Invest. 1995;96: Dolstra H, Fredrix H, Preijers F et al. Recognition of a B cell leukemia-associated minor histocompatibility antigen by CTL. J Immunol. 1997;158: Bonnet D, Warren EH, Greenberg PD, Dick JE, Riddell SR. CD8 + minor histocompatibility antigen-specific cytotoxic T lymphocyte clones eliminate human acute myeloid leukemia stem cells. Proc Natl Acad Sci U S A. 1999;96: Mutis T, Verdijk R, Schrama E et al. Feasibility of immunotherapy of relapsed leukemia with ex vivo-generated cytotoxic T lymphocytes specific for hematopoietic system-restricted minor histocompatibility antigens. Blood. 1999;93: Klein CA, Wilke M, Pool J et al. The hematopoietic system-specific minor histocompatibility antigen HA-1 shows aberrant expression in epithelial cancer cells. J Exp Med. 2002;196: Marijt WA, Heemskerk MH, Kloosterboer FM et al. Hematopoiesis-restricted minor histocompatibility antigens HA-1- or HA-2-specific T cells can induce complete remissions of relapsed leukemia. Proc Natl Acad Sci U S A. 2003;100: van Bergen CA, Kester MG, Jedema I et al. Multiple myeloma-reactive T cells recognize an activation-induced minor histocompatibility antigen encoded by the ATP-dependent interferon-responsive (ADIR) gene. Blood. 2007;109:4089- Chapter 1

22 22 Chapter Holloway PA, Kaldenhoven N, van Dijk M et al. Susceptibility of malignant plasma cells to HA-1 H -specific lysis suggests a role for the minor histocompatibility antigen HA-1 in the graft-versus-myeloma effect. Leukemia. 2004;18: Kawase T, Nanya Y, Torikai H et al. Identification of human minor histocompatibility antigens based on genetic association with highly parallel genotyping of pooled DNA. Blood. 2008;Epub ahead of print: 50. Slager EH, Honders MW, van der Meijden ED et al. Identification of the angiogenic endothelial-cell growth factor-1/ thymidine phosphorylase as a potential target for immunotherapy of cancer. Blood. 2006;107: Warren EH, Vigneron NJ, Gavin MA et al. An antigen produced by splicing of noncontiguous peptides in the reverse order. Science. 2006;313: Torikai H, Akatsuka Y, Miyazaki M et al. The human cathepsin H gene encodes two novel minor histocompatibility antigen epitopes restricted by HLA-A*3101 and -A*3303. Br J Haematol. 2006;134: Hambach L, Nijmeijer BA, Aghai Z et al. Human cytotoxic T lymphocytes specific for a single minor histocompatibility antigen HA-1 are effective against human lymphoblastic leukaemia in NOD/scid mice. Leukemia. 2006;20: Rosinski KV, Fujii N, Mito JK et al. DDX3Y encodes a class I MHC-restricted H-Y antigen that is expressed in leukemic stem cells. Blood. 2008;111: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Spaapen R, van den Oudenalder K, Ivanov R et al. Rebuilding human leukocyte antigen class II-restricted minor histocompatibility antigen specificity in recall antigen-specific T cells by adoptive T cell receptor transfer: implications for adoptive immunotherapy. Clin Cancer Res. 2007;13: de Bueger M., Bakker A, van Rood JJ, van der Woude F, Goulmy E. Tissue distribution of human minor histocompatibility antigens. Ubiquitous versus restricted tissue distribution indicates heterogeneity among human cytotoxic T lymphocyte-defined non-mhc antigens. J Immunol. 1992;149: Wilke M, Dolstra H, Maas F et al. Quantification of the HA-1 gene product at the RNA level; relevance for immunotherapy of hematological malignancies. Hematol J. 2003;4: Pierce RA, Field ED, Mutis T et al. The HA-2 minor histocompatibility antigen is derived from a diallelic gene encoding a novel human class I myosin protein. J Immunol. 2001;167: Spaapen RM, Lokhorst HM, van den Oudenalder K et al. A Single Nucleotide Polymorphism in CD19 Defines a Novel Target for Immunotherapy of B Cell Malignancies with CD4 + Cytotoxic T Cells. ASH Annual Meeting Abstracts. 2007;110: Riddell SR, Bleakley M, Nishida T, Berger C, Warren EH. Adoptive transfer of allogeneic antigen-specific T cells. Biol Blood Marrow Transplant. 2006;12: Schipper RF, van Els CA, D Amaro J, Oudshoorn M. Minimal phenotype panels. A method for achieving maximum population coverage with a minimum of HLA antigens. Hum Immunol. 1996;51: Ivanov R, Aarts T, Hol S et al. Identification of a 40S ribosomal protein S4-derived H-Y epitope able to elicit a lymphoblast-specific cytotoxic T lymphocyte response. Clin Cancer Res. 2005;11: Griffioen M, van der Meijden ED, Honders MW et al. Identification of Phosphatidylinositol 4-Kinase Type II as the First HLA Class II Associated Minor Histocompatibility Antigen Involved in Graft Versus Leukemia Reactivity. ASH Annual Meeting Abstracts. 2007;110: Warren EH, Otterud BE, Linterman RW et al. Feasibility of using genetic linkage analysis to identify the genes encoding T cell-defined minor histocompatibility antigens. Tissue Antigens. 2002;59: Stevanovic S. Antigen processing is predictable: From genes to T cell epitopes. Transpl Immunol. 2005;14: Schuler MM, Donnes P, Nastke MD et al. SNEP: SNP-derived epitope prediction program for minor H antigens. Immunogenetics. 2005;57: Halling-Brown M, Quartey-Papafio R, Travers PJ, Moss DS. SiPep: a system for the prediction of tissue-specific minor histocompatibility antigens. Int J Immunogenet. 2006;33: Larsen MV, Lundegaard C, Lamberth K et al. An integrative approach to CTL epitope prediction: a combined algorithm integrating MHC class I binding, TAP transport efficiency, and proteasomal cleavage predictions. Eur J Immunol. 2005;35: Mutis T. Targeting alloreactive donor T cells to hematopoietic system-restricted minor histocompatibility antigens to dissect graft-versus-leukemia effects from graft-versus-host disease after allogeneic stem cell transplantation. Int J Hematol. 2003;78: Mutis T, Ghoreschi K, Schrama E et al. Efficient induction of minor histocompatibility antigen HA-1-specific cytotoxic T cells using dendritic cells retrovirally transduced with HA-1-coding cdna. Biol Blood Marrow Transplant. 2002;8:

23 General introduction Brossart P, Spahlinger B, Grunebach F et al. Induction of minor histocompatiblity antigen HA-1-specific cytotoxic T cells for the treatment of leukemia after allogeneic stem cell transplantation. Blood. 1999;94: Heemskerk MH, Hoogeboom M, de Paus RA et al. Redirection of antileukemic reactivity of peripheral T lymphocytes using gene transfer of minor histocompatibility antigen HA-2-specific T cell receptor complexes expressing a conserved alpha joining region. Blood. 2003;102: Mommaas B, van Halteren AG, Pool J et al. Adult and cord blood T cells can acquire HA-1 specificity through HA-1 T cell receptor gene transfer. Hematologica. 2005;90: van der Veken LT, Hoogeboom M, de Paus RA et al. HLA class II-restricted T cell receptor gene transfer generates CD4 + T cells with helper activity as well as cytotoxic capacity. Gene Ther. 2005;12: Ivanov R, Hol S, Aarts TI, Hagenbeek A, Ebeling SB. T cell receptor-transgenic primary T cells as a tool for discovery of leukaemia-associated antigens. Clin Exp Immunol. 2006;143: Heemskerk MH, Hoogeboom M, Hagedoorn R et al. Reprogramming of virus-specific T cells into leukemia-reactive T cells using T cell receptor gene transfer. J Exp Med. 2004;199: Morgan RA, Dudley ME, Wunderlich JR et al. Cancer regression in patients after transfer of genetically engineered lymphocytes. Science. 2006;314: Berger C, Jensen MC, Lansdorp PM et al. Adoptive transfer of effector CD8 + T cells derived from central memory cells establishes persistent T cell memory in primates. J Clin Invest. 2008;118: Chakraverty R, Eom HS, Sachs J et al. Host MHC class II+ antigen-presenting cells and CD4 cells are required for CD8-mediated graft-versus-leukemia responses following delayed donor leukocyte infusions. Blood. 2006;108: Gilboa E. DC-based cancer vaccines. J Clin Invest. 2007;117: Rice J, Ottensmeier CH, Stevenson FK. DNA vaccines: precision tools for activating effective immunity against cancer. Nat Rev Cancer. 2008;8: Bendandi M, Rodriguez-Calvillo M, Inoges S et al. Combined vaccination with idiotype-pulsed allogeneic dendritic cells and soluble protein idiotype for multiple myeloma patients relapsing after reduced-intensity conditioning allogeneic stem cell transplantation. Leuk Lymphoma. 2006;47: Dickinson AM, Wang XN, Sviland L et al. In situ dissection of the graft-versus-host activities of cytotoxic T cells specific for minor histocompatibility antigens. Nat Med. 2002;8: Goulmy E, Schipper R, Pool J et al. Mismatches of minor histocompatibility antigens between HLA-identical donors and recipients and the development of graft-versus-host disease after bone marrow transplantation. N Engl J Med. 1996;334: Tseng LH, Lin MT, Hansen JA et al. Correlation between disparity for the minor histocompatibility antigen HA-1 and the development of acute graft-versus-host disease after allogeneic marrow transplantation. Blood. 1999;94: Mutis T, Gillespie G, Schrama E et al. Tetrameric HLA class I-minor histocompatibility antigen peptide complexes demonstrate minor histocompatibility antigen-specific cytotoxic T lymphocytes in patients with graft-versus-host disease. Nat Med. 1999;5: Shlomchik WD, Couzens MS, Tang CB et al. Prevention of graft versus host disease by inactivation of host antigenpresenting cells. Science. 1999;285: Teshima T, Ordemann R, Reddy P et al. Acute graft-versus-host disease does not require alloantigen expression on host epithelium. Nat Med. 2002;8: Spierings E, Hendriks M, Absi L et al. Phenotype frequencies of autosomal minor histocompatibility antigens display significant differences among populations. PLoS Genet. 2007;3:e103. Chapter 1

24 24 Chapter 1

25 2 Toward targeting B cell cancers with CD4+ CTLs: Identification of a CD19-encoded minor histocompatibility antigen using a novel genome-wide analysis Robbert M. Spaapen 1, Henk M. Lokhorst 2, Kelly van den Oudenalder 1, Brith E. Otterud 4, Harry Dolstra 5, Mark F. Leppert 4, Monique C. Minnema 2, Andries C. Bloem 3, Tuna Mutis 1 Departments of 1 Clinical Chemistry and Haematology, 2 Haematology, 3 Immunology, UMCU, NL 4 Department of Human Genetics, University of Utah Medical School, Salt Lake City, USA 5 Central Haematology Laboratory, Radboud UNMC, NL J Exp Med Nov;205:

26 26 Chapter 2 Some minor histocompatibility antigens (mhags) are expressed exclusively on patient hematopoietic and malignant cells, and this unique set of antigens enables specific targeting of hematological malignancies after human histocompatability leucocyte antigen (HLA)- matched allogeneic stem cell transplantation (allo-sct). We report the first hematopoietic mhag presented by HLA class II (HLA-DQA1*05/ B1*02) molecules to CD4 + T cells. This antigen is encoded by a singlenucleotide polymorphism (SNP) in the B cell lineage-specific CD19 gene, which is an important target antigen for immunotherapy of most B cell malignancies. The CD19 L -encoded antigen was identified using a novel and powerful genetic strategy in which zygosity-genotype correlation scanning was used as the key step for fine-mapping the genetic locus defined by pairwise linkage analysis. This strategy was also applicable for genome-wide identification of a wide range of mhags. CD19 L -specific CD4 + T cells provided antigen-specific help for maturation of dendritic cells and for expansion of CD8 + mhag-specific T cells. They also lysed CD19 L -positive malignant cells, illustrating the potential therapeutic advantages of targeting this novel CD19 L -derived HLA class II-restricted mhag. The currently available immunotherapy strategies enable the exploitation of these therapeutic effects within and beyond allo-sct settings. INTRODUCTION Leukemia, lymphoma, and myeloma together account for 500,000 deaths per year worldwide 1. HLA-matched allogeneic stem cell transplantation (allo-sct) is a widely applied immunotherapeutic approach for several of these hematological malignancies. The therapeutic effect of allo-sct is largely mediated by alloreactive donor T cells directed at polymorphic peptides presented by HLA molecules on the recipient s malignant cells 2. These polymorphic peptides, also known as minor histocompatibility antigens (mhags), are frequently derived from cellular proteins encoded by allelic genes on autosomal chromosomes. Although several mhags are expressed ubiquitously, some mhags are exclusively expressed on hematopoietic cells and their malignant counterparts 2-4. Hence, targeting donor T cells toward such hematopoietic mhags is considered an ideal strategy to establish specific antitumor effects after allo-sct 2,4. Because CD8 + T cells are traditionally considered as the effector cells of anti-tumor responses, over the past years the major focus was to identify hematopoietic mhags presented to CD8 + CTLs Nonetheless, several reports, including ours, indicate that not only CD8 + CTLs but also CD4 + T cells may possess immunotherapeutic potential Yet no hematopoietic mhag presented by HLA class II has been identified, partly because the available

27 Identification of a CD19-encoded minor histocompatibility antigen 27 techniques are not well suited for identification of such antigens. More importantly, several of the apparently hematopoietic mhags recognized by CD4 + T cells are not derived from genuine hematopoietic antigens. For instance, the recently identified autosomal mhag presented to CD4 + T cells is derived from the broadly expressed phosphatidylinositol 4-kinase type II β gene 16. We previously isolated an HLA-DQA1*05/B1*02-restricted mhag-specific CD4 + T cell (clone 21) from the PBMCs of a multiple myeloma patient after HLA-identical allo-sct. This clone recognized recipient-derived EBV-transformed B cells (EBVtransformed lymphoblastoid cell lines [EBV-LCLs]) but not the nonhematopoietic fibroblasts and stromal cells, suggesting that its target antigen was encoded by a hematopoietic gene (unpublished data). To identify the mhag recognized by clone 21, we developed a nonlaborious but powerful genetic strategy in which a zygositygenotype correlation analysis was used for fine-mapping of the genomic locus identified by classical pair-wise two-point linkage analysis. The new gene-mapping method was also genomewide applicable for a broad range of mhags. Further investigation on the identified locus revealed that the antigen recognized by clone 21 was encoded by a single-nucleotide polymorphism (SNP) in the B cell lineagespecific CD19 gene, which is a highly important target antigen for immunotherapy of almost all B cell malignancies. The CD19 L -specific CD4 + T cells not only mediated antigen-specific help for the induction and expansion of CD8 + mhag-specific T cells but also displayed antigen-specific and HLA-restricted lysis of CD19 L -positive malignant cells, illustrating the potential therapeutic advantages of targeting this CD19 L -derived HLA class II-restricted mhag. Chapter 2 RESULTS Genetic mapping of the mhag recognized by HLA class II-restricted T cell clone 21 To identify the mhag recognized by clone 21, we started with a genetic approach, the pair-wise two-point linkage analysis. In this method, the genomic locus of the mhag is identified by association of thousands of predefined genetic markers to mhag phenotypes (mhag + or mhag-) in large pedigrees registered in the Centre d Etude du Polymorphisme Humain (CEPH) 17. The CEPH families are suitable for this approach because not only have their genomes been screened for genetic markers but also EBV-LCLs are available from each individual. Upon transduction with the appropriate HLA molecules, these cell lines are used as APCs for mhagspecific T cells to determine the mhag phenotype of the CEPH individuals. Thus, we first tested the reactivity of clone 21 against (HLA-DQA1*05/B1*02 transduced) EBV- LCLs of several CEPH families (Fig. 1A and Fig. S1) and performed the pairwise two-point linkage analysis in which the mhag phenotype data were correlated with predefined genetic markers. Analysis of the data from three families (1331, 1362,

28 Chapter 2 A B lod C D r 2 E CEPH family 1362 Genomic marker Position (cm) Father ( ) Mother ( ) Child ( ) Child ( ) Child ( ) Mb Magnified around r 2 =1 CEPH /+ CEPH /- CEPH /- CEPH /- CEPH /- -/- SC donor SC recipient D16S Pheno./Zyg. +/? D16S rs nd CT nd CT CC D16S rs G G G G nd G G G G D16S D16S rs TT CT CT CT CC D16S rs nd C C nd nd C C D16S RABEP2 rs G G nd nd nd G G D16S D16S rs nd nd Chromosome cm rs Pair-wise two-point linkage analysis Recombinant haplotype analysis Zygosity/genotype correlation analysis rs D16S rs C C C G C G C G G G G G C G D16S rs T T T T T T T T T T D16S rs D16S CD19 Clone #21 recognition rs G G G G G G G G G G rs T T T T T T T T T T Figure 1. Five-step identification of the SNP encoding for the mhag recognized by clone 21. (A) mhag phenotypes of CEPH families (indicated with Utah database ID numbers) were determined using methods described in the Materials and methods section. CEPH family 1362 is depicted as an example (male, square; female, circle; mhag +, black; mhag-, white; undetermined, gray). Phenotypes of families 1331, 1408, and 1416 are given in Figure S1. (B) Genome-wide pairwise twopoint linkage analysis using the mhag phenotypes from families 1331, 1362, and Multiple significant lod scores >3 (at a recombination fraction of Θ = 0.001) were identified on chromosome 16 in the depicted region. (C) Narrowing of the mhag locus using haplotype data from family As depicted, the mhag + children 1189 and 2387 inherited the dark gray recombinant haplotype from the father, who is also mhag +. Thus the mhag locus was narrowed to the 16.8-cM region, which is defined by the shared part of the paternal allele of children 1189 and (D) Zygosity-genotype correlation analysis for fine-mapping the mhag locus. The r 2 values in the y axis represent the correlation between the mhag zygosity of 15 CEPH individuals (Table S1) with the genotypes for 4146 HapMap SNPs in this region. Each bar represents a single SNP. Two SNPs (rs and rs ) with 100% correlation (r 2 = 1) are indicated. (E) The location of rs and rs (both light gray) in the intronic regions of the RABEP2 gene, which is neighbored by the CD19 gene. Also indicated are the nine nonsynonymous or transcription/translating-altering SNPs in these two genes. The mhag phenotypes and zygosities (Pheno./ Zyg.) and the SNP genotypes for five informative CEPH individuals, the SC donor, and the SC recipient are depicted (-, deletion of the base pair). Only rs matched exactly with the phenotypes.

29 Identification of a CD19-encoded minor histocompatibility antigen 29 and 1408; Fig. 1A and Fig. S1) revealed a significant linkage between the mhag phenotypes and a large cluster of markers on chromosome 16, with multiple lod scores >3 (Θ = 0.001; Fig. 1B). According to the CEPH database, two children (1189 and 2387) in a fourth family (1416) displayed a crossing over or recombinant haplotype in this region. As depicted in Figure 1C, the mhag in this family was inherited together with the indicated paternal allele because the mhag child 1197 inherited the other allele from the mhag + father. Thus, because the two children with recombinant haplotypes were mhag +, the mhag gene should be located in the shared part of their paternal allele, between markers D16S3093 and D16S3034. This area was 16.8 cm on the Marshfield Map and contained 270 genes according to National Center for Biotechnology Information Map Viewer. Chapter 2 Development of a novel strategy for fine-mapping of the mhag locus The identified mhag locus in chromosome 16 contained too many genes to be analyzed with previously reported strategies 8,9,17. Therefore, we sought an alternative strategy. Over the last decade, millions of SNPs have been genotyped in CEPH individuals by the International HapMap Project ( to facilitate the identification of polymorphic genes associated with human diseases 18. We hypothesized that direct correlation of mhag phenotypes with the HapMap SNP genotypes might be useful to fine-map the genetic locus of our mhag. From the previous linkage analyses, we already knew the mhag phenotype of 23 CEPH individuals included in the HapMap databases. For 15 of these individuals, we also could deduce the mhag zygosity (+/+, +/-, or -/-) from the inheritance pattern of the mhag in the pedigrees (Fig. S1 and Table S1). Thus, using a specific software that can correlate the zygosity information with the SNP genotypes, we correlated the mhag zygosity information of 15 CEPH individuals with the 4146 HapMap SNPs present in the defined region. Despite the seemingly too low numbers of CEPH individuals entered in the analysis, we found a 100% correlation between the mhag zygosities and the genotypes of two intronic SNPs (rs and rs ; Fig. 1D). According to HapMap, these two SNPs were located within the same linkage disequilibrium (LD) block, posing the possibility that the SNP encoding for our mhag could also be located in this block. Identification of the CD19 L -derived mhag Within the LD block defined by rs and rs , RABEP2 and CD19 were the only genes with hematopoietic-restricted expression. Therefore, we explored whether RABEP2 and CD19 genes contained nonsynonymous or transcription/ translation-altering SNPs that were linked with the SNPs rs and rs Genotyping of five representative CEPH family members for nine additional SNPs in RABEP2 and CD19 (Fig. 1E) revealed a single SNP, rs , that was fully matching with the LD block defined by rs and rs The substitution of guanine for cytosine involved in this SNP causes a change of the valine (V) at

30 30 Chapter 2 position 174 into a leucine (L) within the third exon of CD19 (Fig. 2A). The fact that the SC recipient was positive for both CD19 L - and CD19 V -encoding alleles, whereas the SC-donor lacked the CD19 L -encoding allele (Fig. 1E), substantiated the idea that the mhag could be derived from the CD19 L -allele. Confirming this possibility, clone 21 showed strong reactivity against donor EBV-LCLs transduced with a CD19 L -encoding vector but not with an empty vector (Fig. 2B). Furthermore, clone 21 showed a dose-dependent reactivity against a 15-meric CD19 L peptide PEIWEGEPPCLPPRD, but not against the allelic counterpart peptide, demonstrating the crucial importance of leucine at position 174 for the immunogenicity of the CD19 L -derived mhag (Fig. 2C). Finally, testing the reactivity of clone 21 against the overlapping 15-mer peptides derived from the polymorphic region revealed that only those peptides containing the core sequence WEGEPPCLP were recognized (Fig. 2D). Analysis of eight other CD4 + mhag-specific T cell clones isolated from the same patient revealed the CD19 L specificity of three additional T cell clones (unpublished data). All CD19 L -specific T cell clones used the same TCR Vβ16-CDR3 sequence (unpublished data). We could trace this CD19 L -specific TCR in the patient 3 mo after SCT, concomitant with the occurrence of an acute graft-versus-host disease (GvHD) grade III and a strong anti-tumor response, resulting in significant reduction of serum M protein levels (Fig. 2E). The TCR signal disappeared rapidly after starting immunosuppressive treatment but reappeared soon after tapering immunosuppression, indicating the long term persistence of CD19 L -specific T cells. Discontinuation of immunosuppression led to the development of an extensive and steroid-resistant chronic GvHD, despite the disappearance of CD19 L -specific TCR signal from periphery (Fig. 2E). Genome-wide application of the novel mhag fine-mapping strategy Our novel strategy was highly effective to fine-map the genetic locus of the CD19 L - mhag defined on chromosome 16. Because it would be more convenient if our strategy could be directly executed on a genome-wide basis, we correlated the mhag zygosity of 15 CEPH individuals with the complete set of HapMap SNPs from all chromosomes (public release 21a). This first attempt revealed, in addition to the correct locus, several other 100% false-positive correlations throughout the genome (unpublished data), suggesting that mhag zygosity data from more individuals was required. Therefore, we extracted the mhag zygosity of eight additional CEPH individuals from a recent HapMap release (public release 22), which included the genotypes for the CD19 L -encoding SNP. Indeed, by analyzing the data from 23 CEPH individuals we identified the precise mhag locus (Fig. 3) without any other false-positive 100% correlations, demonstrating the feasibility of genome-wide application of our strategy.

31 Identification of a CD19-encoded minor histocompatibility antigen 31 A exon 5 CD GCCAAAGACCGCCCTGAGATCTGGGAGGGAGAGCCTCCGTGTGTCCCACCGAGGGACAGCCTGAACCAGAGCCTCAGCCAGGACCTC A K D R P E I W E G E P P C V P P R D S L N Q S L S Q D L GCCAAAGACCGCCCTGAGATCTGGGAGGGAGAGCCTCCGTGTCTCCCACCGAGGGACAGCCTGAACCAGAGCCTCAGCCAGGACCTC A K D R P E I W E G E P P C L P P R D S L N Q S L S Q D L 188 rs B LCLs Rt LCLs Do + mock D + CD19 L A K D R P E I WE G E P P C L K D R P E I WE G E P P C L P D R P E I WE G E P P C L P P IFN-γ (ng/ml) R P E I WE G E P P C L P P R P E I WE G E P P C L P P R D C E I WE G E P P C L P P R D S 5 PEIWEGEPPCLPPRD I WE G E P P C L P P R D S L PEIWEGEPPCVPPRD WE G E P P C L P P R D S L N E G E P P C L P P R D S L N Q G E P P C L P P R D S L N Q S 2.5 E P P C L P P R D S L N Q S L P P C L P P R D S L N Q S L S P C L P P R D S L N Q S L S Q C L P P R D S L N Q S L S Q D 0 L P P R D S L N Q S L S Q D L Peptide concentration (μm) IFN-γ (ng/ml) 5 10 IFN-γ (ng/ml) Chapter 2 E Days post allo-sct Donor cl.21 irrel <1 <1 gr.iii 0 ext. agvhd cgvhd cl.21-tcr Vβ16-control Serum M-protein (g/l) 2 1 Prednisone (mg/kg) 0 Figure 2. CD19 L encodes for the mhag recognized by clone 21. (A) The CD19 gene with rs in the third exon encoding a valine (V) to leucine (L) substitution at position 174. (B) IFNγ response of clone 21 to mhag donor (Do) EBV-LCLs transduced either with an empty vector (mock) or with the CD19 L - encoding vector. Response to mhag + recipient (Rt) EBV-LCLs (LCLs) is depicted as positive control. The mean and SEM of three experiments are depicted. (C and D) IFNγ response of clone 21 toward serial concentrations of 15-mer peptides derived either from CD19 L ( ) or from CD19 V ( ; C) or toward CD19 L - derived overlapping 15-mer peptides (D). Donor EBV-LCLs were used as APCs. The core sequence recognized by clone 21 is highlighted in gray. Error bars represent the SEM of triplicate cultures. (E) In vivo presence of CD19 L -specific clone 21. Genomic DNA isolated from patient PBMCs at the indicated days after allo-sct was used to amplify the TCR of clone 21. Vβ16-specific PCR was used as positive control. Genomic DNA from clone 21 (cl.21) and third-party PBMCs (irrel.) were used as positive and negative controls, respectively. Also indicated are the severity of acute and chronic GvHD, the serum M protein levels, and the immunosuppressive prednisone treatment during the monitoring period.

32 32 Chapter r Chromosome r Mb Figure 3. Genome-wide mapping of the CD19 L -mhag locus by zygosity-genotype correlation analysis. The r 2 values on the y axis represent the correlation between the mhag zygosities of 23 CEPH individuals with the genotypes for all HapMap SNPs (public release 21a). 15 of these CEPH individuals are depicted in Table S1. For eight other individuals that were also phenotyped by clone 21 recognition, the zygosity information was derived from the rs genotype information in the latest HapMap release (public release 22; unpublished data). Each bar represents a single SNP; only r 2 values >0.5 are shown. Chromosome 16 was analyzed in more detail (bottom). All r 2 values are depicted. The overall power of the novel mhag fine-mapping strategy To determine whether our method is able to locate the genomic locus of other mhags, we first selected six known clinically relevant mhags (Table S2). After deriving the zygosity of 14 CEPH trios (father-mother-child) for these mhags from the HapMap database, we executed a series of correlation analyses by including data from an increasing number of trios. The results for the representative example mhag HMSD are depicted in Figure 4. Although the analysis of the data from seven trios revealed several loci with 100% correlations, false-positive correlations were effectively reduced by entering data from more trios. Finally, the correct mhag locus was revealed after analyzing the data from 11 trios. Similarly, the genomic locus of the other five mhags could be identified after analysis of, at most, 14 trios (Table 1).

33 Identification of a CD19-encoded minor histocompatibility antigen trios 8 trios 9 trios 10 trios r trios 12 trios 13 trios 14 trios r Genome (Chr.1-22) Genome (Chr.1-22) Genome (Chr.1-22) Genome (Chr.1-22) Figure 4. Retrospective genome-wide mapping of the HMSD-mHag locus by zygosity-genotype correlation analysis. The correlation analysis of the HMSD-mHag, using zygosity data from 7-14 CEPH trios, is depicted as a representative example. On the y axis, r 2 values >0.5 are depicted representing the correlation between the HapMap-derived mhag zygosities of the CEPH individuals (Table S2) and the indicated number of trios with their genotypes for all HapMap SNPs (public release 23). The single r 2 = 1 peak after analysis with 11 trios consisted of seven SNPs within the same LD block. Chapter 2 These successfully mapped mhags displayed frequencies between 35 and 70%. As there are not many known mhags beyond these ranges, in further evaluation, we analyzed 149 nonsynonymous nonlinked HapMap SNPs with frequencies between 5 and 95% (Table S3) as if they are encoding for immunogenic mhags. As depicted in Figure 5A, we successfully mapped 121/124 mhags with frequencies of 10-85%, 4/10 mhags with frequencies of 90%, and 1/15 mhags with frequencies of 5 or 95% using trios. These analyses illustrated that our approach was highly powerful for mapping of mhags with frequencies between 10 and 85%. The power beyond these limits decreased significantly. A potential technical problem of genetic analyses is phenotyping errors. Therefore, we also questioned whether our approach still has any power if the datasets would Table 1. Retrospective genome-wide mapping of known mhag loci by zygosity-genotype correlation analysis mhag SNP Frequency (literature) Frequency (in 14 trios) Number of trios Useful number of individuals Locus: number of r 2 =1 -SNPs Source HMSD rs % 31.0% ref 12 ACC-1Y rs %* 47.6% ref 8 ACC-2 rs %* 47.6% ref 8 ADIR rs % 42.9% ref 20 HA-8 rs %* 64.3% ref 21 HB-1Y rs %* 14.3% ref 22 Results of the correlation analyses for six mhags. Published phenotype frequencies are derived from Spierings et al. 19 or from the indicated reference. Analyses were performed as described in Figure 4. Depicted is the minimal number of trios (Number of trios) resulting in only a single genome locus with 100% correlation, together with the number of CEPH individuals for whom zygosity data was available (Useful number of individuals). Also depicted is the number of SNPs with 100% correlation that was found at this genome locus (Locus: number of r 2 = 1 SNPs).

34 34 Chapter 2 A B # tested # identified # Trios used for genomic mapping # tested # identified # Trios used for genomic mapping Datasets with no phenotyping errors mhag phenotype frequency (%) Datasets with 7-10% false positive phenotypes mhag phenotype frequency (%) Figure 5. Retrospective genomewide mapping of nonsynonymous SNPs with various allele frequencies by zygosity-genotype correlation analysis. Summary of correlation analyses for 149 HapMap SNPs (Table S3). From each SNP one of the alleles was designated to encode for a fictive mhag. For each 5% mhag frequency interval between 5 and 95% 10, 5, or 4 mhags were analyzed. The analyses were executed for a dataset without phenotyping errors (A) or with 7-10% false-positive phenotypes (B). The false-positive phenotypes were introduced in the corresponding datasets by randomly changing -/- typings into +/-, avoiding Mendelian segregation errors. The criterion for positive identification is r 2 = 1 at the genomic locus of the analyzed mhag without any r 2 = 1 false-positive hits at other genomic loci (A), or the criterion is r 2 is above the theoretical r 2 (Fig. S2) at the mhag genomic locus without any false-positive r 2 above this value at irrelevant genomic loci (B). The tables show the number of mhags analyzed for each frequency indicated below in the figures, as well as the number of successfully identified mhags. The figures show the number of used trios for only the successfully mapped mhags. contain 10% false-positive phenotypes. Calculations revealed that such errors cause a certain frequency-dependent reduction in the r 2 values (Fig. S2). Nonetheless, we evaluated whether these calculated r 2 values could be used as threshold to discriminate true and false correlations by reexecuting the analyses with 149 SNPs after introducing 7-10% false-positive phenotypes in the datasets. These analyses resulted in the successful mapping of 96/124 (77%) mhags with frequencies of 10-85% and only 2/10 SNPs with frequencies of 90% (Fig. 5B), indicating that, despite significant power loss at extreme allele frequencies, our approach may still be useful for the genomic mapping of mhags with frequencies between 10 and 85%, even if the datasets contain 10% false-positive phenotypes.

35 Identification of a CD19-encoded minor histocompatibility antigen 35 The immunotherapeutic potential of CD19 L -specific T cells Finally, we explored the possible immunotherapeutic advantages of targeting the CD19 L -encoded HLA class II-restricted antigen by testing the helper and effector functions of clone 21. In a coculture system, antigen-specific activation of clone 21 with the synthetic peptide or with mhag + APCs facilitated antigen-dependent proliferation of a CD8 + T cell clone specific for the mhag LRH-1 9 (Fig. 6A). In another assay, clone 21 induced CD19 L peptide-dependent maturation of HLA-DQ-matched DCs and polarized them to produce significant levels of IL12 (Fig. 6B), which is considered essential for licensing DCs to induce CTL responses 23. Clone 21 also maturated the DCs that were loaded with apoptotic CD19 L -positive EBV-LCLs (Fig. 6C). In evaluation of its effector functions, clone 21 appeared to secrete high levels of granzyme-b and displayed antigen-specific cytotoxic activity against recipientderived but not against donor-derived EBV-LCLs (Fig. 7A and B). Furthermore, when tested against malignant cells derived from various B cell chronic lymphoid leukemia (B-CLL) patients, clone 21 produced significant levels of IFNγ (Fig. 7C) and displayed significant antigen-specific cytotoxic activity only toward HLAmatched and CD19 L -positive malignant cells (Fig. 7D). Collectively, these results indicated that upon adoptive transfer, CD19 L -mhag-specific CD4 + T cells may not only provide effective help to other effector cells but may also directly contribute to an anti-tumor effect by lysing CD19 L -positive tumor cells. Chapter 2 DISCUSSION In this paper, we describe the first genuine hematopoietic mhag presented by HLA class II (HLA-DQA1*05/B1*02) molecules to CD4 + T cells. This antigen is encoded by a SNP in the B cell lineage-specific CD19 gene. To identify the CD19 L -encoded mhag, we developed a powerful genetic strategy, which is nonlaborious and genome-wide applicable to identify a wide range of HLA class I- as well as HLA class II-restricted mhags. What may be the clinical importance of this novel HLA class II-restricted mhag? Like CD20, CD19 is a B cell lineage-specific molecule, with constitutive expression in acute and chronic B cell lymphoid leukemias, B cell lymphomas, and in a subset of acute myeloid leukemias. The expression of CD19 is rarely down-regulated in B cell malignancies. Furthermore, CD19 is not expressed in pluripotent stem cells 24. For these reasons, CD19 is widely considered to be an ideal target for immunotherapy of several B cell malignancies. Immunotherapy strategies based on CD19 antibodies (immunotoxins and chimeric TCRs) have been developed and have used both CD4 + and CD8 + T cells as effector cells 25,26. Numerous investigators have searched for HLA class I CTL epitopes on the CD19 molecule 27. Although these approaches may be very beneficial, we think that in an immunotherapy setting, targeting of CD19 with CD4 + T cells also may provide several additional benefits; in our assays, the CD19 L -

36 36 Chapter 2 A Proliferation (10 3 cpm) B % CD19 Lneg -LCLs CD19 Lpos -LCLs * CD19 Lneg -LCLs + CD19 L -pept. * * * 10 : 0 10 : : 50 CD8 + : cl.21 (x1000 cells) MFI pg/ml CD40L cells cl.21 + CD19 L -pept cl.21 + irr. pept irr.cl. + CD19 L -pept. irr.cl. + irr. pept medium % 30 MFI 1, MFI cl.21 + apcd19 Lpos -LCLs irr.cl. + apcd19 Lpos -LCLs cl.21 + apcd19 Lneg -LCLs CD80 CD83 CD86 HLA-DR CD40 CD80 CD83 CD86 HLA-DR IL-12-p70 C 120 Figure 6. The helper activity of CD19 L -specific clone 21. (A) The LRH-1-specific HLA-B7-restricted CD8 + clone cocultured with CD19 L -negative or -positive HLA-B7/DQA1*05/B1*02 EBV-LCLs (LCLs) in the presence of its own epitope. Different dilutions of irradiated clone 21 were added in the cultures and in some conditions were supplemented with the 15-mer CD19 L peptide. The proliferation of the LRH-1-specific CD8 + clone after 48h is depicted. Error bars represent the SEM of triplicate cultures. The proliferation in the presence of CD19 L -positive EBV-LCLs was significantly higher than CD19 L -negative EBV-LCLs (*, P <0.05). (B and C) Immature DCs generated from HLA-DQ2-matched monocytes loaded with 15-mer CD19 L peptide PEIWEGEPPCLPPRD or irrelevant peptide LPPRDSLNQSLSQDL (irr.pept.; B) or with apoptotic CD19 L -negative EBV-LCLs (apcd19 Lneg -LCLs) transduced with CD19 L (apcd19 Lpos - LCLs; C) were cultured with clone 21 (cl.21). Apoptosis was induced by incubation of EBV-LCLs with FasL. The CD4 + mhag-specific T cell clone 3AB11 (irr.cl.) or CD40L-expressing fibroblasts were used as negative and positive controls, respectively. DC maturation was assessed as described in Materials and methods. specific CD4 + T cells not only mediated potent helper functions to activate DCs and stimulate CD8 + CTLs but also directly functioned as effector cells to lyse HLAmatched CD19 L -positive malignant cells. Thus, immunotherapy with CD19 L -specific

37 Identification of a CD19-encoded minor histocompatibility antigen 37 A D Granzyme B (ng/ml) LCLs Do C DQB1*02 CD19 Lysis (%) LCLs Rt Stimulator cells V L V L mhag - B-CLL (n=2) mhag + B-CLL (n=3) 5:1 (n=2) (n=6) (n=7) (n=3) 0 10:1 20:1 E:T ratio B Lysis (%) IFN-γ (pg/ml) 50:1 LCLs Do 50 LCLs Rt Target cells LCLs Do +peptide 100 * Figure 7. The effector function of CD19 L -mhagspecific clone 21. (A) Granzyme B production of clone 21 in response to recipient (Rt) or donor (Do) EBV-LCLs (LCLs). (B) The lysis of recipient EBV-LCLs or donor EBV-LCLs by clone 21 in the absence or presence of 15-mer CD19 L -peptide at an effector/target ratio of 50:1. The error bars represent the SEM of duplicate cultures. (C) IFNγ response of clone 21 toward CD19 + malignant cells from 18 B-CLL patients. The mhag genotypes of the patients were determined by partial sequencing of the chromosomal DNA extracted from PBMCs. The mean and SEM are shown for the indicated number of patient samples. The difference between the HLAmatched CD19 L -positive patients and the others was statistically significant (*, P <0.05). (D) The lysis of HLA-DQB1*02 and CD19 L -positive (n = 3) and CD19 L -negative (n = 2) B-CLL samples by clone 21 at different effector/target ratios. Error bars indicate the SEM of the different B-CLL samples. Chapter 2 CD4 + CTLs alone or in combination with CD8 + T cells may permit the exploitation of both helper and effector functions whereby more effective and durable antitumor responses may be established. Because CD19 is a B cell-specific molecule, CD19 L -specific therapy may be feasible not only within but also beyond the allo-sct settings. The CD19 L -allele is expressed in 53% of the Caucasian population, and its antigen-presenting molecule HLA-DQA1*05/B1*02 has a frequency of 15.3% in the Caucasian population. Thus, after an HLA-matched SCT, 2.3% of the donorrecipient pairs will be CD19 L mismatched and eligible for a treatment with CD19 L - specific CD4 + T cells or vaccination with peptide-loaded or gene-transfected DCs. This percentage can increase up to 8.1% (15.3 x 53%) in an HLA-matched unrelated donor SCT setting or when patients are not transplanted. In all settings, adoptive immunotherapy with ex vivo-generated CD19 L -specific T cells may be feasible, in particular after generation of CD19 L -specific T cells using the so-called TCR-transfer approach 15.

38 38 Chapter 2 What may be the impact of our mhag identification strategy? Our novel genetic approach involves a correlation analysis between mhag zygosities and HapMap SNP genotypes to locate the genetic locus of the mhag. Our results demonstrate that this approach is genome-wide applicable and able to precisely map the genetic locus of a wide range of mhags with phenotype frequencies of 10-85%. The basic idea behind our strategy is similar to a recently introduced method, which utilizes genome association scans after performing SNP array analysis on DNA samples pooled from mhag + and mhag individuals 28. A major advantage of our strategy is that it covers more SNPs. Although the other method can only analyze 65% of the HapMap SNPs after performing 500,000 SNP arrays, we analyze millions of HapMap SNPs covering the whole genome without even performing an SNP array. Furthermore, our strategy is less laborious. In our analyses, the genome-wide identification of the precise LD block of six known mhags required the mhag phenotyping of CEPH individuals, whereas the other approach had to phenotype at least 100 individuals to identify the locus of these mhags 28. Thus, our method can dramatically reduce the work load and the time required to identify mhags, which are the two most important drawbacks of all current mhag identification strategies. The only apparent limitation of our strategy may be its difficulty in identifying mhags with allele frequencies beyond 10-85%. Nonetheless, it should be emphasized that mhags with very low or high phenotype frequencies are of limited immunotherapeutic value. As illustrated in Figure S3, the chance of an mhag-mismatch between the recipient and donor is <10%, if the mhag phenotype frequency is beyond 26-78% in the HLA-matched sibling setting or beyond 11-89% in the HLA-matched unrelated donor setting. Because the number of eligible patients will also be limited by HLA restriction, mhag frequencies beyond these limits are of little practical value for clinical application. Thus, our strategy, with sufficient power to identify mhags with frequencies of 10-85%, is actually very suitable for the identification of a vast majority of mhags that are of high value for immunotherapy. A frequent problem in genetic linkage analyses is phenotyping errors. In our approach, we devoted an extreme attention to avoid such errors because our analyses are executed with a very limited set of CEPH individuals. As phenotype errors will result in the reduction of correlation coefficient, our standard criterion of r 2 = 1 to discriminate true-positive hits from false-positive ones cannot be applied for datasets containing phenotype errors. Yet our analyses with datasets containing 10% false-positive phenotypes revealed that the reduction in correlation coefficient can be calculated (Fig. S2) and that these calculated values could be used as threshold values to eliminate false hits without dramatically reducing the power of the analyses. Thus, our approach, with theoretically adjusted threshold values, may still be useful for identification of clinically relevant mhags even when phenotyping errors cannot be excluded. Nonetheless, as the power decreases significantly we still think that the best strategy for success is to use all means to avoid phenotype errors.

39 Identification of a CD19-encoded minor histocompatibility antigen 39 Finally, we think that our strategy is universally applicable because there are five ethnic panels with trios genotyped in the Phase III HapMap. Using one of these panels may be sufficient for analysis of a wide range of mhag-specific T cell clones, even if they are obtained from a different population. In fact, using the Caucasoid CEPH panel, we have been able to map the genomic locus of HMSD, ACC-1Y, and ACC-2, mhags which were originally described in the Japanese population. Nonetheless, not all HapMap panels consist of trios, which are required for deducing the zygosity information. Because this may still be a potential limitation for identification of mhags in some ethnic populations, we are currently evaluating the possibility of using unrelated HapMap individuals in our strategy. In conclusion, the first HLA class II-restricted hematopoietic mhag as well as the powerful mhag identification strategy described here can significantly facilitate the application of mhag-based immunotherapy in a broader clinical setting. Chapter 2 Acknowledgments We thank Drs. M. Theobald, J. Kuball, and B. Koeleman for stimulating discussions and critical reading of the manuscript, Dr. F. Koning for providing the HLA-DQ retroviral constructs, Dr. D. Nyholt for providing the standalone version of sssnper, P. de Koning for providing reagents for granzyme-b assays, and R. van Gent for providing B-CLL samples. Materials and methods Cells The HLA-DQA1*05/B1*02-restricted CD4 + T cell clone 21 and the HLA-DP4-restricted CD4 + control clone 3AB11 were previously isolated from a multiple myeloma patient during the development of acute GvHD grade III and a strong GvT response after allo-sct from his HLA-identical brother 29. In brief, after SCT, PBMCs of the recipient were stimulated with irradiated pre-sct PBMCs for 3wk and thereafter expanded with EBV-LCLs derived from pre-sct PBMCs. T cell clones were obtained by limiting dilution. The HLA-B7-restricted CD8 + T cell clone specific for the mhag LRH-1 was described elsewhere 9. T cell clones were expanded using a feeder cell-cytokine mixture as previously described 15. EBV-LCLs were cultured in RPMI-1640, 10% FBS (Integro), and antibiotics. The Phoenix packaging line (-NX-A) and the CD40L-expressing fibroblasts were cultured in DMEM (Invitrogen), 10% FBS, and antibiotics. PBMCs containing 84-99% CD5 + malignant B cells (partly provided by R. van Gent, University Medical Center Utrecht, Utrecht, Netherlands) were obtained from 18 B-CLL patients after written informed consent. Experiments with these human materials were approved by local medical ethical committee of the University Medical Center Utrecht.

40 40 Chapter 2 Immunophenotyping Cells labeled with fluorochrome-conjugated antibodies (all obtained from BD, except the α-δngf-r [culture supernatant of clone 20.4]) were analyzed with a FACSCalibur flow cytometer (BD). FACS data were analyzed with CellQuest software (BD). CD19-derived peptides Commercially synthesized and purified (Pepscan) 15-mer peptides were dissolved in DMSO to 100 mm and diluted in PBS to 6 mm to use in functional assays. Retroviral vectors The LZRS-based retroviral vectors containing HLA-DQA1*0501-IRES-GFP or HLA- DQB1*0201-IRES-ΔNGF-R were provided by F. Koning (Leiden University Medical Center, Leiden, Netherlands). The pmx-cd19174l-ires-gfp vector was generated by cloning the CD19 L gene derived from an expression vector (OriGene Technologies). Generation of retroviral supernatants and retroviral transduction of EBV-LCLs were described elsewhere 15. mhag phenotyping of CEPH families To determine the mhag phenotype (mhag + or mhag-) of the CEPH individuals, their HLA- DQA1*05/B1*02-positive (naturally positive or positive after retroviral transduction) EBV- LCLs were used as APCs for the mhag-specific clone 21 at a 1:1 ratio, in a final volume of 200 µl in U-bottomed 96-well plates. After 18h of incubation, supernatants were collected and tested for the presence of IFNγ using standard ELISA kits. EBV-LCLs were judged mhag + if the mean OD value at 450 nm of triplicate cultures was >0.250, about four times the background OD value. For <10% of the CEPH individuals, the OD value was between 0.1 and In these cases, assays were repeated at least four times to exactly determine the mhag phenotype of the CEPH individual. Individuals remaining doubtful after five attempts were excluded from the analyses. ELISA Cytokine content of cell-free supernatants was determined using commercial ELISA kits according to the manufacturer s instructions (IFNγ, Invitrogen; granzyme-b, Sanquin [provided by P de Koning, University Medical Center Utrecht, Utrecht, Netherlands]; IL12-p70, ebioscience). The SEM of triplicate measurements never exceeded 10%. Two-point pairwise linkage analysis The mhag phenotypes of the CEPH individuals and 533 markers genotyped on CEPH families 1331, 1362, and 1408 were used as input to perform a genome-wide two-point linkage analysis using the subroutine MLINK of the FASTLINK program as previously described 8,9,17,30,31. Recombinant haplotype analysis Genotype data for chromosome 16 were extracted from the CEPH database (version 10; to determine recombinant haplotypes in the CEPH families 32,33. Candidate genes and SNPs were derived from Map Viewer ( and Ensembl (release 45; respectively. Correlation of mhag zygosities with HapMap SNP genotypes The software sssnper (provided by D. Nyholt, Queensland Institute of Medical Research, Brisbane, Australia), a program for identifying genetically indistinguishable SNPs 34, was

41 Identification of a CD19-encoded minor histocompatibility antigen 41 used to analyze the correlation between the mhag zygosities and individual HapMap SNPs. For genome-wide analyses, sssnper was slightly modified. We have shared this modified sssnper at mhag zygosities (+/+, +/-, and -/-) of CEPH individuals were deduced from the Mendelian segregation pattern of the mhag phenotypes in their pedigrees. The HapMap SNP genotypes were downloaded from the HapMap website ( The input file for sssnper was assembled by inserting the deduced mhag zygosities of CEPH individuals as a fictive SNP in the downloaded HapMap file. SNP and TCR sequencing 60 ng of genomic DNA isolated from PBMCs or EBV-LCLs was amplified at 94 C for 5 min for 35 cycles (94 C for 30s, 55 C for 30s, and 72 C for 30s) and 72 C for 10 min using primers (Table S4) and directly sequenced using BigDye Terminator kit (v3.1; Applied Biosystems) in an automated ABI PRISM 3100 sequencer (Applied Biosystems). Chapter 2 Monitoring of the CD19 L -specific TCR The TCRβ chain of clone 21 was amplified from genomic DNA isolated from PBMCs collected periodically after allo-sct using a TCR-specific primer set (Table S4). The specificity of the PCR was controlled using genomic DNA of clone 21 (positive control) and of several irrelevant PBMCs (negative control). T cell help for expansion of CD8 + mhag-specific T cells LRH-1-specific CD8 + T cell clone 9 was stimulated with HLA-matched CD19 L -mhag-negative or -positive EBV-LCLs at a ratio of 1:2.5 in the presence of 5 µm LRH-1 peptide. 30 Gray-irradiated clone 21 was added with or without 0.6 µm of the 15-mer CD19 L -peptide. 3H-Thymidine incorporation was determined after 48h. DC maturation assay Immature DCs, generated from HLA-matched monocytes culturing with GM-CSF and IL4, were maturated by coculture with T cell clones at a DC/T cell ratio of 1:2.5 in the presence of 6 µm of either the CD19 L peptide PEIWEGEPPCLPPRD or irrelevant peptide LPPRDSLNQSLSQDL. In some assays, apoptotic CD19 L -positive EBV-LCLs (apoptosis was induced by incubation with 500 ng/ml FasL for 24h) were added as the source of CD19 L antigen (DC/EBV-LCL ratio, 1:20). As positive control, DCs were cocultured with CD40L-expressing fibroblasts in a DC/ fibroblast ratio of 1:1. After 72h, DC maturation was assessed by determining the percentage of CD80 + and CD83 + cells, the mean fluorescence intensity of CD86, HLA-DR, and/or CD40 on the cells, and the content of IL12 in the culture supernatants. FACS-based cytotoxicity assays The capacity of clone 21 to lyse various mhag + or mhag targets (EBV-LCLs or B-CLL cells) was measured in FACS-based 48h cytotoxicity assays as previously described 15. The SEM of duplicate cultures never exceeded 15%. Statistical analyses Where indicated, differences between groups were analyzed in two tailed Student s t tests with Prism software (GraphPad Software, Inc.). P-values <0.05 were considered significant.

42 42 Chapter 2 References 1. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, CA Cancer J Clin. 2005;55: Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: de Bueger M, Bakker A, van Rood JJ, Van der Woude F, Goulmy E. Tissue distribution of human minor histocompatibility antigens. Ubiquitous versus restricted tissue distribution indicates heterogeneity among human cytotoxic T lymphocytedefined non-mhc antigens. J Immunol. 1992;149: Warren EH, Greenberg PD, Riddell SR. Cytotoxic T-lymphocyte-defined human minor histocompatibility antigens with a restricted tissue distribution. Blood. 1998;91: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Akatsuka Y, Nishida T, Kondo E et al. Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med. 2003;197: Rijke BD, van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: Brickner AG, Evans AM, Mito JK et al. The PANE1 gene encodes a novel human minor histocompatibility antigen that is selectively expressed in B-lymphoid cells and B-CLL. Blood. 2006;107: Warren EH, Vigneron NJ, Gavin MA et al. An antigen produced by splicing of noncontiguous peptides in the reverse order. Science. 2006;313: Kawase T, Akatsuka Y, Torikai H et al. Alternative splicing due to an intronic SNP in HMSD generates a novel minor histocompatibility antigen. Blood. 2007;110: Riddell SR, Bleakley M, Nishida T, Berger C, Warren EH. Adoptive transfer of allogeneic antigen-specific T cells. Biol Blood Marrow Transplant. 2006;12: Jiang YZ, Barrett J. The allogeneic CD4 + T cell-mediated graft-versus-leukemia effect. Leuk Lymphoma. 1997;28: Spaapen R, van den Oudenalder K, Ivanov R et al. Rebuilding Human Leukocyte Antigen Class II-Restricted Minor Histocompatibility Antigen Specificity in Recall Antigen-Specific T Cells by Adoptive T Cell Receptor Transfer: Implications for Adoptive Immunotherapy. Clin Cancer Res. 2007;13: Griffioen M, van der Meijden ED, Slager EH et al. Identification of phosphatidylinositol 4-kinase type II beta as HLA class II-restricted target in graft versus leukemia reactivity. Proc Natl Acad Sci U S A. 2008;105: Warren EH, Otterud BE, Linterman RW et al. Feasibility of using genetic linkage analysis to identify the genes encoding T cell-defined minor histocompatibility antigens. Tissue Antigens. 2002;59: The International HapMap Consortium. The International HapMap Project. Nature. 2003;426: Spierings E, Hendriks M, Absi L et al. Phenotype frequencies of autosomal minor histocompatibility antigens display significant differences among populations. PLoS Genet. 2007;3:e van Bergen CA, Kester MG, Jedema I et al. Multiple myeloma-reactive T cells recognize an activation-induced minor histocompatibility antigen encoded by the ATP-dependent interferon-responsive (ADIR) gene. Blood. 2007;109: Brickner AG, Warren EH, Caldwell JA et al. The immunogenicity of a new human minor histocompatibility antigen results from differential antigen processing. J Exp Med. 2001;193: Dolstra H, de RB, Fredrix H et al. Bi-directional allelic recognition of the human minor histocompatibility antigen HB-1 by cytotoxic T lymphocytes. Eur J Immunol. 2002;32: Cella M, Scheidegger D, Palmer-Lehmann K et al. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via APC activation. J Exp Med. 1996;184: Scheuermann RH, Racila E. CD19 antigen in leukemia and lymphoma diagnosis and immunotherapy. Leuk Lymphoma. 1995;18:

43 Identification of a CD19-encoded minor histocompatibility antigen Kowolik CM, Topp MS, Gonzalez S et al. CD28 costimulation provided through a CD19-specific chimeric antigen receptor enhances in vivo persistence and antitumor efficacy of adoptively transferred T cells. Cancer Res. 2006;66: Brentjens RJ, Latouche JB, Santos E et al. Eradication of systemic B cell tumors by genetically targeted human T lymphocytes co-stimulated by CD80 and interleukin-15. Nat Med. 2003;9: Voss RH, Lotz C, Cellary A, Theobald M. Targeting p53, hdm2, and CD19: vaccination and immunologic strategies. Bone Marrow Transplant. 2000;25 Suppl 2:S43-S Kawase T, Nannya Y, Torikai H et al. Identification of human minor histocompatibility antigens based on genetic association with highly parallel genotyping of pooled DNA. Blood. 2008;111: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Schaffer AA, Gupta SK, Shriram K, Cottingham RW, Jr. Avoiding recomputation in linkage analysis. Hum Hered. 1994;44: Cottingham RW, Jr., Idury RM, Schaffer AA. Faster sequential genetic linkage computations. Am J Hum Genet. 1993;53: Murray JC, Buetow KH, Weber JL et al. A comprehensive human linkage map with centimorgan density. Cooperative Human Linkage Center (CHLC). Science. 1994;265: Dausset J, Cann H, Cohen D et al. Centre d etude du polymorphisme humain (CEPH): collaborative genetic mapping of the human genome. Genomics. 1990;6: Nyholt DR. sssnper: identifying statistically similar SNPs to aid interpretation of genetic association studies. Bioinformatics. 2006;22: Chapter 2

44 44 Chapter 2 Supplemental data CEPH family (1408:01) +/ (1408:02) -/ (1408:13) +/ (1408:12) -/ CEPH family CEPH family Figure S1. mhag phenotypes for the individuals of CEPH families 1408, 1416, and mhag phenotypes for the individuals of CEPH families 1408, 1416, and The male (square) and female (circle) CEPH family members are indicated with their Utah database ID numbers. Individuals with a positive (black), negative (white), or undetermined (gray) phenotype for the mhag recognized by clone 21 are indicated. As an example for Table S1, zygosity information of four individuals from family 1408 was derived from the indicated part of the pedigree and was used as input for the zygosity-genotype correlation analysis (Fig. 1D and 3). Figure S2. The loss of correlation in datasets containing 10% false-positive phenotypes. We constructed a dataset containing only heterozygous (+/-) and homozygous negative (-/-) individuals for various fictive mhags because phenotype/zygosity information for correlation analysis never contains homozygous positive individuals. For each mhag, the frequency of +/- individuals varied from 0-66% because values beyond these are not possible when the zygosity information is deduced from trios. This dataset was correlated with a second dataset, in which 10% false-positive phenotypes were introduced by replacing -/- typings with +/- ones. sssnper software was used to execute the correlation analyses. The results are expressed for +/- frequency in the dataset with 10% false-positive phenotypes. r % +/- in the analysis input

45 Identification of a CD19-encoded minor histocompatibility antigen 45 Chance of mismatch (%) % 26% 78% 89% mhag phenotype frequency (%) Figure S3. The relationship between mhag phenotype frequencies and the chance of mhag-mismatch (patient mhag + /donor mhag ) in an allo-sct setting. The chances are depicted for sibling and MUD transplantation. The plot for siblings was generated using the formula p(mhag-mismatch) = (0.75 [aa] 0.25 [aa] [aa] 1.5 ) 100%, in which [aa] stands for population frequency of mhag individuals. The plot for MUD was generated using the formula p(mhag-mismatch) = [aa] (1 [aa]) 100%. The mhag phenotype frequencies resulting in a mhag-mismatch chance of at least 10% after sibling and MUD SCT are indicated. MUD sibling Chapter 2 Table S1. CEPH individuals used in zygosity-genotype correlation analysis HapMap individual number Corresponding Utah number Clone 21 mhag phenotype Clone 21 mhag zygosity NA homozygous NA homozygous NA homozygous NA homozygous NA homozygous NA heterozygous NA homozygous NA heterozygous NA homozygous NA homozygous NA heterozygous NA homozygous NA homozygous NA heterozygous NA heterozygous HapMap database ID numbers for CEPH individuals are depicted with their corresponding Utah database numbers. Clone 21 mhag phenotype results are indicated. The mhag zygosity of the individuals was deduced from the Mendelian segregation pattern of the mhag phenotypes in their pedigrees (see family 1408 for anexample [Fig. S1]). Information for both mhag phenotype and zygosity (+/+, +/-, or -/-) was submitted for zygosity-genotype correlation analysis (Fig. 3).

46 46 Chapter 2 Table S2. Input phenotypes/zygosities for retrospective genome-wide analysis of six mhags mhag: HMSD mhag: ACC-1Y mhag: ACC-2 mhag: ADIR mhag: HA-8 mhag: HB-1Y SNP: rs SNP: rs SNP: rs SNP: rs SNP: rs SNP: rs HapMap Immunog. SNP allele: A Immunog. SNP allele: A Immunog. SNP allele: A Immunog. SNP allele: A Immunog. SNP allele: C Immunog. SNP allele: T Trio individual Geno. Pheno. Zyg. Geno. Pheno. Zyg. Geno. Pheno. Zyg. Geno. Pheno. Zyg. Geno. Pheno. Zyg. Geno. Pheno. Zyg. NA06993 GG - -/- GG - -/- GG - -/- AA + CC + CC - -/- 1 NA06985 GG - -/- GG - -/- GG - -/- GG - -/- CC + CC - -/- NA06991 GG - -/- GG - -/- GG - -/- AG + +/- CC + CC - -/- NA07034 GG - -/- GG - -/- GG - -/- AG + +/- GG - -/- CC - -/- 2 NA07055 AG + +/- AG + +/- AG + +/- GG - -/- CG + CC - -/- NA07048 GG - -/- GG - -/- GG - -/- GG - -/- CG + +/- CC - -/- NA12144 AA + AG + AG + GG - -/- CC + CC - -/- 3 NA12145 GG - -/- GG - -/- GG - -/- GG - -/- GG - -/- CT + +/- NA10846 AG + +/- AG + +/- AG + +/- GG - -/- CG + +/- CC - -/- NA12056 AG + +/- AG + +/- AG + +/- AG + CC + CC - -/- 4 NA12057 GG - -/- GG - -/- GG - -/- AG + GG - -/- CC - -/- NA10851 GG - -/- GG - -/- GG - -/- NN CG + +/- CC - -/- NA11831 GG - -/- GG - -/- GG - -/- GG - -/- CC + CC - -/- 5 NA11832 GG - -/- AG + AG + GG - -/- GG - -/- CT + NA10855 GG - -/- AG + +/- AG + +/- GG - -/- CG + +/- CT + +/- NA12043 AA + GG - -/- GG - -/- GG - -/- GG - -/- CC - -/- 6 NA12044 GG - -/- AG + AG + AG + +/- CG + CC - -/- NA10857 AG + +/- AG + +/- AG + +/- GG - -/- CG + +/- CC - -/- NA11881 GG - -/- AG + AG + GG - -/- CG + CC - -/- 7 NA11882 GG - -/- AG + AG + GG - -/- CG + CC - -/- NA10859 GG - -/- AA + AA + GG - -/- CG + CC - -/- NA11992 GG - -/- GG - -/- GG - -/- NN CG + CT + +/- 8 NA11993 AG + +/- GG - -/- GG - -/- AG + CG + CC - -/- NA10860 GG - -/- GG - -/- GG - -/- AG + CG + CC - -/- NA11994 GG - -/- GG - -/- GG - -/- GG - -/- CG + +/- CC - -/- 9 NA11995 GG - -/- AG + AG + AG + CG + +/- CC - -/- NA10861 GG - -/- AG + +/- AG + +/- AG + +/- GG - -/- CC - -/- NA12155 GG - -/- GG - -/- GG - -/- GG - -/- CG + +/- CC - -/- 10 NA12156 AA + GG - -/- GG - -/- AG + GG - -/- CT + +/- NA10831 AG + +/- GG - -/- GG - -/- NN GG - -/- CC - -/-

47 Identification of a CD19-encoded minor histocompatibility antigen 47 NA10831 AG + +/- GG - -/- GG - -/- NN GG - -/- CC - -/- NA11829 AG + +/- AG + AG + GG - -/- CG + +/- CC - -/- 11 NA11830 GG - -/- AG + AG + GG - -/- CG + +/- CC - -/- NA10856 GG - -/- AG + AG + GG - -/- GG - -/- CC - -/- NA12003 GG - -/- AG + +/- AG + +/- GG - -/- CG + CC - -/- 12 NA12004 GG - -/- GG - -/- GG - -/- NN CG + CC - -/- NA10838 GG - -/- GG - -/- GG - -/- AG + +/- CC + CC - -/- NA12248 GG - -/- AG + AG + GG - -/- GG - -/- CC - -/- 13 NA12249 GG - -/- AG + AG + NN GG - -/- CC - -/- NA10835 GG - -/- AA + AA + GG - -/- GG - -/- CC - -/- NA11839 AG + GG - -/- GG - -/- AG + GG - -/- CT + +/- 14 NA11840 AG + GG - -/- GG - -/- NN GG - -/- CC - -/- NA10854 AG + GG - -/- GG - -/- AG + GG - -/- CC - -/- Depicted are 14 CEPH trios (father-mother-child), with their HapMap genotypes for SNPs encoding for six previously identified mhags. The mhag phenotype and zygosity were deduced from these genotypes. Individuals carrying the immunogenic SNP-allele were called mhag +. For some individuals, trios did not provide sufficient information to deduce the mhag zygosity (empty fields). The analyses in Figure 4 and Table 1 were executed only with those individuals for whom the mhag zygosity information was available. Immunog, immunogenic; Geno, genotype; Pheno, phenotype; Zyg., zygosity. Chapter 2

48 48 Chapter 2 Table S3. CEPH individuals used in zygosity-genotype correlation analysis SNP Freq. Chr. SNP Freq. Chr. SNP Freq. Chr. SNP Freq. Chr. rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs rs

49 Identification of a CD19-encoded minor histocompatibility antigen 49 Table S4. PCR primers Target Forward primer Reverse primer rs rs rs cctcaagtccccaagattca cttcccacacaatggcacta-3 rs rs ggtggaatgtttcggaccta ggacacagagtcagggggta-3 rs rs ccccgcttaaacccttctta gcctcccttctctcctcct-3 rs gtagagacggggtctcacca tgcatctttccctccttcac-3 rs ccccaggtactcaccatctg aggactgtgccaagcaggt-3 Vβ16-Cβ3 5 - ctgttacattttgtgaaagagtc ctccttcccattcacccaccagctcagctc-3 cl.21-tcr 5 - ctgttacattttgtgaaagagtc gcgtatctaaagtctcttggctg-3 Vβ16 control 5 - ctgttacattttgtgaaagagtc ctaaagtctcttggctgctgg-3 The indicated forward and reverse primers were used for PCR and sequence reactions to identify the genotypes of the SNPs depicted with rs numbers, to determine the Vβ16 sequence of the rearranged TCR gene (Vβ16-Cβ3), or to specifically amplify clone 21 TCR with Vβ16 as a positive control. rs, RefSNP. Chapter 2

50 50 Chapter 2

51 3 Rapid identification of clinically relevant minor histocompatibility antigens via genome-wide zygosity-genotype correlation analysis Robbert M. Spaapen 1, Ron A.L. de Kort 1, Kelly van den Oudenalder 1, Maureen van Elk 1, Andries C. Bloem 2, Henk M. Lokhorst 3, Tuna Mutis 1 Departments of 1 Clinical Chemistry and Haematology, 2 Immunology, 3 Haematology, UMCU, NL Clin Cancer Res Dec; 15:

52 52 Chapter 3 Identification of minor histocompatibility antigens (mhags) with classic methods often requires sophisticated technologies, determination, and patience. We here describe and validate a nonlaborious and convenient genetic approach, based on genome-wide correlations of mhag zygosities with HapMap single-nucleotide polymorphism genotypes, to identify clinically relevant mhags within a reasonable time frame. Using this approach, we sought for the mhag recognized by a HLA-DRB1*1501-restricted T cell clone, isolated from a multiple myeloma patient during a strong graft-versus-tumor effect associated with acute graft-versus-host disease grade 3. In a period of 3 months, we determined the mhag phenotype of 54 HapMap individuals, deduced the zygosity of 20 individuals, defined the mhag locus by zygosity-genotype correlation analyses, tested the putative mhagpeptides from this locus, and finally showed that the mhag is encoded by the arginine (R)-allele of a nonsynonymous single-nucleotide polymorphism in the SLC19A1 gene. We conclude that this powerful and convenient strategy offers a broadly accessible platform toward rapid identification of mhags associated with graft-versus-tumor effect and graft-versus-host disease. Translational Relevance In this study, we have identified a new autosomal HLA class II-restricted minor histocompatibility antigen using a new genetic approach in a time frame of only months. The identified mhag, SLC19A1 R, will be frequently mismatched in an allogeneic stem cell transplantation and is likely to be involved in graft-versus-tumor effect and graft-versus-host disease. Thus, our genetic approach can facilitate rapid and large-scale molecular characterization of novel clinically relevant mhags. Toward meaningful exploitation of our approach, we determined a directed strategy and calculated the number of mhags for wide-scale application of mhag-specific immunotherapy. We included both HLA class I- and HLA class II-restricted mhags because not only CD8 + but also CD4 + T cells contribute to anti-tumor reactions after allogeneic stem cell transplantation. Thus, rapid identification of mhags with our technology and following a directed strategy are relevant approaches toward widescale clinical application of mhag-specific T cell immunotherapy. INTRODUCTION HLA-matched allogeneic stem cell transplantation (allo-sct) is an effective treatment for hematologic malignancies. After allo-sct, donor T cells directed at the minor histocompatibility antigens (mhags) of the recipient mediate graft-versus-

53 Rapid identification of minor histocompatibility antigens 53 mhag specific T cell clone TCR Determine HLA restriction (1 wk) T cell-based mhag-phenotyping of HapMap EBV-LCLs (3-4 wk) IFN-γ GM GM12004 GM EBV-LCLs of HapMap trios Insert HLA restriction element (3-4 wk) Zygosity deduction (1 hr) HLA /? / +/ Download free software & perform zygosity-genotype correlation analysis (4 hr) mhag peptide and/or gene recognition (3-4 wk) Figure 1. Schematic overview of the genomewide mhag identification strategy. The approximate time needed for each single step is indicated. tumor (GvT) effects, but they can also cause graft-versus-host disease (GvHD). mhags are polymorphic peptides derived from intracellular proteins and presented by HLA molecules 1-4. Pioneering studies indicated that mhags expressed exclusively in hematopoietic cells can serve as excellent tools to separate GvT effects from GvHD 5. Currently, one of the important bottlenecks toward broad application of mhag-based immunotherapy strategies is the speed of identifying relevant hematopoietic mhags. Classic identification methods such as peptide elution and cdna library screening are complex, time-consuming, and offer a moderate chance of success 3,4,6. Recently introduced pairwise linkage analyses are also time-consuming, require advanced genetic know-how, and are not always successful in identifying the precise mhag locus We recently identified the CD19 L -encoded mhag by combining pairwise linkage analysis with a novel fine-mapping strategy, called zygosity-genotype correlation analysis 11. In retrospective computational analyses, the genome-wide approach of zygositygenotype correlation analysis seemed powerful enough to be applied as standalone identification strategy for mhags with 10% to 85% population frequency. Because this method was hypothesized to be more rapid and accessible than other (genetic) approaches, we now explored its actual speed and ease, using a CD4 + T cell clone recognizing a novel HLA-DRB1*1501 restricted mhag. Within only 3 months, we succeeded in identifying the HLA class II mhag as a polymorphic peptide encoded by the SLC19A1 R gene. Chapter 3

54 54 Chapter 3 RESULTS Selection of the mhag-specific T cell clone To validate the genome-wide zygosity-genotype correlation approach (Fig. 1) and to evaluate its speed, we used the HLA-DRB1*1501 restricted mhag-specific CD4 + T cell clone 1GF5. The HLA class II-restriction of this clone was ideal for validation of our approach because identification of HLA class II-restricted mhags is usually more difficult as compared with HLA class I-restricted antigens. Furthermore, the mhag could have clinical relevance because the clone 1GF5 was isolated from a multiple myeloma patient during a strong GvT effect associated with acute GvHD grade 3 and showed high IFNγ release as well as strong cytotoxic activity against an allogeneic multiple myeloma cell line in a mhag-specific fashion (Fig. 2A and B). A >1.5 Multiple Myeloma cell lines IFN-γ (ng/ml) HLA-DR15: mhag: nd B Lysis (%) Multiple Myeloma cell lines: HLA-DR15, mhag + HLA-DR15, mhag HLA-DR15 negative 9:1 3:1 1:1 E:T ratio Figure 2. 1GF5 recognizes and lyses multiple myeloma cell line UM9. IFNγ response (A) and cytotoxic activity (B) of 1GF5 against three multiple myeloma cell lines (UM9 [HLA-DRB1*1501, mhag + ], U266 [HLA-DRB1*1501, mhag ], and RPMI [HLA- DRB1*1501 negative]) after 18h (A) and 6h (B) of coincubation using different effector-to-target (E:T) ratios. Columns and points, average of triplicate cultures, which are representative of two independent experiments; bars, SEM. Mapping of the mhag recognized by 1GF5 via mhag zygosity-based genomewide analysis The genetic correlation analyses in our approach use mhag zygosities, which can be deduced using inheritance of mhag phenotypes. Therefore, to identify the mhag recognized by 1GF5, we first determined the mhag phenotypes of several HapMap trios (father-mother-child) by testing the reactivity of 1GF5 toward EBV-LCLs derived from these individuals. In total, we phenotyped 54 EBV-LCLs, of which 42 (78%) were mhag + (Table 1). Using these data, we could deduce the mhag zygosity (+/+, +/-, or -/-) of 20 of 54 individuals (see Table 1 and Figure 1 for an example). Correlating this zygosity information to the genotypes of almost four million SNPs

55 Rapid identification of minor histocompatibility antigens 55 Table 1. Phenotyping and zygosity determination of HapMap individuals HapMap individual Trio relation IFN-γ absorbance value Phenotype Zygosity GM12144 Father GM12145 Mother GM06994 Father /- GM07029 Child /- GM07022 Father GM07019 Child GM06993 Father /- GM06991 Child /- GM07034 Father GM07055 Mother GM07048 Child GM12056 Father GM12057 Mother GM12043 Father GM12044 Mother GM10857 Child GM11881 Father GM11882 Mother GM10859 Child GM11839 Father GM11840 Mother GM10854 Child GM11831 Father GM11832 Mother GM10855 Child GM11829 Father GM11830 Mother GM10856 Child GM11992 Father GM11993 Mother GM10860 Child GM11994 Father GM11995 Mother GM10861 Child GM12264 Father /- GM10863 Child /- GM12154 Father /- GM12236 Mother /- GM10830 Child /- GM12155 Father /- GM12156 Mother /- GM10831 Child /- GM12248 Father /- GM12249 Mother /- GM10835 Child /- GM12003 Father GM12004 Mother /- GM10838 Child /- GM12005 Father GM10839 Child GM12750 Father GM12761 Mother /- GM12752 Child /- GM12815 Mother /- Depicted are the 54 phenotyped HapMap individuals derived from father-mother-child trios. The IFNγ response of clone 1GF5 (average absorbance value of triplicate culture) against their EBV-LCLs (naturally HLA- DRB1*1501-positive or retrovirally transduced) is shown with the subsequently designated phenotypes (absorbance value >0.250 was considered as mhag + ; ref. 11). The mhag zygosity was deduced from the inheritance pattern of the phenotypes within each trio (see Figure 1 for an example). For some individuals, trios did not provide sufficient information to deduce the mhag zygosity (empty fields). The analysis depicted in Figure 3A was executed only with those individuals for whom the mhag zygosity information was available. Chapter 3

56 56 Chapter 3 in the whole genome using a freely available software ( dcch/research/hemato-oncology) revealed 100% correlation (r 2 = 1) with four SNPs (rs , rs , rs , and rs ; Fig. 3A), located within A rs rs rs rs r Chromosome B SLC19A1: MVPSSPAVEKQVPVEPGPDPELRSWR(R/H)LVCYLCFYGFMAQIRPGESFITPYLLG... rs C LCL Rt LCL Do + mock + SLC19A1 R >1.5 IFN-γ (ng/ml) D P V E P G P D P E L R SWR R V E P G P D P E L R SWR R L E P G P D P E L R SWR R L V P G P D P E L R SWR R L V C G P D P E L R SWR R L V C Y P D P E L R SWR R L V C Y L D P E L R SWR R L V C Y L C P E L R SWR R L V C Y L C F E L R SWR R L V C Y L C F Y L R SWR R L V C Y L C F Y G R SWR R L V C Y L C F Y G F SWR R L V C Y L C F Y G F M WR R L V C Y L C F Y G F M A R R L V C Y L C F Y G F M A Q R L V C Y L C F Y G F M A Q I >1.5 IFN-γ (ng/ml) Figure 3. Genome-wide mapping of the SLC19A1 R -encoded mhag recognized by 1GF5. A, zygositygenotype correlation analysis was done with the 20 known mhag zygosities shown in Table 1. Each column represents a single SNP. The r 2 values represent the correlation between the mhag zygosities of the EBV-LCLs and their genotypes for ~ HapMap SNPs. Only r 2 values above 0.5 are shown. The four SNPs (rs , rs , rs , and rs ) with 100% correlation (r 2 = 1.0) are indicated. B, the segment of SLC19A1 in which the SNP rs encodes for a histidine (H) to an arginine (R) substitution at amino acid position 27. C, IFNγ response (arbitrary units) of 1GF5 to mhag donor (Do) EBV-LCLs (LCL) transduced either with an empty vector (mock) or with the full-length SLC19A1 R -encoding vector. Response to mhag + recipient (Rt) EBV-LCLs is depicted as positive control. D, IFNγ response of 1GF5 toward donor EBV-LCLs loaded with SLC19A1 R -derived overlapping 15-mer peptides. C and D, horizontal columns, mean of triplicate wells; bars, SEM. Representative results from two different experiments.

57 Rapid identification of minor histocompatibility antigens 57 one linkage disequilibrium block on chromosome 21. This indicated that this linkage disequilibrium block could contain the SNP encoding for the mhag. Supporting this idea, the four SNPs also showed 100% correlation with the mhag phenotypes of the remaining 34 mhag + EBV-LCLs whose mhag zygosities could not be deduced. Identification of the SLC19A1 R -encoded mhag One of the four 100% correlating SNPs (rs ) was nonsynonymous, encoding for a histidine (H) to an arginine (R) substitution at position 27 of SLC19A1 (Fig. 3B), suggesting that SLC19A1 R could be encoding the mhag. Supporting this possibility, the original SCT patient, but not the donor, carried the SLC19A1 R -encoding allele (data not shown), and donor EBV-LCLs were recognized by 1GF5 on transduction with the SLC19A1 R gene but not with an empty vector (Fig. 3C). Finally, 1GF5 recognized several synthetic 15-mer peptides containing the SLC19A1 R -derived sequence RLVCYLC, illustrating that the HLA class II-restricted mhag was indeed a polymorphic peptide derived from SLC19A1 R (Fig. 3D). Analysis of all other CD4 + mhag-specific T cell clones isolated from the patient revealed the recognition of SLC19A1 R peptides by five additional clones using at least two different T cell receptors. All these clones also recognized the mhag + multiple myeloma cell line UM9 (data not shown). Chapter 3 DISCUSSION In this study, we describe and validate a nonlaborious and convenient genetic approach to identify clinically relevant mhags. Using this genome-wide zygositygenotype correlation approach, we identified the mhag recognized by a CD4 + T cell clone as a polymorphic peptide derived from the SLC19A1 R gene, which encodes a transmembrane protein functioning as a transporter for natural folate compounds 12. The SNP encoding for the SLC19A1 R -mhag is polymorph in all HapMap populations, with a phenotype frequency between 39% and 88% ( implying that a mismatch for this new mhag will be frequently encountered in an allo-sct setting. It is likely that this HLA class II-restricted mhag can contribute to the GvT effect after allo-sct because multiple SLC19A1 R -mhag-specific T cells were isolated from our patient who manifested a strong GvT effect, and all of these T cells were capable of recognizing a mhag + multiple myeloma cell line. To date, two other autosomal HLA class II-restricted mhags are known: the ubiquitously expressed PI4K2BS-mHag presented by HLA-DQ6 and the hematopoietic restricted CD19 L - mhag presented by HLA-DQ2 11,13. Both mhags are suggested to be involved in GvT effects. On the other hand, because microarray data (biogps.gnf.org) indicate that SLC19A1 is expressed not only in hematopoietic cells but also in lung and liver, the mhag derived from SLC19A1 may also contribute to the development of GvHD after allo-sct.

58 58 Chapter 3 In a retrospective evaluation, we determined that the whole mhag identification procedure took only 3 months. It should be stressed that the whole procedure for the identification of the SLC19A1 R -mhag was extremely rapid because one of the SNPs identified by the correlation analysis encoded for the searched mhag. We think that such cases will be encountered frequently because HapMap includes 25% to 35% of common SNP variation in the human genome 14. However, even if the searched SNP is not identified directly, additional exploration may not take too long because the regions identified with this method usually contain only a few candidate genes 11,15. Thus, the strategy is indeed much faster than many other strategies described thus far and therefore likely to overcome the most important bottleneck toward efficient identification of clinically relevant mhags. Technically, in our strategy, it is important to realize that errors in mhag phenotyping will negatively influence the outcome of the analysis. Although our previous calculations indicate that 10% of phenotyping errors can be tolerated, the analyses will become more complicated, and therefore phenotyping should be done with great care. Still, the correlation analyses may fail in ~7% of the cases because HapMap estimates that their SNPs can currently capture ~93% of all SNPs with minor allele frequency of >0.05 within CEU populations 14. Furthermore, our approach may be less suitable for the identification of mhags that are only frequent in limited ethnical populations because our approach is based on the use of trios, and the International HapMap Project does not include trios for all ethnicities. With the rapid identification of mhags now possible, the remaining question is How many new mhags presented by which HLA molecules should be identified for widescale application of mhag-based immunotherapy? Several mhags identified thus far have little significant clinical value because they are presented by infrequent HLA-alleles 6,16. It seems easy to tackle this issue by focusing on frequent HLA-A and HLA-B alleles. However, focusing on both HLA types does not increase population coverage significantly because frequent HLA-A and HLA-B alleles are inherited together as common haplotypes 17,18. Therefore, focusing either on the most frequent HLA-A alleles or on the most frequent HLA-B alleles is a better strategy to make significant progress. HLA-A alleles seem a better choice for different populations. For instance, it is known that >92.5% of the European population expresses at least one of the HLA-A1, -A2, -A3, -A11, and -A24 alleles 17. In contrast, more than 10 HLA-B alleles are necessary to achieve a similar coverage in the same population. Similarly, for HLA class II-alleles, >95% of the population is covered by only three HLA-DP alleles (HLA-DPB1*0201, HLA-DPB1*0401, and HLA-DPB1*0402). We calculated that approximately 14 mhags per aforementioned HLA-A or HLA-DP allele are needed to establish at least one mhag-mismatch in >75% of sibling transplantations (Fig. 4A and B). This number decreases to eight in case of a matched unrelated donor (MUD) setting and only to a couple if MUDs could be a priori selected for a mhag-mismatch (Fig. 4A and B). Thus, broad application of

59 Rapid identification of minor histocompatibility antigens 59 A 100 HLA class I % of patients with eligible donor # of identified HLA-A1/A2/A3/A11/A24 restricted mhags 100 B % of patients with eligible donor HLA class II Sibling MUDdonor random MUDdonor a priori Chapter # of identified HLA-DP2/DP4.1/DP4.2 restricted mhags Figure 4. The coverage of the European population eligible for mhag-based immunotherapy. The histogram represents the coverage of European patients eligible for mhag-based immunotherapy (patient mhag + /donor mhag ) per five or six mhags each presented by respectively one of the most frequent (A) HLA-A alleles (HLA-A1, HLA-A2, HLA-A3, HLA-A11, HLA-A24) or (B) HLA-DP alleles (HLA-DPB1*0201, HLA-DPB1*0401, HLA-DPB1*0402). Coverage percentages are shown for sibling transplantation (dark gray), random MUD transplantation (dark gray plus light gray), or transplantation from an a priori MUD (dark gray plus light gray plus white). Calculations were based on three assumptions: (a) The identified mhags will have a population frequency of 10% to 85% (limits of our strategy; ref. 11). (b) Frequency distribution of mhags is similar to that of SNPs described by HapMap (phase I + II; ref. 17). (c) mhags are randomly identified and not linked to each other. The mhag-mismatch chances for sibling and MUD transplantations were calculated as described (11). HLA coverage for the European population was calculated using a published online tool (22). The dashed line is drawn at 75%. 60 mhag-based strategies will be possible with, in total, ~40 HLA-A or ~24 HLA-DP restricted mhags, and we believe that our identification strategy can significantly contribute to achieve this goal.

60 60 Chapter 3 Materials and methods Cells The HLA-DRB1*1501 restricted CD4 + T cell clone 1GF5 was previously isolated from a multiple myeloma patient 19. It was expanded using a feeder cell cytokine mixture as previously described 20. EBV-transformed lymphoblastoid cell lines (EBV-LCLs) and the Phoenix packaging line were cultured in RPMI 1640 and DMEM (Invitrogen), respectively, both supplemented with 10% fetal bovine serum (Integro) and antibiotics. T cell-mediated cytotoxicity assay Serial dilutions of effector T cell clone 1GF5 were incubated with luciferase-transduced multiple myeloma cell lines in the presence of 125 μg/ml beetle luciferin (Promega) in white opaque flat-bottomed 96-well plates (Costar). After 6h, the light signal emitted from surviving multiple myeloma cells was determined using a luminometer (Molecular Devices), and the percentage lysis was calculated compared with medium control (set to 0%) as described 21. Retroviral vectors and virus production The pmx-hla-drb1*1501-ires-gfp and the pmx-slc19a1 27R -IRES-GFP vectors were generated by cloning commercially synthesized genes, HLA-DRB1*1501 and SCL19A1 27R (GenScript), into the pmx-vector as described 20. Generation of retroviral supernatants and retroviral transductions were described elsewhere 20. SLC19A1-derived peptides Commercially synthesized and purified 15-mer peptides (Pepscan) were dissolved in DMSO to 100 mmol/l and diluted in PBS to 6 mmol/l for use in functional assays. mhag phenotyping of HapMap EBV-LCLs The phenotyping procedure has been described previously 11. In short, HLA-DRB1*1501- positive (naturally positive or positive after retroviral transduction) EBV-LCLs from HapMap individuals were used as antigen-presenting cells to stimulate T cell clone 1GF5. IFNγ release in supernatants was determined using ELISA (Invitrogen). EBV-LCLs were judged mhag + if the mean absorbance value at 450 nm of triplicate cultures was >0.250, about three times the background absorbance value, as described 11. Genome-wide zygosity-genotype correlation analysis mhag zygosities (+/+, +/-, or -/-) of HapMap individuals were deduced from the Mendelian segregation pattern of the mhag phenotypes in the father-mother-child trios as previously described 11. Zygosity-genotype correlation analysis was done using a modified version of the open source software sssnper 22 with embedded HapMap single-nucleotide polymorphism (SNP) genotypes for the CEU population (downloaded from 14,23. The complete Linux-based analysis package is available online for download ( nl/subsite/dcch/research/hemato-oncology). Information about the SLC19A1 gene and relevant SNPs was derived from Ensembl (

61 Rapid identification of minor histocompatibility antigens 61 References 1. den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Wang W, Meadows LR, den Haan JM et al. Human H-Y: a male-specific histocompatibility antigen derived from the SMCY protein. Science. 1995;269: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: Spaapen R, Mutis T. Targeting hematopoietic-specific minor histocompatibility antigens to distinguish graft-versustumor effects from graft-versus-host disease. Best Pract Res Clin Haematol. 2008;21: Akatsuka Y, Nishida T, Kondo E et al. Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med. 2003;197: Rijke BD, van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: Gubarev MI, Jenkin JC, Leppert MF et al. Localization to chromosome 22 of a gene encoding a human minor histocompatibility antigen. J Immunol. 1996;157: Gubarev MI, Jenkin JC, Otterrud BE et al. Localization to chromosome 11 of a gene encoding a human minor histocompatibility antigen. Exp Hematol. 1998;26: Spaapen RM, Lokhorst HM, van den Oudenalder K et al. 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. 2008;205: Huang Y. Pharmacogenetics/genomics of membrane transporters in cancer chemotherapy. Cancer Metastasis Rev. 2007;26: Griffioen M, van der Meijden ED, Slager EH et al. Identification of phosphatidylinositol 4-kinase type II beta as HLA class II-restricted target in graft versus leukemia reactivity. Proc Natl Acad Sci U S A. 2008;105: Frazer KA, Ballinger DG, Cox DR et al. A second generation human haplotype map of over 3.1 million SNPs. Nature. 2007;449: Kawase T, Nannya Y, Torikai H et al. Identification of human minor histocompatibility antigens based on genetic association with highly parallel genotyping of pooled DNA. Blood. 2008;111: Spierings E, Hendriks M, Absi L et al. Phenotype frequencies of autosomal minor histocompatibility antigens display significant differences among populations. PLoS Genet. 2007;3:e Bui HH, Sidney J, Dinh K et al. Predicting population coverage of T cell epitope-based diagnostics and vaccines. BMC Bioinformatics. 2006;7: Schipper RF, van Els CA, D Amaro J, Oudshoorn M. Minimal phenotype panels. A method for achieving maximum population coverage with a minimum of HLA antigens. Hum Immunol. 1996;51: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Spaapen R, van den Oudenalder K, Ivanov R et al. Rebuilding Human Leukocyte Antigen Class II-Restricted Minor Histocompatibility Antigen Specificity in Recall Antigen-Specific T Cells by Adoptive T Cell Receptor Transfer: Implications for Adoptive Immunotherapy. Clin Cancer Res. 2007;13: Brown CE, Wright CL, Naranjo A et al. Biophotonic cytotoxicity assay for high-throughput screening of cytolytic killing. J Immunol Methods. 2005;297: Nyholt DR. sssnper: identifying statistically similar SNPs to aid interpretation of genetic association studies. Bioinformatics. 2006;22: The International HapMap Consortium. The International HapMap Project. Nature. 2003;426: Chapter 3

62 62 Chapter 3

63 4 Genome-wide prediction of minor histocompatibility antigens based on differential HLA-binding of polymorphic peptides Robbert M. Spaapen 1, David Gestel 3, Maureen van Elk 1, Henk M. Lokhorst 2, Can Keşmir 3, Tuna Mutis 1 Departments of 1 Clinical Chemistry and Haematology, 2 Haematology, UMCU, NL 3 Department of Theoretical Biology and Bioinformatics, UU, NL Submitted

64 64 Chapter 4 The curative graft-versus-tumor (GvT) effect after allogeneic stem cell transplantation is mainly mediated by donor-derived T cells recognizing minor histocompatibility antigens (mhags) presented by malignant cells. Conceptually, induction of specific GvT is possible by targeting of donor T cells toward mhags expressed solely in the hematopoietic cells of the recipient. Hence, broad applicability of mhag-based immunotherapy is dependent on the identification of a large array of clinically relevant mhags that display balanced population frequency and are presented by a specific set of common HLA molecules. We here propose a mhag-specific reverse immunology strategy, which offers a systematic approach to identify such mhags. Unique in our strategy is the a priori selection of all polymorphisms that display phenotype frequency between 26-78%, and the use of a mhag-specific algorithm, which analyses polymorphic peptides for differential HLAbinding, since differential surface expression appears a common feature of mhags. The selected peptides are further screened for proper proteasomal cleavage and efficient TAP translocation. We developed an algorithm to perform semi-automated predictions and comparisons of the polymorphic peptides. Genome-wide application of this strategy revealed 941 putative mhags that can be presented by HLA-A1, -A2, -A3, -A11, or -A24, the only five HLA class I molecules offering a population coverage of more than 92.5%. Using publicly available micro array datasets, we determined that 124 of these putative mhags display a hematopoietic restricted expression pattern, while the others are expressed in at least one non-hematopoietic tissue. The set of hematopoietic restricted mhags represents the most probable candidate list for mhag-based immunotherapy. INTRODUCTION Allogeneic stem cell transplantation (allo-sct) followed by antigen-specific cellular immunotherapy is an appealing approach to achieve long term remissions in hematological malignancies. Ideal target antigens for such strategies are hematopoietic restricted minor histocompatibility antigens (mhags). These are HLA-bound polymorphic peptides derived from hematopoietic genes and are presented by normal and malignant cells of the patient to the donor-derived T cells 1-4. Identification of a large array of hematopoietic mhags is therefore a high priority for the broad clinical applicability of mhag-based immunotherapy. To date, most mhags have been identified by the so called forward immunology approaches, such as peptide elution, cdna library screening or genetic linkage analyses, which all attempt to identify the target antigen of mhag-specific T cell clones isolated from

65 Genome wide mhag predictions 65 allo-sct patients Although these T cell-to-antigen based strategies have been highly successful until now, their general drawback is the lack of control on HLA restriction, frequency and tissue distribution of mhags recognized by the generated T cells. A better control of these parameters is a key aspect for future identification of clinically relevant mhags, which are defined by hematopoietic restricted expression, a balanced population frequency and presentation by a specific set of HLA-alleles 12. While forward immunology hampers at this point, opportunities are offered by the reverse immunology approach. Thanks to the recent advances in human genomics, this gene-to-t cell approach allows a priori selection of genes containing nonsynonymous coding single nucleotide polymorphisms (nssnps) with a balanced frequency Similarly, bioinformatics tools are now powerful enough to be specifically employed for the selection of peptides binding to HLA antigens of choice 16. Furthermore, hematopoietic mhags can be separated from candidate mhags based on genome-wide gene expression arrays in a broad number of tissues. Indeed, the reverse immunology has been proven to be a powerful method for the identification of HLA class I-restricted tumor associated antigens in the last decade 17,18. Nonetheless, classical reverse immunology strategies, which do not include mhag-specific features, seem not very effective for the identification of mhags, because only a few mhags have been identified by reverse immunology until now To improve the reverse immunology algorithms for the prediction of mhags, we here evaluated the specific features of known mhags and observed a common, and probably an important feature of mhags. This feature is the differential surface expression of mhags compared to their non-immunogenic allelic counterparts, which has been found in 12 out of 16 adequately studied mhags until now. Since differential HLA-binding of the allelic peptides contributes to this phenomenon, we developed a semi-automated procedure to search the whole genome for allelic peptides that display differential HLA binding. This semi-automated procedure uses the well known and powerful prediction programs NetMHC and NetCHOP and TAP affinity matrices to predict HLA-binding, proteasome mediated cleavage and TAP translocation respectively 16, Our genome-wide analyses revealed 941 polymorphic peptides that display a balanced population frequency and have differential binding affinity for HLA-A1, -A2, -A3, -A11 and -A24 molecules, which all together cover more than 92.5% of the Caucasoid population. 124 of these antigens display a hematopoietic restricted expression, representing the most probable candidates for mhag-based immunotherapy. Chapter 4

66 66 Chapter 4 RESULTS Known HLA class I mhags are often generated by differential antigen processing To improve the classical reverse immunology strategies with mhag-specific features, we first made a literature- and prediction-based inventory of all 26 currently known autosomal mhags with respect to their biochemical and immunologic properties (Table 1). There were no common features with respect to the gene origin or peptide sequences. However, it was remarkably apparent that merely two mhags (BCL2A1 and HB-1) were expressed in a bidirectional way, i.e. both allelic peptides were expressed on the cell surface as immunogenic mhags 5,8,9,19. In contrast, twelve (plus predictions for another four) mhags displayed a differential surface expression, meaning that the immunogenic peptide, but not the allelic peptide was expressed on the cell surface. This was caused by several mechanisms such as differential transcription 10,25 or translation 6,28 of polymorphic genes, differential proteasomemediated cleavage 30,31, differential TAP translocation 32 or differential HLA binding/ stability of the allelic peptides 40,41. Strikingly, this feature appeared also crucial for the existence of at least three mhags, HA-2, HA-3 and HA-8, since T cells specific for these mhags were also able to recognize synthetic peptides derived from the non-immunogenic alleles 30,32,35. For the remaining 6 mhags the differential antigen processing or bidirectional recognition were not determined. Thus, differential surface expression of allelic peptides is a frequently encountered and probably important property of mhags, since T cells are not always able to discriminate between single amino acid differences. This prompted us to include this common property of mhags in the reverse immunology strategies. A new mhag prediction algorithm based on differential HLA binding Since not all mechanisms involved in the differential surface expression of antigens could currently be combined in an integrated prediction strategy, we developed a mhag-specific algorithm by taking only the differential HLA-binding into account. Thus, we developed an semi-automated procedure which compares the protein sequences of polymorphic genes for the presence of nonameric peptides displaying differential affinity for HLA-A1, -A2, -A3, -A11 and -A24 according to artificial neural networks from the NetMHC 3.0 Server 16,22. The choice of these 5 HLA antigens was based on our observations that >92.5% of the Caucasoid population expresses at least one of these HLA antigens 12. Only those peptide pairs were selected for which the predicted IC 50 binding value of one allele was <500 nm (is log-score >0.426) AND the other allele >1000 nm (is log-score <0.362) (Fig. 1). Subsequently, the algorithm subjected all selected HLA-binding candidate mhag-peptides to C-terminal proteasomal cleavage predictions using the neural network trained NetChop 3.0, discarding peptides with a C-terminal cleavage < For selection ofthe remaining peptides, weight matrix based predictions for TAP transport efficiency were used, retaining only peptides with a score >

67 Genome wide mhag predictions 67 Table 1. Current known mhags are often differentially intracellular processed Levels of differential peptide processing Yet unknown HLA-binding affinity Translation Transcription C-terminal or internal proteasome cleavage TAP transport TCR affinity mhag name HLA restriction a Peptide sequence b C-term prot. score b TAP score bc HLA binding score (log) b Source ACC6 B*4402/03 MEIFIEVFSHF 0,96-0,39 ref UGT2B17 A*0206 CVATMIFMI 0,10 0,80 0,46 ref UGT2B17 A*2902 AELLNIPFLY 0,97-0,20 ref UGT2B17 B*4403 AELLNIPFLY 0,97-0,56 ref LRH-1 B*0702 TPNQRQNVC 0,13-0,36 0,46 ref PANE1 A*0301 RVWDLPGVLK 0,56-0,71 ref ACC4 A*3303 WATLPLLCAR 0,27-0,30 ref 29 WATLPLLCAG 0,04-0,05 ACC5 A*3101 ATLPLLCAR 0,27 1,57 0,76 ref 29 ATLPLLCAG 0,04-1,31 0,06 HA-1 B*4001 KECVLHDDL 0,49 0,98 0,56 ref 20 KECVLRDDL 0,18 0,98 0,52 HA-3 d A*0101 VTEPGTAQY 0,98 2,87 0,59 ref 30 VMEPGTAQY 0,98 3,01 0,33 SP110 d A*0301 SLPRGTSTPK - - 0,67 ref 31 SLPGGTSTPK HA-8 A*0201 RTLDKVLEV 0,97 0,53 0,68 ref 32 PTLDKVLEV 0,97-0,19 0,26 HA-1 A*0201 VLHDDLLEA 0,96-0,43 0,75 ref 33 VLRDDLLEA 0,94-0,41 0,61 HA-1 A*0206 VLHDDLLEA 0,96-0,43 0,63 ref 34 VLRDDLLEA 0,94-0,41 0,53 HA-2 A*0201 YIGEVLVSV 0,97 0,27 0,78 ref 7, 35 YIGEVLVSM 0,97 0,26 0,63 TRIM22 A*0201 MAVPPCCIGV 0,96-0,33 ref 36 MAVPPCRIGV 0,90-0,21 ACC1C A*2402 DYLQYVLQI 0,91 0,45 0,74 ref 9 ACC1Y A*2402 DYLQCVLQI 0,77 0,45 0,75 ref 5 HB-1H B*4402/03 EEKRGSLHVW 0,90-0,36 ref 8 HB-1Y B*4402/03 EEKRGSLYVW 0,89-0,37 ref 19 ACC2 B*4403 KEFEDDIINW 0,41-0,54 ref 5 KEFEDGIINW 0,63-0,52 C19ORF48 A*0201 CIPPDSLLFPA 0,81-0,34 ref 37 CIPPDTLLFPA 0,86-0,38 HER-2/NEU A*0201 ATLERPKTL 0,97 1,07 0,15 ref 21 ATLERAKTL 0,96 1,07 0,13 LB-ADIR-1F A*0201 SVAPALALSPA 0,89-0,39 ref 38 SVAPALALFPA 0,91-0,53 LB-ECGF-1 B*0702 RPHAIRRPLAL 0,92-0,81 ref 39 RPRAIRRPLAL 0,92-0,85 SLC1A5 B*4002 AEATANGGLAL 0,96-0,40 ref 26 AEPTANGGLAL 0,96-0,35 a For B*4402/03 binding mhags, only scores for B*4403 are depicted b Bold fields indicate a difference between the two alleles c TAP scores were calculated only for 9-mers d HA-3 and SP110 are generated through differential internal proteasome cleavage Chapter 4

68 68 Chapter 4 Differential mhag prediction and ranking Since our semi-automated procedure allowed for high-throughput analyses, we set out to perform a genome-wide analysis using the SNP and transcript data from public databases. These databases contain almost 16 million SNPs throughout the whole human genome, of which almost 100,000 are nssnps (Fig. 1). We extracted only 15,872,232 SNPs (Ensembl) Non-synonymous coding (Ensembl) 97,571 SNPs MAF > 0.1 (covering mhag freq of 19-99%) (HapMap) 9,692 SNPs Transcript mining (Ensembl) All possible SNP-containing nonamer-pairs (algorithm) 17,952 transcripts 161,568 nonamer-pairs (SNP generates two alleles) HLA-binding -A1/-A2/-A3/-A11/-A24 (algorithm/netmhc) Binding: log-score > Non-binding: log-score < Nonamers binding to HLA none one one both C-terminal prot. cleavage (algorithm/netchop) TAP translocation (algorithm/matrix) Frequency check (manually) Score > 0.5 Score > % > mhag freq > 78% Putative mhags: HLA-A*0101: 47 HLA-A*0201: 293 HLA-A*0301: 183 HLA-A*1101: 282 HLA-A*2402: 136 Tissue distribution (micro-array data) Discard all genes that are non-hematopoietic 124 putative genuine hematopoietic mhags Figure 1. A schematic overview of genome-wide mhag predictions based on differential HLA-binding of polymorphic peptides. The exact numbers of nssnps, transcripts, peptide pairs and predicted mhags are shown. Furthermore the selection procedure is depicted including cut-off values and a schematic view of the HLA-binding selection, retaining peptide pairs for which only one of the allelic peptides was predicted to bind to HLA.

69 Genome wide mhag predictions 69 those nssnps with a minor allele frequency (MAF) of 0.1 in the CEU population 13,15, thereby covering all putative mhags with a population frequency between 19-99% as calculated based on Mendelian p 2 +2pq+q 2 distribution. For each of these nssnps, we downloaded the corresponding transcript(s), totalling 17,952 transcripts that were used for input in the algorithm (Fig. 1). All output peptides were subjected to an extra allele frequency selection retaining only putative mhag-peptides, for which the allele of the nssnp that was not encoding for the mhag had a frequency between , corresponding to clinically relevant mhag population frequencies of 26%-78% 42. Finally the total number of putative mhags predicted to be presented by HLA-A1, -A2, -A3, -A11 and -A24 were respectively 47, 293, 183, 282 and 136 (Fig. 1). We ranked these peptides according to their NetCTL scores, which is a combined score for HLA-binding, proteasome cleavage and TAP transport 43. Identification of the putative mhags with strict hematopoietic expression Clinical application of a mhag is not only dependent on its population frequency and HLA restriction, but also on its expression in different tissues. Therefore we extracted tissue distribution information of the genes encoding for the putative mhags from a publicly available comprehensive gene expression database 44,45. We grouped the tissues in the dataset in four relevant categories (normal hematopoietic cells; hematopoietic-derived tumor cells; GvHD organs; non-gvhd organs) and validated the tissue expression profiling approach using currently known mhags and using genes well known to be expressed by either the hematopoietic or nonhematopoietic system (Table S1). Thus next, we analyzed the expression profiles of putative mhags with available probes in the dataset, in total 418 of the 941. We categorized the putative mhags based on their profile and determined that a vast number of putative mhags (124) had a strictly hematopoietic expression profile as visualized in a heat map (Table 2). However, towards revealing the clinical relevancy of all of these hematopoietic restricted mhags, further exploration of the genes expressed by normal hematopoietic cells but seemingly not in the tumors is necessary, since only a limited number of tumor cell lines and no primary tumors were included in the dataset. Because proper tissue expression data for 523 of the putative mhags was lacking, there may be many more clinically relevant mhags within our total predicted dataset. Chapter 4 DISCUSSION We here provide a new reverse prediction strategy using a mhag-specific algorithm based on actual mhag processing properties. By genome-wide application of this algorithm on a specific set of polymorphisms with population frequencies between 26%-78%, we identified 941 putative mhags, presented by 5 different HLA molecules covering >92.5% of the Caucasoid population. Our tissue distribution analyses reveal that at least 124 of these putative mhags may be associated with GvT.

70 70 Chapter 4 Table 2. Putative hematopoietic mhags derived from a genome-wide reverse analysis Antigen processing predictions for both allelic peptides Tissue expression Amino acid Putative mhag Allele predicted to bind MHC Allele predicted not to bind MHC Hematopoietic Non-hematopoietic polymorphism frequency NetCHOP TAP NetMHC NetCHOP TAP NetMHC Normal Tumor a GvHD Non-GvHD Ensembl gene Gene symbol Chrom nssnp Peptide CTLscore Peptide CTLscore HLA-A1 predictions ENSG PKHD1L1 8 rs S R 73,5 GSSEATYAY 0,98 1,23 0,58 2,65 GRSEATYAY 0,98 1,30 0,09 0,58 7,14 7,01 1,15 1,13 ENSG CCDC99 5 rs Y H 47,8 QLEKSVSIY 0,83 1,24 0,51 2,37 QLEKSVSIH 0,06-0,26 0,10 0,41 3,77 10,32 1,09 2,31 ENSG IFI44L 1 rs T I 38,9 DTTSITERY 0,96 1,13 0,51 2,35 DITSITERY 0,96 1,20 0,17 0,93 21,70 84,29 1,02 2,75 ENSG DNHD1 11 rs V E 65,5 SSDTDNAEV 0,95 0,03 0,46 2,10 SSDTDNAEE 0,05-0,79 0,20 0,80 1,07 5,81 1,14 1,15 HLA-A2 predictions ENSG CEP110 9 rs L P 61,9 KLLQDLISL 0,96 0,50 0,82 1,39 KPLQDLISL 0,96 0,36 0,10 0,31 11,84 6,18 1,03 1,30 ENSG NIACR2 12 rs L F 41,7 FLLPLGIIL 0,82 0,32 0,82 1,36 FFLPLGIIL 0,60 0,38 0,29 0,54 247,63 3,43 3,23 1,68 ENSG BPI 20 rs K E 73,5 KLQPYFQTL 0,97 0,48 0,75 1,29 ELQPYFQTL 0,97 0,35 0,34 0,67 105,08 1,49 1,70 1,25 ENSG PPIL5 14 rs W R 42,0 TLQKSLWSL 0,97 0,43 0,74 1,28 TLQKSLRSL 0,95 0,43 0,34 0,67 12,36 17,60 1,14 1,31 ENSG CEP110 9 rs L P 61,9 LLQDLISLI 0,50 0,27 0,79 1,26 PLQDLISLI 0,49 0,06 0,36 0,61 11,84 6,18 1,03 1,30 ENSG TUBGCP6 22 rs L S 44,2 KLYWTHGYV 0,96 0,21 0,68 1,17 KSYWTHGYV 0,92 0,22 0,36 0,68 13,76 1,51 1,10 1,10 ENSG ANKZF1 2 rs L P 25,7 SLIPDSAIV 0,96 0,20 0,67 1,15 SPIPDSAIV 0,97 0,06 0,05 0,23 12,26 6,84 1,04 1,25 ENSG PIH1D1 19 rs7462 L P 48,7 LMVAMPLLL 0,80 0,48 0,66 1,13 LMVAMPLLP 0,03 0,11 0,16 0,25 2,68 4,85 2,70 2,52 ENSG CYP20A1 2 rs L F 72,7 KLTPVSAQL 0,98 0,52 0,59 1,06 KLTPVSAQF 0,97 1,21 0,14 0,42 8,07 1,75 1,06 1,89 ENSG HEATR1 1 rs V A 55,0 SILQEHIGV 0,85 0,24 0,61 1,05 SILQEHIGA 0,49-0,13 0,32 0,54 3,64 9,50 1,59 1,74 ENSG ZNF rs K E 32,7 KVQEKVHSV 0,96 0,25 0,60 1,05 EVQEKVHSV 0,96 0,12 0,20 0,44 9,95 1,89 1,25 2,02 ENSG CNOT3 19 rs42318 L S 48,7 SLAHSQASL 0,98 0,45 0,59 1,04 SSAHSQASL 0,97 0,46 0,14 0,37 5,24 9,94 3,00 2,81 ENSG CARD6 5 rs L S 25,7 CLFSTFPQL 0,93 0,50 0,59 1,04 CLFSTFPQS 0,06-0,89 0,24 0,32 4,37 0,98 2,00 2,26 ENSG NLRC5 16 rs L P 26,8 LLTSSLLDL 0,92 0,40 0,58 1,03 LPTSSLLDL 0,95 0,26 0,02 0,18 97,08 17,05 1,52 2,35 ENSG C2orf24 2 rs L P 65,5 SLANVSSCL 0,98 0,39 0,58 1,02 SPANVSSCL 0,98 0,25 0,03 0,20 13,24 7,28 2,25 2,21 ENSG OSBPL10 3 rs N D 76,1 NLVHAIESL 0,96 0,52 0,56 1,01 DLVHAIESL 0,96 0,37 0,22 0,49 18,85 5,70 2,82 2,21 ENSG TIMELESS 12 rs L P 73,3 LLLWLQNCL 0,92 0,44 0,57 1,00 PLLWLQNCL 0,92 0,23 0,20 0,45 7,30 11,22 1,05 1,34 ENSG TBRG4 7 rs L P 31,0 HLPGSLMEL 0,98 0,40 0,54 0,97 HLPGSLMEP 0,24 0,03 0,11 0,21 2,51 8,76 1,78 2,67 ENSG TRAF3IP3 1 rs Q E 48,3 QLQAKIECL 0,97 0,40 0,52 0,94 ELQAKIECL 0,97 0,30 0,26 0,54 142,85 16,10 1,41 2,97 ENSG USP18 22 rs M T 59,8 YMIRVKDSL 0,61 0,51 0,55 0,94 YTIRVKDSL 0,49 0,45 0,18 0,36 1,81 12,70 1,05 1,41 ENSG NLRP11 19 rs L P 56,6 LQAANILLL 0,93 0,48 0,52 0,93 LQAANILLP 0,04 0,11 0,10 0,15 1,05 14,54 1,14 2,49 ENSG CRNKL1 20 rs L F 78,4 LVLAWSSQL 0,96 0,42 0,51 0,92 LVLAWSSQF 0,92 1,11 0,07 0,30 5,14 2,88 1,44 3,27 ENSG CEP rs M V 28,3 GMTSGLGSV 0,95 0,12 0,52 0,92 GVTSGLGSV 0,94 0,13 0,24 0,51 7,14 6,55 1,19 1,40 ENSG CCDC99 5 rs Y H 47,8 SIYTPVVSL 0,98 0,60 0,48 0,88 SIHTPVVSL 0,98 0,53 0,30 0,62 3,77 10,32 1,09 2,31 ENSG PIGM 1 rs L F 40,7 WLIGQAMWL 0,52 0,46 0,51 0,87 WFIGQAMWL 0,33 0,53 0,07 0,18 3,87 3,83 1,12 1,21 ENSG C2orf24 2 rs L P 65,5 SVPQCLPSL 0,83 0,43 0,48 0,86 SVPQCLPSP 0,03 0,06 0,06 0,10 13,24 7,28 2,25 2,21 ENSG PARP9 3 rs Y C 67,3 EMYEAHAWI 0,81 0,20 0,47 0,84 EMCEAHAWI 0,38 0,13 0,23 0,40 19,27 11,81 2,00 2,25 ENSG SLC22A4 5 rs F L 69,0 TVLIGIFTL 0,98 0,50 0,44 0,83 TVLIGILTL 0,98 0,50 0,32 0,66 14,30 1,28 1,33 1,95 ENSG NLRC5 16 rs L P 26,8 ALSPPGHLL 0,98 0,51 0,44 0,83 ALSPPGHLP 0,49 0,13 0,05 0,15 97,08 17,05 1,52 2,35 ENSG BLMH 17 rs V I 55,8 AVLEQEPIV 0,71 0,29 0,48 0,83 AVLEQEPII 0,52 0,38 0,34 0,60 5,09 7,25 2,26 2,33 ENSG SFI1 22 rs Q R 69,0 ILHAEQQLL 0,79 0,51 0,45 0,82 ILHAERQLL 0,65 0,51 0,30 0,56 44,00 8,40 1,71 2,19 ENSG ATR 3 rs Q R 26,5 KLKVFQEFL 0,92 0,40 0,43 0,80 KLKVFREFL 0,84 0,40 0,29 0,57 8,87 4,75 1,27 2,14 ENSG NOM1 7 rs12919 M V 75,9 GMLREGLKL 0,92 0,42 0,43 0,80 GVLREGLKL 0,90 0,43 0,15 0,39 4,03 2,94 1,25 1,33

71 Genome wide mhag predictions 71 HLA-A3 predictions ENSG BPI 20 rs K E 73,5 KVTNSVSSK 0,90 0,22 0,72 1,51 KVTNSVSSE 0,04-0,66 0,11 0,18 105,08 1,49 1,70 1,25 ENSG LAMA1 18 rs K E 57,5 LLILRAIPK 0,83 0,27 0,68 1,41 LLILRAIPE 0,03-0,61 0,08 0,12 1,03 49,08 1,15 1,74 ENSG BCL2A1 15 rs Y C 54,5 RLAQDYLQY 0,96 1,37 0,62 1,37 RLAQDYLQC 0,03 0,11 0,07 0,15 91,52 65,83 1,87 1,96 ENSG EMR1 19 rs K Q 39,3 SVLNERFFK 0,50 0,22 0,66 1,32 SVLNERFFQ 0,03-0,03 0,14 0,26 31,51 12,24 1,10 1,35 ENSG C19orf60 19 rs K E 74,3 GTGSSPSRK 0,85 0,10 0,62 1,31 GTGSSPSRE 0,03-0,78 0,03 0,03 4,93 2,14 1,97 2,64 ENSG SEC31B 10 rs S P 34,8 ASLIPPLKK 0,79 0,27 0,61 1,27 APLIPPLKK 0,94 0,12 0,23 0,58 4,18 1,31 1,15 1,87 ENSG PPP1R2P4 13 rs R G 51,3 RVSLPTSPR 0,91 0,80 0,57 1,25 RVSLPTSPG 0,26-0,45 0,09 0,19 132,69 68,38 1,10 1,86 ENSG SYTL3 6 rs L Q 35,0 SLSKLQWQK 0,89 0,23 0,58 1,24 SQSKLQWQK 0,66 0,25 0,32 0,71 29,19 1,06 1,13 1,22 ENSG NEIL3 4 rs Q H 33,9 AQYSSPELK 0,79 0,38 0,58 1,23 AHYSSPELK 0,86 0,35 0,24 0,59 1,48 10,51 1,08 1,47 ENSG VWDE 7 rs K Q 69,7 KVETQFVNK 0,80 0,27 0,57 1,20 KVETQFVNQ 0,06 0,02 0,12 0,24 1,05 5,84 1,27 2,30 ENSG SIGLEC1 20 rs K R 66,4 HLQVKYAPK 0,67 0,16 0,57 1,18 HLQVKYAPR 0,37 0,60 0,30 0,64 4,15 1,51 2,60 1,30 ENSG RABEPK 9 rs K R 25,7 GVGPSRVQK 0,91 0,17 0,54 1,16 GVGPSRVQR 0,70 0,61 0,28 0,67 1,57 3,96 3,08 2,93 ENSG PCSK5 9 rs Q E 58,4 GTYYEKQTK 0,95 0,26 0,53 1,15 GTYYEKETK 0,92 0,26 0,34 0,80 5,76 1,26 1,18 1,96 ENSG PIGN 18 rs H D 41,7 RDYKHNIKK 0,75 0,28 0,54 1,15 RDYKDNIKK 0,77 0,28 0,35 0,80 3,80 3,61 1,11 1,58 ENSG CHID1 11 rs6682 V A 75,9 QVSEHFFEY 0,98 1,36 0,49 1,13 QASEHFFEY 0,98 1,32 0,25 0,68 3,91 3,95 3,48 2,23 ENSG C5orf45 5 rs10277 Q R 72,6 AQFVLPPRK 0,92 0,32 0,48 1,05 ARFVLPPRK 0,97 0,38 0,08 0,32 4,70 2,15 1,20 2,42 ENSG ZNF rs K E 32,7 NSFSKVQEK 0,79 0,26 0,49 1,05 NSFSEVQEK 0,70 0,26 0,35 0,77 9,95 1,89 1,25 2,02 ENSG MTRR 5 rs10380 Y H 29,5 YLFRKELRY 0,97 1,26 0,44 1,05 YLFRKELRH 0,51-0,24 0,30 0,62 7,38 4,09 1,40 3,21 ENSG SP110 2 rs M T 70,8 MTLGELLKR 0,91 0,65 0,46 1,04 TTLGELLKR 0,90 0,62 0,35 0,83 78,56 18,39 2,57 1,87 ENSG C11orf82 11 rs S R 61,9 SSLSVTPQR 0,90 0,68 0,45 1,03 SRLSVTPQR 0,96 0,75 0,06 0,29 10,80 4,48 1,59 2,31 ENSG APOL1 22 rs M I 38,1 GITSSTMDY 0,95 1,24 0,43 1,02 GITSSTIDY 0,95 1,24 0,35 0,86 2,43 4,65 2,31 2,25 ENSG C2orf3 2 rs S N 72,6 GSGSSKVKK 0,78 0,07 0,48 1,02 GNGSSKVKK 0,61-0,02 0,14 0,35 3,24 6,05 1,00 1,71 ENSG ERAP1 5 rs27044 Q E 43,3 RMLRSQLLL 0,79 0,51 0,45 0,98 RMLRSELLL 0,66 0,51 0,33 0,74 9,54 3,39 1,20 2,39 ENSG TBC1D8 2 rs R G 31,2 STSRPLVFR 0,59 0,68 0,46 0,98 STSRPLVFG 0,08-0,57 0,05 0,08 7,60 1,00 1,32 3,32 ENSG VISA 20 rs K Q 26,5 LTSSGHQEK 0,97 0,13 0,43 0,97 LTSSGHQEQ 0,61-0,12 0,04 0,16 3,23 3,76 1,11 1,46 ENSG CALCOCO2 17 rs10278 A P 51,7 SASPLSIKK 0,94 0,27 0,43 0,96 SPSPLSIKK 0,97 0,11 0,18 0,49 9,46 5,80 1,29 2,78 ENSG TTLL4 2 rs Q E 60,3 TVPATPPQK 0,80 0,23 0,44 0,95 TVPATPPEK 0,83 0,23 0,33 0,76 2,27 6,35 0,94 1,41 ENSG FASTKD3 5 rs K R 29,2 RQPEPFGVK 0,81 0,36 0,43 0,95 RQPEPFGVR 0,55 0,80 0,20 0,51 3,22 6,88 1,67 2,03 ENSG HICE1 19 rs R G 55,0 KADSSGVRK 0,59 0,22 0,45 0,94 KADSSGVGK 0,36 0,22 0,35 0,72 1,05 6,46 1,13 1,13 ENSG NLRC5 16 rs K Q 39,8 AAKFLGFRK 0,82 0,35 0,43 0,94 AAKFLGFRQ 0,12 0,10 0,05 0,12 97,08 17,05 1,52 2,35 ENSG KIAA rs R L 36,3 NMHNLLIHR 0,67 0,75 0,43 0,94 NMHNLLIHL 0,95 0,52 0,07 0,30 12,74 1,05 1,16 1,18 HLA-A11 predictions ENSG C5orf45 5 rs10277 Q R 72,6 AQFVLPPRK 0,92 0,32 0,80 1,34 ARFVLPPRK 0,97 0,38 0,33 0,66 4,70 2,15 1,20 2,42 ENSG TBC1D8 2 rs R G 31,2 STSRPLVFR 0,59 0,68 0,80 1,31 STSRPLVFG 0,08-0,57 0,16 0,22 7,60 1,00 1,32 3,32 ENSG BPI 20 rs K E 73,5 KVTNSVSSK 0,90 0,22 0,78 1,31 KVTNSVSSE 0,04-0,66 0,12 0,15 105,08 1,49 1,70 1,25 ENSG C11orf82 11 rs S R 61,9 SSLSVTPQR 0,90 0,68 0,73 1,25 SRLSVTPQR 0,96 0,75 0,14 0,39 10,80 4,48 1,59 2,31 ENSG NLRC5 16 rs K Q 39,8 AAKFLGFRK 0,82 0,35 0,74 1,25 AAKFLGFRQ 0,12 0,10 0,16 0,26 97,08 17,05 1,52 2,35 ENSG PPP1R2P4 13 rs R G 51,3 RVSLPTSPR 0,91 0,80 0,70 1,21 RVSLPTSPG 0,26-0,45 0,09 0,14 132,69 68,38 1,10 1,86 ENSG VWDE 7 rs K Q 69,7 KVETQFVNK 0,80 0,27 0,72 1,20 KVETQFVNQ 0,06 0,02 0,17 0,27 1,05 5,84 1,27 2,30 ENSG C19orf60 19 rs K E 74,3 GTGSSPSRK 0,85 0,10 0,72 1,20 GTGSSPSRE 0,03-0,78 0,08 0,08 4,93 2,14 1,97 2,64 ENSG C14orf rs S N 53,1 SIYNWTEEY 0,97 1,47 0,64 1,17 NIYNWTEEY 0,97 1,47 0,35 0,75 6,62 2,12 0,95 3,00 ENSG LAMA1 18 rs K E 57,5 LLILRAIPK 0,83 0,27 0,68 1,16 LLILRAIPE 0,03-0,61 0,07 0,08 1,03 49,08 1,15 1,74 ENSG ARHGAP25 2 rs S R 38,3 QSMVQELRK 0,50 0,27 0,68 1,11 QRMVQELRK 0,83 0,35 0,15 0,36 237,53 4,85 1,49 1,40 ENSG VISA 20 rs K Q 26,5 LTSSGHQEK 0,97 0,13 0,64 1,10 LTSSGHQEQ 0,61-0,12 0,08 0,21 3,23 3,76 1,11 1,46 ENSG HYLS1 11 rs R C 41,6 AATAFTHIR 0,95 0,78 0,61 1,09 AATAFTHIC 0,14 0,14 0,05 0,10 3,38 40,04 1,09 1,63 ENSG ZNF rs K E 32,7 NIVQNSFSK 0,82 0,35 0,64 1,09 NIVQNSFSE 0,03-0,53 0,05 0,06 9,95 1,89 1,25 2,02 ENSG ZFYVE26 14 rs Y C 63,7 AVLSCAVAY 0,76 1,34 0,58 1,05 AVLSCAVAC 0,02 0,08 0,07 0,12 11,49 2,02 1,37 2,25 Chapter 4

72 72 Chapter 4 ENSG IMMT 2 rs S P 67,8 SASQLQKQK 0,73 0,22 0,59 0,99 PASQLQKQK 0,61-0,01 0,21 0,40 4,26 8,66 2,60 2,16 ENSG HPSE 4 rs K R 39,6 YLNGRTATK 0,98 0,14 0,54 0,95 YLNGRTATR 0,97 0,58 0,26 0,56 23,34 1,35 1,31 1,34 ENSG ATXN7 3 rs M V 54,0 PAMNNVHMK 0,95 0,07 0,53 0,94 PAVNNVHMK 0,96 0,07 0,34 0,66 19,79 1,96 1,14 2,35 ENSG TELO2 16 rs Q R 65,1 TQPGFILLR 0,91 0,64 0,51 0,93 TRPGFILLR 0,97 0,70 0,12 0,36 4,49 1,67 1,12 2,37 ENSG C2orf3 2 rs S N 72,6 GSGSSKVKK 0,78 0,07 0,54 0,92 GNGSSKVKK 0,61-0,02 0,13 0,29 3,24 6,05 1,00 1,71 ENSG LLGL2 17 rs R P 47,3 AATGVHIER 0,70 0,77 0,51 0,90 AATGVHIEP 0,05 0,17 0,05 0,10 11,81 1,43 2,93 2,24 ENSG STEAP4 7 rs G D 45,0 GIIIIAIHR 0,58 0,68 0,52 0,90 DIIIIAIHR 0,51 0,62 0,23 0,45 5,24 1,33 1,18 2,32 ENSG CCDC rs W R 30,1 RLDKVRWKK 0,93 0,30 0,49 0,89 RLDKVRRKK 0,85 0,30 0,32 0,62 7,32 8,06 1,09 1,12 ENSG FASTKD3 5 rs K R 29,2 RQPEPFGVK 0,81 0,36 0,50 0,88 RQPEPFGVR 0,55 0,80 0,25 0,49 3,22 6,88 1,67 2,03 ENSG MCTP2 15 rs R L 36,3 RSTIAFAHR 0,95 0,77 0,47 0,87 RSTIAFAHL 0,98 0,55 0,05 0,25 6,88 1,75 1,10 2,42 ENSG SIGLEC1 20 rs K R 66,4 HLQVKYAPK 0,67 0,16 0,51 0,86 HLQVKYAPR 0,37 0,60 0,24 0,45 4,15 1,51 2,60 1,30 ENSG VWDE 7 rs F C 67,9 GYTGRRFQK 0,97 0,15 0,47 0,86 GYTGRRCQK 0,96 0,15 0,23 0,50 1,05 5,84 1,27 2,30 ENSG IIP45 1 rs K E 55,8 FSEESAVPK 0,66 0,12 0,50 0,86 FSEESAVPE 0,03-0,76 0,03 0,01 3,08 4,03 1,36 1,52 ENSG MYO15B 17 rs R K 58,6 HTFLQRSHY 0,97 1,21 0,43 0,85 HTFLQKSHY 0,97 1,21 0,35 0,73 8,05 1,32 1,08 1,72 ENSG LY9 1 rs V M 73,5 TVYAQVFNL 0,97 0,61 0,45 0,85 TMYAQVFNL 0,97 0,60 0,32 0,65 23,66 3,81 1,07 1,63 ENSG TIMELESS 12 rs Q R 68,1 LQELYLANK 0,65 0,16 0,48 0,82 LRELYLANK 0,91 0,22 0,09 0,28 7,30 11,22 1,05 1,34 ENSG NRG1 8 rs Q R 67,0 SPALPPQLK 0,97 0,08 0,44 0,81 SPALPPRLK 0,95 0,08 0,36 0,68 14,47 1,32 2,98 2,43 ENSG PITRM1 10 rs I V 57,5 TTISVSFLL 0,57 0,41 0,46 0,79 TTVSVSFLL 0,69 0,41 0,34 0,62 4,94 4,44 2,52 3,29 ENSG CCDC rs W R 30,1 KVRWKKEEK 0,84 0,27 0,43 0,79 KVRRKKEEK 0,74 0,27 0,34 0,64 7,32 8,06 1,09 1,12 ENSG TMCO4 1 rs K Q 61,9 TEFMAGLVK 0,65 0,18 0,44 0,77 TEFMAGLVQ 0,05-0,07 0,05 0,07 1,28 3,50 1,34 1,29 HLA-A24 predictions ENSG PRR5 22 rs L P 29,2 YYEKIFSTL 0,98 0,39 0,73 1,72 YYEKIFSTP 0,87 0,01 0,20 0,57 35,14 1,50 3,17 3,18 ENSG MTRR 5 rs10380 Y H 29,5 RYFLKHGIL 0,93 0,71 0,68 1,62 RHFLKHGIL 0,86 0,60 0,16 0,51 7,38 4,09 1,40 3,21 ENSG SIPA1 11 rs F S 28,3 AFPPVPPAL 0,94 0,48 0,63 1,50 ASPPVPPAL 0,94 0,43 0,23 0,66 20,36 11,51 1,69 1,87 ENSG AMICA1 11 rs V A 36,6 YYHKLRMSV 0,96 0,31 0,60 1,44 YYHKLRMSA 0,86-0,05 0,34 0,85 172,21 1,01 1,10 1,21 ENSG TMC8 17 rs I N 71,4 SYLRVNVLI 0,73 0,40 0,61 1,43 SYLRVNVLN 0,06-0,40 0,17 0,34 20,18 6,98 1,11 1,46 ENSG MYO19 17 rs F L 64,5 RFWPLGLVL 0,96 0,67 0,57 1,39 RLWPLGLVL 0,97 0,61 0,19 0,57 4,42 2,93 1,13 1,13 ENSG TUBGCP6 22 rs L S 44,2 EYLSFRDKL 0,75 0,44 0,56 1,32 EYLSFRDKS 0,04-0,95 0,18 0,35 13,76 1,51 1,10 1,10 ENSG TBCD 17 rs Y C 57,1 MYAHTHLRL 0,98 0,54 0,53 1,30 MCAHTHLRL 0,91 0,43 0,07 0,31 5,73 11,49 1,66 1,47 ENSG ZFYVE26 14 rs Y C 63,7 AYDKEGWQY 0,96 1,39 0,49 1,27 ACDKEGWQY 0,67 1,28 0,06 0,30 11,49 2,02 1,37 2,25 ENSG PITRM1 10 rs I V 32,7 PYVPLFCSI 0,96 0,23 0,50 1,23 PYVPLFCSV 0,97 0,13 0,28 0,75 4,94 4,44 2,52 3,29 ENSG UNQ rs V A 54,9 KVPMVIMDW 0,98 0,50 0,49 1,22 KAPMVIMDW 0,98 0,46 0,36 0,93 1,02 6,93 1,12 1,42 ENSG PLXDC2 10 rs V I 54,0 TVGATTTQF 0,72 1,08 0,49 1,20 TIGATTTQF 0,64 1,08 0,32 0,83 50,78 1,03 1,84 3,28 ENSG ESPNL 2 rs V A 44,8 AYTPALKTV 0,98 0,32 0,49 1,20 AYTPALKTA 0,95-0,05 0,28 0,73 4,83 7,58 2,42 3,16 ENSG CRNKL1 20 rs F L 75,7 LVLAWSSQF 0,92 1,11 0,46 1,18 LVLAWSSQL 0,96 0,42 0,23 0,65 5,14 2,88 1,44 3,27 ENSG UTRN 6 rs T A 32,4 RTMEEWRQF 0,82 1,16 0,45 1,14 RAMEEWRQF 0,81 1,20 0,28 0,78 10,24 1,49 1,06 1,59 ENSG CYP20A1 2 rs F L 78,2 KLTPVSAQF 0,97 1,21 0,44 1,14 KLTPVSAQL 0,98 0,52 0,17 0,54 8,07 1,75 1,06 1,89 ENSG MYO15B 17 rs W R 69,0 CWPGLRNEL 0,81 0,48 0,45 1,11 CRPGLRNEL 0,96 0,46 0,10 0,37 8,05 1,32 1,08 1,72 ENSG DIDO1 20 rs L S 28,3 LYKCSGKYL 0,93 0,55 0,43 1,09 LYKCSGKYS 0,04-0,84 0,13 0,25 9,19 3,65 1,23 1,40 ENSG APOB48R 16 rs A P 66,1 PFPKQAQVL 0,93 0,30 0,44 1,09 PFPKQPQVL 0,95 0,30 0,32 0,84 25,23 1,48 2,16 1,39 ENSG APOL3 22 rs S R 26,5 GFQGISQSL 0,97 0,35 0,43 1,08 GFQGISQRL 0,96 0,35 0,32 0,83 9,46 5,17 1,97 2,36 a Only hematopoietic tumor cell lines, of which only few were included in the dataset (see Materials and Methods) =0-3.5x median of all tissues for the specific probe(s) Abbreviations: Chrom, chromosome; nssnp, nonsynonymous single nucleotide polymorphism =3.5-5x median of all tissues for the specific probe(s) =5-8x median of all tissues for the specific probe(s) =8-12x median of all tissues for the specific probe(s) =12-20x median of all tissues for the specific probe(s) >20x median of all tissues for the specific probe(s)

73 Genome wide mhag predictions 73 While reverse immunology is conceptually an ideal method to identify clinically relevant mhags, currently the most important bottleneck of the reverse strategy is the generation of peptide-specific T cell responses. Given the difficulty and labor intensity of this final and crucial step, we believe that utmost attention should be given to eliminate the false-positive predictions in the reverse strategies. This is the main reason why we include mhag-specific features in the reverse strategy. Currently there are other methods claiming the prediction of putative mhags (SNEP, sipep, Peptidecheck) These methods do not take differential surface expression into account and assume that all properly processed HLA-binding polymorphic peptides are putative mhags. The fact that only 2 of the currently known mhags identified so far display bilateral expression, suggests that proper surface expression of both allelic peptides may negatively influence their immunogenicity. Thus, it may be very well possible that a vast majority of the predictions of these available strategies are false-positive. In contrast, by taking differential antigen processing into account, we may have decreased the chance for false-positive predictions significantly, thus identifying the best possible mhag candidates throughout the whole genome. We of course realize that our selection displays a strong underestimation of the number of mhag candidates, because differential HLA-binding is not the only mechanism for differential surface expression. Thus, it is theoretically possible to improve our strategy by including differential transcription/translation processes, proteasome cleavage and differential TAP translocation algorithms. Nonetheless, although we identified only a fraction of possible candidates we believe that our strategy is a significant improvement toward a reliable reverse identification strategy. Chapter 4 The obvious next step of our work is confirming the existence and subsequently the clinical impact of these putative mhags by experimental approaches. One such an approach is the generation of mhag-specific T cells from a naive T cell pool derived from a mhag healthy individual using peptide loaded monocyte-derived dendritic cells. This strategy was successful in the past for generating different tumor antigen-specific T cells, and also mhag-specific T cells It may also be considered to induce mhag-specific T cells by immunization of HLA-transgenic mice with the putative mhag-peptides. TCRs derived from these mhag-specific murine T cells could be valuable tools for adoptive immunotherapy. An alternative approach is to isolate mhag-peptide-specific T cells from already immunized post allo-sct PBMCs of appropriate recipients where the recipient but not the donor is genetically positive for the putative mhag. Since this alternative approach may be hampered by the immunodominance of other mhags, it must be necessary to include multiple recipients in the analysis. Any of these approaches still requires recognition of mhag + targets by the peptide-specific T cells as evidence that the candidate peptides can be naturally processed and bind to HLA molecules to be expressed on the cell surface as a mhag.

74 74 Chapter 4 In summary, all T cell generation approaches still contain drawbacks, but these limitations are counterbalanced by the major advantage of having a user defined HLA restriction, mhag frequency and tissue distribution. Finally, it needs to be emphasized that a concerted and systematic action of researchers, dedicated to bring mhag-based immunotherapy into clinical practice, is necessary to test a large number of possible candidates and we believe that the first extensive list of putative mhags presented in this work can provide a basis for these actions. METHODS Prediction algorithms The HLA binding affinity of polymorphic peptides was predicted using artificial neural networks from the NetMHC 3.0 Server 16,22,49. The HLA binding affinity of both allelic peptides was compared. Peptides with a predicted binding affinity of IC 50 -value <500 nm were considered as binders; peptides with IC 50 -value >1000 or above were considered as non-binders. Allelic peptides displaying differential HLA-binding were screened for proper C-terminal cleavage by proteasomes and TAP transport efficiency using the neural network trained NetChop 3.0 and 3.1 and weight matrices respectively 23,24. For all putative mhags, we determined the NetCTL 1.2 scores, which are combined scores for HLA-binding, proteasome cleavage and TAP transport 43. SNPs and genes Information concerning SNP type, SNP frequencies, gene and protein sequences, were downloaded from Ensembl and HapMap databases ( and org) 13,15. Tissue distribution of putative mhags Tissue distribution of genes was derived from the publicly available micro-array dataset downloaded from biogps.gnf.org 44,45. The median value of each probe in the dataset was calculated with Microsoft Excel. All probes revealing no score throughout all tested tissues of more than 3.5x the median for that probe were considered as unreliable and therefore discarded. Tissues were divided in four groups: hematopoietic normal (BDCA4 + dentritic cells, bonemarrow, CD14 + monocytes, CD19 + B cells, CD33 + myeloid cells, CD34 + cells, CD4 + T cells, CD56 + NK cells, CD8 + T cells, whole blood), hematopoietic tumor cell lines (X721, K562, HL60, MOLT4, Daudi, Raji.) GvHD organs (bronchial epithelial cells, kidney, liver, lung, skin), and non-hematopoietic non-gvhd organs (adipocyte, adrenal cortex, adrenal gland, amygdala, appendix, atrioventricular node, cardiac myocytes, caudate nucleus, cerebellum peduncles, cerebellum, ciliary ganglion, cingulated cortex, dorsal root ganglion, globus pallidus, heart, hypothalamus, medulla oblongata, occipital lobe, olfactory bulb, pancreas, pancreatic islet, parietal lobe, pituitary, pons, prefrontal cortex, salivary gland, skeletal muscle, smooth muscle, spinal cord, subthalamic nucleus, superior cervical ganglion, temporal lobe, thalamus, tongue, trachea, trigeminal ganglion, whole brain). Maximum score within each group was compared to the median of all tissues and visualized in a heat map.

75 Genome wide mhag predictions 75 REFERENCES 1. Feng X, Hui KM, Younes HM, Brickner AG. Targeting minor histocompatibility antigens in graft versus tumor or graft versus leukemia responses. Trends Immunol. 2008;29: Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: Riddell SR, Berger C, Murata M, Randolph S, Warren EH. The graft versus leukemia response after allogeneic hematopoietic stem cell transplantation. Blood Rev. 2003;17: Spaapen R, Mutis T. Targeting hematopoietic-specific minor histocompatibility antigens to distinguish graft-versustumor effects from graft-versus-host disease. Best Pract Res Clin Haematol. 2008;21: Akatsuka Y, Nishida T, Kondo E et al. Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med. 2003;197: de Rijke B, van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Kawase T, Nannya Y, Torikai H et al. Identification of human minor histocompatibility antigens based on genetic association with highly parallel genotyping of pooled DNA. Blood. 2008;111: Murata M, Warren EH, Riddell SR. A human minor histocompatibility antigen resulting from differential expression due to a gene deletion. J Exp Med. 2003;197: Wang W, Meadows LR, den Haan JM et al. Human H-Y: a male-specific histocompatibility antigen derived from the SMCY protein. Science. 1995;269: Spaapen RM, de Kort RAL, van den Oudenalder K et al. Rapid Identification of Clinical Relevant Minor Histocompatibility Antigens via Genome-Wide Zygosity-Genotype Correlation Analysis. Clinical Cancer Research. 2009;15: Smedley D, Haider S, Ballester B et al. BioMart--biological queries made easy. BMC Genomics. 2009;10: Hubbard TJ, Aken BL, Ayling S et al. Ensembl Nucleic Acids Res. 2009;37:D690-D Frazer KA, Ballinger DG, Cox DR et al. A second generation human haplotype map of over 3.1 million SNPs. Nature. 2007;449: Buus S, Lauemoller SL, Worning P et al. Sensitive quantitative predictions of peptide-mhc binding by a Query by Committee artificial neural network approach. Tissue Antigens. 2003;62: Paschen A, Eichmuller S, Schadendorf D. Identification of tumor antigens and T cell epitopes, and its clinical application. Cancer Immunol Immunother. 2004;53: Kessler JH, Melief CJ. Identification of T cell epitopes for cancer immunotherapy. Leukemia. 2007;21: Dolstra H, de RB, Fredrix H et al. Bi-directional allelic recognition of the human minor histocompatibility antigen HB-1 by cytotoxic T lymphocytes. Eur J Immunol. 2002;32: Mommaas B, Kamp J, Drijfhout JW et al. Identification of a novel HLA-B60-restricted T cell epitope of the minor histocompatibility antigen HA-1 locus. J Immunol. 2002;169: Wenandy L, Kollgaard T, Letsch A et al. The 1170 A-P single-nucleotide polymorphism (SNP) in the Her-2/neu protein (HER2) as a minor histocompatibility antigen (mhag). Leukemia. 2009;23: Nielsen M, Lundegaard C, Worning P et al. Reliable prediction of T cell epitopes using neural networks with novel sequence representations. Protein Sci. 2003;12: Peters B, Bulik S, Tampe R, Van Endert PM, Holzhutter HG. Identifying MHC class I epitopes by predicting the TAP transport efficiency of epitope precursors. J Immunol. 2003;171: Nielsen M, Lundegaard C, Lund O, Kesmir C. The role of the proteasome in generating cytotoxic T cell epitopes: insights obtained from improved predictions of proteasomal cleavage. Immunogenetics. 2005;57: Kawase T, Akatsuka Y, Torikai H et al. Alternative splicing due to an intronic SNP in HMSD generates a novel minor histocompatibility antigen. Blood. 2007;110: Kamei M, Nannya Y, Torikai H et al. HapMap scanning of novel human minor histocompatibility antigens. Blood. 2009;113: Terakura S, Murata M, Warren EH et al. A single minor histocompatibility antigen encoded by UGT2B17 and presented by human leukocyte antigen-a*2902 and -B*4403. Transplantation. 2007;83: Chapter 4

76 76 Chapter Brickner AG, Evans AM, Mito JK et al. The PANE1 gene encodes a novel human minor histocompatibility antigen that is selectively expressed in B-lymphoid cells and B-CLL. Blood. 2006;107: Torikai H, Akatsuka Y, Miyazaki M et al. The human cathepsin H gene encodes two novel minor histocompatibility antigen epitopes restricted by HLA-A*3101 and -A*3303. Br J Haematol. 2006;134: Spierings E, Brickner AG, Caldwell JA et al. The minor histocompatibility antigen HA-3 arises from differential proteasome-mediated cleavage of the lymphoid blast crisis (Lbs) oncoprotein. Blood. 2003;102: Warren EH, Vigneron NJ, Gavin MA et al. An antigen produced by splicing of noncontiguous peptides in the reverse order. Science. 2006;313: Brickner AG, Warren EH, Caldwell JA et al. The immunogenicity of a new human minor histocompatibility antigen results from differential antigen processing. J Exp Med. 2001;193: den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Torikai H, Akatsuka Y, Miyauchi H et al. The HLA-A*0201-restricted minor histocompatibility antigen HA-1H peptide can also be presented by another HLA-A2 subtype, A*0206. Bone Marrow Transplant. 2007;40: Pierce RA, Field ED, Mutis T et al. The HA-2 minor histocompatibility antigen is derived from a diallelic gene encoding a novel human class I myosin protein. J Immunol. 2001;167: Wolfel C, Lennerz V, Lindemann E et al. Dissection and molecular analysis of alloreactive CD8 + T cell responses in allogeneic hematopoietic stem cell transplantation. Cancer Immunol Immunother. 2008;57: Tykodi SS, Fujii N, Vigneron N et al. C19orf48 encodes a minor histocompatibility antigen recognized by CD8 + cytotoxic T cells from renal cell carcinoma patients. Clin Cancer Res. 2008;14: van Bergen CA, Kester MG, Jedema I et al. Multiple myeloma-reactive T cells recognize an activation-induced minor histocompatibility antigen encoded by the ATP-dependent interferon-responsive (ADIR) gene. Blood. 2007;109: Slager EH, Honders MW, van der Meijden ED et al. Identification of the angiogenic endothelial-cell growth factor-1/ thymidine phosphorylase as a potential target for immunotherapy of cancer. Blood. 2006;107: Spierings E, Gras S, Reiser JB et al. Steric hindrance and fast dissociation explain the lack of immunogenicity of the minor histocompatibility HA-1Arg Null allele. J Immunol. 2009;182: Nicholls S, Piper KP, Mohammed F et al. Secondary anchor polymorphism in the HA-1 minor histocompatibility antigen critically affects MHC stability and TCR recognition. Proc Natl Acad Sci U S A. 2009;106: Spaapen RM, Lokhorst HM, van den Oudenalder K et al. 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. 2008;205: Larsen MV, Lundegaard C, Lamberth K et al. Large-scale validation of methods for cytotoxic T-lymphocyte epitope prediction. BMC Bioinformatics. 2007;8: Wu C, Orozco C, Boyer J et al. BioGPS: an extensible and customizable portal for organizing and querying gene annotation resources. Genome Biology. 2009;10:R Su AI, Wiltshire T, Batalov S et al. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci U S A. 2004;101: DeLuca DS, Eiz-Vesper B, Ladas N, Khattab BA, Blasczyk R. High-throughput minor histocompatibility antigen prediction. Bioinformatics. 2009;25: Halling-Brown M, Quartey-Papafio R, Travers PJ, Moss DS. SiPep: a system for the prediction of tissue-specific minor histocompatibility antigens. Int J Immunogenet. 2006;33: Schuler MM, Donnes P, Nastke MD et al. SNEP: SNP-derived epitope prediction program for minor H antigens. Immunogenetics. 2005;57: Lundegaard C, Lund O, Nielsen M. Accurate approximation method for prediction of class I MHC affinities for peptides of length 8, 10 and 11 using prediction tools trained on 9mers. Bioinformatics. 2008;24:

77 Genome wide mhag predictions 77 Supplemental data Table S1. Validation of the tissue distribution analysis using currently known mhags and genes well known to be expressed by either the hematopoietic or non-hematopoietic system Tissue Expression Gene Entrez # Probes in Hematopoietic Non-hematopoietic Expression ratio Ensembl gene symbol gene ID dataset a Normal Tumor GvHD Non-GvHD hemat./non-hemat. Known HLA class I restricted mhag encoding genes b ENSG BCL2A ,52 65,83 1,87 1,96 47 ENSG HB na na na na ENSG HMHA ,74 60,10 22,52 5,73 25 c ENSG HMSD na na na na ENSG MYO1G ,01 55,56 1,99 1,10 39 ENSG P2X ,10 379,71 1,02 3, ENSG PANE ,24 146,34 1,09 1,59 92 ENSG SP ,56 18,39 2,57 1,87 31 ENSG ADIR ,73 7,46 4,78 1,71 2 ENSG AKAP ,95 3,22 5,96 2,17 2 ENSG C19ORF ,38 49,96 9,45 5,87 5 ENSG CTSH ,44 10,13 44,85 2,20 1 ENSG DBY ,58 14,31 0,88 2,98 10 ENSG DFFRY ,18 1,34 1,02 2,25 5 ENSG ECGF ,99 1,48 6,08 3,08 9 ENSG ERBB ,72 1,39 10,71 4,45 1 d ENSG KIAA ,65 40,70 6,13 5,31 7 ENSG RPS4Y ,36 14,46 3,21 7,77 2 ENSG SLC1A ,80 8,61 2,39 3,61 2 ENSG SMCY ,16 3,05 1,59 2,81 8 ENSG TMSB4Y na na na na ENSG TRIM ,47 31,63 2,19 9,36 7 ENSG UGT2B ,19 2,60 1,09 5,14 1 ENSG UTY na na na na Well known genuine hematopoietic genes ENSG CD1C ,11 4,62 1,68 1,46 6 ENSG CD1D ,92 3,78 1,07 1,90 65 ENSG CD ,37 3,56 2,81 3,36 14 ENSG CD ,61 6,51 1,51 1,35 12 ENSG CD3D ,85 149,14 2,28 2,76 92 ENSG CD3E ,36 116,53 1,09 1, ENSG CD3G ,99 24,64 1,58 2,32 17 ENSG CD ,24 1,35 3,52 3,41 14 ENSG CD ,32 1,91 1,73 2,49 45 ENSG CD8A ,40 1,27 1,67 1,62 37 ENSG CD8B ,04 1,31 1,08 1,54 41 ENSG CD ,36 89,05 1,05 2,73 57 ENSG CD ,75 44,08 1,33 7,99 54 Chapter 4 Well known genuine liver or lung tissue genes ENSG AHSG ,27 1, ,89 1,34 1/900 ENSG ALB ,24 1,31 877,47 15,59 1/672 ENSG SFTPA2B ,28 1, ,65 31,29 1/2393 ENSG SFTPB ,28 1,41 796,37 1,79 1/564 ENSG SFTPC ,36 1, ,48 2,72 1/2346 ENSG SFTPD ,44 1,36 382,02 1,50 1/266 a Number of probes within the micro-array dataset. Values for the four tissue groups are averages of the indicated number of probes b The first 8 genes are well described and generally thought to encode for strictly hematopoietic mhags c HA-1 is well described as hematopoietic restricted mhag. The GvHD tissue expression is caused by a 22 fold difference in lung expression compared to the median, however expression in hematopoietic cells was even 22 times higher than expression in the lung d ERBB2 or HER2/NEU is a well known tumor antigen expressed at low levels in normal tissue but overexpressed in many cancers. A likely reason why no expression was seen in the hematopoietic tumors, is that the number of tumors included in this dataset was very limited =0-3.5x median of all tissues for the specific probe(s) =3.5-5x median of all tissues for the specific probe(s) =5-8x median of all tissues for the specific probe(s) =8-12x median of all tissues for the specific probe(s) =12-20x median of all tissues for the specific probe(s) >20x median of all tissues for the specific probe(s)

78 78 Chapter 4

79 5 Rebuilding human leukocyte antigen class II-restricted minor histocompatibility antigen specificity in recall antigen-specific T cells by adoptive T cell receptor transfer: Implications for adoptive immunotherapy Robbert M. Spaapen 1, Kelly van den Oudenalder 1, Roman Ivanov 1, Andries C. Bloem 2, Henk M. Lokhorst 3, Tuna Mutis 1 Departments of 1 Clinical Chemistry and Haematology, 2 Immunology, 3 Haematology, UMCU, NL Clin Cancer Res Jul;13:

80 80 Chapter 5 Donor T cells directed to hematopoietic minor histocompatibility antigens (mhags) are appealing tools for adoptive immunotherapy of hematological malignancies after allogeneic stem cell transplantation (allo-sct). Toward the development of a convenient strategy for ex vivo generation of human leukocyte antigen (HLA) class II-restricted mhag-specific T cells, we evaluated the feasibility of rebuilding mhagspecific T cell functions in donor-derived recall antigen-specific T cells via T cell receptor (TCR) transfer. TCR α- and β-chains of an HLA- DPB1*0401-restricted T cell clone recognizing a multiple myelomaassociated mhag were retrovirally transferred into a tetanus toxoid (TT)-specific clone derived from the original stem cell donor. TCR double-transduced cells were compared with the parent mhag- and TTspecific clones for antigen specificity, cytokine secretion, and cytotoxic activity and were analyzed for their in vitro expansion capacity in a TT- or mhag-specific fashion. mhag-tcr-transduced TT-specific cells displayed both TT and mhag specificity. Similar to the parent cells, they secreted Th-1 cytokines and exerted significant cytotoxic activity against TT-pulsed or mhag + target cells, including multiple myeloma cells. A 4wk expansion of TCR-transduced cells via the TT-specific TCR had no negative influence on the mhag-specific cytotoxic activity and resulted in 10- to 100-fold better cell yields as compared with mhagspecific expansion. HLA class II-restricted, mhag-specific effector functions can be successfully reconstructed in donor-derived TTspecific T cells via TCR-transfer. Effective expansion of these T cells via TT-specific TCRs illustrate the suitability of this strategy for ex vivo expansion and possibly for in vivo TT-specific reboosting of HLA class II-restricted immunotherapeutic T cells. INTRODUCTION Minor histocompatibility antigens (mhags) are MHC-bound polymorphic peptides derived from intracellular proteins. With these properties, mhags constitute the main targets of donor T cells mediating the curative graft-versus-tumor (GvT) effect and graft-versus-host disease (GvHD) after human leukocyte antigen (HLA)-matched allogeneic stem cell (SC) transplantation (allo-sct) 1. From the clinical point of view, a major breakthrough of the past decade was the discovery of mhags expressed exclusively on hematopoietic cells 2-4. In several in vitro studies, cytotoxic T cells (CTL) directed against hematopoietic mhags effectively eliminated mhag + leukemia and myeloma cells, but spared nonhematopoietic cells 5-8. These studies suggested the possibility to separate GvT from GvHD by treatment of recipients with donor T cells directed at hematopoietic mhags. Subsequently, we and others have shown

81 mhag-tcr transfer into tetanus toxoid specific T cells 81 the feasibility of this concept by ex vivo generation of mhag-specific CTLs from unprimed SC donors using dendritic cells (DCs) pulsed with mhag-peptides or transduced with mhag cdna Nevertheless, such DC-based in vitro CTL induction protocols from unprimed donors require repeated stimulations of T cells in long term cultures and are therefore not always successful for the generation of large quantities of mhag-specific T cells within a reasonable time period 12. Furthermore, long term ex vivo culturing may also impair the in vivo survival capacity of adoptively transferred T cells. Finally, because such protocols require biochemical identification of mhags, they cannot be applied for the ex vivo generation of donor CTLs against hematopoietic mhags defined only by cellular assays. One such novel mhag is recognized by the HLA-DPB1*0401-restricted cytotoxic CD4 + T cell clone 3AB11 isolated from a multiple myeloma patient after HLA-identical sibling SCT. We have previously shown the hematopoietic-restricted and myeloma-associated expression of this mhag by determining the target cell specificity of clone 3AB Toward tackling the current drawbacks of the mhag-specific adoptive immunotherapy, in particular for HLA class II-restricted mhags, we here investigated the feasibility and efficacy of introducing the HLA class II-restricted mhag-specific TCR into donor s preactivated T cells via TCR-transfer. Although several previous studies simply introduced HLA class I- or class II-restricted TCRs into randomly expanded T cells 14-17, we refrained to follow such a strategy because adoptive transfer of randomly activated donor T cells into recipients involves a high risk for GvHD induction. Considering requirements such as ease of generation and expandability, we chose to introduce mhag-specific TCRs into recall antigen (tetanus toxoid [TT])- specific T cells. To test the feasibility of our hypothesis, we inserted TCR α- and β-chains of the HLA-DPB1*0401-restricted, mhag-specific T cell clone 3AB11 into a TT-specific clone derived from the original SC donor. After purification, TCR doubletransduced cells were compared with the parent mhag- and TT-specific clones for antigen-specific activity, cytokine secretion profile, and cytotoxic activity, and the TCR-transduced cells were analyzed for their efficacy of in vitro expansion in a TTor mhag-specific fashion. Chapter 5 RESULTS Selection of a donor-derived TT-specific T cell clone for retroviral TCR-transfer Because virtually all individuals receive tetanus vaccinations, we expected to easily generate TT-specific T cell from the healthy SC donor. Indeed, a 6-day stimulation protocol followed by limiting dilution revealed several CD4 + TT-specific T cell clones. To show the proof of principle of our hypothesis, a TT-specific, easily expandable T cell clone, N TT 3CB10, was selected for TCR-transfer. This clone recognized TT in an HLA-DRB1*1501-restricted fashion because its TT-specific IFNγ response was

82 82 Chapter 5 A EBV-LCL d B C EBV-LCL d TT EBV-LCL p EBV-LCL p TT + anti-dp + anti-dq + anti-dr HLA-DRB1*typing of EBV-LCL: 0301/ / / anti-dr Lysis (%) >2000 IFN-γ (pg/ml) IFN-γ (pg/ml) 5:1 20:1 E:T ratio EBV-LCL p EBV-LCL d EBV-LCL d TT Figure 1. HLA-DRB1*1501-restricted TT-specific IFNγ production and cytotoxic activity of donorderived T cell clone N TT 3CB10. (A) IFNγ secretion of clone N TT 3CB10 toward patient and donor EBV- LCLs (EBV-LCL p; EBV-LCL d) alone or loaded with TT. Where indicated, antigen-presenting EBV-LCLs were preincubated with α-hla-dp, -DQ, and -DR antibodies for 1h. Antibodies remained present in the assay. Experiments were done in triplicate, and SEM did not exceed 10%. (B) IFNγ secretion of clone N TT 3CB10 against TT-pulsed EBV-LCLs sharing only one of the HLA-DRB1 alleles expressed by the donor (HLA-DRB1*0301 or *1501). Antigen-presenting EBV- LCLs were additionally preincubated with α-hla-dr. Cultures were done in triplicate; SEM never exceeded 10%. (C) cytotoxic activity of clone N TT 3CB10 against unpulsed patient EBV-LCLs ( ), donor EBV-LCLs ( ), and donor EBV-LCLs pulsed with TT ( ). The cytotoxic activity was measured at the indicated effector-to-target (E:T) cell ratios in a 48h FACSbased cytotoxicity assay. All measurements were done in duplicate; SEM did not exceed 15%. blocked by α-hla-dr antibodies and was only present against HLA-DRB1* APCs (Fig. 1A and B). Clone N TT 3CB10 showed also significant TT-dependent cytotoxic activity against TT-pulsed donor-derived EBV-LCLs (Fig. 1C). Dual antigen specificity of TT-specific T cells transferred with mhag-tcr Clone N TT 3CB10 was retrovirally transduced with the TCR α- and β-chains of mhagspecific T cell clone 3AB11. After a brief expansion period, the TCR α and β doubletransduced cells were enriched to >88% purity by FACS sorting of cells expressing egfp and ΔNGF-R. However, as illustrated in Figure 2A, isolation of mhag-specific T cells required a second sorting of cells with the highest expression levels of egfp and ΔNGF-R. Only these cells produced significant levels of IFNγ in response to mhag + EBV-LCLs. These results were consistent with previous studies 18 and suggested that multiple copies of the introduced TCR chains were required for adequate surface expression of mhag-specific TCR heterodimers. In a detailed analysis, the latter cells responded also significantly toward the mhag + MM cell line UM9, but showed no response against mhag EBV-LCLs from donor (Fig. 2B), illustrating the

83 mhag-tcr transfer into tetanus toxoid specific T cells 83 A 1.3% 88.1% FACS sort R4 R3 R2 R1 B no stimulus EBV-LCL d FACS sort: R1 R2 R3 R4 EBV-LCL p EBV-LCL d TT IFN-γ (pg/ml) egfp NGF-R > mhag + EBV-LCL MM cell line UM >2000 IFN-γ (pg/ml) Figure 2. Selection of dual mhag- and TT-specific T cells by multiple sorting of the mhag-tcr-transduced TT-specific T cells. (A) double FACS sorting of egfp + and ΔNGF-R + TCR-transduced TT-specific T cells. After the first FACS sort, double-transduced cells were enriched up to 88%, but did not respond to mhag + cells (data not shown). Second, double-positive cells were separated into four different fractions based on expression levels of marker genes as indicated (R1 to R4). IFNγ response of each of the four sorted fractions against mhag + EBV-LCLs was determined. (B) IFNγ secretion of the double-transduced T cells with the highest egfp/δngf-r expression (R4) alone and against donor- and recipient-derived EBV-LCLs, the mhag + multiple myeloma cell line UM9, and against donor EBV-LCLs loaded with TT. All assays were done in triplicate; SEM of the results did not exceed 10%. 0 successful induction of mhag specificity in TT-specific T cells by TCR-transfer. TCRtransduced cells also produced high levels of IFNγ against TT-loaded donor EBV- LCLs, demonstrating that they maintained TT specificity (Fig. 2B). Cytokine profiles of the mhag-tcr-transduced and the original T cell clones To determine whether TCR-transduced cells maintained effector functions of the original mhag-specific and TT-specific T cells, we first compared their cytokine secretion profiles. The original mhag-specific and TT-specific T cell clones displayed a Th1-like cytokine profile indicated by the production of high levels of Chapter 5 Table 1. The IFNγ and IL4 secretion profiles of the original mhag- and TTspecific and TCR-transduced T cell clones APC Tetanus Toxoid T cell clone IFN-γ IL-4 mhag + - TCR-transduced > mhag + - 3AB11 > mhag - + TCR-transduced > mhag - + TT > anti-cd3/cd28 beads - TCR-transduced > anti-cd3/cd28 beads - 3AB11 > anti-cd3/cd28 beads - TT > Cytokines were measured in cell-free supernatants after stimulation of T cells with the indicated APCs or with α-cd3/cd28 beads in triplicate (in pg/ml). SEM did not exceed 10%. The lower detection levels for IFNγ and IL4 were 2 pg/ml.

84 84 Chapter 5 IFNγ, but little or no IL4 in response to antigen-specific or α-cd3/cd28-mediated stimulations (Table 1). Both cells produced also significant levels of TNFα (>1,000 pg/ml) and IL10 ( pg/ml), but only minimal levels of IL2, IL5, and IL13 (data not shown). Similarly, TCR-transduced cells produced IFNγ in response to mhag, TT- or nonspecific stimulations, revealing the maintenance of the original Th1-like cytokine profile (Table 1). TNFα and IL10 production levels of the TCR-transduced cells were also similar to the parental cells. Cytotoxic activity of the mhag-tcr-transduced and the original T cell clones From the immunotherapeutic point of view, perhaps the most important effector function of both CD8 + as well as CD4 + mhag-specific T cells is the cytotoxic activity against relevant target cells, in particular tumor cells. Therefore, next to determining the cytokine profiles, we also compared TCR-transduced cells with the parental T cells for cytotoxic capacity against patient or donor EBV-LCLs and against a mhag + MM cell line UM9. Because the cytotoxic activity of CD4 + cells may not always be visible in short assays, we tested the cytotoxic activities in 5h as well as in 48h assays. As expected, the original TT-specific clone displayed significant TTspecific cytotoxic activity in 5h assays. The lysis levels of TT-pulsed target cells increased to ~75% in 48h (Fig. 3A ). The mhag-specific T cell clone 3AB11 lysed two mhag + EBV-LCLs and MM cells in 5h assays (Fig. 3B, top). However, there were subtle differences. Although the mhag + EBV-LCLs derived from an unrelated patient were lysed about 50%, EBV-LCLs of the original MM patient and the mhag + A Lysis (%) h N TT 3CB10 B Lysis (%) h 3AB11 C Lysis (%) TCR 3AB11 -transduced N 100 TT 3CB10 5h 50 EBV-LCL p EBV-LCL d EBV-LCL d TT EBV-LCL mhag + MM cell line UM9 0 5:1 20:1 E:T ratio 0 5:1 20:1 E:T ratio 0 5:1 20:1 E:T ratio h h h Lysis (%) 50 Lysis (%) 50 Lysis (%) :1 20:1 E:T ratio 0 5:1 20:1 E:T ratio 0 5:1 20:1 E:T ratio Figure 3. Comparison of the cytotoxic activities of the original mhag- and TT-specific and TCR-transduced T cell clones. The cytotoxic activities of (A) the original TT-specific clone (N TT 3CB10), (B) the mhagspecific clone (3AB11), and (C) the TCR-transduced cells were compared in 5h 51 Cr release assays and in 48h FACS-based cytotoxicity assays against the indicated target cells at two different effector-to-target ratios (E:T ratio). All 5h and 48h assays were done in triplicate and duplicate, respectively; SEM did not exceed 15%.

85 mhag-tcr transfer into tetanus toxoid specific T cells 85 UM9 myeloma cells were lysed about 20%. Nonetheless, in 48h assays, the lysis levels of all targets were significantly higher and reached to maximum levels of 85% (Fig. 3B, bottom). The TCR-transduced cells displayed both mhag-specific and TTspecific cytotoxic activity similar to the parent cells. Most importantly, the lysis of recipient s EBV-LCLs was 45% to 60% and MM UM9 cells 40% in 48h assays (Fig. 3C, bottom), indicating that at long term, the TCR-transduced cells exerted effective cytotoxic activity against mhag + malignant B or plasma cells. Efficient expansion of the mhag-tcr-transduced clone with the original TT antigen One of the important reasons for inserting a TCR into recall antigen-specific T cells was the assumption that the recall antigen specificity could be exploited for efficient ex vivo expansion of TCR-transferred T cells. We therefore investigated the feasibility and efficacy of expanding TCR-transduced TT-specific cells in a TT-specific fashion and compared this with expansion in a mhag-specific fashion. To this end, the TCR-transduced cells were stimulated either by TT-loaded mhag EBV-LCLs or by unloaded mhag + EBV-LCLs. The cell cultures were supported by the addition of IL2; the stimulations were repeated at 14d intervals, and the expanded cells were tested for mhag- and TT-specific cytotoxic activities. In independent experiments, the TT-specific expansion rates of cells varied between 10- and 100-fold, sufficient for the generation of large numbers of cells for adoptive immunotherapy within four weeks. A representative experiment is depicted in Figure 4: four weeks of expansion of the TCR-transferred T cells in a TT-specific fashion revealed an effective and 15-fold better expansion as compared with mhag-specific stimulation (Fig. 4A). More important, TCR-transduced cells stimulated in a TT-specific fashion showed cytotoxic activity against mhag + target cells at all time points, illustrating the feasibility of expansion of TCR-transferred cells via the intrinsic TT-specific TCR without reducing their immunotherapeutic functions (Fig. 4B). On the other hand, cells stimulated via the transduced mhag-specific TCR did not expand at all and showed poor specific lysis, suggesting a functional deficit when using the introduced TCR for expansion (Fig. 4C). Chapter 5 DISCUSSION We evaluated the feasibility of transfer of a HLA class II-restricted, mhag-specific TCR into recall antigen-specific CD4 + CTLs. Our results show that TT-specific T cells can be readily isolated from peripheral blood, and that they are suitable target cells for adoptive TCR-transfer. Upon transduction with mhag-specific TCRs, the cells display dual TT and mhag specificity and maintain the cytokine profile of the original TT- and mhag-specific T cells. Importantly, these dual-specific T cells displayed effective mhag-dependent cytotoxic activity against myeloma cells and could be successfully expanded through triggering via the TT-specific TCR.

86 86 Chapter 5 A Fold expansion Expansion: mhag-specific TT-specific B Lysis (%) Culture time (days) TT-specific expansion Target: EBV-LCL mhag + EBV-LCL mhag - EBV-LCL mhag - TT C Lysis (%) Target: Culture time (days) mhag-specific expansion EBV-LCL mhag + EBV-LCL mhag - EBV-LCL mhag - TT Culture time (days) Figure 4. TT- or mhag-specific expansion of TCRtransduced T cells. (A) fold expansion of TCRtransduced T cells stimulated with only mhag + EBV- LCLs ( ) or TT-pulsed mhag EBV-LCLs ( ) for 28d. Antigen-specific stimulations were repeated at day 14. The cultures were supplemented with IL2 (120 IU/ ml) twice a week after 3d of each stimulation. B and C, cytotoxic activity of TCR-transduced T cells stimulated in a TT-specific (B) or mhag-specific fashion (C) at days 0, 21, and 28. The cytotoxic activities were determined in 5h 51 Cr release assays by incubating the indicated target cells with effector T cells at E:T ratios of 5:1 and 20:1. Shown are the results only for E:T ratios of 5:1. Tests were done in triplicate; SEM did not exceed 15%. These results show the suitability of this strategy for ex vivo generation and expansion of donor-derived HLA class II-restricted mhag-specific T cells. To our knowledge, our study is the first attempt to rebuild the T cell specificity toward an HLA class II-restricted potential immunotherapeutic antigen into TT-specific T cells. Unlike several previous studies, we deliberately refrained from using randomly activated donor T cells as targets for adoptive TCR due to a number of reasons. First of all, using randomly activated donor T cells as targets for therapeutic TCRs in an allo- SCT setting introduces a high risk for severe GvHD. Because recipient and donor are mismatched for several mhags, GvHD might be mediated by insertion of TCRs into immunogenic mhag-specific T cells. Theoretically, such a risk can be avoided if recipient s cells are used as targets for TCR-transfer. Unfortunately, however, in several cases, like in our case, it may not be possible to access recipient T cells after SCT. Besides this logistic drawback, inserting TCRs into polyclonally activated T cells may introduce other problems such as the risk for introducing TCRs into potential immune deviating Th-2 cells or regulatory T cells. Illustrating this possibility,

87 mhag-tcr transfer into tetanus toxoid specific T cells 87 a recent study reported that upon TCR transduction into randomly activated T cells, the cytokine profile of the original T cells was altered from a Th1-like into Th2- like profile with high production of IL In contrast, targeting TCRs into antigenspecific T cells with a known cytokine profile may reduce such risks. Furthermore, the latter strategy may also reduce the risk of formation of undesired self-reactive TCRs through hybrid formation between endogenous and transduced TCR chains 20. More importantly, the transfer of specific receptors into T cells with known virus specificity or allospecificity may permit the maintenance of the dual antigen-specific T cells via triggering through endogenous TCRs, as shown by a number of previous studies As suggested in these studies, the use of cytomegalovirus- and EBVspecific T cells as targets for TCR-transfer can permit continuous in vivo stimulation of transduced cells by these latent viruses 21,22. In the case of influenza-specific T cells, however, it seems possible to boost TCR-transduced cells by influenza vaccinations 24. Although we acknowledge these reports, because we primarily aimed at finding a suitable target cell for HLA class II-restricted TCRs, we chose CD4 + TT-specific T cells because the expression of the CD4 molecule is generally required to increase the avidity of TCR-MHC class II/peptide interactions 25. In fact, our results reveal that cytotoxic TT-specific cells can serve as appropriate targets for TCR-transfer: after the retroviral transduction of mhag-specific TCRs, the cells displayed dual antigen specificity and could be effectively expanded via stimulation through TT-specific TCRs, without losing the dual antigen specificity. In fact, our results suggest that TT-specific stimulation is probably the best way to expand these cells because stimulation through transgenic mhag-specific TCRs resulted only poor expansion of the cells, which also displayed diminished mhag-specific effector functions. Naturally, one obvious advantage of inserting the therapeutic mhag-specific TCRs into TT-specific cells is the possibility to boost these dual-specific T cells in vivo by vaccination of the recipients with TT. Because a TT-specific recall response can readily be maintained by TT boosters in healthy individuals, we think that a similar strategy may also be sufficient for the in vivo persistence of dual-specific T cells in recipients of allo-sct. However, it may also be considered to booster the recipients with TT-loaded donor-dc vaccinations if simple TT boosters seem insufficient to ensure persistence of dual-specific T cells. Chapter 5 In this study, we inserted the mhag-specific TCRs in a well-defined, TT-specific T cell clone to show the proof of the principle of our strategy. We acknowledge, however, that in a clinical setting, it may be more convenient to use polyclonal TTspecific T cells rather than T cell clones as targets for TCR-transfer. Therefore, a possible improvement of our methodology can be the direct selection of IFNγproducing TT-specific T cells using clinically applicable immunomagnetic bead-

88 88 Chapter 5 based IFNγ capture assays 7. Selection of IFNγ-secreting cells may also have the advantage that TCRs will be preferentially targeted into CTLs because it is known that IFNγ secretion and cytotoxic activity of CD4 + cells are well correlated 26,27. In conclusion, we here present a feasible strategy to generate immunotherapeutic T cells directed against HLA class II-restricted mhags. Because this strategy is suitable for hematopoietic mhags that could be defined only by cellular assays, it may expand the application arena of mhag-specific immunotherapy for leukemia and myeloma patients without the need for biochemical or genetic identification of mhags. Acknowledgments We thank Dr. Saskia Ebeling for supervision of TCR cloning. METHODS Cell culture and T cell clones EBV-LCLs and the multiple myeloma (MM) cell line UM9 13 were cultured in RPMI 1640 (Invitrogen), supplemented with 10% fetal bovine serum (FBS; Integro BV) and antibiotics (100 units/ml penicillin, 100 μg/ml streptomycin; both Life Technologies). The amphotropic Phoenix packaging cell line (ϕ-nx-a) was cultured in DMEM (Invitrogen) containing 10% FBS and antibiotics. The mhag-specific clone 3AB11 The isolation and functional characterization of HLA class II-restricted, mhag-specific CD4 + T cell clone 3AB11 were reported previously 13. Briefly, 3AB11 was isolated from the peripheral blood mononuclear cells (PBMCs) of a MM patient during the development of acute GvHD grade III and a strong GvT response after allo-sct from his HLA-identical brother. 3AB11 was cytotoxic and produced IFNγ in response to mhag + MM cells, B cells, T cells, monocytes, but not to nonhematopoietic stromal cells derived from mhag + individuals. The TT-specific T cell clone N TT 3CB10 This clone was derived from the HLA-identical SC donor using standard limiting dilution techniques. Briefly, PBMCs were cultured in the presence of TT (15 LF/ml; NVI) and received 60 IU/ml interleukin 2 (IL2; Chiron B.V.) at day 5. Proliferating cells were cloned at day 6 by limiting dilution at 0.3 cell per well using a feeder cell-cytokine-antigen mixture consisting of 30 Gy irradiated PBMCs from three random donors (10 6 cells/ml), 50 Gy irradiated EBVtransformed B cells (EBV-LCLs) from two random donors (10 5 cells/ml), phytohemagglutinin (100 ng/ml; Murex Diagnostics), TT (1.5 LF/ml), and IL2 (120 IU/ml). Culture medium was RPMI 1640 supplemented with 10% pooled human serum (HS), glutamine, and antibiotics. T cell clones, expanded by the addition of IL2 (120 IU/ml), were tested for TT specificity after a second round of stimulation with the above-described feeder cell cytokine mixture. Selected TT-specific clones were further expanded by restimulation with the above-described

89 mhag-tcr transfer into tetanus toxoid specific T cells 89 feeder cell-cytokine-antigen mixture and were cryopreserved until use. In some experiments, T cells were stimulated in a truly mhag- or TT-specific fashion using only mhag + EBV-LCLs or TT-pulsed (7.5 LF/ml for 48-72h) mhag EBV-LCLs as antigen presenting cells (APCs), respectively. Cloning of mhag-specific TCR α- and β-chains into retroviral vectors cdna derived from clone 3AB11 was amplified in reverse transcription-pcr using primers covering the entire repertoire of known TCR α- and β-chains. Two in-frame gene transcripts, TCR AV21S1 and TCR BV3S1, were identified and independently cloned into two separate pmx retroviral vectors containing the truncated nerve growth factor receptor (ΔNGF-R) and the green fluorescent protein (GFP) as marker genes, respectively. The genes were inserted downstream to the internal ribosome entry sequence, which ensures coexpression of the transgene and the marker gene. The vectors were then transfected into ϕ-nx-a packaging cells, and the retroviral supernatants were generated as described elsewhere 19. Functionality of the retroviral constructs encoding the TCR α- and β-chains was confirmed by transduction of the TCR-deficient Jurkat clone 76, followed by analysis of TCR α/β cell surface expression (data not shown). Retroviral transduction of TT-specific T cells Before retroviral transductions,tt-specific T cells were activated with the above-described feeder-cytokine mix for 48h. Activated TT cells were then transduced with TCR α and TCR β vectors using RetroNectin (Takara)-coated plates as described elsewhere with some minor adjustments 19. Briefly, a mixture of TCR α and TCR β vectors (0.75 ml each) was incubated in RetroNectin-coated 24-well plates. After 1h, 0.5 ml of the viral supernatant was replaced with 1 ml of T cell suspension (10 6 cells/ml) in RPMI supplemented with 10% HS and 240 IU/ml IL2. After 24h incubation at 37 C and 5% CO2, the T cells were harvested, and the retroviral transduction was repeated using fresh viral supernatants. Sixteen hours after the second transduction, the cells were washed and expanded with 10% HS and 120 IU/ml IL2 in RPMI. Retroviral transduction efficiency was determined by fluorescence-activated cell sorting (FACS) analysis 48h after last transduction. TCR double-transduced cells were then purified by FACS sorting of egfp + and ΔNGF-R + cells using a FACSAria sorter (Becton Dickinson). Immunophenotyping Cell surface expression of various receptors were determined by FACS analysis using FITC-, PE-, or APC-conjugated monoclonal antibodies (mab). All mabs except α-δngf-r (clone 20.4, culture supernatant) and PE-conjugated goat-anti-mouse immunoglobulin (SBA) were purchased from BD Biosciences. Where indicated, propidium iodide (PI) was added as viable dye. Flow cytometry was done using a FACSCalibur flow cytometer; the data were analyzed using the CellQuest software (both BD Biosciences). Chapter 5 Cytokine secretion profile of T cell clones To induce cytokine secretion, parental and TCR-transduced T cell clones were stimulated in a mhag- or TT-specific fashion using appropriate EBV-LCLs as APCs. For TT-specific stimulations, mhag EBV-LCLs were pulsed with 7.5 LF/ml TT for 48 to 72h. Where indicated, EBV-LCLs were preincubated with α-hla-dp, -DQ or -DR antibodies (10 μg/ml; BD Biosciences) for 1h at 37 C. T cells and APCs were then coincubated at a 1:1 ratio in a final volume of 200 μl in U-bottomed 96-well plates for 24h. The cytokines released in the

90 90 Chapter 5 culture supernatants were determined using standard sandwich ELISA kits (IL2, IL4, IL5, IL13, tumor necrosis factor-α [TNFα], IFNγ [Biosource], IL10 [Sanquin]). The SEM of the triplicate measurements never exceeded 10%. 51 Cr release-based 5h cytotoxicity assays Serial dilutions of effector T cells were incubated with 51 Cr (50 μci; Na 2 51 CrO 4 )-labeled target cells in a final volume of 200 μl RPMI/10% FBS in 96-well U-bottomed microtiter plates (Costar 3799) at 37 C. Spontaneous and maximal 51 Cr releases were determined by incubating target cells with medium alone and with 0.1% Triton X-100 in PBS, respectively. After 5h, cell-free supernatants were harvested to determine their 51 Cr content by γ-counting on a Cobra autogamma betaplate reader (Packard). The mean percentage of specific lysis in triplicate wells was calculated as follows: % specific lysis = (mean experimental release mean spontaneous release)/(mean maximal release mean spontaneous release) 100%. The SEM of triplicate cultures never exceeded 15%. FACS-based 48h cytotoxicity assays To determine the lytic activity of T cell clones on B cells and on MM cells at long term, we slightly modified a previously described FACS-based cytotoxicity assay 28. T cell clones and target cells (EBV-LCLs or MM cells) were coincubated at different ratios in a final volume of 200 μl RPMI/10% FBS at 37 C. After 48h, the wells were resuspended thoroughly, 150 μl of cell suspension was drawn and double stained with CD3-FITC plus CD19-APC or CD138- APC antibodies depending on the target cell. After 20 min incubation at room temperature, 2000 Flow-count fluorospheres (Beckman Coulter B.V.) and PI (0.1 mg/ml) were added. The samples were measured immediately on a FACSCalibur flow cytometer to determine the counts of viable CD19 + or CD138 + cells. In each sample, the counts of CD19/CD138 + target cells were normalized to the counts of fluorospheres, and the % specific lysis induced by T cell clones was then calculated as follows: % specific lysis = ([normalized counts of control viable target cells) (normalized counts of coincubation viable target cells])/(normalized counts of control viable target cells) 100%. The SEM of duplicate cultures never exceeded 15%.

91 mhag-tcr transfer into tetanus toxoid specific T cells 91 REFERENCES 1. Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Dolstra H, Fredrix H, Preijers F et al. Recognition of a B cell leukemia-associated minor histocompatibility antigen by CTL. J Immunol. 1997;158: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Holloway PA, Kaldenhoven N, van DM et al. Susceptibility of malignant plasma cells to HA-1 H -specific lysis suggests a role for the minor histocompatibility antigen HA-1 in the graft-versus-myeloma effect. Leukemia. 2004;18: Marijt WA, Heemskerk MH, Kloosterboer FM et al. Hematopoiesis-restricted minor histocompatibility antigens HA-1- or HA-2-specific T cells can induce complete remissions of relapsed leukemia. Proc Natl Acad Sci U S A. 2003;100: Rijke BD, van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: Mutis T, Verdijk R, Schrama E et al. Feasibility of immunotherapy of relapsed leukemia with ex vivo-generated cytotoxic T lymphocytes specific for hematopoietic system-restricted minor histocompatibility antigens. Blood. 1999;93: Brossart P, Spahlinger B, Grunebach F et al. Induction of minor histocompatiblity antigen HA-1-specific cytotoxic T cells for the treatment of leukemia after allogeneic stem cell transplantation. Blood. 1999;94: Dolstra H, de Rijke B, Fredrix H et al. Bi-directional allelic recognition of the human minor histocompatibility antigen HB-1 by cytotoxic T lymphocytes. Eur J Immunol. 2002;32: Mutis T, Ghoreschi K, Schrama E et al. Efficient induction of minor histocompatibility antigen HA-1-specific cytotoxic T cells using dendritic cells retrovirally transduced with HA-1-coding cdna. Biol Blood Marrow Transplant. 2002;8: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Fujio K, Misaki Y, Setoguchi K et al. Functional reconstitution of class II MHC-restricted T cell immunity mediated by retroviral transfer of the alpha beta TCR complex. J Immunol. 2000;165: Stanislawski T, Voss RH, Lotz C et al. Circumventing tolerance to a human MDM2-derived tumor antigen by TCR gene transfer. Nat Immunol. 2001;2: Tsuji T, Yasukawa M, Matsuzaki J et al. Generation of tumor-specific, HLA class I-restricted human Th1 and Tc1 cells by cell engineering with tumor peptide-specific T cell receptor genes. Blood. 2005;106: van der Veken L, Hoogeboom M, de Paus RA et al. HLA class II-restricted T cell receptor gene transfer generates CD4 + T cells with helper activity as well as cytotoxic capacity. Gene Ther. 2005;12: Viola A, Lanzavecchia A. T cell activation determined by T cell receptor number and tunable thresholds. Science. 1996;273: Ivanov R, Hol S, Aarts TI, Hagenbeek A, Ebeling SB. T cell receptor-transgenic primary T cells as a tool for discovery of leukaemia-associated antigens. Clin Exp Immunol. 2006;143: Willemsen RA, Weijtens ME, Ronteltap C et al. Grafting primary human T lymphocytes with cancer-specific chimeric single chain and two chain TCR. Gene Ther. 2000;7: Heemskerk MH, Hoogeboom M, Hagedoorn R et al. Reprogramming of virus-specific T cells into leukemia-reactive T cells using T cell receptor gene transfer. J Exp Med. 2004;199: Rossig C, Bollard CM, Nuchtern JG, Rooney CM, Brenner MK. Epstein-Barr virus-specific human T lymphocytes expressing antitumor chimeric T cell receptors: potential for improved immunotherapy. Blood. 2002;99: Kershaw MH, Westwood JA, Hwu P. Dual-specific T cells combine proliferation and antitumor activity. Nat Biotechnol. 2002;20: Cooper LJ, Al-Kadhimi Z, Serrano LM et al. Enhanced antilymphoma efficacy of CD19-redirected influenza MP1- specific CTLs by cotransfer of T cells modified to present influenza MP1. Blood. 2005;105: Swain SL. T cell subsets and the recognition of MHC class. Immunol Rev. 1983;74: Chapter 5

92 92 Chapter Mutis T, Cornelisse YE, Ottenhoff TH. Mycobacteria induce CD4 + T cells that are cytotoxic and display Th1-like cytokine secretion profile: heterogeneity in cytotoxic activity and cytokine secretion levels. Eur J Immunol. 1993;23: Rotteveel FT, Kokkelink I, van Lier RA et al. Clonal analysis of functionally distinct human CD4 + T cell subsets. J Exp Med. 1988;168: Ozdemir O, Ravindranath Y, Savasan S. Cell-mediated cytotoxicity evaluation using monoclonal antibody staining for target or effector cells with annexinv/propidium iodide colabeling by fluorosphere-adjusted counts on three-color flow cytometry. Cytometry A. 2003;56:

93 mhag-tcr transfer into tetanus toxoid specific T cells 93 Chapter 5

94 94 Chapter 5

95 6 Induction of potent in vivo anti-tumor effects by native and TCRredirected CD4 + cytotoxic human T lymphocytes directed at a single minor histocompatibility antigen Robbert M. Spaapen 1, Kelly van den Oudenalder 1, Richard W.J. Groen 2, Teun Guichelaar 1, Maureen van Elk 1, Tineke Aarts-Riemens 1, Andries C. Bloem 2, Gert Storm 3, Anton C. Martens 2, Henk M. Lokhorst 4, Tuna Mutis 1 Departments of 1 Clinical Chemistry and Haematology, 2 Immunology, 4 Haematology, UMCU, NL 3 Department of Pharmaceutics, UU, NL Submitted

96 96 Chapter 6 Although many human tumor-specific CD4 + T cells display strong in vitro cytotoxic capacities, little is known about their potential as sole therapeutic tools in cancer treatment. Therefore we here explored the efficacy of immunotherapy with sole CD4 + cytotoxic human T cells directed at a single HLA class II-restricted minor histocompatibility antigen (mhag). Rag2 -/- γc -/- mice carrying bioluminescence detectable, mhag + human multiple myeloma tumors were treated with native mhag-specific T cells or with Tetanus Toxoid (TT)-specific T cells that were genetically engineered to express the mhag-specific TCR. Effective mhag-specific anti-tumor responses were achieved after three injections of native or TCR-transduced mhag-specific T cells. The therapeutic effect was confined to tumors located in the bone marrow, was less prominent at higher tumor-load and was absent after injection of a lower T cell-dose, revealing the critical importance of tumor location and T cell-dose to tumorload ratio for the clinical outcome. Finally, the long term persisting dual-specific T cells could be boosted by in vivo stimulation via the TT-specific TCR, implicating the potential advantages of immunotherapy with TCR-redirected dualspecific CD4 + T cells. In conclusion, this study demonstrates for the first time the feasibility of immunotherapy of bone marrow tumors with sole CD4 + human T cells directed to a single tumor associated (mh) antigen. Introduction The immune system is armed with various cellular tools to effectively combat against cancer. Over the past decades, the potency of these immunotherapeutic tools was demonstrated not only by successful treatment of experimental animal tumors with adoptively transferred T cells, but also by the induction of long term remissions in leukemia patients treated with allogeneic stem cell transplantation (allo-sct) and donor lymphocyte infusions (DLI) 1. To date, it is evident that both CD8 + and CD4 + T cells are essential components of anti-tumor immunity. Nonetheless, a vast majority of novel tumor treatment strategies is based on CD8 + cytotoxic T cells, mainly because CD8 + T cells are considered as the effector cells of these responses and CD4 + T cells solely as helper cells 2-6. Despite the frequently reported observations that CD4 + murine T cells can mediate effective anti-tumor immunity in the absence of CD8 + T cells 7-11, there are only a few investigators proposing the use of CD4 + T cells as sole immunotherapeutic tools. Although it is well-documented that significant clinical responses can be achieved by the application of CD8-depleted DLI 12-16, CD4 + T cells are currently only exploited in a few immunotherapy strategies, mainly to induce and maintain CD8 + cytotoxic T cells

97 Immunotherapy with mhag-specific CD4 + T-cells 97 Set out to explore the direct anti-tumor effects of CD4 + human T cells and their suitability for adoptive immunotherapy of hematological malignancies in the allo- SCT setting, we have previously shown that several CD4 + T cells directed at tumor associated minor histocompatibility antigens (mhags) possess excellent capacities to kill tumor cells in vitro 20,21. Towards clinical application of such T cells, we have also demonstrated that these potent anti-tumor activities can be transferred into recall antigen (Tetanus Toxoid; TT)-specific CD4 + T cells by the well-known T cell receptor (TCR) transfer approach. The dual antigen specificity allowed effective in vitro expansion of these CD4 + T cells via stimulation of the intrinsic TT-specific TCR, suggesting the possibility to boost such T cells also in vivo via similar stimulation strategies 21. Hence, we now addressed the feasibility and efficacy of adoptive immunotherapy with native as well as with TCR-transduced dual-specific CD4 + T cells directed at a single mhag expressed on multiple myeloma (MM) cells. Administration of both native and TCR-redirected T cells into NK, B, and T cell deficient and macrophage-depleted Rag2 -/- gc -/- mice, which carried bioluminescence detectable MM tumors resulted in impressive mhag-specific and T cell-dose dependent antitumor effects in the bone marrow. Furthermore, the in vivo persisting dual-specific T cells could be boosted by administration of TT-loaded APCs, demonstrating for the first time the feasibility and the potential effect of adoptive immunotherapy with native or TCR-redirected dual-specific CD4 + T cells. Results In vivo anti-tumor effects of native CD4 + T cells directed at a single mhag In previous studies, we have shown the potent in vitro cytotoxic activity of mhagspecific CD4 + T cell clone 3AB11 against the mhag + MM cell line UM9 21. To determine its in vivo therapeutic potential, we cultured this cytotoxic CD4 + T cell clone in large quantities and administered it into macrophage-depleted Rag2 -/- γc -/- mice, which carried bioluminescence imaging (BLI)-detectable UM9-luc-eGFP tumors mainly in the bone marrow. The TT-specific oligoclonal T cell line CTL TT was used as control. While a single injection of 3AB11 at a low dose of cells/mice had no effect, a high dose therapy, consisting of three injections of cells/ mice with two day intervals, rapidly reduced the tumor-load below detection levels, demonstrating for the first time the feasibility of establishing effective anti-tumor responses by adoptive transfer of CD4 + human T cells targeting a single (minor H) antigen (Fig. 1A). No tumor reduction was detected after administration of control TT-specific CTL TT, demonstrating the antigen specificity of the anti-tumor effect. We also determined the impact of tumor-load on the outcome of CD4 + T cell therapy by treating mice carrying a three-fold higher tumor-load with high dose 3AB11 injections. Compared to the untreated group, 3AB11 treatment significantly delayed the MM growth, however no remission was reached (Fig. 1B), illustrating the importance of an optimal T cell-dose to tumor-load ratio for a successful treatment outcome. Chapter 6

98 98 Chapter 6 A Total tumor load per mouse (AU) No treatment CTL TT high dose 3AB11 low dose 3AB11 high dose *** *** ** * * B No treatment 3AB11 high dose * * * * Time (days) Time (days) Figure 1. In vivo anti-tumor effects of a native CD4 + T cell clone 3AB11 directed at a single mhag. (A) Rag2 -/- γc -/- mice with established UM9-luc-eGFP tumors in the bone marrow were depleted of macrophages using CL2MDP liposomes one day prior to either treatment with a single i.v. injection of 3AB11 ( cells/mouse, designated as low dose) (n=6; ) or with 3 consecutive i.v. injections of 3AB11 (3 injections of cells/mouse was designated as high dose) (n=5; ). Controls were untreated mice (n=4; ) and mice treated with high dose CTL TT (n=4; ). The curves represent the growth of UM9-luc-eGFP in the mice, measured by BLI of the ventral side (AU, arbitrary units represent the normalized photon emission counts). (B) Treatment efficacy of mice with a three times higher tumor-load as quantified by BLI. Mice were left untreated (n=5; ) or treated with high dose 3AB11 (n=7; ) one day after depletion of macrophages by CL2MDP liposome injections. In both graphs, T cell injections are depicted with arrows. Error bars represent the SEM. Differences compared to the untreated group were analyzed using t tests (*, P <0.05; **, P <0.01; ***, P <0.001). In vivo anti-tumor effects of dual antigen-specific TCR-transduced CD4 + T cells An attractive way to generate sufficient number of antigen-specific T cells for adoptive T cell therapy is the transfer of antigen-specific TCR into other T cells 22. We have previously shown that the TCR of clone 3AB11 can be successfully inserted in TT-specific CD4 + T cells, allowing efficient expansion via stimulation through the TT-specific TCR without altering the mhag-specific cytotoxic activity against the MM cell line UM9 21. Thus, we also investigated the feasibility of adoptive immunotherapy with dual-specific CD4 + T cells in vivo using a TCR transduced TT-specific oligoclonal T cell line (CTL TT -TCR), which displayed in vitro mhag-specific cytotoxic activity against UM9-luc-eGFP (Fig. 2A and B). Similar to the observations using the parental clone 3AB11, injection of a high dose of CTL TT -TCR into mice carrying established UM9-luc-eGFP tumors induced significant tumor reduction. The anti-tumor effect was mhag-specific, since no anti-tumor response was induced by injection of the untransduced CTL TT (Fig. 2C). These findings illustrated the feasibility and efficacy of adoptive immunotherapy solely with TT-specific T cells redirected at a single HLA class II-restricted tumor associated antigen.

99 Immunotherapy with mhag-specific CD4 + T-cells 99 A egfp CTL TT CTL TT -TCR C No treatment CTL TT high dose CTL TT -TCR high dose B Specific lysis (%) CTL TT CTL TT -TCR NGF-R Total tumor load per mouse (AU) *** *** ** *** 1:1 3:1 9:1 27:1 E:T ratio Time (days) Figure 2. In vitro and in vivo anti-tumor effects of dual TT- and mhag-specific TCR-transduced CD4 + T cells. (A) egfp and ΔNGF-R expression on the untransduced TT-specific cells CTL TT (left panel) and on TCR-transduced cells after FACS sorting (CTL TT -TCR; right panel). (B) Specific lysis of UM9-luc-eGFP by CTL TT ( ) or CTL TT -TCR ( ) at indicated effector to target (E:T) ratios after co-incubation for 26 hours. Results represent the mean percentage lysis and error bars indicate the SEM of triplicate wells. Similar results were obtained in two independent assays. (C) Rag2 -/- γc -/- mice with established UM9-luc-eGFP tumors in bone marrow were treated with high dose CTL TT (n=5; ) or CTL TT -TCR (n=6; ) or not treated (n=5; ) one day after CL2MDP liposome pretreatment. Arrows represent days of i.v. T cell injections. Mean and SEM of tumor photon emission are shown per group. Statistical differences were calculated compared to the untreated group (**, P <0.01; ***, P <0.001). In vivo boostability of long term persisting dual antigen-specific T cells A potential advantage of inserting therapeutic TCRs into TT-specific T cells is the possibility to boost these therapeutic T cells via their intrinsic TCRs. To test the feasibility of this idea we injected mice with TT-loaded EBV transformed B cells (EBV-LCLs; i.v.) 43 days after the initial CTL TT -TCR administration (Fig. 3A). Subcutaneous injections of TT-loaded or unloaded EBV-LCLs and UM9 cells ( ) were also carried out to visualize the possible infiltration of boosted T cells into these areas using in vivo fluorescence imaging (FLI), a technique sensitive enough for the tracking of α-hucd4-alexa-700 labeled T cells at these sites (Fig. S1). Nine days after the boost injections, clear accumulation of T cell-specific signals was detected at the s.c. injection sites of TT-loaded EBV-LCLs and UM9 cells but not of unloaded EBV-LCLs (Fig. 3B). This demonstrated the presence of a long term memory pool of dual-specific T cells that could be readily boosted via TT-specific TCRs and could home to tumor sites. Chapter 6 Challenges for tumor immunotherapy with T cells: extramedullary tumor outgrowth Clinical treatment of MM by DLI is often complicated by extramedullary relapses. Similar to this clinical scenario, long term monitoring of our mice successfully treated

100 100 Chapter 6 A 10 6 Tumor load (AU) UM9 CTL TT -TCR TT boost i.v. & s.c. FLI α-cd4-alexa-700 i.p Time (days) B Fluorescence Image (FLI) TT-loaded mhag EBV-LCL Unloaded mhag EBV-LCL mhag + UM9 TT-loaded mhag EBV-LCL Figure 3. In vivo boosting of dual antigenspecific T cells. (A) Overview of tumor growth and T cell (CTL TT -TCR) injections in the mouse depicted in (B). At day 57, TT-loaded EBV-LCLs ( ) were i.v. injected, and TT-loaded or unloaded EBV-LCLs and UM9 cells ( ) were s.c. injected at indicated locations (B). Nine days later, T cells in vivo labeled with α-hucd4-alexa-700 antibody were visualized on the dorsal side of anesthetized mice using FLI. Images are representative for three indepedent experiments with similar results. with high dose native mhag-specific CD4 + T cells revealed progressive outgrowth of MM, which seemed predominantly located outside the bone marrow (Fig. 4A). Confirming this observation, locus-specific quantification of the BLI data showed that the vast majority of original tumor loci in the bone marrow remained myeloma free for a period of at least five weeks, while extramedullary tumors were progressive Figure 4. Extramedullary tumor outgrowth is in vivo but not in vitro resistant to T cell mediated kill. (A) Representative examples of tumor outgrowth over time in untreated mice (lower panel) and mice treated with high dose 3AB11 (upper panel). Arrows indicate extramedullary tumor loci in the treated mice. Note the outgrowth of extramedullary tumor in treated mice, while the original tumor sites in the bone marrow remain tumor free. (B) Locus-specific photon emission was quantified for extramedullary sites ( ) and for the bone marrow ( ) in mice with recurrent tumor growth after treatment (treatment at indicated days). (C) 12wk after T cell treatment, mice were dissected to visualize extramedullary tumor burdens at luciferase positive locations. Shown pictures are representative examples for subcutaneous (upper panel) and ovaria (lower panel) tumor relapses, present in respectively 40% and 80% of treated mice. (D) Single cell suspensions of UM9-luc-eGFP tumors were derived from different treatment groups and used as targets for 3AB11 in a luciferase based killing assay. Results represent the mean lysis percentage of triplicate wells after 48 hours. Error bars indicate the SEM. Similar results were obtained in two independent assays. (E) Treatment of extramedullary relapses by additional injections of 3AB11. Established UM9-luc-eGFP tumors in bone marrow of Rag2 -/- γc -/- mice were treated with high dose 3AB11 (n=4; ). 22 days after primary treatment, the mice were again injected i.v. with high dose 3AB11. T cell injections are depicted by arrows. Mean and SEM of tumor photon emission are shown.

101 Immunotherapy with mhag-specific CD4+ T-cells 101 A 3AB11 (high dose) Untreated Time (weeks) B C Tumor load (AU/cm2) 104 Bone marrow Extramedullary ventral view 103 dorsal view Time (days) UM9-luc Target: cell line 40 CTLTT 3AB11 E 105 3AB11 high dose (2 treatments) 3AB11 Effector: Chapter 6 Tumor load per mouse (AU) CTLTT Effector: Specific lysis (%) D :1 5:1 E:T ratio 20: Time (days) 40 50

102 102 Chapter 6 (Fig. 4B). Dissection of sacrificed mice revealed that such tumors progressed mainly in ovaria and at s.c. sites (Fig. 4C). Since these results could reflect a (therapy-induced) resistance toward T cell mediated cytotoxicity, we dissected extramedullary tumors from treated and untreated mice and used their single cell suspensions as targets for 3AB11. However, all extramedullary tumor cells derived from both treated and untreated mice were efficiently killed by 3AB11 (Fig. 4D), ruling out the possibility of a (therapy-induced) resistance of extramedullary tumors. Thus, we attempted to treat the extramedullary recurrences occurring after a high dose treatment by additional injections of 3AB11 ~22 days after primary treatment. Yet, no visible tumor regression was detected (Fig. 4E), indicating that the CD4 + T cells were either not capable of infiltrating these extramedullary tumor sites or were inactivated in these sites by local regulatory mechanisms. Discussion We here show the achievement of potent in vivo anti-tumor effects by adoptive immunotherapy with sole CD4 + cytotoxic human T cells recognizing a single MMassociated mhag. We also demonstrate that similar potent anti-tumor effects can be achieved using recall antigen (TT)-specific CD4 + cytotoxic T cells that are genetically engineered to express the TCR recognizing the same single MM-associated mhag. Thus our results reveal for the first time the proof of principle of adoptive T cell therapy of a hematological malignancy with dual-specific CD4 + cytotoxic T cells. The dual-specific CD4 + T cells were not only able to reduce tumor-load, but also established a long term in vivo memory and could be readily boosted via their recall antigen-specific TCR, indicating the potential advantages of this strategy. Although, the impact of TT boosts on the anti-tumor effect in our model still needs to be determined, we consider these results as significant advances towards feasible and effective adoptive CD4 + immunotherapy with dual antigen-specific T cells. While several clinical and experimental studies demonstrated the importance of CD4 + T cells in anti-tumor immunity, a general entertained idea is that CD4 + T cells mediate their anti-tumor activities mainly via activation of and support to other immune cells 8-10, In addition to these helper functions, Th1 like CD4 + T cells often display cytotoxic activity in vitro and may contribute to anti-tumor effects via this effector function. In fact, our current data generated in a humanized murine model completely lacking NK, T and B cells and depleted of macrophages prior to T cell infusions are likely to reflect the direct cytotoxic activity of CD4 + T cells against tumor cells. Thus, with their cytotoxic effector functions, next to the well documented helper activities, native or TCR-transduced CD4 + T cells are highly valuable cellular tools for immunotherapy and therefore deserve to be included in clinical trials alone or in combination with other effector cells.

103 Immunotherapy with mhag-specific CD4 + T-cells 103 Our results also illustrate the crucial aspects of immunotherapy with such T cells: in our experiments consistent with previously published data from adoptive CD8 + T cell transfer approaches an optimal T cell-dose to tumor-load ratio was crucial for the successful therapeutic outcome. Increasing the T cell-dose and administration frequency may be associated not only with logistic drawbacks but also with undesired side effects. Therefore, the best way towards effective CD4 + and/or CD8 + T cell mediated tumor therapy seems by minimizing tumor-load by surgery, chemo and/or radiotherapy prior to T cell administration. Unlike several other tumor models, our bioluminescence model allows detection of individual tumor loci at various tissues. This enabled us to demonstrate a clear discrepancy between bone marrow and extramedullary tumor sites with respect to their response to CD4 + T cell therapy. While bone marrow tumors were highly susceptible to therapy, tumors progressed in the extramedullary sites despite additional CD4 + T cell treatments. Since T cells could home to extramedullary sites (see Figure 3) and since extramedullary tumors showed no ex vivo resistance toward T cell mediated lysis (see Figure 4), likely possibilities for tumor escape in the extramedullary sites are the impaired infiltration of T cells into these sites or the inactivation of T cells due to local secretion of inhibitory factors. Such an inhibitory factor could be the vascular endothelial growth factor (VEGF), because its expression in ovarian cancer shows an inverse correlation with the number of T cells infiltrating into the tumor microenvironment and it inhibits T cell effector functions 30,31. In fact many other soluble factors such as TGFβ, PGE2 and IL10 as well as inhibitory ligands such as PD-L1/2, CTLA-4 ligands, FASL and TRAIL have been shown to contribute to T cell suppression within the tumor microenvironment To our knowledge, a tissue-specific regulation of anti-tumor immunity has not been reported yet. Thus detailed investigation of the underlying mechanisms of extramedullary MM outgrowth in our model can provide new insights into tissue-specific regulatory mechanisms and seems also relevant for clinical trials, since the relapse of MM at extramedullary sites is a known complication in patients treated with DLI In conclusion, the effective clearance of bone marrow tumors with native and TCRtransduced CD4 + T cells directed at a single minor H antigen indicates that CD4 + T cells are suitable tools for the treatment of marrow-born hematological tumors and encourages the evaluation of their immunotherapeutic potency in clinical phase I/ II trials. Chapter 6 Acknowledgements We thank Louis van Bloois of the Department of Pharmaceutics, University of Utrecht, for generating CL2MDP liposomes.

104 104 Chapter 6 METHODS Cells The human HLA-DPB1*0401-restricted mhag-specific CD4 + T cell clone 3AB11 and TTspecific CD4 + T cell clones were previously described in detail 21,39. Clone 3AB11 recognizes a yet unknown mhag with hematopoietic restricted tissue distribution 39. The oligoclonal TT-specific cell line, CTL TT, was established by mixing four TT-specific CD4 + T cell clones (N TT 3AC6, N TT 3AG10, N TT 1A3, N TT 1E10) in equal proportions. All T cells were expanded using a feeder cell-cytokine mixture as described 21. Luciferase transduced human MM cell line UM9-luc-eGFP was described elsewhere 40. UM9-luc-eGFP and EBV-LCLs were cultured in RPMI-1640 (Invitrogen) supplemented with 10% FBS (Integro) and antibiotics. Retroviral vectors and transduction of T cells The retroviral pmx vectors, TCR α-ires-δngf-r and TCR β-ires-egfp, carrying the TCR α and -β chains of clone 3AB11 were described previously 21. The generation of retroviral supernatants, the retroviral transductions of TT-specific T cells and the FACS sorting (BD) based on egfp and ΔNGF-R expression were also described 21. Mice Rag2 -/- gc -/- mice were bred and housed Specified-Pathogen-Free at the Central Animal Facility of the University of Utrecht 41. All animal experiments were conducted according to the Dutch Law on Animal Experiments with permission from the local Ethical Committee for Animal Experimentation. Transplantation and in vivo monitoring of tumor cells Female mice, 9-14wk of age received UM9-luc-eGFP cells via the tail vein one day after sublethal irradiation (350 cgy). Tumor growth was quantitatively monitored after i.p. injection of beetle luciferin (2.5 mg; Promega) by either Biospace (Biospace Lab, France) or Roper (Roper Scientific, USA) bioluminescence imaging (BLI) systems 40. Arbitrary photon counts of both imaging systems were normalized for comparison of different experiments. In untreated mice, the UM9-luc-eGFP cells grow as a typical MM with early phase growth in bone marrow followed by some metastatic growth at extramedullary foci from 8 weeks on. Adoptive transfer and in vivo monitoring of T cells At different levels of tumor intensity, mice were macrophage depleted by i.v. injection of fresh 2-chloromethyl biphosphonate (CL2MDP) liposomes as described previously 42. One, three and six days later mice were i.v. injected with CD4 + mhag-specific T cell clone 3AB11, TCR transduced CTL TT (CTL TT -TCR) or the parental control cell line CTL TT. In vivo monitoring of CTL TT -TCR in the mice was carried out by fluorescence imaging (FLI; Biospace Lab, France) up to seven days following i.p. injection with 50 µg α-hucd4-alexa-700 (ITK, The Netherlands). Two days prior to antibody injection some mice were boosted via injection of TT-loaded EBV-LCLs ( iv, sc). Cytotoxicity assay UM9-luc-eGFP-derived tumors were dissected from various foci of sacrificed mice. Single cell suspensions of these tumors were then used as target cells in ex vivo luciferase-based cytotoxicity assays in white opaque flat bottom 96-well plates (Costar). Effector CD4 + T cells were added at different effector to target (E:T) ratios in the presence of 125 µg/ml beetle

105 Immunotherapy with mhag-specific CD4 + T-cells 105 luciferin (Promega). At 26 and 48 hours of culture, the light signal emitted from surviving UM9- luc-egfp cells was measured using a luminometer (Molecular Devices). The percentage lysis was calculated relative to medium control as described 43. Tumor cells from BM and from some extramedullary lesions could not be subjected to cytotoxicity assays as they did not survive a few hours of ex vivo culture. Statistical analyses Unpaired two-tailed student s t tests were used to analyze differences between groups (GraphPad Prism software). P-values <0.05 were considered significant. Chapter 6

106 106 Chapter 6 References 1. Kolb HJ. Graft-versus-leukemia effects of transplantation and donor lymphocytes. Blood. 2008;112: Bevan MJ. Helping the CD8 + T cell response. Nat Rev Immunol. 2004;4: Shedlock DJ, Shen H. Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science. 2003;300: Bourgeois C, Veiga-Fernandes H, Joret AM, Rocha B, Tanchot C. CD8 lethargy in the absence of CD4 help. Eur J Immunol. 2002;32: Surman DR, Dudley ME, Overwijk WW, Restifo NP. Cutting edge: CD4 + T cell control of CD8 + T cell reactivity to a model tumor antigen. J Immunol. 2000;164: Nishimura T, Iwakabe K, Sekimoto M et al. Distinct role of antigen-specific T helper type 1 (Th1) and Th2 cells in tumor eradication in vivo. J Exp Med. 1999;190: Kamiryo Y, Eto M, Yamada H et al. Donor CD4 T cells are critical in allogeneic stem cell transplantation against murine solid tumor. Cancer Res. 2009;69: Perez-Diez A, Joncker NT, Choi K et al. CD4 cells can be more efficient at tumor rejection than CD8 cells. Blood. 2007;109: Hung K, Hayashi R, Lafond-Walker A et al. The central role of CD4 + T cells in the antitumor immune response. J Exp Med. 1998;188: Ossendorp F, Mengede E, Camps M, Filius R, Melief CJ. Specific T helper cell requirement for optimal induction of cytotoxic T lymphocytes against major histocompatibility complex class II negative tumors. J Exp Med. 1998;187: Muranski P, Restifo NP. Adoptive immunotherapy of cancer using CD4 + T cells. Curr Opin Immunol. 2009;21: Baron F, Siquet J, Schaaf-Lafontaine N et al. Pre-emptive immunotherapy with CD8-depleted donor lymphocytes after CD34-selected allogeneic peripheral blood stem cell transplantation. Hematologica. 2002;87: Champlin R, Ho W, Gajewski J et al. Selective depletion of CD8 + T lymphocytes for prevention of graft-versus-host disease after allogeneic bone marrow transplantation. Blood. 1990;76: Champlin R, Jansen J, Ho W et al. Retention of graft-versus-leukemia using selective depletion of CD8-positive T lymphocytes for prevention of graft-versus-host disease following bone marrow transplantation for chronic myelogenous leukemia. Transplant Proc. 1991;23: Nimer SD, Giorgi J, Gajewski JL et al. Selective depletion of CD8 + cells for prevention of graft-versus-host disease after bone marrow transplantation. A randomized controlled trial. Transplantation. 1994;57: Soiffer RJ, Alyea EP, Hochberg E et al. Randomized trial of CD8 + T cell depletion in the prevention of graft-versushost disease associated with donor lymphocyte infusion. Biol Blood Marrow Transplant. 2002;8: Liu Z, Noh HS, Chen J et al. Potent tumor-specific protection ignited by adoptively transferred CD4 + T cells. J Immunol. 2008;181: Ostrand-Rosenberg S. CD4 + T lymphocytes: a critical component of antitumor immunity. Cancer Invest. 2005;23: Melief CJ, Van Der Burg SH. Immunotherapy of established (pre)malignant disease by synthetic long peptide vaccines. Nat Rev Cancer. 2008;8: Spaapen RM, Lokhorst HM, van den Oudenalder K et al. 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. 2008;205: Spaapen R, van den Oudenalder K, Ivanov R et al. Rebuilding Human Leukocyte Antigen Class II-Restricted Minor Histocompatibility Antigen Specificity in Recall Antigen-Specific T Cells by Adoptive T Cell Receptor Transfer: Implications for Adoptive Immunotherapy. Clin Cancer Res. 2007;13: Morgan RA, Dudley ME, Wunderlich JR et al. Cancer regression in patients after transfer of genetically engineered lymphocytes. Science. 2006;314: Symons HJ, Levy MY, Wang J et al. The allogeneic effect revisited: exogenous help for endogenous, tumor-specific T cells. Biol Blood Marrow Transplant. 2008;14: De Veerman M, Heirman C, Van MS et al. Retrovirally transduced bone marrow-derived dendritic cells require CD4 + T cell help to elicit protective and therapeutic antitumor immunity. J Immunol. 1999;162: Chakraverty R, Eom HS, Sachs J et al. Host MHC class II+ antigen-presenting cells and CD4 cells are required for CD8-mediated graft-versus-leukemia responses following delayed donor leukocyte infusions. Blood. 2006;108:

107 Immunotherapy with mhag-specific CD4 + T-cells Corthay A, Skovseth DK, Lundin KU et al. Primary antitumor immune response mediated by CD4 + T cells. Immunity. 2005;22: Abad JD, Wrzensinski C, Overwijk W et al. T cell receptor gene therapy of established tumors in a murine melanoma model. J Immunother. 2008;31: Hambach L, Nijmeijer BA, Aghai Z et al. Human cytotoxic T lymphocytes specific for a single minor histocompatibility antigen HA-1 are effective against human lymphoblastic leukaemia in NOD/scid mice. Leukemia. 2006;20: Hambach L, Vermeij M, Buser A et al. Targeting a single mismatched minor histocompatibility antigen with tumorrestricted expression eradicates human solid tumors. Blood. 2008;112: McArdle SE, Rees RC. Tumor immunology: new perspectives. Cancer Res. 2009;69: Mulligan JK, Day TA, Gillespie MB, Rosenzweig SA, Young MR. Secretion of vascular endothelial growth factor by oral squamous cell carcinoma cells skews endothelial cells to suppress T cell functions. Hum Immunol. 2009;70: Smyth MJ, Godfrey DI, Trapani JA. A fresh look at tumor immunosurveillance and immunotherapy. Nat Immunol. 2001;2: Gajewski TF, Meng Y, Blank C et al. Immune resistance orchestrated by the tumor microenvironment. Immunol Rev. 2006;213: Blank C, Gajewski TF, Mackensen A. Interaction of PD-L1 on tumor cells with PD-1 on tumor-specific T cells as a mechanism of immune evasion: implications for tumor immunotherapy. Cancer Immunol Immunother. 2005;54: Zomas A, Stefanoudaki K, Fisfis M, Papadaki T, Mehta J. Graft-versus-myeloma after donor leukocyte infusion: maintenance of marrow remission but extramedullary relapse with plasmacytomas. Bone Marrow Transplant. 1998;21: Minnema MC, van de Donk NW, Zweegman S et al. Extramedullary relapses after allogeneic non-myeloablative stem cell transplantation in multiple myeloma patients do not negatively affect treatment outcome. Bone Marrow Transplant. 2008;41: Terpos E, Rezvani K, Basu S et al. Plasmacytoma relapses in the absence of systemic progression post-high-dose therapy for multiple myeloma. Eur J Haematol. 2005;75: Zeiser R, Deschler B, Bertz H, Finke J, Engelhardt M. Extramedullary vs medullary relapse after autologous or allogeneic hematopoietic stem cell transplantation (HSCT) in multiple myeloma (MM) and its correlation to clinical outcome. Bone Marrow Transplant. 2004;34: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Rozemuller H, van der Spek E, Bogers-Boer LH et al. A bioluminescence imaging based in vivo model for preclinical testing of novel cellular immunotherapy strategies to improve the graft-versus-myeloma effect. Hematologica. 2008;93: Weijer K, Uittenbogaart CH, Voordouw A et al. Intrathymic and extrathymic development of human plasmacytoid dendritic cell precursors in vivo. Blood. 2002;99: van Rijn RS, Simonetti ER, Hagenbeek A et al. A new xenograft model for graft-versus-host disease by intravenous transfer of human peripheral blood mononuclear cells in Rag2-/- gammac-/- double-mutant mice. Blood. 2003;102: Brown CE, Wright CL, Naranjo A et al. Biophotonic cytotoxicity assay for high-throughput screening of cytolytic killing. J Immunol Methods. 2005;297: Chapter 6

108 108 Chapter 6 Supplemental data PBS T cells 1*10 6 T cells 3*10 6 T cells 10*10 6 Figure S1. In vivo FLI for visualization of s.c. located T cells. T cells (CTL TT -TCR) were s.c. injected at indicated locations on the dorsal side of Rag2 -/- γc -/- mice (n=2). PBS was injected as control. The T cells were then in vivo labeled via i.p. injection of α-hucd4- ALEXA-700 antibody. After one day, the T cells were visualized in anesthetized mice using FLI.

109 Immunotherapy with mhag-specific CD4 + T-cells 109 Chapter 6

110 110 Chapter 6

111 General discussion 7

112 112 Chapter 7 Every year, around 700,000 new cases of leukemia, lymphoma, and multiple myeloma are diagnosed worldwide. These hematological malignancies together account for about 500,000 deaths per year 1. Although these cancers can be temporary eradicated by standard treatments including chemotherapeutic agents, HLA-matched allogeneic stem cell transplantation (allo-sct) is the only curative option for a considerable percentage of patients 2-5. The therapeutic effect of this treatment is mainly induced by donor T cells directed at minor histocompatibility antigens (mhags) that are expressed on the tumor cells. Merely 15 years ago, the biochemical nature of mhags was identified as peptides presented by HLA molecules on the cell surface towards T cells. These peptides are derived from cellular proteins that are polymorphic between donor and patient, causing that the patient-derived peptides are recognized as foreign by donor T cells. The polymorphic character of such a mhag-encoding protein is always caused by a polymorphism in the DNA, usually a single nucleotide polymorphism (SNP). Unfortunately, next to the powerful graft-versus-tumor (GvT) effect, allo-sct is also associated with a transplant-related mortality and morbidity due to acute and chronic graft-versus-host disease (GvHD). Like GvT, GvHD is also caused by donor T cells specific for mhags. In contrast to GvT associated mhags that are solely presented by tumor and hematopoietic cells to the donor T cells, GvHD associated mhags are (also) presented by vital non-hematopoietic cells. Therefore, specific targeting of hematopoietic and tumor expressed mhags seems to be an ideal strategy to separate the GvT from GvHD after allo-sct. An important step towards separating GvT from GvHD using immunotherapeutic intervention strategies is the identification of mhag-encoding sequences with a clinical relevance. Therefore, we describe in this chapter our definition of clinically relevant mhags, the identification of such mhags, and the pre-clinical progress towards mhag-based immunotherapy. Clinically relevant mhags: high population coverage and hematopoietic expression Over the last 15 years, more than 30 polymorphic proteins encoding for one or more mhags were identified (Table 1) Of these mhags, only few can be nominated as clinically relevant, meaning that they can be used in a broad group of patients as immunotherapeutic tools. There are two major factors that determine whether a mhag is broadly applicable in the clinic. First, targeting a mhag is only relevant for a patient if specifically the tumor is attacked (GvT effect) but not the normal tissues (GvHD), and second, clinical applicability naturally depends on the percentage of mhag + patients with a suitable donor. In Chapters 2, 3 and 4 of this thesis, we extensively discuss that these properties are crucial for the inclusion of mhags in immunotherapeutic strategies.

113 General discussion 113 Clinically relevant tissue distribution of mhags While several mhag-encoding proteins are expressed ubiquitously, some are exclusively expressed by hematopoietic cells and their malignant counterparts (Table 1) 2,50,51. Such tissue-specific expression profiles suggest that after allo-sct donor T cells directed at ubiquitous mhags can directly induce GvT and GvHD, and T cells specific for hematopoietic mhags can induce GvT (without GvHD) via direct antigen dependent tumor lysis. Thus, an immunotherapeutic approach targeting only hematopoietic mhags seems ideal for specific induction of GvT effects without the occurrence of detrimental GvHD (Chapter 1). Hence, many researchers have deliberately focused on identifying hematopoietic mhags by pre-selection of T cell clones based on their ability to recognize hematopoietic and tumor cells, but not fibroblasts 50, Moreover, some investigators tend to consider mhags as clinically relevant only based on hematopoietic restricted T cell reactivity in such in vitro tissue distribution assays. However, we think that this may be a premature and risky conclusion, because 1) it is almost impossible to include all GvHD tissues in such a screening, and 2) the functional expression of non-hematopoietic mhags may be greatly influenced by in vivo environmental factors like inflammation or by use of therapeutic agents 55. Therefore, we only consider a mhag as hematopoietic, if after the actual identification it is shown to be restricted to the hematopoietic compartment at the molecular level. Still a limited tissue screening can be useful as a first crude selection of T cell clones, because subsequent classical mhag identification methods can be very expensive and laborious. However, considering the speed and ease of our new mhag identification strategy (Chapter 3), a prescreening may be omitted because the generation and testing of tissue-derived cell lines also requires a considerable investment. Furthermore, such a crude pre-screening for hematopoietic specificity may be even a larger investment than identification of additional T cell clones using our strategy, since it allows for simultaneous identification of multiple mhags recognized by T cell clones with the same HLA restriction. As discussed in Chapter 4 and below, a reverse identification strategy completely circumvents tissue screening and is therefore in theory the most efficient methodology towards identification of hematopoietic mhags. Clinically relevant mhag and HLA -frequencies Broad clinical applicability of any given mhag is also dependent on the population frequencies of both the particular mhag polymorphism and the corresponding HLA molecule presenting the mhag. An ideal mhag is positive in many patients, and negative in their HLA-matched donors, because a large population of patients will then be eligible for mhag-based immunotherapy after allo-sct. However, for several mhags identified so far, the mhag population frequencies are unfavorable (e.g. >85% or <15%), attributing only little clinical value to these mhags (Table 1). To elucidate the boundaries of clinical relevancy, in Chapter 2 we calculate to what extent population frequencies influence the eligibility of patients for mhag-based Chapter 7

114 114 Chapter 7 Table 1. Currently known mhags ordered by clinical relevancy, determined by tissue distribution, mhag frequency and HLA restriction. Patients with mhag Gene Tissue Polymorphism expression ab mhag frequency HLA restriction HLA Patients frequency c (%) donor (%) d Source HER-2/NEU HER2 hematopoietic rs ,3 A* ,0 25,1 6,40 ref 6 HA-1 HMHA1 hematopoietic rs ,8 A* ,0 27,6 6,32 ref 7 CD19 CD19 hematopoietic rs ,4 DQB1* ,9 18,1 4,61 ref 8,Chapter 2 HB-1Y HB-1 hematopoietic rs ,3 B*4402/03 27,6 12,8 3,70 ref 9 LRH-1 P2X5 hematopoietic rs ,0 B* ,7 13,9 3,49 ref 10 ACC6 HMSD hematopoietic rs ,3 B*4402/03 27,6 10,0 3,37 ref 11 ACC1Y BCL2A1 hematopoietic rs ,5 A* ,5 7,2 2,08 ref 12 SP110 SP110 hematopoietic rs ,3 A* ,3 22,4 2,01 ref 13 LB-PTK2B-1T PTK2B hematopoietic rs ,0 DRB3* ,8 10,4 1,78 ref 14 HA-2 MYO1G hematopoietic SNP 94,5 A* ,0 44,4 1,60 ref 15,16 HA-1 HMHA1 hematopoietic rs ,8 B*4001 9,6 5,7 1,29 ref 17 ACC2 BCL2A1 hematopoietic rs ,5 B*4403 9,5 4,2 1,26 ref 12 PANE1 PANE1 hematopoietic rs ,6 A* ,3 24,3 1,16 ref 18 LB-LY75-1K LY75 hematopoietic rs ,0 DRB1*1301 8,8 2,9 1,03 ref 14 HB-1H HB-1 hematopoietic rs ,8 B*4402/03 27,6 26,2 0,90 ref 19 ACC1C BCL2A1 hematopoietic rs ,0 A* ,5 14,4 0,65 ref 20 HA-1 HMHA1 hematopoietic rs ,8 A*0206 0,2 0,1 0,02 ref 21 HY (SMCY) SMCY ubiquitous Y 50,0 A* ,0 23,5 6,38 ref 22 HY (UTY) UTY ubiquitous Y 50,0 A* ,0 23,5 6,38 ref 23 C19ORF48 C19ORF48 ubiquitous rs ,3 A* ,0 27,4 6,33 ref 24 LB-ADIR-1F ADIR ubiquitous rs ,0 A* ,0 17,4 5,79 ref 25 HA-8 KIAA0020 ubiquitous rs ,2 A*0201/02 47,1 33,1 5,65 ref 26 HY (DBY) DBY ubiquitous Y 50,0 DQB1*0501/02 30,9 15,5 4,20 ref 27 HY (DFFRY) DFFRY ubiquitous Y 50,0 A* ,1 15,1 4,09 ref 28,29 HY (SMCY) SMCY ubiquitous Y 50,0 B* ,7 12,8 3,48 ref 30 LB-MTHFD1-1Q MTHFD1 ubiquitous rs ,0 DRB1* ,1 17,8 3,24 ref 14 HY (UTY) UTY ubiquitous Y 50,0 B* ,2 11,1 3,01 ref 31

115 General discussion 115 SLC19A1 SLC19A1 ubiquitous rs ,0 DRB1* ,2 23,7 2,69 ref 32,Chapter 3 LB-MR1-1R MR1 ubiquitous rs ,0 DRB3* ,0 6,0 2,33 ref 14 HA-3 AKAP13 ubiquitous rs ,0 A* ,1 26,2 2,10 ref 33 HY (RPS4Y) RPS4Y ubiquitous Y 50,0 DRB3* ,1 6,6 1,78 ref 34 LB-PI4K2B-1S PI4K2B ubiquitous rs ,1 DQB1* ,5 11,3 1,77 ref 35 LB-ECGF-1 ECGF1 ubiquitous SNP 11,4 B* ,7 2,9 1,32 ref 36 HY (UTY) UTY ubiquitous Y 50,0 B*4001 9,6 4,8 1,31 ref 37 HY (DDX3Y) DBY ubiquitous Y 50,0 B*2705 5,5 2,8 0,75 ref 38 TRIM22 TRIM22 ubiquitous SNP 2,7 A* ,0 1,3 0,62 ref 39 ACC5 CTSH ubiquitous rs ,0 A*3101 5,2 0,6 0,26 ref 40 UGT2B17 UGT2B17 ubiquitous CNV 96,5 B*4403 9,5 9,2 0,22 ref 41 UGT2B17 UGT2B17 ubiquitous CNV 96,5 A*2902 7,4 7,1 0,17 ref 42 HY (UTY) UTY ubiquitous Y 50,0 B*5201 1,2 0,6 0,16 ref 43 HY (RPS4Y) RPS4Y ubiquitous Y 50,0 B*5201 1,2 0,6 0,16 ref 44 SLC1A5 SLC5A1 ubiquitous rs ~25,0 B*4002 1,6 0,4 0,15 ref 45 HY (TMSB4Y) TMSB4Y ubiquitous Y 50,0 A*3303 0,8 0,4 0,10 ref 46 ACC4 CTSH ubiquitous rs ,0 A*3303 0,8 0,1 0,04 ref 40 UGT2B17 UGT2B17 ubiquitous CNV 96,5 A*0206 0,2 0,2 0,00 ref 45 a Hematopoietic: gene is expressed in at least one hematopoietic lineage and not in non-hematopoietic tissues under non-inflammatory conditions ubiquitous: gene is expressed in at least one non-hematopoietic tissue b frequencies in the Caucasoid population are derived from (a combination of) HapMap, NCBI, reference 47, or the original reference c DRB3-frequencies were derived from reference 48; other HLA frequencies are in the European population and derived from reference 49 d Mismatch percentages were calculated using the formulas described in Chapter 2 Abbreviations: CNV, copy number variation; SNP, single nucleotide polymorphism without rs-number; Y, derived from the Y-chromosome Chapter 7

116 116 Chapter 7 therapy. Our calculations show that if the mhag population frequency is between 26% and 78%, the chance of a mhag-mismatch between a stem cell recipient and his HLA-matched sibling donor is more than 10%. Since the number of therapy eligible patients is also limited by HLA restriction, we consider mhag frequencies beyond these boundaries (i.e. <26% and >78%) of little value for clinical application. Indeed, although several known mhags have a suitable frequency, they do not have significant clinical value because they are presented by infrequent HLA molecules (Table 1). Thus, the ultimate strategy towards broad application of mhags in the future is to aim at the identification of mhags restricted to HLA elements with a high population coverage. Therefore we also address the most efficient combinations of HLA-alleles leading to high population coverage in Chapter 3. The best strategy seems focusing on identification of only mhags presented by the most frequent HLA-A, -B, -DP, -DQ and -DR alleles. However, combination of different classes of HLA molecules is definitely not an efficient strategy. In contrast, the best strategy to reach a population coverage of >92.5% with a minimal number of HLA class I- or II-alleles is the combination of respectively HLA-A1, -A2, -A3, -A11, -A24, or only HLA-DP2 and -DP4 alleles. Thus concentrating new mhag identification efforts on these alleles seems the most efficient way to broaden the clinical applicability of mhag-based immunotherapy. Methods for efficient identification of clinically relevant mhags It is clear that the current set of identified clinically relevant mhags is not large enough to apply mhag-based immunotherapy in a broad cohort of patients (Table 1). Thus identification of novel clinically relevant mhags is required. As discussed in Chapter 3, more than 40 new HLA class I- or 24 class II-restricted mhags are needed to have at least one mhag-mismatch in a reasonable population (>75%) after allo-sct from an HLA-matched unrelated donor. Basically, identification of novel mhags can be performed using classical forward strategies such as our newly introduced forward strategy (Chapters 2 and 3), or using reverse strategies (Chapter 4). Forward strategies aim at the identification of mhags recognized by existing T cell clones, while the reverse strategy uses algorithms to predict putative mhags that still need verification via generation of a specific T cell response (Fig. 1). Classical forward strategies: successful, but time-consuming During the last fifteen years, several forward strategies were introduced for mhag identification. The first mhags were identified using high performance liquid chromatography (HPLC) fractionated peptide pools eluted from mhag + cells. A final peptide fraction loaded on mhag cells that was still recognized by a mhagspecific T cell clone was sequenced using mass spectrometry 15,30. Although very elegant, these separation and identification techniques require highly specialized skills, only mastered by a few laboratories in the world. Therefore, other laboratories utilized cdna library pools derived from mhag + cells that were transfected in mhag

117 General discussion 117 PBMC from patient after allo-sct T cell specific for unknown mhag mhag identification Identified mhag Forward identification Reverse identification Epitope prediction algorithms combined with databases Putative mhag PBMC from donor or allo-sct patient T cell specific for mhag Figure 1. Schematic overview of the principles and workflow of a forward versus a reverse mhag identification strategy. Forward strategies aim at the identification of mhags recognized by existing T cell clones, while the reverse strategy focuses at the identification of T cell clones recognizing predicted mhags. cells as screening target for mhag-specific T cells 19,42. And a variant of this cdna library screening was used specifically for the identification of HLA class II-restricted mhags 14,35. Furthermore pairwise genetic linkage analyses or whole genome association scans were used to determine the genomic location of mhags 10,12,20. However, also cdna library screening and conventional genetic analyses are timeconsuming and require many appropriate laboratory skills. Zygosity-genotype correlation analysis: rapid and accessible forward identification To tackle the speed and specialty issues of all previous forward strategies, we introduce in Chapters 2 and 3 a new approach based on genome-wide zygositygenotype correlation analyses and utilizing EBV transformed B cell lines (EBV- LCLs) that are extensively SNP-genotyped by the International HapMap Project. This approach is not only generally accessible and very rapid (3 months from T cell to mhag; Chapter 3), but also guarantees a success-rate of about 90% to identify mhags with a frequency between 10-85% (Chapters 2 and 3). Obviously these limits fit well with the defined boundaries for the clinically relevant mhags mentioned above. Furthermore, the method allows analysis of multiple T cell clones simultaneously towards identification of several mhags in a short time frame. Thus, the zygosity-genotype correlation strategy is probably the most appropriate approach developed until now for the high-speed identification of clinically relevant mhags. Chapter 7

118 118 Chapter 7 The reverse strategy is in theory ideal for clinically relevant mhag identification Although high-speed mhag identification is now possible, forward strategies are by definition not the most ideal strategies to identify clinically relevant mhags, because the parameters defining clinical relevancy can not be controlled when generating mhag-specific T cell clones from allo-sct patients. Indeed, despite major investments, several mhags identified using any forward method do not have a clinically relevant frequency, HLA restriction or tissue distribution. Towards forward identification of mhags with clinical relevancy, the following strategic steps may be taken. It is obvious to generate T cell clones only from patients with preferable HLA types. After generation, T cell clones may be selected for a favorable mhag frequency determined by testing against a panel of unrelated EBV- LCLs (preferably included in the HapMap project). But still, as described above, the tissue distribution of a mhag can not be conclusively determined prior to a forward identification procedure. These disadvantages can be overcome with the reverse identification strategy. With the help of prediction algorithms for intracellular antigen processing, putative mhags can be identified in genes with nonsynonymous coding polymorphisms (Chapter 4). Since these predictions can be executed for any given polymorphic gene, a putative mhag will be already clinically relevant if the input gene is hematopoietic restricted. Furthermore, since for many SNPs frequencies are now available in public databases, it is possible to select only those polymorphisms with appropriate frequency. Finally, predictions can be executed for each desired HLAallele. Thus in theory, the reverse strategy is ideal for the identification of clinically relevant mhags. Many putative mhags were previously predicted by several researchers, all assuming that any peptide sequence containing a nonsynonymous polymorphism which is properly processed by the intracellular antigen processing machinery is a putative mhag In Chapter 4, we describe a different prediction strategy that distinguishes itself from current mhag prediction approaches, as it selects putative mhags based only on differential antigen processing induced by a nonsynonymous polymorphism. As discussed there, we believe that our more convergent mhag prediction strategy which actually is centered around the differential processing origin of the majority of current mhags allows focusing on only a few, but the most likely mhag candidates. Nonetheless, the power and efficacy of this method still needs to be proven, because the actual bottleneck of reverse immunology strategy appears to be the subsequent generation of mhag-specific T cells from PBMCs of allo-sct patients or healthy donors. Therefore, only little success is achieved using the reverse approach to date 6,9,17, despite intense efforts of different groups (unpublished communications). We and others generated some high-affinity T cell lines specific for predicted mhags, but unfortunately the T cells were not capable of recognizing naturally processed mhag-peptides (unpublished data and

119 General discussion 119 communications). Thus, more efficient and rapid methods are required to screen the immunogenicity of predicted putative mhags. Therefore future research will reveal whether the reverse identification strategy is in practice as ideal as in theory. Guided mhag identification towards broad clinical applicability To summarize, it is evident that the previous more or less random identification of mhags paying not enough attention to frequency or HLA restriction was not the most straight forward strategy to gain population coverage and broaden clinical applicability. Therefore, future mhag identification efforts should focus on mhags with population frequencies between 26% and 78% and presented on HLA-A1, -A2, -A3, -A11, -A24, -DP2 and -DP4 alleles, with the goal to reach a maximal coverage of patients. Since the forward identification strategies do not allow for identification of mhags solely with a hematopoietic tissue distribution, reverse identification seems a better approach. But as long as the drawbacks of T cell generation for reverse mhag identification are not tackled, the new genetic strategy described in this thesis can be used for high-throughput identification of new mhags towards broad applicability of mhag-based immunotherapy. HLA class II-restricted mhags: powerful tools for immunotherapy CD4 + and CD8 + T cells specific for mhags are potent mediators of anti-tumor effects, because of their highly immunogenic alloreactive character 2-5. As described in Chapter 1, there are different strategies for mhag-based immunotherapy: DNA, peptide or dendritic cell vaccination or adoptive T cell transfer. The latter is currently poorly implemented in clinical protocols because of major difficulties to generate large numbers of mhag-specific T cells from healthy donors. For both vaccination and adoptive transfer strategies, investigators focus mainly on CD8 + T cells because these cells are considered as the most efficient effector cells of anti-tumor immunity. Minor histocompatibility antigen-specific CD4 + T cells have never been tested as primary and sole therapy in preclinical models, mainly because CD4 + T cells are generally believed to function as mediators of anti-tumor effects via cellcell contact and cytokine-mediated support of other immune cells, like CD8 + T cells, macrophages or NK cells In this thesis, we however focus more on CD4 + T cells because our previous studies revealed that several patients responding to allo-sct generated a sizeable CD4 + T cell response against mhags and these T cells could be readily isolated and cloned 53,66. B cell-specific CD19 encodes for a clinically relevant HLA class II-restricted mhag In order to address the clinical impact of HLA class II-restricted mhags in this thesis, we have given specific attention for the identification of such mhags. Employing our new methodology, we have been able to identify the first genuine hematopoietic mhag restricted by an HLA class II molecule. This antigen is derived from the B cell-specific CD19 protein which is expressed in almost all B cell malignancies, Chapter 7

120 120 Chapter 7 except multiple myeloma. Furthermore, with respect to population frequency and HLA restriction, this new mhag fulfills all requirements to be a clinically relevant mhag (Table 1). The CD4 + T cell clones recognizing this new antigen are cytotoxic against CD19 L expressing malignant cells, suggesting that this new HLA class II-restricted antigen has a high therapeutic potential. Next, we identified another HLA class II-restricted mhag encoded by the SLC19A1 gene. However, although we never encountered recognition of non-hematopoietic targets by the SLC19A1- specific T cells, we did not prove that SLC19A1 is hematopoietic restricted at a molecular level. Moreover, public databases indicate a moderate expression of this gene in liver and lung tissues, hence we did not test these tissues in further detail. Preclinical exploration of mhag-specific CD4 + T cells: potent anti-tumor effects To explore the clinical impact of CD4 + T cells we tested another highly interesting T cell clone, 3AB11, in a preclinical adoptive T cell approach. By adopting novel strategies to generate large cell numbers required for adoptive immunotherapy described for CD8 + T cell clones 67, we rebuilded the mhag specificity of this T cell clone into tetanus toxoid (TT)-specific T cells using T cell receptor (TCR) transfer (Chapter 5). We preferred to use TT-specific cells as TCR targets, since in vivo administered T cells can be boosted in a controlled fashion using TT injections, in contrast to CMV-specific T cells that were studied by other investigators 67. Indeed, we showed that the TCR-redirected CD4 + T cells were still expandable to large cell numbers via triggering their TT-specific TCR, while maintaining their anti-tumor cytotoxic capacity (Chapter 5). Chapter 6 shows that both original, and more importantly TCR-redirected dualspecific CD4 + T cells can in addition to a well-described helper function kill bone marrow tumors effectively in a direct fashion in vivo. Also very fascinating is that we targeted only a single antigen with these CD4 + T cells, suggesting that adoptive immunotherapy targeting even a sole HLA class II-restricted mhag may be effective in patients with marrow born tumors. Although CD4 + T cells received only little attention in the past, the tendency is currently shifting towards these intriguing cells as important additions to current CD8 + adoptive strategies. The results presented in Chapter 5 and 6 contribute to the understanding of the potency of mhag-specific CD4 + T cells and to the development of such adoptive strategies. Future perspectives for CD4 + mhag-specific T cells All our in vitro and in vivo experiments with the mhag-specific CD4 + T cell clone 3AB11, which recognizes only hematopoietic cells, indicate its clinical relevancy. Nonetheless, the gene encoding for this mhag is yet unknown. Thus a highly important future objective will be the identification of the mhag recognized by 3AB11 using our new zygosity-genotype correlation analysis. Our preliminary genetic experiments indicate that the antigen is most probably encoded by a gene

121 General discussion 121 on chromosome 9. It should be noted however that this mhag is estimated to be expressed in only 10-15% of Caucasoid patients, thus testing of all HapMap trios is essential for successful identification of this mhag. Next we consider the CD19 L -derived mhag as a highly clinically relevant mhag that deserves to be tested in the (pre)clinical setting. Thus currently, a very intriguing step towards clinical application of the CD19 L -derived mhag is to investigate whether we can functionally transfer the CD19 L -specific TCR into recall antigen-specific T cells, and whether TCR-redirected T cells can induce a potent anti-tumor effect in a preclinical B cell tumor model. T cell persistence after adoptive transfer is often limited 68-70, but we showed in Chapter 6 that dual mhag- and TT-specific CD4 + T cells were long term persisting in vivo and were readily recruited after TT boosts. We could not investigate the anti-tumor effect after these TT boosts, because at that specific time the tumorload was too high for intervention by T cells. Thus, a new experimental setup with lower tumor loads can address the clinical impact of boosting dual antigen-specific T cells. In such a setting, it will be fascinating to investigate whether the in vivo persistence and anti-tumor activity of dual-specific CD4 + human T cells can be improved by using different T cell subpopulations as target for TCR-transfer (central memory versus effector memory 71 ) and different culture conditions for expansion while preserving the phenotype and function (IL2 versus IL7 versus IL15 72,73 ). Chapter 7

122 122 Chapter 7 References 1. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, CA Cancer J Clin. 2005;55: Goulmy E. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive immunotherapy. Immunol Rev. 1997;157: Spaapen R, Mutis T. Targeting hematopoietic-specific minor histocompatibility antigens to distinguish graft-versustumor effects from graft-versus-host disease. Best Pract Res Clin Haematol. 2008;21: Feng X, Hui KM, Younes HM, Brickner AG. Targeting minor histocompatibility antigens in graft versus tumor or graft versus leukemia responses. Trends Immunol. 2008;29: Riddell SR, Berger C, Murata M, Randolph S, Warren EH. The graft versus leukemia response after allogeneic hematopoietic stem cell transplantation. Blood Rev. 2003;17: Wenandy L, Kollgaard T, Letsch A et al. The 1170 A-P single-nucleotide polymorphism (SNP) in the Her-2/neu protein (HER2) as a minor histocompatibility antigen (mhag). Leukemia. 2009;23: den Haan JM, Meadows LM, Wang W et al. The minor histocompatibility antigen HA-1: a diallelic gene with a single amino acid polymorphism. Science. 1998;279: Spaapen RM, Lokhorst HM, van den Oudenalder K et al. 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. 2008;205: Dolstra H, de RB, Fredrix H et al. Bi-directional allelic recognition of the human minor histocompatibility antigen HB-1 by cytotoxic T lymphocytes. Eur J Immunol. 2002;32: de Rijke B, van Horssen-Zoetbrood A, Beekman JM et al. A frameshift polymorphism in P2X5 elicits an allogeneic cytotoxic T lymphocyte response associated with remission of chronic myeloid leukemia. J Clin Invest. 2005;115: Kawase T, Akatsuka Y, Torikai H et al. Alternative splicing due to an intronic SNP in HMSD generates a novel minor histocompatibility antigen. Blood. 2007;110: Akatsuka Y, Nishida T, Kondo E et al. Identification of a polymorphic gene, BCL2A1, encoding two novel hematopoietic lineage-specific minor histocompatibility antigens. J Exp Med. 2003;197: Warren EH, Vigneron NJ, Gavin MA et al. An antigen produced by splicing of noncontiguous peptides in the reverse order. Science. 2006;313: Stumpf AN, van der Meijden ED, van Bergen CA et al. Identification of 4 new HLA-DR-restricted minor histocompatibility antigens as hematopoietic targets in antitumor immunity. Blood. 2009;114: den Haan JM, Sherman NE, Blokland E et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science. 1995;268: Pierce RA, Field ED, Mutis T et al. The HA-2 minor histocompatibility antigen is derived from a diallelic gene encoding a novel human class I myosin protein. J Immunol. 2001;167: Mommaas B, Kamp J, Drijfhout JW et al. Identification of a novel HLA-B60-restricted T cell epitope of the minor histocompatibility antigen HA-1 locus. J Immunol. 2002;169: Brickner AG, Evans AM, Mito JK et al. The PANE1 gene encodes a novel human minor histocompatibility antigen that is selectively expressed in B-lymphoid cells and B-CLL. Blood. 2006;107: Dolstra H, Fredrix H, Maas F et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med. 1999;189: Kawase T, Nannya Y, Torikai H et al. Identification of human minor histocompatibility antigens based on genetic association with highly parallel genotyping of pooled DNA. Blood. 2008;111: Torikai H, Akatsuka Y, Miyauchi H et al. The HLA-A*0201-restricted minor histocompatibility antigen HA-1H peptide can also be presented by another HLA-A2 subtype, A*0206. Bone Marrow Transplant. 2007;40: Meadows L, Wang W, den Haan JM et al. The HLA-A*0201-restricted H-Y antigen contains a posttranslationally modified cysteine that significantly affects T cell recognition. Immunity. 1997;6: Randolph SS, Warren EH, Riddell SR. Identification of Novel Minor Histocompatibility Antigens Encoded by the Y Chromosome Gene, UTY. Blood. 2002;100:615a. 24. Tykodi SS, Fujii N, Vigneron N et al. C19orf48 encodes a minor histocompatibility antigen recognized by CD8 + cytotoxic T cells from renal cell carcinoma patients. Clin Cancer Res. 2008;14: van Bergen CA, Kester MG, Jedema I et al. Multiple myeloma-reactive T cells recognize an activation-induced minor histocompatibility antigen encoded by the ATP-dependent interferon-responsive (ADIR) gene. Blood. 2007;109:

123 General discussion Brickner AG, Warren EH, Caldwell JA et al. The immunogenicity of a new human minor histocompatibility antigen results from differential antigen processing. J Exp Med. 2001;193: Vogt MH, van den Muijsenberg JW, Goulmy E et al. The DBY gene codes for an HLA-DQ5-restricted human malespecific minor histocompatibility antigen involved in graft-versus-host disease. Blood. 2002;99: Vogt MH, de Paus RA, Voogt PJ, Willemze R, Falkenburg JH. DFFRY codes for a new human male-specific minor transplantation antigen involved in bone marrow graft rejection. Blood. 2000;95: Pierce RA, Field ED, den Haan JM et al. Cutting edge: the HLA-A*0101-restricted HY minor histocompatibility antigen originates from DFFRY and contains a cysteinylated cysteine residue as identified by a novel mass spectrometric technique. J Immunol. 1999;163: Wang W, Meadows LR, den Haan JM et al. Human H-Y: a male-specific histocompatibility antigen derived from the SMCY protein. Science. 1995;269: Warren EH, Gavin MA, Simpson E et al. The human UTY gene encodes a novel HLA-B8-restricted H-Y antigen. J Immunol. 2000;164: Spaapen RM, de Kort RAL, van den Oudenalder K et al. Rapid Identification of Clinical Relevant Minor Histocompatibility Antigens via Genome-Wide Zygosity-Genotype Correlation Analysis. Clinical Cancer Research. 2009;15: Spierings E, Brickner AG, Caldwell JA et al. The minor histocompatibility antigen HA-3 arises from differential proteasome-mediated cleavage of the lymphoid blast crisis (Lbs) oncoprotein. Blood. 2003;102: Spierings E, Vermeulen CJ, Vogt MH et al. Identification of HLA class II-restricted H-Y-specific T-helper epitope evoking CD4 + T-helper cells in H-Y-mismatched transplantation. Lancet. 2003;362: Griffioen M, van der Meijden ED, Slager EH et al. Identification of phosphatidylinositol 4-kinase type II beta as HLA class II-restricted target in graft versus leukemia reactivity. Proc Natl Acad Sci U S A. 2008;105: Slager EH, Honders MW, van der Meijden ED et al. Identification of the angiogenic endothelial-cell growth factor-1/ thymidine phosphorylase as a potential target for immunotherapy of cancer. Blood. 2006;107: Vogt MH, Goulmy E, Kloosterboer FM et al. UTY gene codes for an HLA-B60-restricted human male-specific minor histocompatibility antigen involved in stem cell graft rejection: characterization of the critical polymorphic amino acid residues for T cell recognition. Blood. 2000;96: Rosinski KV, Fujii N, Mito JK et al. DDX3Y encodes a class I MHC-restricted H-Y antigen that is expressed in leukemic stem cells. Blood. 2008;111: Wolfel C, Lennerz V, Lindemann E et al. Dissection and molecular analysis of alloreactive CD8 + T cell responses in allogeneic hematopoietic stem cell transplantation. Cancer Immunol Immunother. 2008;57: Torikai H, Akatsuka Y, Miyazaki M et al. The human cathepsin H gene encodes two novel minor histocompatibility antigen epitopes restricted by HLA-A*3101 and -A*3303. Br J Haematol. 2006;134: Terakura S, Murata M, Warren EH et al. A single minor histocompatibility antigen encoded by UGT2B17 and presented by human leukocyte antigen-a*2902 and -B*4403. Transplantation. 2007;83: Murata M, Warren EH, Riddell SR. A human minor histocompatibility antigen resulting from differential expression due to a gene deletion. J Exp Med. 2003;197: Ivanov R, Hol S, Aarts T et al. UTY-specific TCR-transfer generates potential graft-versus-leukaemia effector T cells. Br J Haematol. 2005;129: Ivanov R, Aarts T, Hol S et al. Identification of a 40S ribosomal protein S4-derived H-Y epitope able to elicit a lymphoblast-specific cytotoxic T lymphocyte response. Clin Cancer Res. 2005;11: Kamei M, Nannya Y, Torikai H et al. HapMap scanning of novel human minor histocompatibility antigens. Blood. 2009;113: Torikai H, Akatsuka Y, Miyazaki M et al. A novel HLA-A*3303-restricted minor histocompatibility antigen encoded by an unconventional open reading frame of human TMSB4Y gene. J Immunol. 2004;173: Spierings E, Hendriks M, Absi L et al. Phenotype frequencies of autosomal minor histocompatibility antigens display significant differences among populations. PLoS Genet. 2007;3:e Ogata T, Gregoire L, Goddard KA et al. Evidence for association between the HLA-DQA locus and abdominal aortic aneurysms in the Belgian population: a case control study. BMC Med Genet. 2006;7: Bui HH, Sidney J, Dinh K et al. Predicting population coverage of T cell epitope-based diagnostics and vaccines. BMC Bioinformatics. 2006;7: Warren EH, Greenberg PD, Riddell SR. Cytotoxic T-lymphocyte-defined human minor histocompatibility antigens with a restricted tissue distribution. Blood. 1998;91: Chapter 7

124 124 Chapter de Bueger M, Bakker A, van Rood JJ, Van der Woude F, Goulmy E. Tissue distribution of human minor histocompatibility antigens. Ubiquitous versus restricted tissue distribution indicates heterogeneity among human cytotoxic T lymphocyte-defined non-mhc antigens. J Immunol. 1992;149: Faber LM, van Luxemburg-Heijs SA, Veenhof WF, Willemze R, Falkenburg JH. Generation of CD4 + cytotoxic T-lymphocyte clones from a patient with severe graft-versus-host disease after allogeneic bone marrow transplantation: implications for graft-versus-leukemia reactivity. Blood. 1995;86: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. A class II-restricted cytotoxic T cell clone recognizes a human minor histocompatibility antigen with a restricted tissue distribution. Br J Haematol. 2005;128: Dolstra H, Fredrix H, Preijers F et al. Recognition of a B cell leukemia-associated minor histocompatibility antigen by CTL. J Immunol. 1997;158: Kloosterboer FM, van Luxemburg-Heijs SA, van Soest RA et al. Up-regulated expression in nonhematopoietic tissues of the BCL2A1-derived minor histocompatibility antigens in response to inflammatory cytokines: relevance for allogeneic immunotherapy of leukemia. Blood. 2005;106: DeLuca DS, Eiz-Vesper B, Ladas N, Khattab BA, Blasczyk R. High-throughput minor histocompatibility antigen prediction. Bioinformatics. 2009;25: Schuler MM, Donnes P, Nastke MD et al. SNEP: SNP-derived epitope prediction program for minor H antigens. Immunogenetics. 2005;57: Halling-Brown M, Quartey-Papafio R, Travers PJ, Moss DS. SiPep: a system for the prediction of tissue-specific minor histocompatibility antigens. Int J Immunogenet. 2006;33: Chakraverty R, Eom HS, Sachs J et al. Host MHC class II+ antigen-presenting cells and CD4 cells are required for CD8-mediated graft-versus-leukemia responses following delayed donor leukocyte infusions. Blood. 2006;108: Corthay A, Skovseth DK, Lundin KU et al. Primary antitumor immune response mediated by CD4 + T cells. Immunity. 2005;22: De Veerman M, Heirman C, van Meirvenne S et al. Retrovirally transduced bone marrow-derived dendritic cells require CD4 + T cell help to elicit protective and therapeutic antitumor immunity. J Immunol. 1999;162: Hung K, Hayashi R, Lafond-Walker A et al. The central role of CD4 + T cells in the antitumor immune response. J Exp Med. 1998;188: Ossendorp F, Mengede E, Camps M, Filius R, Melief CJ. Specific T helper cell requirement for optimal induction of cytotoxic T lymphocytes against major histocompatibility complex class II negative tumors. J Exp Med. 1998;187: Perez-Diez A, Joncker NT, Choi K et al. CD4 cells can be more efficient at tumor rejection than CD8 cells. Blood. 2007;109: Symons HJ, Levy MY, Wang J et al. The allogeneic effect revisited: exogenous help for endogenous, tumor-specific T cells. Biol Blood Marrow Transplant. 2008;14: Holloway PA, Kaldenhoven N, Kok-Schoemaker HM et al. Antigens shared by malignant plasma cells and normal B cells may be involved in graft versus myeloma. Clin Exp Immunol. 2003;131: Heemskerk MH, Hoogeboom M, Hagedoorn R et al. Reprogramming of virus-specific T cells into leukemia-reactive T cells using T cell receptor gene transfer. J Exp Med. 2004;199: Berger C, Turtle CJ, Jensen MC, Riddell SR. Adoptive transfer of virus-specific and tumor-specific T cell immunity. Curr Opin Immunol. 2009;21: Hambach L, Nijmeijer BA, Aghai Z et al. Human cytotoxic T lymphocytes specific for a single minor histocompatibility antigen HA-1 are effective against human lymphoblastic leukaemia in NOD/scid mice. Leukemia. 2006;20: Hambach L, Vermeij M, Buser A et al. Targeting a single mismatched minor histocompatibility antigen with tumorrestricted expression eradicates human solid tumors. Blood. 2008;112: Berger C, Jensen MC, Lansdorp PM et al. Adoptive transfer of effector CD8 + T cells derived from central memory cells establishes persistent T cell memory in primates. J Clin Invest. 2008;118: Kaneko S, Mastaglio S, Bondanza A et al. IL7 and IL15 allow the generation of suicide gene-modified alloreactive self-renewing central memory human T lymphocytes. Blood. 2009;113: Wang LX, Li R, Yang G et al. Interleukin-7-dependent expansion and persistence of melanoma-specific T cells in lymphodepleted mice lead to tumor regression and editing. Cancer Res. 2005;65:

125 General discussion 125 Chapter 7

126 126 Chapter 7

127 Nederlandse samenvatting

128 128 Nederlandse samenvatting Elk jaar worden wereldwijd ongeveer mensen gediagnosticeerd met leukemie, lymfoom of multipel myeloom (verschillende vormen van bloedkanker). Daarnaast sterven er jaarlijks circa mensen aan deze ziektes. Ook al kan bloedkanker vaak (tijdelijk) gestabiliseerd of genezen worden door behandeling met chemo- en/of bestralingstherapieën, toch lijkt een allogene stamcel-transplantatie (allo-sct), ook wel beenmerg-transplantatie genoemd, de enige optie met een aanzienlijke kans op echte genezing. Dat komt omdat het transplantaat van een geschikte donor niet alleen uit stamcellen bestaat, maar ook uit T cellen, die normaal gesproken een belangrijke rol spelen in de afweer tegen bacteriën en virussen (Box 1). Box 1. T cellen: een belangrijk onderdeel van ons afweersysteem T cellen zijn een type witte bloedcel, die op hun buitenzijde een eiwit hebben, de T cel receptor (TCR), dat lichaamsvreemde eiwitfragmenten (peptiden) herkent. Elke T cel receptor herkent maar één peptide. Omdat het menselijke repertoire bestaat uit miljoenen T cellen met ieder een unieke T cel receptor, is de mens toch uitgerust met een goed (afweer)systeem om lichaamsvreemde peptiden te detecteren. Het gespecialiseerde eiwit HLA, aanwezig op bijna alle andere cellen, laat deze peptiden voortdurend aan de T cellen zien (peptidepresentatie), waarna alle cellen die lichaamsvreemde peptiden presenteren door de T cellen worden opgeruimd. Ondanks dat de donor en patiënt wel gematcht zijn voor een identieke HLA-typering, zijn ze niet 100% hetzelfde in alle peptiden die door het HLA gepresenteerd worden. Een positief gevolg hiervan is dat de T cellen van de donor na transplantatie een afweerreactie bij de patiënt kunnen veroorzaken en dan alle resterende tumorcellen opruimen. Deze afweerreactie wordt ook wel het graft-versus-tumor (GvT) effect genoemd. Helaas gaat deze gunstige reactie vaak gepaard met afweerreacties gericht tegen andere gezonde lichaamscellen van de patiënt (graft-versushost ziekte, GvHD) met in het ergste geval de dood tot gevolg. De peptiden die gepresenteerd worden in de patiënt maar niet in de donor, en dus na de transplantatie een belangrijke rol spelen in afweerreacties tegen gezonde en/of kankercellen van de patiënt, heten minor histocompatibility antigenen (mhags). Om in de toekomst een veilige en genezende stamcel-transplantatie te realiseren als behandeling voor bloedkanker, is het noodzakelijk om de graft-versus-tumor effecten te scheiden van de graft-versus-host ziekte. In theorie is dit mogelijk door het transplantaat te ontdoen van alle T cellen, gevolgd door infusie van specifiek die donor T cellen die mhags herkennen op tumorcellen en bloedcellen van de patiënt, maar niet op zijn andere lichaamscellen (hematopoietische mhags genoemd). Dit zou tot gevolg hebben dat de tumor- en bloedcellen van de patiënt opgeruimd

129 Nederlandse samenvatting 129 worden zonder dat vitale organen worden aangevallen. Dat de bloedcellen van de patiënt opgeruimd worden is niet erg, omdat de patiënt door de transplantatie voorzien is van een nieuw bloedsysteem ontstaan uit de stamcellen van de donor. Deze nieuwe donor-afgeleide bloedcellen worden natuurlijk niet aangevallen door de donor T cellen, omdat ze door hen niet als lichaamsvreemd worden beschouwd. Om na stamcel-transplantatie T cellen te kunnen infuseren die alleen hematopoietische mhags herkennen, is het van groot belang om dit soort mhags op moleculair niveau te karakteriseren. In de afgelopen 15 jaar zijn er pas 17 hematopoietische mhags als zodanig geïdentificeerd (zie tabel op pagina 114). Er zullen nog tientallen nieuwe hematopoietische mhags ontdekt moeten worden om meer dan 90% van de bloedkanker patiënten te kunnen behandelen met mhag-specifieke T cellen. De identificatie-methodes die tot nu toe gebruikt zijn, zijn tijdrovend (ongeveer 2 tot 6 jaar per identificatie) en vereisen een zeer specialistisch laboratorium. Naast het feit dat er nog niet voldoende hematopoietische mhags bekend zijn, staan ook de verschillende behandelmethoden gebaseerd op hematopoietische mhags nog in de kinderschoenen. Vandaar dat het doel van dit proefschrift tweeledig is: 1) Het identificeren en vergemakkelijken van de identificatie van nieuwe hematopoietische mhags. 2) Het onderzoeken van de therapeutische capaciteit van een subcategorie T cellen, gekenmerkt door expressie van het eiwit met de naam CD4, ten opzichte van gekweekte kankercellen in het laboratorium, en ten opzichte van plasmacel kanker (multipel myeloom) in een preklinisch muizenmodel. Identificatie van de CD19 L -mhag: een nieuw identificatie strategie en een nieuw doelwit voor therapie In Hoofdstuk 2 wordt de identificatie beschreven van een nieuwe hematopoietische mhag, CD19 L genaamd. Omdat deze mhag goed gepresenteerd wordt door verschillende soorten bloedkanker die uit B cellen zijn ontstaan, is het een belangrijk nieuw doelwit voor potentiële behandeling van deze categorie kankers na stamcel-transplantatie. Voor de identificatie van de CD19 L -mhag hebben we een nieuwe strategie toegepast, welke na enkele aanpassingen het identificeren van meer mhags relatief eenvoudig heeft gemaakt. Deze zogeheten zygositeitgenotype correlatie analyse is ook zeer snel (mogelijk binnen 3 maanden) zoals we in Hoofdstuk 3 laten zien met de karakterisering van een andere nieuwe mhag (SLC19A1 R ). We denken dat deze nieuwe mhag minder therapeutische kracht heeft dan de CD19 L -mhag, omdat de SLC19A1 R -mhag waarschijnlijk ook op gezonde cellen gepresenteerd kan worden aan T cellen.

130 130 Nederlandse samenvatting De huidige identificatie van nieuwe mhags heeft niet gestructureerd plaatsgevonden, daarom wordt in Hoofdstuk 3 en 7 beschreven op welke wijze identificatie van nieuwe hematopoietische mhags dient te gebeuren om in de toekomst wel significante vorderingen te maken richting mhag-gebaseerde therapie voor een grote patiëntengroep. De meest ideale mhag identificatie strategie staat nog in de kinderschoenen In Hoofdstuk 4 hebben we bovenstaande ideeën als uitgangspunt gebruikt voor het stroomlijnen van een andere identificatie strategie. Deze strategie, gebaseerd op voorspellingen door computerprogramma s, is in theorie de meest ideale voor de identificatie van specifiek de therapeutisch interessante hematopoietische mhags, maar is in de praktijk zeer moeilijk uit te voeren gebleken. Door gebruik te maken van 1) kenmerken van de tot nu toe bekende mhags, 2) de nieuwste databases en 3) de nieuwste computerprogramma s, hebben we een nieuwe lijst mogelijke mhags samengesteld met, volgens ons, de beste kandidaten uit heel het menselijke genoom. Momenteel wordt onderzocht of de kandidaten van deze lijst ook daadwerkelijk mhags zijn en daarna gebruikt kunnen worden als doelwit voor de behandeling van kanker. CD4-positieve T cellen zijn een goede kandidaat voor behandeling van kanker Voor een behandeling van kanker met T cellen hebben onderzoekers zich tot nu toe vooral gericht op T cellen die het eiwit CD8 tot expressie brengen, omdat deze T cellen volgens decennia oude ideeën echte killers zijn. Een andere subgroep van T cellen brengt geen CD8, maar CD4 tot expressie. Deze T cellen zijn van oudsher altijd beschouwd als hulpjes voor de CD8-positieve T cellen, zonder dat ze zelf direct de kankercellen aan zouden vallen. Echter, in de afgelopen 10 jaar is gebleken dat ook de CD4-positieve T cellen kankercellen kunnen killen. Omdat het genereren van grote hoeveelheden (CD8- en CD4-positieve) T cellen voor infusie na stamcel-transplantatie moeilijk is, kan gebruik gemaakt worden van zogenaamde T cel receptor transfer, zoals ook in Hoofdstuk 5 beschreven is. Daar hebben we namelijk de T cel receptor van een CD4-positieve mhag-specifieke T cel geïsoleerd. Vervolgens hebben we deze T cel receptor in andere CD4-positieve T cellen geplaatst, die normaal gesproken de afweerreactie tegen tetanus verzorgen. Deze gemodificeerde T cellen konden daarna reageren tegen zowel tetanus (met hun originele T cel receptor) als de mhag (met hun geïntroduceerde T cel receptor), met als belangrijk voordeel dat ze veel beter tot grote hoeveelheden te expanderen zijn. Tot slot laten we in Hoofdstuk 6 zien dat dergelijke dubbel-specifieke T cellen niet alleen kankercellen in een kweekschaaltje kunnen opruimen, maar ook aanwezige myeloomcellen in een muizenmodel. De myeloomcellen in deze muizen konden

131 Nederlandse samenvatting 131 licht uitzenden als gevolg van een chemische reactie wat we konden detecteren met een zeer gevoelige camera (zie foto s op pagina 101). In dit muizenmodel laten we overtuigend zien dat de CD4-positieve T cellen vooral effect hadden op de myeloomcellen in het beenmerg (dijbeenderen van de muis). De myeloom die uiteindelijk buiten het beenmerg weer uitgroeide, konden we niet behandelen met extra T cel injecties, maar bleek na isolatie uit de muis in een kweekschaaltje wel weer vatbaar voor therapie met T cellen. Deze resultaten zijn zeer vergelijkbaar met het ziekteverloop van multipel myeloom in patiënten, wat dit model dus uitermate geschikt maakt voor verder onderzoek naar terugkerende myeloom. Conclusies De resultaten in dit proefschrift hebben op verschillende fronten bijgedragen in de richting van een verbeterde therapie voor bloedkanker. Als vervolg op de identificatie van de nieuwe CD19 L -mhag, wordt momenteel een klinische trial opgezet ter behandeling van B cel kanker na stamcel-transplantatie. Een belangrijke bijdrage in dit proefschrift is ook de methode die we geïntroduceerd hebben voor identificatie van nieuwe mhags. Het verder ontwikkelen van de (in theorie ideale) identificatie strategie voor mhags zal zijn waarde in de toekomst nog moeten bewijzen. De geleverde bewijzen voor de therapeutische inzetbaarheid van gemanipuleerde tetanus-reactieve T cellen, die het CD4 eiwit tot expressie brengen, zijn een aanleiding om dit soort T cellen toe te voegen aan de huidige behandelmethoden. Tot slot geven we in de algemene discussie (Hoofdstuk 7) duidelijke richtlijnen over hoe identificatie en klinische toepassing van mhags vervolgd dient te worden.

132 132 Nederlandse samenvatting

133 Dankwoord

134 134 Dankwoord Dankwoord Graag wil ik enkele mensen (Tabel 1) met een persoonlijk tintje bedanken (Fig. 1). Tabel 1. Bedankt op alfabetische volgorde. Bedankt. Persoon Relatie Bedankt voor 1 Iedereen ja het met oprechte interesse van voor tot achter lezen van dit proefschrift Tuna co-promotor dit alles 2 Henk promotor dit alles 3 Kelly paranimf de snelle -reacties die nodig waren op de zware thesis-momenten Michaël paranimf hetzelfde als hier direct boven Kelly ex-collega, analist een geweldige, gezellige en vruchtbare tijd in het lab, het zijn van mijn baas Maureen collega, analist je enorme inzet, begrip en vriendelijkheid, en ook een vruchtbare tijd 4 4 Amber nichtje je ondeugende blik Andries indirecte collega 5 die wijdopenstaande deur, de hulp en de leuke grappen Anja indirecte collega 6 het altijd warme welkom als ik even kwam buurten Anke aap vele wijze raad als medeslachtoffer van de wetenschap Annet indirecte collega 6 gezellige praatjes op het lab Anton collega, post-doc je vriendelijkheid naar het lab toe Anton indirecte collega 5 mooie natuurfoto's, muizen en aanbevelingen Antoni Chicago-genoot een onvergetelijke vakantie en je Lost-verslaving Apen vrienden de onuitputtelijke bron van weekenden, spellen, feesten, hulp en geluk Arjan indirecte collega 6 de voetbalorganisatie, ook al deed ik niet vaak mee Arnold indirecte collega 6 de hulp die ik af en toe nodig had B&D-ers 7 vrienden van vriendin alle leuke feestjes en gesprekken Bastiaan broertje het altijd klaarstaan en je liefde voor onze familie Berris collega, analist je pittige eigen mening en je hulp wanneer nodig Bianka collega, analist je gezelligheid en Duitse bieren Brigit indirecte collega 6 je brede kennis en je hulp met het overtuigen van de sponsor Brigita vriendin van broertje je gezelligheid en je warme hartelijkheid Carin indirecte collega 6 de hulp met het zoeken naar enveloppen die op waren Caro vrouw van Greg je gastvrijheid en de leuke introductie met je moeder Casper aap de pompen Cat aap je super enthousiasme en je enorme energievoorraad Cees indirecte collega 6 je beide benen, die nooit van de grond komen Christiaan aap de pompen Claartje aap de geweldige avondjes in de kroegen en onvoorwaardelijke vriendschap Coen indirecte collega 6 likkende, maar wetenschappelijke woorden onder het genot van een biertje Cor indirecte collega 6 je altijd jolige stemming en vrolijke voorkomen Daniël aap de pompen Daniël aap de pompen David aap de pompen Dianne indirecte collega 6 de gezelligheid op de borrels Didy moeder van vriendin hetzelfde als Joop en voor het happen op mijn opmerkingen Eefje fluimbeer je naïeve en daarom zeer grappige opmerkingen en reacties Eelo indirecte collega 6 je optimisme en leuke zinloze flauwekul Ellen ex-collega, AIO de leuke gesprekken en het vergeten van de laatste 4 letters van mijn naam Ellen indirecte collega 8 al je telefoontjes dat er beenmerg was Ellis ex-indirecte collega 5 de gezelligheid jaren geleden Eric indirecte collega 5 korte samenwerking en interessante brainstorms Erik indirecte collega 6 de dokterspraatjes en de voetballessen op de retraites Esther indirecte collega 6 de gezelligheid bij jullie in de AIO-kamer Evelien collega, AIO je medeleven en brede interesse

135 Dankwoord 135 Evelyn indirecte collega 6 voor je "smetvrees", heel vermakelijk Familie familie de supergezellige dagen, weekenden en feesten Flip indirecte collega 6 de wijsheden, de open deur en de Keystoneverhalen Frank fluimbeer het introduceren van de Opelclub en de snowboardlessen Frank ex-indirecte collega 5 de wijze adviezen over carrière en buitenland Fred nieuwe collega het dragen van een hoed Gerrit indirecte collega 5 het cell-sorting Greg ex-nieuwe collega de open armen, je inboedel en de gezelligheid Gwen indirecte collega 6 de leuke gesprekken Harry indirecte collega 6 het vertrouwen in onze helaas tot nog toe niks opleverende collaboratie Henk indirecte collega 5 het begin van mijn loopbaan en de hulp waar nodig Henk-Jan indirecte collega 5 de betaalde tol en de pokerlessen van een echte blufkoning Henny indirecte collega 5 de droge grappen Irma indirecte collega 6 de hulp wanneer ik het nodig had Jan aap de pompen Jan Willem indirecte collega 6 de verhalen op de retraite onder het genot van een biertje Jan-Willem fluimbeer de ongekende hoeveelheid lol Jelmer broertje van vriendin de (tafel)tennislessen en de biertjes in Tilburg Jerney indirecte collega 6 je vriendelijkheid en je interesse in Eelo Jeroen fluimbeer voor de gezellige biertjes bij jullie thuis en op wintersport Joan aap de pompen Johan collega, post-doc het (hopelijk succesvol) vervolgen van stukken van mijn onderzoek Joop vader van vriendin de open deur, de warmte en gezelligheid Joost fluimbeer de squashlessen en de vele wintersportfoto's Jorinde fluimbeer je soms onbeholpen lompe uitspattingen, heel grappig Joris aap de pompen Joukje indirecte collega 6 de hulp met verschikkelijke bureaucratische flauwekul Julie collega, post-doc de boeiende gesprekken die we hadden Jürgen indirecte collega 5 de zeer wijze raad voor wetenschap en buitenland Justin nieuwe collega het garanderen van mijn succes Kasper indirecte collega 9 het in de gaten houden van Ruud Kees fluimbeer de wijze indruk die je maakt terwijl je slap ouwehoert Kees indirecte collega 5 de inspirirende plannen voor een levenslange samenwerking Koen aap de pompen Koos indirecte collega 5 het cell-sorting Kristan fluimbeer de Biesbosch kanotochtjes en de skilessen Kristan indirecte collega 5 je kritiek tijdens werkbesprekingen en hulp voor de luciferase assay Laurens neefje de mooie tekeningen Lijnie ex-collega, analist je altijd vrolijke aanwezigheid op het lab LUMCers de concurrenten de heel erg leuke en soms pittige gesprekken op alle congressen Maarten collega, analist je oprechtheid en cynische grappen Maarten ex-indirecte collega 6 het gezellige praatje als ik je nog eens tegenkwam Madelon fluimbeer je begrip dat T cellen en niet DCs de toekomst zijn ;-) Margriet aap je enthousiasme en gezelligheid tijdens het apenweekend Marieke indirecte collega 5 de borrels georganiseerd door ons samen als Jordan borrelcommissie Marije student je expressieve enthousiasme en doorzettingsvermogen Marijn neefje de lieve glimlach over de webcam Marjolein zus onze leuke contacten, ook al is dat niet wekelijks Mark indirecte collega 6 je fraaie schaamteloze uitspraken Marnix aap de pompen Martha aap je enorme enthousiasme en zin in alles wat de apen doen Martin student je chaos, je nobelprijs-aspiraties en je humor

136 136 Dankwoord Martin indirecte collega 6 je slappe geouwehoer en een zweverige ervaring Martine indirecte collega 6 je adviezen en behulpzame opmerkingen Matin ex-nieuwe collega je slappe lach acties Matthijs aap de pompen Matthijs fluimbeer de meest ongelukkige, onnozele, maar o zo grappige opmerkingen Matthijs indirecte collega 10 de organisatie van de beste AIO-retraite ever Mercedes nieuwe collega het sluitend maken van de mechanistische cirkel Michael collega, AIO de letterlijke warmte en het geouwehoer op werk en congressen Michaël fluimbeer de 1x per jaar droge opmerking op fluimbeerfeestjes Michaël indirecte collega 11 de kopjes koffie en de antilichamen Michel aap de pompen Michelle nieuwe collega je altijd aardige interesse Mieke ex-indirecte collega 5 je nog altijd oprechte interesse in Nederland Miel aap de pompen Monique collega, arts je leuke interesse in experimenten en je hart voor onderzoek Myrte ex-tc de Uithofgenoot de gezelligheid en maffia-introductie, helemaal te gek NCMLSers de concurrenten de fijne samenwerkingen (die helaas iedere keer gestrand zijn) Nicoline fluimbeer je interesse en inlevingsvermogen Niels collega, arts de opwindende allo-sct discussie op Kos Olga ex-tc de Uithofgenoot de kubb-lessen en je heerlijke recht-door-zee mening Overige indirecte collega's 8 de hulp Overige indirecte collega's 12 de vriendelijke hulp wanneer nodig Overige 13 indirecte collega's 5 de gezelligheid en gulle giften (materieel en immaterieel) Overige 14 (ex-)indirecte collega's 6 de gezelligheid Pamela aap de gezelligheid en gastvrijheid bij jullie thuis Paul indirecte collega 5 je hulp bij mijn KWF-aanvraag Pieter ex-indirecte collega 15 de granzyme informatie en de biertjes op congressen Richard indirecte collega 6 de interesse en vriendelijkheid Richard indirecte collega 5 de extra uurtjes in de disco's in New Orleans Rob indirecte collega 6 vriendelijk en geïnteresseerd zijn Rolf indirecte collega 6 de aardige maar korte gesprekken af en toe Ron student de pokerinstructies en het gehoorzamen van mijn bevelen op het lab Ronan ex-indirecte collega 6 je Franse lessen in het kweeklab Roos ex-indirecte collega 5 de gezelligheid en het schoonvegen van de labtafels met je rok Ruud indirecte collega 9 je aardige en brutale opmerkingen Sabien aap het overheerlijke gebak dat iedere keer weer klaarstaat op de high tea Samantha indirecte collega 5 de gezellige tijd jaren geleden en verdere hulp wanneer nodig Sanne niet mijn student de gezelligheid en je vertrouwen in mijn kennis Sanne fluimbeer je gezelligheid en het volwassen maken van Jan-Willem (hoogste tijd) Sanne ex-tc de Uithofgenoot de speciale relatie die wij samen hebben, schat Saskia indirecte collega 5 je huis voor een weekje Selena fluimbeer het doen van de afwas na gezellige dinertjes bij jullie thuis Serge aap de pompen Silvie indirecte collega 6 je enthousiasme en vrolijkheid Sosja aap je gezelligheid bij ieder feestje Sponsoren sponsoren (zie voorin) hun zeer welkome bijdrage Suus fluimbeer je enthousiasme over de Wii Suus indirecte collega 5 je wijze kledingtips en je optimistische gezelligheid Suzanne ex-indirecte collega 6 het organiseren van gezellige momenten Teun collega, post-doc onze oneindig lange gesprekken en wonderbaarlijke ideeën Tineke collega, analist je enthousiaste en lachwekkende verhalen over je familie Tirza aap het goedkeuren van pompen Tom ex-indirecte collega 5 al je zwetspraatjes over respect in het C1-lab Tom nieuwe baas het leuke onderzoek dat ik nu mag doen 16

137 Dankwoord 137 Tomasz collega, analist je grappige, goed bedoelde gekheid Vechtlusters 17 tennisteam de flauwe grappen op en naast de baan Vera student de gezelligheid, je inzet en je snelle pipetteerwerk Viviènne ex-indirecte collega 5 je eerlijke uitstraling en de leuke tijd in het lab lang geleden VSCers 18 voetbalteam de lol onder de douche, maar ook op het veld en in de kantine Walter indirecte collega 5 het af en toe gezellig komen buurten sinds jullie in het WKZ zitten Wiard zwager het mee helpen opeten van alle restjes (en ook de niet-restjes) Wouter aap de pompen Wouter de grote baas het bij elkaar krijgen en houden van de afdeling X neefje/nichtje tja... nog niks Ya Ping indirecte collega 6 het voorlezen van de tekst op de powerpoint dia's Yan nieuwe collega het proeven van Bailey's Yonathan fluimbeer je gezelligheid en interesse in Nederlandse cultuur Leo vader je adequate hulp, advies, luisterend oor en onvoorwaardelijke steun Willy moeder je adequate hulp, advies, luisterend oor en onvoorwaardelijke steun Tamara vriendin jouw liefde, je vertrouwen, je hulp, je monchoutaart, je creativiteit en je geduld 1 Naast de bedankjes wens ik iedereen heel veel succes met zijn bezigheden en plannen 2 De goede, begripvolle, koppige, vrije, gezellige begeleiding. Je magnetische werking als we klaar waren met een werkbespreking of discussie. De gezelligheid, eventueel onder het genot van een biertje, zowel op het lab als bij congressen. Zowel het vertrouwen in als het wantrouwen van mijn ideeën. Het kritisch bekijken van je eigen, maar ook zeker van andermans resultaten. Het nuanceren van mijn stelligheid. Het aan de man brengen van jouw ongekende scala van (niet-bestaande) spreekwoorden en gezegdes. En nog veel meer. Bedankt! 3 De onuitputtelijke en oprechte interesse en het aanhoudende enthousiasme over alle uitgevoerde experimenten. Het kritisch meedenken met alle experimenten, het proberen te begrijpen van ingewikkelde computer-programma s (bijvoorbeeld voor database onderzoek, voor genetic linkage analyse of voor het maken van eenvoudige presentaties). De (voor mij) zware bergetappe op Kos en het lachend maken van filmpjes met je mobiel op de top. Het warme hart dat je een ieder toedraagt. En nog veel meer. Bedankt! 4 Zowel wetenschappelijk als persoonlijk 5 Van de afdeling Immunologie 6 Van mijn afdeling, het LKCH 7 Sandra, Mark, Emile, Aafke, Carline en alle anderen 8 Van het Hoeklab 9 Van het Stamcellab 10 Van de afdeling Virologie 11 Van het Van Bleeklab 12 Van het Stamcellab en het HLAlab 13 Van de afdeling Immunologie, waar er te veel van rondlopen om alle namen hier op te schrijven 14 In willekeurige volgorde: Miranda, Marie-José, Martijn, Valentina, Vivianne, Carina, Chantal, Claudia, Mirjam, Vivian, Jelle, Sandra, Eszter, Caroline, Marloes, Menno, Henk-Jan, Mark, Çetin, Thijs, Eva, Ton, Anton, Peter, Hans, Ron en alle anderen 15 Van de afdeling Pathologie 16 Sorry Carolien 17 In willekeurige volgorde: Sjoerd, Marcel, Martijn, Erik, Arnoud en Imre 18 In willekeurige volgorde: met name Michel, Jan-Willem, Nils, Maarten en Michaël Figuur 1. Bedankt.