uva-dare (digital academic repository) studies on the ... · uva-dare is a service provided by the...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Studies on the immune system in CLL

Mous, R.

Link to publication

Citation for published version (APA):Mous, R. (2008). Studies on the immune system in CLL.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 11 Sep 2020

https://dare.uva.nl/personal/pure/en/publications/studies-on-the-immune-system-in-cll(6d5c9f37-0646-4e4e-93d2-6b8a80bf3f16).html


Rogier Mous

Studies on the immune system in CLLThesis, University of AmsterdamISBN 978-90-8891-0623

© 2008 Rogier Mous, Amsterdam, The NetherlandsPrinted by BOX Press, Oisterwijk

The studies described in this thesis were performed at the Department of Hematology and the Laboratory of Experimental Immunology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

The printing of this thesis was financially supported by: MRC-Holland, Roche, Sanquin, University of Amsterdam


ACADEMISCH PROEFSCHRIFT

Ter verkrijging van de graad van doctoraan de Universiteit van Amsterdamop gezag van de Rector Magnificus

prof. dr. D.C. van den Boomten overstaan van een door het college voor promoties ingestelde

commissie, in het openbaar te verdedigen in de Aula der Universiteit

op

vrijdag 3 oktober 2008, te 10.00 uur

door

Rogier Mous

geboren te Amsterdam

Promotiecommissie

Promotores: prof. dr. M.H.J. van Oers prof. dr. R.A.W. van Lier

Co-promotor: dr. E.F. Eldering

Overige leden: prof. dr. C.J.M. van Noesel prof. dr. J.P. Medema prof. dr. A. Hagenbeek prof. dr. R.J.M. ten Berge prof. dr. J.H.F. Falkenburg dr. J.B.A.G. Haanen

Faculteit der Geneeskunde, Universiteit van Amsterdam

voor mijn oudersvoor Mariekepara Joana

Contents

Chapter 1: Introduction: CLL, a paradigm of tumor-associated 9 immune dysfunction Chapter 2: Enhanced survival of regulatory T cells in CLL 31

Chapter 3: Detection of p53 dysfunction in chronic 57 lymphocytic leukemia cells via multiplex quantification of p53 target gene induction

Chapter 4: Redirection of CMV specific CTL towards CLL 71 via CD20 targeted HLA/CMV complexes

Chapter 5: Adequate synapse formation between leukemic 91 B cells and effector T cells following stimulation with artificial TCR ligands

Chapter 6: Granzyme B induced apoptosis is enhanced by 115 small molecule inhibitors of XIAP

Chapter 7: Summary and discussion 133

Samenvatting voor niet-ingewijden 151 Dankwoord

List of publications

Curriculum vitae

CLL: a paradigm of tumor-associated immune dysfunction

Rogier Mous1, René AW van Lier2 and Marinus HJ van Oers1

Departments of 1Hematology and 2Experimental Immunology Academic Medical Center, Amsterdam

1

11

Introduction

Abstract

Disorders of the immune system occur in a large range of malignancies. Whereas some of the abnormalities may affect different components of host defense against pathogens, others may contribute to the survival or even ex-pansion of tumor cells. In B cell chronic lymphocytic leukemia (CLL) various immune disorders may be observed in the course of the disease. In CLL, an increased amount of regulatory T cells, skewing of the immune sys-tem towards Th2 and down modulation of costimulatory molecules combined with enhanced surface expression of CD200 by tumor cells may all serve as a protective mechanism against tumor-directed immune responses. Other co-inciding immune disorders (like autoimmunity and hypogammaglobulinae-mia) also have an effect on host immunity but offer no obvious advantage for the tumor cells. Strikingly, amidst all these impairments, immunity against chronic (herpes) viruses like cytomegalovirus (CMV) and Epstein Barr Virus (EBV) is con-served in CLL patients. Studying these unique T cell populations in CLL may provide new insights into the requirements for the maintenance of immunity against chronic (herpes) viruses. Moreover, these virus-specific T cells may also hand us a powerful tool for active immunotherapy for CLL if they can be redirected against tumor cells.

12

Introduction

Cancer is one of the most important causes of death in the western world (WHO 2002). For many types of cancer, curative treatment is still lacking. Fortunately, recent research has led to the development of new therapeu-tic tools that are claimed to have a stronger and more tumor-specific effect. Whereas some of these new agents attack the weak spots of the cancer cells, others aim at directing the attention of the immune system towards tumor cells(27;49;69). Under normal circumstances however, the immune system should be able to eliminate malignant cells without medical intervention. Therefore, the development of a malignancy might also be seen as a failure of the immune system. Moreover, many malignancies are accompanied by various deregulations of the immune system that are harmful for the health of the patient but do not have a clear benefit for tumor cell survival. From this perspective, studying the immune system of cancer patients might not only give us more insight into the mechanism of tumor development but might also lead to the development of new therapeutic strategies. In this thesis, we have studied chronic lymphocytic leukemia (CLL), an example of a malignant disease that is accompanied by various immune disorders.

CLL: no curative treatment

CLL is the leukemic malignancy with the highest prevalence in Western Eu-rope. At present however, there is no cure for the disease. Despite the fact that the first results of reduced intensity stem cell transplantation (RIST) look promising(19;58), this treatment has one major drawback: since CLL mainly affects elderly people many of them are not eligible for this procedure due to age limitations for this therapy. Thus, an urgent need exists for the develop-ment of new therapies that effectively attack tumor cells and that have few side effects. To develop these strategies, more knowledge on tumor biology needs to be acquired. In most cases, CLL is marked by a slow accumulation of mature CD19/CD5/CD23 positive B-lymphocytes. It is thought that the accumulation of malig-nant B cells may result from disturbed apoptosis due to overexpression of anti-apoptotic B cell leukemia-2 (Bcl-2)(29), which was recently linked to mutations or deletions in a region on chromosome 13 containing microRNA

Chapter 1

13

genes(13). Besides Bcl-2 overexpression, CLL cells also use other mechanisms to protect themselves against apoptosis. Mutations or deletions of tumor-sup-pressor gene p53 (occurring in about 5-10% of all patients at diagnosis)(14) or ataxia-telangiectasia mutated gene (ATM; occurring in 10-15% of all patients at diagnosis)(2), which exerts its function upstream of p53, are correlated with an adverse prognosis and resistance to chemotherapy(26), since the most fre-quently used therapeutic agents depend on a functional p53 pathway in the tumor cells(54;63). Moreover, the prevalence of p53 mutations is even higher among patients that relapse after initial chemotherapeutic treatment, possibly due to prior DNA damage or selection for p53 dysfunctional clones(61;65). Thus, it goes without saying that there is an urgent need for therapeutic ap-proaches that act in a p53 independent way. Besides protection against apoptosis, CLL has also found ways to protect itself against the immune system. Since the immune system is capable of detect-ing and eliminating malignant cells, altering host immunity can also offer a potential benefit for tumor cells. In CLL, most protection mechanisms against tumor surveillance affect the generation of adequate tumor-directed T cell re-sponses. This is thought to be either mediated via changes within CLL cells which protect them against T cell-induced apoptosis, or to be contributed by changes within different T cell populations that may be induced by the tumor cells(25). Both aspects will be further addressed in the next paragraphs.

Tumor cells are protected against T cell attack

Altered expression of surface moleculesTumor cells use various tricks to escape immune surveillance. One of these mechanisms is disabling recognition by immune cells through downmodula-tion of various surface molecules. The most obvious way to do this is to reduce MHC class I and II expression thereby preventing the presentation of peptides derived from tumor-specific antigens(38;53;70). That this is a very efficient way to prevent immune surveillance is supported by the fact that many viruses use the same strategy to hide from virus-specific immune cells(4;57). Therefore, it is not surprising that oncogenic transformation by viruses is sometimes ac-companied by downmodulation of antigen presenting molecules (9;55). In CLL, there is no evidence for virus infection of tumor cells. Nevertheless, CLL cells have a lower surface expression of class I antigens than their non-malig-

Introduction

14

nant counterparts. In addition, activation of anti-tumor immune responses is also prevented by reduced surface expression of costimulatory molecules on CLL cells. It has been postulated that a chronic deficiency of CD40 ligand on T cells in CLL patients is responsible for the downmodulation of immune-stimulatory molecules on the CLL cells(15). Studies by Kato et al(35) showed that CD40 ligation through adenoviral transduction of CLL cells with CD40L resulted in upregulation of surface expression of costimulatory molecules, MHC class I and death receptor molecules. Moreover, transduced tumor cells were able to induce autologous T cell responses in vitro. Later, promising re-sults were obtained when CD40L transduced tumor cells were reinfused in CLL patients. This resulted in an increased tumor-specific T cell activity and, in some cases, reduction of the tumor burden(72). At present, more clinical studies with CD40L transduced tumor cell reinfusion are being performed. The results of these studies should point out whether resupplying CD40L is a feasible approach to enhance tumor cell immunogenicity in vivo. Recently it has been discovered that CLL, as well as a subset of multiple my-eloma, has increased surface expression of CD200, a membrane glycoprotein which shares extracellular domains with T cell regulation molecules like CD2, CD80 and CD86 (44;46). The function of this molecule was discovered only a few years ago. Animal studies pointed out that experimentally induced au-toimmune disease had an earlier onset in CD200 -/- mice than in wild type animals(31). The results of this study suggest that CD200 has a down-modula-tory effect on immune activation. To exert this effect, CD200 is dependent on interaction with its receptor, CD200R(74). This implies that, in order to have a down-modulating effect on immune activation, cells that express CD200 need to be in direct contact with cells that carry the receptor. Whereas CD200 is ubiquitously expressed, the expression of CD200R, is mainly restricted to cells of the myeloid lineage, like dendritic cells, monocytes and macrophages(73). Therefore, if CD200 indeed has an immune-modulatory effect in CLL, the most likely place for this to occur is in the lymph nodes, where dendritic cells are present. Recently, mouse studies have demonstrated that tumor cells with increased CD200 surface expression indeed appeared to be protected against immune attack(37), possibly due to an inhibitory effect on APC (which have high CD200R surface expression). Notably, CD200 blocking antibodies com-pletely abrogated the immune-modulatory effect of CD200 on tumor cells, resulting in efficient tumor clearance. It is questionable however whether ma-nipulating CD200 ligation can be used in a clinical setting, since it will have to

Chapter 1

15

be highly tumor cell-specific in order to avoid massive immune activation and maybe even auto-immunity.

Intrinsic factors protect CLL cells against CTL induced apoptosisWhen, despite all changes in the expression of surface molecules, a tumor cell is recognized by a T cell, subsequent killing of this tumor cell may be impaired in vivo because the tumor cells have adopted various ways to protect themselves against T cell-induced apoptosis. When a cytotoxic T cell (CTL) attacks a tar-get cell, it releases the content of cytotoxic granules that is subsequently taken up by the target cell, possibly via endocytosis. Once inside the target cell, the granular content is released into the cytoplasm of the target cell and apoptosis of the target cell is induced(16). The initiators of apoptosis are the enzymes from the granzyme family(10), of which granzyme B is the best-studied mem-ber. Granzyme B is capable of cleaving the pro-apoptotic molecule Bid, which in turn translocates to the mitochondria and initiates apoptosis through mi-tochondrial depolarization and subsequent caspase activation(5). Tumor cells may protect themselves against granzyme B induced apoptosis via different mechanisms. First, many tumor cells express high levels of serpin protease inhibitor PI-9, which serves as an inhibitor of granzyme B(45). Recent data demonstrate that CLL cells also may contain very high levels of PI-9 (M. Bots, personal communication) and therefore may be more resistant to granzyme B induced apoptosis. Furthermore, the overexpression of anti-apoptotic Bcl-2 may prevent mitochondrial depolarization in CLL cells upon CLT attack(18), but whether this also protects them against CTL induced apoptosis in vivo re-mains elusive. Finally, CLL as well as other types of cancer also expresses high levels of x-chromosome-linked inhibitor of apoptosis protein (XIAP) (60;68), a molecule which can inhibit both mitochondrial-initiated apoptosis (through inhibition of caspase 9) and the activation of caspase-3. Recent research has demonstrated that XIAP indeed protects tumor cells against effector cell in-duced apoptosis(32). Moreover, the development of small molecule inhibitors of XIAP now enables interfering with the function of XIAP in a clinical set-ting(59). These inhibitors have demonstrated to sensitize tumor cells for both cytotoxic drugs(59) and Fas ligation in vitro(33). At present, the capacity of these compounds to sensitize tumor cells to CTL mediated killing is being explored (this thesis).

Introduction

16

Cytokine milieu inhibits Th1 responsesThe production and secretion of cytokines allows cells to influence immune responses in their proximity but may also exert systemic effects. For example, the combination of the entity of the pathogen and the type of cytokines re-leased can determine whether Th1 or Th2 type helper T cells will predominate in an immune activation. Thus, the balance between Th1 and Th2 cytokines affects the way the immune system is able to respond to certain antigenic stimuli. It has been observed that CLL patients have elevated serum levels of IL-6 and IL-10(20). Since both stimulatory and inhibitory influences of IL-10 on CLL tumor cell growth and survival have been described(21;36), the effect of increased levels of IL-10 on tumor cells in vivo remains uncertain. Besides through increased serum levels of IL-6 and IL-10, the cytokine balance may be further pivoted towards Th2 because of reduced availability of IL-2 for T cells, caused by absorption by CLL cells (which express the IL-2R)(22). It is tempting to speculate that the disturbed cytokine balance in CLL renders pa-tients more susceptible to infections and provides a protective environment for tumor cells against tumor specific T cell responses, but at present there is no strong experimental evidence supporting this theory.Various reports address the role of CLL cells in the production of above-men-tioned cytokines. CLL cells are capable of producing IL-6(8) and IL-10(64), but whether they are the main source of serum IL-10 in vivo is not known. In contrast to these Th2 type cytokines, CLL cells have also been demonstrated to produce interferon gamma (IFN-γ), which subsequently may serve as an autocrine survival factor(12). This rather confusing finding raises the question whether altering the disturbed Th1/Th2 balance in CLL, for example through administration of cytokines, can have a beneficial effect on either malignant cells or on T cell dysfunctions. Identification of the cells responsible for the production of excessive amounts of Th2 cytokines in CLL however may give important clues as how to restore the Th1/Th2 balance.

Chapter 1

17

T cells: the good or the bad guys?

Tumor-specific T cellsIn healthy individuals, immune responses should only be elicited against non-self antigens. Despite this principle protects us against auto-immunity, this also implies that it may be difficult to raise tumor-specific immune responses, since tumor cells frequently highly resemble their non-malignant counter-parts. To become a tumor cell, a cell has to acquire a number of mutations, deletions or duplications of genes to escape internal control mechanisms that normally prevent oncogenesis(48). These changes within the tumor cell lead to an altered protein content which may be noticeable for CD8 positive T cells because peptides of the mutated or differentially expressed proteins are pre-sented on the cell surface in the context of MHC class I molecules(43;62). It has been postulated that in CLL, tumor-specific T cell clones are present that demonstrate cytotoxic potential against autologous tumor cells in vitro(42). Nevertheless, without in vitro manipulation, these clones are not capable of clearing a substantial amount of tumor cells in vivo, which may be explained in several ways. As described earlier in this chapter, CLL cells have a reduced expression of costimulatory molecules and increased surface expression of CD200, which may directly or indirectly affect the activation of tumor-spe-cific T cell responses. This is supported by the finding that T cells from CLL patients may be impaired in their capacity to differentiate towards a Th1 phe-notype, a feature which can also be induced in T cells from healthy donors through direct contact with CLL cells(25). The latter suggests a role for surface antigens in this process, but the molecules involved in this CLL-induced T cell suppression remain to be identified. Alternatively, the activation of tumor-di-rected T cell responses in CLL may be inhibited by regulatory T cells. These cells are discussed in the next paragraph.

Increased number of regulatory T cellsIt has been described that in CLL, as well as in other malignancies, there are increased numbers of regulatory T cells (Treg)(6). Since animal studies have indicated that Treg depletion can result in augmented tumor-rejection(66), it is possible that the increased amount of Treg in CLL patients may interfere with adequate tumor surveillance by tumor-specific T cells. At present, it is not clear what causes the increased amount of Treg in cancer patients. The results from a study by Beyer et al show that this T cell population in multi-

Introduction

18

ple myeloma patients is largely expanded after emerging from the thymus(7), suggesting that the expansion of this population is antigen-driven(71). This may imply the involvement of a tumor antigen in the formation and expan-

Figure 1. Mechanisms of escape from immune surveillance by CLL. Schematic over-view of possible interactions between CLL cells and the immune system in a lymphoid environment (lymph node or bone marrow) that either directly inhibit (tumor directed) T cell responses or interfere with adequate T cell activation by dendritic cells (DC). CD200R, CD200 receptor; CTL, cytotoxic T lymphocyte; IFN- γ, interferon-gamma; IL-10, interleukin-10; MHC-I, Major Histocompatibility Complex class I; TCR, T cell receptor; TH, helper T cell

sion of these Treg, but in the case of CLL may also very well be the result of the long-term exposure to chronic herpes viruses, like cytomegalovirus (CMV) or Epstein-Barr virus (EBV) that are carried by a large fraction of the CLL patient population(34). However, if Treg protect tumor cells against T cell at-tack, they may form an interesting target for therapy. Various studies suggest that Treg are very sensitive to apoptosis-inducing agents like agonistic Fas an-tibodies(23) and chemotherapeutic drugs(67). The latter is confirmed by the

TH

CLL

CD200

CD200R

CTL

CLL

regulatory T cells

?

IL-10

tumor-specificT cells

costimulatorymolecules

DC

IFN-γγγγ

?

TCR

MHC-I

Chapter 1

19

finding that in CLL, treatment with fludarabine results in a reduced number of Treg that are less suppressive than before treatment(6). Thus, the treatment of CLL with chemotherapeutic agents may serve as a double-edged sword.

Virus-specific T cellsThe knowledge of all above described mechanisms of defense against CTL attack and the possible ways to manipulate them remains trivial if a powerful tumor-specific T cell population is lacking. In CLL however there is one spe-cific T cell population that might be assigned to fulfill this role. During more advanced stages of the disease, CLL patients may suffer from various upper airway bacterial infections. The reactivation of CMV and EBV however is rarely observed in CLL patients. Nevertheless, the prevalence of CMV and EBV seropositivity within the CLL patient population is high(40), mainly because of the age of these patients. This demonstrates that the im-munity against this chronic herpes virus is very well maintained amidst all immune deregulations that occur in CLL. The efficiency of the suppression of CMV reactivation is supported by the observations that treatment of CLL patients with alemtuzumab frequently results in exacerbations of CMV(39). Since alemtuzumab targets all lymphocytes, it is likely that the observed reac-tivations of CMV under alemtuzumab treatment are caused by T cell deple-tion(11). The CD8 positive CTL population that is associated with chronic CMV infec-tion is very unique in that it possesses an effector-memory (EM) phenotype. This EM phenotype is marked by surface expression of CD45Ra and the lack of both CD27 and CD28 expression combined with high content of granzyme B and perforin(1;28). This allows CMV specific CTL to exert their cytotoxic function without the need of antigen presenting cell interference and makes them independent of costimulatory signals. During chronic CMV infection, the total CD8 positive T cell pool frequently consists for an important part of these EM T cells(24). Remarkably, in CMV seropositive CLL patients, this population is highly increased compared to healthy CMV seropositive indi-viduals(40). It is not clear why the CMV specific CD8 population is expanded in CLL, but apparently these T cells manage to prevent CMV reactivation. The capacity to induce target cell death without pre-stimulation and inde-pendence of costimulation makes CMV specific CTL an ideal weapon for ac-tive anti-tumor immunotherapy. Moreover, the use of CMV specific CTL in a therapeutic setting might provide a solution to the problem of therapy resis-

Introduction

20

tance due to p53 dysfunction since these cells induce apoptosis of target cells in a p53-independent manner. Different groups have appreciated the qual-ity of these T cells and have developed strategies to redirect the attention of these T cells towards tumor cells(30;34). Unfortunately, one major drawback of most active anti-tumor immunotherapy approaches is the requirement of ex vivo manipulation of patient cells. Recent studies have indicated that this may be circumvented through targeting of viral peptides presented in MHC class I (MHC-I) molecules to tumor cells in vivo(50;56). These peptide-loaded MHC-I molecules can be targeted to tumor cells via a single-chain antibody fragment that is specific for a desired tumor-antigen (figure 2). In case of CLL,

Figure 2. CMV specific CTL can be redirected against CLL. A streptavidin bound CD20 single chain antibody fragment (scFvCD20) is used to target CLL cells with biotinylated MHC class I molecules containing peptides from the immunodominant CMVpp65 protein. CMV specific cytotoxic T lymphocytes can recognize these targeted tumor cells and will be triggered to excrete cytotoxic molecules (of which granzyme B and perforin are the most important). These molecules subsequently induce death of the targeted CLL cell. CMV, cytomegalovirus; CTL, cytotoxic T lymphocyte; GrB, granzyme B; MHC-I, Major Histocompatibility Complex class I; TCR, T cell receptor.

CLL

CMV-specific

CTL

CD20

scFvCD20

MHC-I

CMVpp65

TCR

perforin

GrB

Chapter 1

21

it was demonstrated that the cytolytic capacity of CMV-specific CTL could be efficiently redirected against autologous CLL cells that were targeted with MHC-I molecules containing CMVpp65 peptide via a CD20 single-chain an-tibody fragment(47). Although further testing should provide insight in the efficiency of tumor cell targeting and killing in vivo, this strategy looks prom-ising for the development of an active immunotherapy for CLL, as the CMV specific CTL may not be hampered in their cytolytic response by the lack of costimulatory molecules on CLL. Moreover, as this antigen-specific targeting method avoids ex vivo manipulation, it seems perfectly applicable in a clini-cal setting. With growing knowledge on tumor-specific antigens, it should be possible to adapt this approach to other tumor types via tailor-made single-chain antibody fragments(17;51;52). Therefore, redirecting the power of CMV specific CTL towards tumor cells has the potency to become a feasible clinical approach towards active immunotherapy beyond CLL.

Concluding remarks

CLL is an example of a tumor that manages to escape immune surveillance through various mechanisms. On one hand, changes in the expression of apoptosis-regulating molecules protect the tumor cells against either CTL- or drug-induced cell death. On the other hand, both the tumor cells them-selves and changes in the immune system of CLL patients may suppress the induction of tumor-directed T cell responses. This combination unfortunately also hampers the development of successful immunotherapeutic strategies to-wards CLL. Recent research has focused on manipulating CLL cells to facili-tate killing by tumor-specific CTL or cytostatic drugs. Alternatively, attention is directed to manipulating T cell populations to elicit T cell driven anti-tumor responses. Possibly, combining both strategies may have a synergistic effect and therefore result in a therapy that is highly tumor-cell specific at cost of low side effects.

Introduction

22

Aim of this thesis

CLL is a malignancy that is accompanied by various immune disorders and drug-resistance mechanisms that severely complicate the treatment of the dis-ease. The objectives of this thesis are (1) to provide more insight into CLL associated immune dysfunction, (2) to improve the understanding of the of genes involved in apoptosis regulation that leads to drug-resistance and (3) to develop new strategies that either circumvent or overcome resistance to treat-ment. In chapter 2 of this thesis, we address the increased number of regulatory T cells (Treg) in CLL, which might protect CLL cells against tumor surveillance. In this perspective, two possible mechanisms that might facilitate Treg accu-mulation are explored: antigen stimulation and resistance to apoptosis. Chapter 3 of this thesis addresses another important problem complicating the treatment of CLL: drug resistance due to dysfunction of the p53 pathway. Unfortunately, the current available technique (FISH) often fails to detect p53 dysfunction in clinical samples at diagnosis. It is therefore explored in chapter 3 whether MLPA (a semi-quantative PCR assay) may serve as a tool to in-crease the detection rate of p53 dysfunction in CLL samples. The last chapters of this thesis focus on the development of a therapeutic strat-egy that acts in a p53-independent manner. It is investigated whether the redi-rection of virus-specific cytotoxic T lymphocytes towards CLL cells may serve as an active immunotherapy for this leukemic malignancy. In chapter 4, we describe a targeted complex that can be used to redirect CMV specific CTL to-wards tumor cells. Subsequently, the efficacy of this complex to activate virus-specific CTL and induce tumor cell killing is investigated by a detailed study of (mechanisms) of T cell activation (chapter 5). In chapter 6, we address the use of small molecule inhibitors of XIAP as therapeutic agent in combination with immunotherapy. To this end, it is investigated whether CLL cells (which are resistant to apoptosis) can be sensitized to CTL mediated killing via small molecule inhibitors of XIAP. Finally, in chapter 7, we discuss the relevance of the data described in chapters 2-6 as to the future approach towards treatment of CLL.

Chapter 1

23

Reference List

1. Appay V, Dunbar PR, Callan M, et al. Memory CD8+ T cells vary in differentiation pheno-type in different persistent virus infections. Nat Med. 2002;8:379-385. 2. Austen B, Powell JE, Alvi A, et al. Mutations in the ATM gene lead to impaired overall and treatment-free survival that is independent of IGVH mutation status in patients with B-CLL. Blood. 2005;106:3175-3182. 3. Baldini L, Cro L, Cortelezzi A, et al. Immunophenotypes in “classical” B-cell chronic lym-phocytic leukemia. Correlation with normal cellular counterpart and clinical findings. Can-cer. 1990;66:1738-1742. 4. Barnes PD, Grundy JE. Down-regulation of the class I HLA heterodimer and beta 2-mi-croglobulin on the surface of cells infected with cytomegalovirus. J Gen Virol. 1992;73 ( Pt 9):2395-2403. 5. Barry M, Heibein JA, Pinkoski MJ, et al. Granzyme B short-circuits the need for caspase 8 activity during granule-mediated cytotoxic T-lymphocyte killing by directly cleaving Bid. Mol Cell Biol. 2000;20:3781-3794. 6. Beyer M, Kochanek M, Darabi K, et al. Reduced frequencies and suppressive function of CD4+CD25hi regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood. 2005;106:2018-2025. 7. Beyer M, Kochanek M, Giese T, et al. In vivo peripheral expansion of naive CD4+CD25high FoxP3+ regulatory T cells in patients with multiple myeloma. Blood. 2006;107:3940-3949. 8. Biondi A, Rossi V, Bassan R, et al. Constitutive expression of the interleukin-6 gene in chronic lymphocytic leukemia. Blood. 1989;73:1279-1284. 9. Boshkov LK, Macen JL, McFadden G. Virus-induced loss of class I MHC antigens from the surface of cells infected with myxoma virus and malignant rabbit fibroma virus. J Immunol. 1992;148:881-887. 10. Bots M, Medema JP. Granzymes at a glance. J Cell Sci. 2006;119:5011-5014. 11. Brett S, Baxter G, Cooper H, et al. Repopulation of blood lymphocyte sub-populations in rheumatoid arthritis patients treated with the depleting humanized monoclonal antibody, CAMPATH-1H. Immunology. 1996;88:13-19. 12. Buschle M, Campana D, Carding SR, et al. Interferon gamma inhibits apoptotic cell death in B cell chronic lymphocytic leukemia. J Exp Med. 1993;177:213-218.

Introduction

24

13. Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A. 2002;99:15524-15529. 14. Callet-Bauchu E, Salles G, Gazzo S, et al. Translocations involving the short arm of chro-mosome 17 in chronic B-lymphoid disorders: frequent occurrence of dicentric rearrange-ments and possible association with adverse outcome. Leukemia. 1999;13:460-468. 15. Cantwell M, Hua T, Pappas J, et al. Acquired CD40-ligand deficiency in chronic lympho-cytic leukemia. Nat Med. 1997;3:984-989. 16. Catalfamo M, Henkart PA. Perforin and the granule exocytosis cytotoxicity pathway. Curr Opin Immunol. 2003;15:522-527. 17. Cheung NK, Modak S, Lin Y, et al. Single-chain Fv-streptavidin substantially improved therapeutic index in multistep targeting directed at disialoganglioside GD2. J Nucl Med. 2004;45:867-877. 18. Chiu VK, Walsh CM, Liu CC, et al. Bcl-2 blocks degranulation but not fas-based cell-me-diated cytotoxicity. J Immunol. 1995;154:2023-2032.

19. Dreger P, Brand R, Milligan D, et al. Reduced-intensity conditioning lowers treatment-related mortality of allogeneic stem cell transplantation for chronic lymphocytic leukemia: a population-matched analysis. Leukemia. 2005;19:1029-1033.

20. Fayad L, Keating MJ, Reuben JM, et al. Interleukin-6 and interleukin-10 levels in chron-ic lymphocytic leukemia: correlation with phenotypic characteristics and outcome. Blood. 2001;97:256-263. 21. Fluckiger AC, Durand I, Banchereau J. Interleukin 10 induces apoptotic cell death of B-chronic lymphocytic leukemia cells. J Exp Med. 1994;179:91-99. 22. Foa R, Giovarelli M, Jemma C, et al. Interleukin 2 (IL 2) and interferon-gamma produc-tion by T lymphocytes from patients with B-chronic lymphocytic leukemia: evidence that normally released IL 2 is absorbed by the neoplastic B cell population. Blood. 1985;66:614-619. 23. Fritzsching B, Oberle N, Eberhardt N, et al. In contrast to effector T cells, CD4+CD25+FoxP3+ regulatory T cells are highly susceptible to CD95 ligand- but not to TCR-mediated cell death. J Immunol. 2005;175:32-36. 24. Gamadia LE, Rentenaar RJ, Baars PA, et al. Differentiation of cytomegalovirus-specific CD8(+) T cells in healthy and immunosuppressed virus carriers. Blood. 2001;98:754-761.

Chapter 1

25

25. Gorgun G, Holderried TA, Zahrieh D, et al. Chronic lymphocytic leukemia cells induce changes in gene expression of CD4 and CD8 T cells. J Clin Invest. 2005;115:1797-1805. 26. Grever MR, Lucas DM, Dewald GW, et al. Comprehensive assessment of genetic and mo-lecular features predicting outcome in patients with chronic lymphocytic leukemia: results from the US Intergroup Phase III Trial E2997. J Clin Oncol. 2007;25:799-804.

27. Hachem A, Gartenhaus RB. Oncogenes as molecular targets in lymphoma. Blood. 2005;106:1911-1923. 28. Hamann D, Baars PA, Rep MH, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med. 1997;186:1407-1418. 29. Hanada M, Delia D, Aiello A, et al. bcl-2 gene hypomethylation and high-level expression in B-cell chronic lymphocytic leukemia. Blood. 1993;82:1820-1828. 30. Heemskerk MH, Hoogeboom M, Hagedoorn R, et al. Reprogramming of virus-spe-cific T cells into leukemia-reactive T cells using T cell receptor gene transfer. J Exp Med. 2004;199:885-894. 31. Hoek RM, Ruuls SR, Murphy CA, et al. Down-regulation of the macrophage lineage through interaction with OX2 (CD200). Science. 2000;290:1768-1771. 32. Kashkar H, Seeger JM, Hombach A, et al. XIAP targeting sensitizes Hodgkin lymphoma cells for cytolytic T-cell attack. Blood. 2006;108:3434-3440.

33. Kater AP, Dicker F, Mangiola M, et al. Inhibitors of XIAP sensitize CD40-activated chronic lymphocytic leukemia cells to CD95-mediated apoptosis. Blood. 2005;106:1742-1748. 34. Kater AP, Remmerswaal EB, Nolte MA, et al. Autologous cytomegalovirus-specific T cells as effector cells in immunotherapy of B cell chronic lymphocytic leukaemia. Br J Haematol. 2004;126:512-516. 35. Kato K, Cantwell MJ, Sharma S, et al. Gene transfer of CD40-ligand induces autologous immune recognition of chronic lymphocytic leukemia B cells. J Clin Invest. 1998;101:1133-1141. 36. Kitabayashi A, Hirokawa M, Miura AB. The role of interleukin-10 (IL-10) in chronic B-lymphocytic leukemia: IL-10 prevents leukemic cells from apoptotic cell death. Int J Hematol. 1995;62:99-106.

37. Kretz-Rommel A, Qin F, Dakappagari N, et al. CD200 Expression on Tumor Cells Sup-presses Antitumor Immunity: New Approaches to Cancer Immunotherapy. J Immunol. 2007;178:5595-5605.

Introduction

26

38. Lassam N, Jay G. Suppression of MHC class I RNA in highly oncogenic cells occurs at the level of transcription initiation. J Immunol. 1989;143:3792-3797. 39. Lundin J, Kimby E, Bjorkholm M, et al. Phase II trial of subcutaneous anti-CD52 mono-clonal antibody alemtuzumab (Campath-1H) as first-line treatment for patients with B-cell chronic lymphocytic leukemia (B-CLL). Blood. 2002;100:768-773. 40. Mackus WJ, Frakking FN, Grummels A, et al. Expansion of CMV-specific CD8+CD45RA+. Blood. 2003;102:1057-1063. 41. Majolini MB, D’Elios MM, Galieni P, et al. Expression of the T-cell-specific tyrosine kinase Lck in normal B-1 cells and in chronic lymphocytic leukemia B cells. Blood. 1998;91:3390-3396. 42. Mayr C, Bund D, Schlee M, et al. MDM2 is recognized as a tumor-associated antigen in chronic lymphocytic leukemia by CD8+ autologous T lymphocytes. Exp Hematol. 2006;34:44-53. 43. Mayr C, Bund D, Schlee M, et al. Fibromodulin as a novel tumor-associated antigen (TAA) in chronic lymphocytic leukemia (CLL), which allows expansion of specific CD8+ autologous T lymphocytes. Blood. 2005;105:1566-1573. 44. McWhirter JR, Kretz-Rommel A, Saven A, et al. Antibodies selected from combinatorial libraries block a tumor antigen that plays a key role in immunomodulation. Proc Natl Acad Sci U S A. 2006;103:1041-1046.

45. Medema JP, de JJ, Peltenburg LT, et al. Blockade of the granzyme B/perforin pathway through overexpression of the serine protease inhibitor PI-9/SPI-6 constitutes a mechanism for immune escape by tumors. Proc Natl Acad Sci U S A. 2001;98:11515-11520. 46. Moreaux J, Hose D, Reme T, et al. CD200 is a new prognostic factor in multiple myeloma. Blood. 2006;108:4194-4197. 47. Mous R, Savage P, Remmerswaal EB, et al. Redirection of CMV-specific CTL towards B-CLL via CD20-targeted HLA/CMV complexes. Leukemia. 2006;20:1096-1102. 48. Murray AW. Creative blocks: cell-cycle checkpoints and feedback controls. Nature. 1992;359:599-604. 49. O’Neill DW, Adams S, Bhardwaj N. Manipulating dendritic cell biology for the active im-munotherapy of cancer. Blood. 2004;104:2235-2246.

Chapter 1

27

50. Ogg GS, Dunbar PR, Cerundolo V, et al. Sensitization of tumour cells to lysis by virus-spe-cific CTL using antibody-targeted MHC class I/peptide complexes. Br J Cancer. 2000;82:1058-1062. 51. Osbourn JK, Field A, Wilton J, et al. Generation of a panel of related human scFv antibod-ies with high affinities for human CEA. Immunotechnology. 1996;2:181-196. 52. Pavlinkova G, Batra SK, Colcher D, et al. Constructs of biotin mimetic peptide with CC49 single-chain Fv designed for tumor pretargeting. Peptides. 2003;24:353-362. 53. Restifo NP, Esquivel F, Kawakami Y, et al. Identification of human cancers deficient in antigen processing. J Exp Med. 1993;177:265-272. 54. Rosenwald A, Chuang EY, Davis RE, et al. Fludarabine treatment of patients with chronic lymphocytic leukemia induces a p53-dependent gene expression response. Blood. 2004;104:1428-1434.

55. Rotem-Yehudar R, Winograd S, Sela S, et al. Downregulation of peptide transporter genes in cell lines transformed with the highly oncogenic adenovirus 12. J Exp Med. 1994;180:477-488. 56. Savage P, Cowburn P, Clayton A, et al. Anti-viral cytotoxic T cells inhibit the growth of cancer cells with antibody targeted HLA class I/peptide complexes in SCID mice. Int J Can-cer. 2002;98:561-566. 57. Scheppler JA, Nicholson JK, Swan DC, et al. Down-modulation of MHC-I in a CD4+ T cell line, CEM-E5, after HIV-1 infection. J Immunol. 1989;143:2858-2866.

58. Schetelig J, Thiede C, Bornhauser M, et al. Evidence of a graft-versus-leukemia ef-fect in chronic lymphocytic leukemia after reduced-intensity conditioning and allogeneic stem-cell transplantation: the Cooperative German Transplant Study Group. J Clin Oncol. 2003;21:2747-2753. 59. Schimmer AD, Welsh K, Pinilla C, et al. Small-molecule antagonists of apoptosis suppres-sor XIAP exhibit broad antitumor activity. Cancer Cell. 2004;5:25-35. 60. Schliep S, Decker T, Schneller F, et al. Functional evaluation of the role of inhibitor of apoptosis proteins in chronic lymphocytic leukemia. Exp Hematol. 2004;32:556-562. 61. Shanafelt TD, Witzig TE, Fink SR, et al. Prospective evaluation of clonal evolution during long-term follow-up of patients with untreated early-stage chronic lymphocytic leukemia. J Clin Oncol. 2006;24:4634-4641.

Introduction

28

62. Siegel S, Wagner A, Kabelitz D, et al. Induction of cytotoxic T-cell responses against the oncofetal antigen-immature laminin receptor for the treatment of hematologic malignancies. Blood. 2003;102:4416-4423. 63. Silber R, Degar B, Costin D, et al. Chemosensitivity of lymphocytes from patients with B-cell chronic lymphocytic leukemia to chlorambucil, fludarabine, and camptothecin analogs. Blood. 1994;84:3440-3446. 64. Sjoberg J, guilar-Santelises M, Sjogren AM, et al. Interleukin-10 mRNA expression in B-cell chronic lymphocytic leukaemia inversely correlates with progression of disease. Br J Hae-matol. 1996;92:393-400. 65. Sturm I, Bosanquet AG, Hermann S, et al. Mutation of p53 and consecutive selective drug resistance in B-CLL occurs as a consequence of prior DNA-damaging chemotherapy. Cell Death Differ. 2003;10:477-484. 66. Sutmuller RP, van Duivenvoorde LM, van EA, et al. Synergism of cytotoxic T lympho-cyte-associated antigen 4 blockade and depletion of CD25(+) regulatory T cells in antitumor therapy reveals alternative pathways for suppression of autoreactive cytotoxic T lymphocyte responses. J Exp Med. 2001;194:823-832. 67. Taams LS, Smith J, Rustin MH, et al. Human anergic/suppressive CD4(+)CD25(+) T cells: a highly differentiated and apoptosis-prone population. Eur J Immunol. 2001;31:1122-1131. 68. Tamm I, Kornblau SM, Segall H, et al. Expression and prognostic significance of IAP-fam-ily genes in human cancers and myeloid leukemias. Clin Cancer Res. 2000;6:1796-1803. 69. Van DA, Gao L, Stauss HJ, et al. Antigen-specific cellular immunotherapy of leukemia. Leukemia. 2005;19:1863-1871. 70. Versteeg R, Kruse-Wolters KM, Plomp AC, et al. Suppression of class I human histocom-patibility leukocyte antigen by c-myc is locus specific. J Exp Med. 1989;170:621-635. 71. Vukmanovic-Stejic M, Zhang Y, Cook JE, et al. Human CD4+ CD25hi Foxp3+ regu-latory T cells are derived by rapid turnover of memory populations in vivo. J Clin Invest. 2006;116:2423-2433. 72. Wierda WG, Cantwell MJ, Woods SJ, et al. CD40-ligand (CD154) gene therapy for chronic lymphocytic leukemia. Blood. 2000;96:2917-2924.

73. Wright GJ, Cherwinski H, Foster-Cuevas M, et al. Characterization of the CD200 receptor family in mice and humans and their interactions with CD200. J Immunol. 2003;171:3034-3046.

Chapter 1

29

74. Wright GJ, Puklavec MJ, Willis AC, et al. Lymphoid/neuronal cell surface OX2 glycopro-tein recognizes a novel receptor on macrophages implicated in the control of their function. Immunity. 2000;13:233-242.

Introduction

2Enhanced survival and increased

formation of regulatory T cells in CLLRogier Mous1*, Margot Jak1*, Ester BM Remmerswaal2, René Spijker1, Annelieke Jaspers2, Adriana Yagüe1, Eric Eldering2, René AW van Lier2

and Marinus HJ van Oers1


*both authors contributed equally to this study

Submitted

33

Abstract

Recently it has been described that chronic lymphocytic leukemia (CLL) pa-tients have increased numbers of regulatory T cells (Treg), but the cause of this expansion is unknown. In the present study, we analysed the mechanism behind Treg expansion in CLL. Neither analysis of the T cell receptor (TCR) repertoire nor CD45 iso-form expression of Treg from CLL patients provided evidence for chronic antigenic stimulation. We found evidence however for increased formation of Treg via CD70 costimulation. CD40 ligand activated CLL cells (a putative model of lymph node CLL cells) strongly induced the formation of Treg via CD27-CD70 costimulation. RT-MLPA expression analysis of 34 apoptosis-regulating genes showed that in comparison to other CD4+ T cells, Treg of both healthy individuals and CLL patients had a high expression of pro-apoptotic Noxa and a low expression of anti-apoptotic Bcl-2. However, Bcl-2 levels of Treg in CLL patients were significantly higher than in healthy individuals. Finally, at the functional level, Treg from CLL patients were more resistant to drug-induced apoptosis than Treg from healthy individuals. In conclusion, Treg in CLL may accumulate both by increased formation, fa-cilitated by CD27-CD70 interaction in the lymph node proliferation centres, and decreased sensitivity to apoptosis.

Increased numbers of Treg in CLL

34

Introduction

Chronic lymphocytic leukemia (CLL) is characterized by the slow accumula-tion of mature CD5+/CD19+ B cells. Notably, CLL patients frequently also have increased numbers of circulating CD4+ and CD8+ T cells(26;31). So far there are no data supporting CLL-specificity of these expanded T cell populations. In contrast, we have shown that CLL patients can have increased numbers of CMV-specific CD8+ memory effector cells(19). Recently it has been described that CLL patients have increased numbers of CD4+/CD25bright regulatory T cells (Treg)(4), which possess normal sup-pressive capacity. So far, the mechanism of this Treg expansion in CLL is not known. In healthy adults, the majority of Treg are generated in the periph-ery (CD4+/CD25bright/Foxp3+/CD45R0+)(1) in contrast to naturally occurring Treg (CD4+/CD25bright/Foxp3+/CD45RA+) which are generated in the thymus. It seems likely that adaptive CD4+ Treg in adults are continuously produced from the memory CD4+ T cell pool(37), since most Treg possess a memory phenotype (CD45R0+) and the T cell receptor repertoire of Treg shows a high homology to that of the CD4+ memory T cell pool. However, the turnover of Treg appears to be much faster than that of memory T cells(37) and since in healthy individuals there is no steady increase in Treg numbers the increased proliferation rate apparently is counterbalanced by regulated apoptosis of Treg (11;33). Thus, the increase of Treg in CLL might be due either to increased proliferation, decreased apoptosis or both. In the present study we have examined potential mechanisms of increased Treg numbers in CLL. Possible expansion by chronic antigenic stimulation was evaluated by analysis of the T cell receptor repertoire and the expression of differentiation markers on Treg. Furthermore, we investigated whether CLL cells themselves play a stimulatory role in the formation of Treg. Finally, al-terations in apoptosis of Treg were studied by expression profiling of 34 apop-tosis regulating genes as well as by assessment of Treg sensitivity to cytotoxic drugs.

Chapter 2

35

Methods

Cells from CLL patients and healthy individualsPeripheral blood was drawn from CLL patients (diagnosed according to the NCI-WG guidelines) as well as from healthy volunteers. PBMC were isolated and either used immediately or stored in liquid nitrogen. During all in vitro experiments, cells were maintained in Iscove’s modified Dulbecco medium (IMDM: Gibco Life technology, Paisley, USA) supplemented with 10% heat-inactivated fetal calf serum, penicillin, gentamycin and β-mercaptoethanol. All PMBC samples contained at least 90% CD5+/CD19+ cells as assessed via flow cytometry. The studies were approved by the Ethical Review Board of the Institute and from all participants written informed consent was obtained. Patient characteristics are listed in table 1.

Flow cytometryPBMC were stained using antibodies against CD4, CD8, CD5, CD19, CD25, CD45RA or CD45R0 and CD127 (all Becton Dickinson, San Jose, CA) or with antibodies against CD95 (IQ products, Groningen, the Netherlands). For in-tracellular staining, cells were fixed and permeabilised (eBioscience, San Di-ego, CA) and subsequently stained for Bcl-2 (Dako, Glostrup, Denmark), Ki-67 (BD Pharmingen, San Jose, CA) and Foxp3 (eBioscience, San Diego, CA). Antibody stained cell samples were then analysed by flow cytometry.

Isolation of T cell populationsThawed PMBC from either healthy individuals or CLL patients were stained with antibodies against CD4, CD25 and CD127 (all Becton Dickinson, San Jose, CA). Subsequently, Treg (CD4+/CD25bright/CD127low) and non-Treg CD4+ T cells (CD4+/CD25-/CD127+) were obtained by cell sorting (FACS Aria, Bec-ton Dickinson, San Jose, CA). The isolated cells were immediately lysed to prepare RNA or perform protein isolation. Treg enriched populations con-tained approximately 80% CD4+/Foxp3+ cells as assessed by flow cytometry.

Analysis of Vβ repertoireRNA isolated from sorted T cells was subjected to template switch-anchored reverse transcriptase–polymerase chain reaction (RT-PCR) by using Super Smart PCR cDNA Synthesis Kit (BD Biosciences Clontech, Palo Alto, CA). Vβ PCR was performed on amplicons as described previously (21). For the spec-


36

tratyping, samples were mixed with Genescan-500 ROX size standards and run on an ABI 3100 capillary sequencer (Applied Biosystems, Warrington, United Kingdom) in Genescan mode.

In vitro CD40 ligand stimulation of CLL cellsPBMC from CLL patients (> 90% CD5+/CD19+ cells) were stimulated with CD40 ligand (CD40L) transfected NIH3T3 (3T40L) cells as described previ-ously(16). Briefly, CLL cells were added to 6-well plates coated with gamma irradiated (30 Gy) CD40L transfected NIH3T3 cells. Non-transfected 3T3 cells were used as negative controls. After 2 days, the CLL cells were gently removed from the fibroblast layer and used in further experiments.

Cell stimulation cultures (CSC)CSC were performed with 3T3- or 3T40L-stimulated CLL cells (APC) and PMBC of a healthy individual or (autologous) CLL patient (responders) in a 1:1 ratio (2 x 105 stimulators: 2 x 105 responders). Cells were cultured in 96-wells plates (Costar, Corning Inc., NY, USA) in the presence of soluble CD3 mAb (clone CLB-T3.4/E)(34) and in the presence or absence of a blocking CD70 mAb (clone CLB-2F2)(14). After 4 days cells were harvested and Foxp3 expression was analyzed by flow cytometry as described above.

Reverse transcription–multiplex ligation-dependent probe amplification assayReverse transcription–multiplex ligation-dependent probe amplification assay (RT-MLPA) procedure was performed as described previously(10). Briefly, 100 ng total RNA as obtained from sorted T cell populations was re-verse transcribed using a gene-specific probe mix. The resulting cDNA was annealed overnight at 60°C to the MLPA probes. Annealed oligonucleotides were covalently linked by Ligase-65 (MRC, Amsterdam, The Netherlands) at 54°C. Ligation products were amplified by polymerase chain reaction (PCR; 33 cycles, 30 seconds at 95°C, 30 seconds at 60°C, and 1 minute at 72°C) us-ing one unlabeled and one 6-carboxy-fluorescein–labelled primer (10 pM). PCR products were size separated on an ABI 3100 capillary sequencer in the presence of 1 pM ROX 500 size standard (Applied Biosystems, Warrington, United Kingdom). Results were analyzed using the programs Genescan analy-sis and Genotyper (Applied Biosystems). Category tables containing the area for each assigned peak (scored in arbitrary units) were compiled in Genotyper

Chapter 2

37

and exported for further analysis with Excel spreadsheet software (Microsoft, Redman, WA). Data were normalized by setting the sum of all signals at 100% and expressing individual peaks relative to the 100% value. The thus obtained expression levels of all tested genes in Treg populations (see isolation of T cell populations) were then compared to the levels found in the naïve-memory cell population and reflected as relative expression (gene expression in Treg set as 1).

Quantative PCR analysis of Noxa expression20 ng of RNA extracted from sorted cell populations (see analysis of Vβ rep-ertoire) was used to synthetize cDNA with superscript II reverse transcriptase (Invitrogen, Carlsbad, CA). From these cDNA pools, specific targets were am-plified by PCR performed with Lightcycler FastStart DNA Master SYBR Green I (Roche Diagnostics, Indianapolis, IN), using the sense and antisense Noxa primers 5’-GGAGATGCCTGGGAAGAAGG-3’ and 5’-TCAGGTTCCT-GAGCAGAAGAG-3’ and the 18S primers 5’-GGACAACAAGCTCCGT-GAAGA-3’ and 5’-CAGAAGTGACGCAGCCCTCTA-3’ respectively. The re-sults were normalized to 18S. Thus obtained values for Treg were set as 1 and compared to values of non-Treg (relative expression).

Western blottingWestern blotting was performed as described previously(20). Protein samples were separated by 13% sodium dodecyl sulfate–polyacrylamide gel electro-phoresis followed by Western blotting. Blots were probed with the following antibodies or antisera: polyclonal anti-Mcl-1 (BD Pharmingen, San Jose, CA), monoclonal anti-Noxa (Imgenex, San Diego, CA), anti-Bcl-XL (Transduction Laboratories, Lexington, KY), rabbit-anti-Bcl-2 (Alexis Biochemicals, San Diego, CA) or antiserum to β-actin (Santa Cruz Biotechnology, Santa Cruz, CA).

Drug sensitivity assaysPBMC of both CLL patients and healthy individuals were incubated with vari-ous concentrations of Roscovitine (Sigma Aldrich, St Louis, MO), fludarabine (Sigma-Aldrich, St Louis, MO) or agonistic Fas antibody CH11 (Beckman Coulter, Fullerton, CA). After 24 hours, cells were fixed and permeabilised and stained for CD3, CD4, CD25 (all Beckton Dickinson) and Foxp3 (eBio-science) to determine the percentage of regulatory T cells (Treg) out of the


38

total CD4 population. The obtained values were then normalized by calculat-ing the percentage of Treg remaining after drug stimulation compared to non-stimulated as follows: (Foxp3+/(CD4+)drug stimulated)/(Foxp3+/(CD4+)non-stimulated) x 100%.Alternatively, drug-treated PBMC samples were fixed and analyzed for the presence of fragmented DNA (permeabilization buffer containing 0.1mM EDTA, 10mg ml-1 propidium iodide and 50mg ml-1 RNase-I) or cleaved cas-pase-3 (BD Pharmingen, San Jose, CA) within the Foxp3 positive and nega-tive population.

StatisticsThe two-tailed Mann-Whitney U test was used to analyze differences between 2 groups. Alternatively, the Wilcoxon matched paired test was used to analyze differences between paired samples. P-values < 0.05 were considered statisti-cally significant.

Chapter 2

39

ID gender age (yr) Rai stage IgVH muta-tion status

prior therapy

1 m 60 2 mut no Tx2 m 86 2 mut no Tx3 m 52 0 unmut no Tx

4* f 28 4 unmut F, CA during analysis5 m 65 1 unmut no Tx6 f 65 1 unmut F during analysis7 m 52 0 unmut no Tx8 m 84 0 mut no Tx9 m 83 0 unmut no Tx

10* m 35 2 unmut F,C,R > 1yr11 m 62 2 mut no Tx12 f 75 0 mut no Tx13 m 74 0 mut no Tx14 f 59 0 mut no Tx15 m 73 1 mut no Tx16 f 73 1 mut CA < 1yr17 f 59 1 ND no Tx18 m 78 2 unmut CA < 1yr19 f 42 2 mut CA < 1yr20 f 62 0 mut no Tx21 m 54 2 mut no Tx

Patient characteristics including gender, age, Rai stage, mutation of IgVH genes and prior therapy. (f= female, m = male, mut = mutated IgVH genes, unmut = unmu-tated IgVH genes, ND = not determined, C= cyclophosphamide, CA= chlorambu-cil, F= fludarabine, R= rituximab, no Tx = no therapy). Patients indicated with an asterisk (*) showed highly progressive disease.

Table 1. Patient characteristics.


40

Results

CLL patients have increased numbers of regulatory T cells In agreement with our previous studies in CLL patients we found increased numbers of CMV-specific CD8+/CD45RA+/CD27- cells(19) but no increase in the CMV-associated CD4+/CD27-/CD28- T cell population(35) (data not shown). In addition, we observed an increase in numbers of CD4+/CD25bright/CD127low T cells (figure 1A). This phenotype has been associated with Foxp3 expression and regulatory function(17;30). Indeed, counterstaining with Foxp3 antibody showed predominantly Foxp3 positive T cells within the CD4+/CD25bright/CD127low population (figure 1B), thereby confirming recent findings that CLL patients have increased numbers of regulatory T cells(4) (Treg).

No evidence for predominant antigen involved in Treg formationA possible mechanism behind Treg formation and/or maintenance is (chronic) antigenic stimulation. Recently it has been demonstrated that upon antigenic stimulation, a limited number of Treg clones arises with the same T cell recep-tor (TCR) as the antigen-specific T cells from the effector cell pool(37). Thus,

A B

HD CLL0

50

100

150

200

250

x 10

4 CD

4+ /CD

25br

ight

/IL7R

low

ml-1 p = 0.0173

CD

25

Foxp

3

IL7R

R1 (9%)

R1CD4

82%

isotype

Figure 1. Analysis of T cell populations in CLL patients. CD4+ T cells of CLL patients (CLL) were analyzed by flow cytometry and compared to T cells of age-matched healthy individuals (HD). A Absolute numbers of CD4+/CD25bright/IL7Rlow T cells. B Foxp3 counterstaining of CD4+/C25bright/IL7Rlow population of a CLL patient. After gating on CD4+ lymphocytes, CD25 was plotted against IL7R (left dot plot). The CD25bright/IL7Rlow population is indicated with a region mark (R1). Within R1, the percentage of Foxp3 positive cells was determined (right dot plot). Dot plots are representative for both CLL patients and healthy individuals; the percentage of CD-25bright/IL7Rlow within the CD4+ T cell population varied between donors.

Chapter 2

41

Figure 2. T cell recep-tor (TCR) repertoire analysis and phenotype of Treg from CLL pa-tients. A and B CD4+/CD25bright/IL7Rlow (Treg) and CD4+/CD25-/IL7R+ (CD4; non-Treg T cells) from 2 CLL patients (CLL) and 2 healthy individuals (HD) were sorted by FACS. RNA isolated from these T cell populations was used as input for Vβ PCR assays. The figures displayed are representative for both CLL patients and both healthy individuals, re-spectively. A Vβ repertoire of each T cell population (specified on the left). The bands on gel represent the product of each individual Vβ PCR (see bottom of each gel). The markers on the

right indicate the area in which the product for each PCR is expected. * size marker.B Spectratyping of 3 randomly chosen Vβ families. Each peak represents a CDR3 region with a certain length. On top of the picture, the different Vβ families are indicated; the sorted T cell populations are indicated on the left.

an antigenic “fingerprint” is present within both the effector T cell pool and the Treg population. To see whether a predominant antigen may be involved in the formation or maintenance of the Treg population in CLL patients, we screened the TCR repertoire of CLL patients (n=2) as well as the TCR reper-toire recovered from Treg of healthy individuals (n=2) and compared this rep-ertoire to that of non-Treg CD4+ T cells (combination of naïve and memory


42

CD4 cells) from the same individuals. We observed that the complete range of Vβ genes was used in Treg of both healthy individuals and CLL patients (figure 2A). Fragment length analysis (spectratyping) of 3 randomly chosen Vβ family PCR products showed similar peak patterns for both non-Treg T cells and Treg (figure 2B) in all individuals, making the involvement of a pre-dominant antigen in Treg formation in CLL patients less likely. Only within the Vβ11 family, a discrepant peak was observed in Treg compared to the non-Treg CD4+ T cell population, but this peak was present in Treg of both CLL patients and healthy individuals. Next, we examined the “antigen experience” of Treg by determining the sur-face expression of CD45R0. In line with previous studies(37), we found that the vast majority of the Treg in these adult individuals have an antigen-ex-perienced phenotype when compared to non-Treg T cells, as characterized by surface expression of CD45R0(37) (figure 2C). We observed no difference in the percentage of CD45R0 positive Treg between healthy individuals and CLL patients. Finally, we noted that CMV seropositivity did not influence the percentage of CD45R0 positive Treg (figure 2D).

Figure 2. C PBMC from CLL patients (n=21) and healthy individuals (n=6) were stained for CD3, CD4, Foxp3 and CD45R0. The percentage of CD45R0+ cells within the Treg (CD3+/CD4+/Foxp3+) and non-Treg CD4 (CD3+/CD4+/Foxp3-) population is plot-ted. D The percentage of CD45R0+ cells within Treg and non-Treg CD4 populations of CLL patients as depicted in figure 2C, separated according to CMV serology.

Chapter 2

43

CD40 ligand-stimulated CLL cells induce Foxp3 expression in CD4+ T cells in a CD70 dependent mannerRecently, it has been shown that CD70+ non-Hodgkin’s lymphoma (NHL) B cells can induce Treg via CD70 costimulation(40). In contrast to periph-eral blood CLL cells, CD40 ligand (CD40L) stimulated CLL cells (which resemble CLL cells from a lymph node environment(32)) have high CD70 surface expression(16;28). Therefore, we hypothesized that in a lymph node environment CLL cells might facilitate the formation of Treg. To test this, we performed cell stimulation cultures (CSC) using CLL cells that were pre-stim-ulated with 3T3 or CD40L transduced 3T3 cells (3T40L; see methods) as APC and PBMC from a healthy individual or autologous PBMC from CLL patients as responder cells. All CSC were performed in the presence of mitogenic CD3 mAb (see methods). After 4 days, the cells were harvested and analyzed by flow cytometry. After CD40L-stimulation of CLL cells, both the percentage of CD5+/CD19+/CD70+ cells and CD70 mean fluorescence intensity (MFI) strongly increased (figure 3A). Strikingly, we observed that CD40L-stimu-lated CLL cells augmented Foxp3 expression in CD4+ T cells of a healthy in-dividual. Moreover, this augmentation could be blocked by CD70 antibodies (2F2; figure 3B). To test if this effect was also present in an autologous setting, CD40L-stimulated CLL cells were used to stimulate autologous T cells. Also here, CD40L-stimulated CLL cells augmented Foxp3 expression in autologous CD4+ T cells (n=4), which could again be blocked by CD70 antibodies (figure 3C).

High expression of Noxa and low Bcl-2 characterize pro-apoptotic profile of TregSince Treg have been described to be highly susceptible to apoptosis(33), we investigated whether this might be related to the expression of pro- or anti-apoptotic molecules. To establish the ‘apoptotic profile’ of Treg, both non-Treg CD4+ T cells and Treg of 3 healthy individuals were sorted based on IL2R and IL7R expression(30). RNA was extracted from these T cell populations and used as input for RT-MLPA expression analysis to evaluate the expression lev-els of 34 apoptosis-regulating genes. We found that overall expression profiles in Treg were very similar to those in non-Treg CD4+ T cells (n=3; figure 4A). However, 2 genes had a markedly different expression in Treg. First, the levels of the pro-apoptotic BH3-only molecule Noxa were considerably increased in Treg as compared to non-Treg CD4+ T cells (2.89 fold increase; p=0.02). This


44

Figure 3. Cell Stimulation Cultures (CSC) and Foxp3 induction. CSC were performed with 3T3 or 3T40L stimulated CLL cells (APC) and PMBC of a healthy individual (HD) or autologous PBMC from CLL patients (responders) in a 1:1 ratio. Cells were cultured in the presence of soluble CD3 mAb and in the presence or absence of a blocking CD70 mAb (2F2). Foxp3 expression was analyzed after 4 days. A CD70 expression on CD5+/CD19+ cells before (t=0) and after 2 days co-culture with 3T3 or 3T40L. Left: percentage CD5+/CD19+/CD70+ cells ± SEM. Right: CD70 expres-sion of CD5+/CD19+/CD70+ cells. Data are presented as mean fluorescence intensity (MFI) ± SEM. Top right: CD70 expression; overlay of 3T3 stimulated CLL cells (light grey line) and 3T40L stimulated CLL cells (black line). Gated on CD5+/CD19+ cells. B CSC with 3T3 stimulated CLL (3T3-CLL) or 3T40L stimulated CLL cells (3T40L-CLL) as APC and PBMC of HD as responders in presence (grey bars) or absence (white bars) of 2F2 (n=3). Data are presented as percentage CD4+/Foxp3+ cells (mean ± SEM). C CSC of 3T3 stimulated CLL (3T3-CLL) or 3T40L stimulated CLL cells (3T40L-CLL) and autologous PBMC of CLL patient in presence (grey bars) or absence (white bars) of 2F2 (n=4). Data are presented as percentage CD4+/Foxp3+ cells (mean ± SEM).

finding was confirmed by quantitative PCR analysis (figure 4B). Secondly, Bcl-2 expression was found to be significantly decreased in Treg as compared to non-Treg CD4+ T cells (3.02 fold decrease; p=0.01). Protein analysis sub-sequently confirmed the elevated expression levels of Noxa and low expres-

Chapter 2

45

sion of Bcl-2 in Treg (figure 4C). Thus, Treg seem to have a unique apoptotic profile which suggests enhanced susceptibility to apoptosis induction. Indeed, spontaneous apoptosis in purified Treg was higher than in non-Treg CD4+ T cells (figure 4D). Moreover, when sorted Treg cells were treated overnight with a moderate concentration of Roscovitine (a drug that preferably induces apoptosis in cells with high Noxa levels(13)) , they were much more sensitive to apoptosis induction than non-Treg CD4+ T cells (figure 4D and E).

Treg of CLL patients are protected against apoptosis Since Treg have a highly apoptosis-prone gene expression profile, the increased number of Treg in CLL might be caused by small alterations in expression lev-els of apoptosis-regulating genes. In line with this assumption, we found that Treg from CLL patients express higher levels of Bcl-2 than Treg from healthy individuals (figure 5A). This was confirmed at the protein level by intracel-lular staining (figure 5B). Although Foxp3-/CD4+ T cells of CLL patients also demonstrated high expression of Bcl-2 when compared to healthy individuals, the Bcl-2Treg : Bcl-2non-Treg expression ratio was higher in CLL patients (mean ± SEM: 0.79 ± 0.14) than in healthy individuals(0.62 ± 0.07; p = 0.0008). This supports the notion that Treg of CLL patients might be relatively protected against apoptosis, since the elevated expression of Bcl-2 observed in Treg from CLL patients might serve to counterbalance the high expression of Noxa (fig-ure 4A and B). We were not able to elucidate the mechanism behind high Bcl-2 levels in Treg from CLL patients and we also did not observe (CD70 induced) upregulation of Bcl-2 in CD4+ T cells during CSC experiments (data not shown). In addition, Treg of CLL patients seemed to have increased ex-pression of inhibitor of apoptosis protein 1 (IAP1), a gene that has been impli-cated in apoptotic responses to TNF(36;38). On the other hand, we observed that Treg of CLL patients had higher expression of Fas/CD95 (figure 5C), a molecule that has been implied in activation-induced cell death in T cells(24). Finally, Treg of CLL patients displayed lower cycling activity as assessed by the percentage of Ki-67 positive cells (figure 5D).To test the potential functional consequences of these changes, we compared Treg from CLL patients and healthy individuals for sensitivity to drug-in-duced apoptosis. PBMC from CLL patients (n=6) and healthy individuals (n=5) were incubated with cytotoxic drugs and monitored for the percentage of CD25bright/Foxp3+ cells within the total CD4+ T cell population (figure 5E). In Treg from CLL patients we observed a strongly decreased apoptosis induc-


46

Bcl

-W

Bcl

2

Bcl

-XL A1

MC

L1-L

B2M

Pum

a

Bim

BA

D

Nox

a

Bid

HR

K

BM

F

BIK NIX

NIP

3

MA

P1 FLT

Bak

Bax

1

Bax

2

Bcl

-Rm

b

Bcl

-G

MC

L1-S

GU

S

NIA

P

IAP1

XIA

P

SUR

VIVI

N

APO

LLO

N

LIVI

N

APA

F-XL PI-9

Flip

AIF

APA

F

DIA

BLO

PAR

N

0

1 0

2 0Tr e g

CD4

BC L - 2 - lik e BH3 - o n ly Ba x - lik e IA P M is c e lla n e o u s

rela

tive

expr

essi

on (%

)

( n =3 )

A

Noxa

Actin

Treg CD4

1 2

Bcl-2

Bcl-XL

Mcl-1

1 + 2HD:

C

p = 0.02p = 0.01

HD1 HD2 HD30.0

0.5

1.0

TregCD4

Nox

a(r

elat

ive

expr

essi

on)

B

91%

CD4

Treg

non-treated Roscovitine

FSC

70%

SS

C

63% 21%

E

0 12.50.0

2.5

5.0

7.5

10.0

CD4TregRoscovitine (uM)

% c

leav

ed c

aspa

se-3

0 12.50.0

2.5

5.0

7.5

10.0

12.5

Roscovitine (µµµµM)

% D

NA

frag

met

atio

n

D

Chapter 2

47

Figure 4. Apoptosis-regulating genes in Treg. CD4+ T cell populations of three healthy individuals (HD) enriched for either Treg (CD25bright/IL7Rlow) or non-Treg CD4+ T cells (CD25-/IL7R+) were isolated by flow cytometry (see methods) and lysed to obtain RNA and protein content. A relative expression levels of apoptosis-regulating genes, measured via RT-MLPA (n=3; see methods). Bars graph represents the mean ± SEM. B quantative PCR analysis of Noxa expression in the sorted cell populations. The results are presented as relative expression compared to the expression levels in Treg (black bars); the results were normalized by setting the expression levels obtained for regulatory T cells as 1. C western blot of protein lysates from Treg and non-Treg CD4+ T cells of two healthy individuals (the protein lysates of Treg from the two healthy individuals were pooled to obtain sufficient protein concentration for western blotting). Actin is used as a loading control. D Treg and non-Treg CD4+ T cells (isolated as described above) were incubated for 24 hours in the presence of CDK inhibitor Roscovitine (12.5μM). The viable cells were subsequently identified by flow cytometry (forward-sideward scatter). E PBMC of a healthy individual were incubated with Roscovitine (12.5μM) for 24 hours. Subse-quently, the cells were analyzed for apoptosis parameters (caspase-3 cleavage and DNA fragmentation). A Foxp3 antibody was used to identify regulatory T cells.

tion by Roscovitine (a drug that acts via the Noxa-Mcl1 axis(13)), Fas ligation or fludarabine when compared to Treg of healthy individuals. Together, these data support the hypothesis that Treg of CLL patients accumulate through reduced apoptosis rather than by increased proliferation.

Discussion

In the present study we investigated the mechanisms behind the expansion of Treg in CLL. We observed that Treg of CLL patients as well as Treg of healthy individuals predominantly have the phenotype of adaptive Treg (CD4+/CD25bright/Foxp3+/CD45R0+). Although we did not find evidence for a pre-dominant (tumor) antigen driving Treg expansion in CLL, our experiments suggest that T cell stimulation in a CLL lymph node environment might result in increased formation of Treg via CD27-CD70 costimulation. Furthermore, we observed that Treg (compared to non-Treg CD4+ T cells) have a pro-apop-totic phenotype characterized by high levels of Noxa and low expression of Bcl-2. Nevertheless, Treg of CLL patients seem to be less sensitive to apoptosis induction than Treg from healthy individuals, possibly via increased expres-sion of Bcl-2.


48

0

25

50

75

100

0 12.5 25 50

**

Roscovitine (µµµµM)

Treg

(% o

f non

-tre

ated

)

0

25

50

75

100

0 25 50 100

*

fludarabine (µµµµM)

Treg

(% o

f non

-tre

ated

)

0

25

50

75

100

0 0.2 1

*

αααα-Fas (µµµµg ml-1)

Treg

(% o

f non

-tre

ated

)

HDCLL

E

Bcl-2

HD CLL HD CLL0.0

0.5

1.0

CD4 Treg

rela

tive

expr

essi

on

Noxa

HD CLL HD CLL0.0

2.5

5.0

CD4 Treg

rela

tive

expr

essi

on

Mcl-1

HD CLL HD CLL0.0

0.5

1.0

CD4 Tregre

lativ

e ex

pres

sion

IAP1

HD CLL HD CLL0

1

2

CD4 Treg

rela

tive

expr

essi

on

C D

HD CLL HD CLL0

25

50

75

TregCD4

CD

95 (M

FI)

HD CLL HD CLL0

10

20

30

40

CD4 Treg

% K

i-67+ c

ells

B

HD CLL HD CLL0

100

200

300

400

TregCD4

Bcl

-2 (M

FI )

* ****** ***

A

****

* *******

* p < 0.05*** p < 0.0001

Figure 5 Analysis of apoptosis sensitivity of Treg from CLL patients. A RNA isolated from Treg and non-Treg CD4+ T cells (CD4) of three CLL patients (CLL) and two healthy individuals (HD) was used as input for RT-MLPA (see meth-ods). Graphs compose a selection of 34 apoptosis-regulating genes of which expression levels were obtained. The results are plotted as relative expression, which is calculated as follows: expressionTreg population/ expressionnon-Treg T cell population. Bars rep-resent the mean relative expression; error bars indicate range. B Bcl-2 staining of PBMC from CLL patients (n=21) and healthy individuals (n=11). CD4 = CD3+/CD4+/Foxp3-; Treg = CD3+/CD4+/Foxp3+. Data are presented as MFI; samples were standardized by using an isotype-matched control antibody. C CD95 expression on T cells from CLL pa-tients (n=19) and healthy individuals (n=5). Data are presented as MFI; samples were standardized using an isotype-matched control antibody.

Chapter 2

49

D percentage of Ki-67+ cells in Treg and non-Treg CD4 from CLL patients (n=19) and healthy individuals (n=10). Cutoff for Ki-67 staining was determined using an isotype-matched control antibody. E PMBC from CLL patients (n=6) and healthy individuals (n=6) were incubated with various concentrations of Roscovitine, Fas-agonistic antibody CH-11 (α-Fas) and fluda-rabine. After 24 hours, the cells were harvested and analyzed by flow cytometry. Finally, the percentage of Treg that remained compared to non-treated samples was calculated

Our finding that Treg in CLL patients are predominantly CD45R0+ and use a similar T cell receptor (TCR) repertoire to non-Treg CD4+ T cells is in line with a recent study which indicates that Treg arise continuously from the memory T cell pool upon antigenic stimulation(37). The latter also makes it tempting to speculate about the involvement of a predominant tumor antigen/peptide in the formation of the increased numbers of Treg in CLL patients. Neverthe-less, the TCR repertoire analysis performed in the current study on Treg from CLL patients did not support this possibility. Moreover, also CMV seroposi-tivity did not influence the percentage of CD45R0+ Treg, making a role for this antigen (which has been demonstrated to influence the CD8+ T cell repertoire in CLL(19)) in the formation of Treg in CLL unlikely.Recent studies show that NHL B cells are powerful inducers of Treg(23;40). Yang et al. propose that this Treg induction takes place in the lymph node via CD70 ligation by CD70+ NHL B cells. The results from our current study suggest that in CLL this might also be the case: CD40L stimulated CLL cells (resembling CLL cells from a lymph node environment(32)) enhanced the formation of Treg upon TCR stimulation (figure 3B and C), and this effect could be blocked by CD70 antibodies. Therefore, it seems that CD27-CD70 co-stimulation may be an important step in the formation of Treg in B cell malignancies. Moreover, if indeed the increased number of Treg in CLL can be explained by CD70 ligation by CLL cells in vivo, this might also explain why these Treg have increased surface expression of CD95, since it is known that CD95 expression is upregulated on CD70 co-stimulated T cells(3). Analysis of number and phenotype of Treg in bone marrow of CLL patients showed no difference with peripheral blood Treg (data not shown), which indicates that the bone marrow is probably not the primary site of Treg formation in CLL. Altogether, we consider it more likely that Treg formation in CLL takes place in lymph nodes where CLL cells may function as professional APC to induce Treg, possibly by CD27-CD70 co-stimulation. The observation that the high-


50

est frequencies of Treg occur in CLL patients with extended disease(4) (and thus with more lymphoid tissue) supports this hypothesis.Besides via increased formation in the lymph nodes, our data also suggest that Treg in CLL may accumulate via decreased sensitivity to apoptosis. The latter may strongly influence the rapid turnover of Treg in vivo, which ac-cording to our findings seems to be facilitated by an altered balance between two molecules involved in apoptosis regulation: Noxa and Bcl-2. Therefore, the observed increased expression of anti-apoptotic Bcl-2 in CLL Treg may counterbalance the high expression of pro-apoptotic Noxa. Previously, we demonstrated that upon T cell activation, Noxa is expressed and may play an important role in determining the size of activated T cell populations(2). Bcl-2 and Noxa are not direct binding partners(5;7), hence a direct reciprocal effect of their altered expression is unlikely. Rather, the effects on changes in apoptosis thresholds probably occur indirectly via shared binding partners, such as Mcl-1 and Bim(13). In addition, the increased expression of Bcl-2 also seems to protect CLL Treg against CD95 ligation (the mechanism via which Treg are thought to be eliminated in vivo (11)). Increased frequencies of Treg occur in many types of cancer(6;18;39). There is evidence that the presence of Treg in the tumor microenvironment may affect antitumor responses and promote disease progression(8;9;22). This may also be the case in CLL. Thus, targeting Treg in CLL might influence the course of the disease. Interestingly, one of the drugs used in the first line treatment of CLL, fludarabine, has been reported to reduce frequencies of Treg and af-fect their suppressive capacity(4). Thus, the effect of this drug on CLL might partially be contributed to its effect on Treg. Moreover, in active immuno-therapy it has been shown that blockade of CTLA-4 potentates anti-tumor T-cell responses, possibly by selective targeting of antitumor Treg(15;27) . Our data suggest that Treg are very sensitive to Roscovitine, a cyclin dependent kinase inhibitor that targets Mcl-1 and therefore preferably induces apoptosis in cells with high levels of its binding partner Noxa(13). By selectively target-ing Treg and inducing apoptosis in CLL cells(12), Roscovitine could therefore be a potent adjuvant drug in active immunotherapy. Alternatively, in view of the relatively high Bcl-2 expression in CLL Treg, it would also be interest-ing to monitor Treg frequencies and suppressive capacity in CLL patients that are being treated with oblimersen, a therapeutical Bcl-2 antisense oligonucle-otide(25;29). In conclusion, Treg in CLL patients appear to accumulate through increased

Chapter 2

51

formation, facilitated by CD70 ligation by tumor cells in the lymph nodes as well as by decreased sensitivity to apoptosis. Since the increased number of Treg might be considered to negatively affect the course of the disease, target-ing either one of the above-mentioned mechanisms may provide additional strategies in the treatment of CLL.

Acknowledgements

We are indebted to our patients for their commitment to this study. We thank the hematologists of the Department of Hematology of Meander Medical Centre, Amersfoort, The Netherlands, for referral of patients.


52

Reference List

1. Akbar AN, Vukmanovic-Stejic M, Taams LS, et al. The dynamic co-evolution of memory and regulatory CD4+ T cells in the periphery. Nat Rev Immunol. 2007;7:231-237.

2. Alves NL, Derks IA, Berk E, et al. The Noxa/Mcl-1 axis regulates susceptibility to apoptosis under glucose limitation in dividing T cells. Immunity. 2006;24:703-716.

3. Arens R, Baars PA, Jak M, et al. Cutting edge: CD95 maintains effector T cell homeostasis in chronic immune activation. J Immunol. 2005;174:5915-5920.

4. Beyer M, Kochanek M, Darabi K, et al. Reduced frequencies and suppressive function of CD4+CD25hi regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood. 2005;106:2018-2025.

5. Certo M, Del GM, V, Nishino M, et al. Mitochondria primed by death signals determine cellular addiction to antiapoptotic BCL-2 family members. Cancer Cell. 2006;9:351-365. 6. Cesana GC, DeRaffele G, Cohen S, et al. Characterization of CD4+CD25+ regulatory T cells in patients treated with high-dose interleukin-2 for metastatic melanoma or renal cell carcinoma. J Clin Oncol. 2006;24:1169-1177.

7. Chen L, Willis SN, Wei A, et al. Differential targeting of prosurvival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function. Mol Cell. 2005;17:393-403.

8. Clarke SL, Betts GJ, Plant A, et al. CD4+CD25+FOXP3+ regulatory T cells suppress anti-tumor immune responses in patients with colorectal cancer. PLoS ONE. 2006;1:e129. 9. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells in ovarian carci-noma fosters immune privilege and predicts reduced survival. Nat Med. 2004;10:942-949. 10. Eldering E, Spek CA, Aberson HL, et al. Expression profiling via novel multiplex assay allows rapid assessment of gene regulation in defined signalling pathways. Nucleic Acids Res. 2003;31:e153. 11. Fritzsching B, Oberle N, Eberhardt N, et al. In contrast to effector T cells, CD4+CD25+FoxP3+ regulatory T cells are highly susceptible to CD95 ligand- but not to TCR-mediated cell death. J Immunol. 2005;175:32-36. 12. Hahntow IN, Schneller F, Oelsner M, et al. Cyclin-dependent kinase inhibitor Roscovitine induces apoptosis in chronic lymphocytic leukemia cells. Leukemia. 2004;18:747-755.

Chapter 2

53

13. Hallaert DY, Spijker R, Jak M, et al. Crosstalk among Bcl-2 family members in B-CLL: seliciclib acts via the Mcl-1/Noxa axis and gradual exhaustion of Bcl-2 protection. Cell Death Differ. 2007;14:1958-1967. 14. Hintzen RQ, Lens SM, Beckmann MP, et al. Characterization of the human CD27 ligand, a novel member of the TNF gene family. J Immunol. 1994;152:1762-1773. 15. Hodi FS, Butler M, Oble DA, et al. Immunologic and clinical effects of antibody blockade of cytotoxic T lymphocyte-associated antigen 4 in previously vaccinated cancer patients. Proc Natl Acad Sci U S A. 2008;105:3005-3010. 16. Kater AP, Evers LM, Remmerswaal EB, et al. CD40 stimulation of B-cell chronic lympho-cytic leukaemia cells enhances the anti-apoptotic profile, but also Bid expression and cells re-main susceptible to autologous cytotoxic T-lymphocyte attack. Br J Haematol. 2004;127:404-415. 17. Liu W, Putnam AL, Xu-Yu Z, et al. CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells. J Exp Med. 2006;203:1701-1711.

18. Liyanage UK, Moore TT, Joo HG, et al. Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocar-cinoma. J Immunol. 2002;169:2756-2761. 19. Mackus WJ, Frakking FN, Grummels A, et al. Expansion of CMV-specific CD8+CD45RA+CD27- T cells in B-cell chronic lymphocytic leukemia. Blood. 2003;102:1057-1063. 20. Mackus WJ, Kater AP, Grummels A, et al. Chronic lymphocytic leukemia cells display p53-dependent drug-induced Puma upregulation. Leukemia. 2005;19:427-434. 21. Maslanka K, Piatek T, Gorski J, et al. Molecular analysis of T cell repertoires. Spectratypes generated by multiplex polymerase chain reaction and evaluated by radioactivity or fluores-cence. Hum Immunol. 1995;44:28-34. 22. Miller AM, Lundberg K, Ozenci V, et al. CD4+CD25high T cells are enriched in the tumor and peripheral blood of prostate cancer patients. J Immunol. 2006;177:7398-7405. 23. Mittal S, Marshall NA, Duncan L, et al. Local and systemic induction of CD4+CD25+ regulatory T-cell population by non-Hodgkin lymphoma. Blood. 2008;111:5359-5370. 24. Miyawaki T, Uehara T, Nibu R, et al. Differential expression of apoptosis-related Fas anti-gen on lymphocyte subpopulations in human peripheral blood. J Immunol. 1992;149:3753-3758.


54

25. O’Brien S, Moore JO, Boyd TE, et al. Randomized phase III trial of fludarabine plus cyclo-phosphamide with or without oblimersen sodium (Bcl-2 antisense) in patients with relapsed or refractory chronic lymphocytic leukemia. J Clin Oncol. 2007;25:1114-1120. 26. Porakishvili N, Roschupkina T, Kalber T, et al. Expansion of CD4+ T cells with a cytotoxic phenotype in patients with B-chronic lymphocytic leukaemia (B-CLL). Clin Exp Immunol. 2001;126:29-36. 27. Quezada SA, Peggs KS, Curran MA, et al. CTLA4 blockade and GM-CSF combination immunotherapy alters the intratumor balance of effector and regulatory T cells. J Clin Invest. 2006;116:1935-1945. 28. Ranheim EA, Cantwell MJ, Kipps TJ. Expression of CD27 and its ligand, CD70, on chronic lymphocytic leukemia B cells. Blood. 1995;85:3556-3565. 29. Schlagbauer-Wadl H, Klosner G, Heere-Ress E, et al. Bcl-2 antisense oligonucleotides (G3139) inhibit Merkel cell carcinoma growth in SCID mice. J Invest Dermatol. 2000;114:725-730. 30. Seddiki N, Santner-Nanan B, Martinson J, et al. Expression of interleukin (IL)-2 and IL-7 receptors discriminates between human regulatory and activated T cells. J Exp Med. 2006;203:1693-1700. 31. Serrano D, Monteiro J, Allen SL, et al. Clonal expansion within the CD4+CD57+ and CD8+CD57+ T cell subsets in chronic lymphocytic leukemia. J Immunol. 1997;158:1482-1489. 32. Smit LA, Hallaert DY, Spijker R, et al. Differential Noxa/Mcl-1 balance in peripheral ver-sus lymph node chronic lymphocytic leukemia cells correlates with survival capacity. Blood. 2007;109:1660-1668. 33. Taams LS, Smith J, Rustin MH, et al. Human anergic/suppressive CD4(+)CD25(+) T cells: a highly differentiated and apoptosis-prone population. Eur J Immunol. 2001;31:1122-1131. 34. Uss E, Rowshani AT, Hooibrink B, et al. CD103 is a marker for alloantigen-induced regu-latory CD8+ T cells. J Immunol. 2006;177:2775-2783. 35. van Leeuwen EM, Remmerswaal EB, Heemskerk MH, et al. Strong selection of virus-spe-cific cytotoxic CD4+ T-cell clones during primary human cytomegalovirus infection. Blood. 2006;108:3121-3127.

36. Vince JE, Wong WW, Khan N, et al. IAP antagonists target cIAP1 to induce TNFalpha-dependent apoptosis. Cell. 2007;131:682-693.

Chapter 2

55

37. Vukmanovic-Stejic M, Zhang Y, Cook JE, et al. Human CD4+ CD25hi Foxp3+ regu-latory T cells are derived by rapid turnover of memory populations in vivo. J Clin Invest. 2006;116:2423-2433. 38. Wang L, Du F, Wang X. TNF-alpha induces two distinct caspase-8 activation pathways. Cell. 2008;133:693-703. 39. Wolf AM, Wolf D, Steurer M, et al. Increase of regulatory T cells in the peripheral blood of cancer patients. Clin Cancer Res. 2003;9:606-612. 40. Yang ZZ, Novak AJ, Ziesmer SC, et al. CD70+ non-Hodgkin lymphoma B cells induce Foxp3 expression and regulatory function in intratumoral CD4+CD25 T cells. Blood. 2007;110:2537-2544.


3Detection of p53 dysfunction

in chronic lymphocytic leukemia cells via multiplex quantification of

p53 target gene induction Rogier Mous1, Annelieke Jaspers2, Marinus H.J. van Oers1,

Arnon P. Kater1 and Eric Eldering2


Altered version submitted to Leukemia

59

Abstract

In CLL, dysfunction of the p53 tumour suppressor pathway is associated with impaired therapeutic response and poor prognosis. Although FISH reliably detects gross chromosomal aberrations at 17p, it has become clear that this does not fully concur with p53 functional status. In order to develop a gen-erally applicable and standardised method to assess p53 function in clinical samples, we here evaluate analysis of p53-responsive gene transcripts Puma, p21 and Bax upon ionizing radiation using multiplex ligation dependent probe amplification (MLPA). For 26 cases of CLL, MLPA analysis was com-pared with FISH and alternative read-outs of p53 function: CD95 expression by FACS and p21 and p53 protein expression by Western blot. Threshold val-ues could be set for Puma and p21 gene induction that reliably discriminated p53 function. We conclude that MLPA analysis for p53 function shows prom-ise for further evaluation in relation to clinical course in larger cohorts.

MLPA analysis for p53 function

60

Introduction Chronic lymphocytic leukemia (CLL) is a heterogeneous disease with variable clinical course. Deletions of the long arm of chromosome 11 (11q-) and espe-cially the short arm of chromosome 17 (17p-) are the strongest independent prognostic factors associated with adverse outcome (8). 17p- corresponds with loss of one allele encoding the tumour suppressor gene p53. Deletions of the long arm of chromosome 11 imply loss of the ataxia telangiectasia (ATM) gene, an upstream mediator in the p53 signalling pathway. Both 11q- and 17p- are associated with dysfunction of the p53 tumor suppressor pathway (11). These chromosomal abnormalities may play an important role in relapse upon therapy, since the prevalence of 17p- and 11q- increases from 5-15% at diagnosis to over 50% during relapse (3).Currently, fluorescence in situ hybridisation (FISH) is the most widely used technique to detect chromosomal anomalies in CLL samples. Although recent studies demonstrate that deletions of one p53 or ATM alleles are frequently accompanied by mutation of the other allele, the exact overlap between the patient populations defined by del17p and p53 dysfunction is poorly defined. In fact, novel data revealed that mutations in the p53 locus are not restricted to patients with 17p- (6;12). It would therefore be clinically relevant to test the activity of the p53 pathway in a functional assay, in addition to FISH analy-sis.Measuring p53 and/or p21 protein levels by western blot following ion-izing radiation (IR) is an accepted semi-quantitative means to detect p53 (dys)function in tumor samples(11;12). Recently, alternatives to this technique have been described: 1. Staining of the surface molecule CD95 following IR (4;5) , 2. Intracellular staining of p53 and p21 via flow cytometry following IR (4), 3. FASAY analysis (10) and 4. The combination of etoposide and nut-lin-3a as an alternative to IR to induce p53 activity (2). Previously, we evalu-ated the apoptosis gene profile in CLL samples after fludarabine incubation. It was demonstrated that induction of expression of the p53 responsive gene Puma strictly relied on functional p53 (9). In the current study, the potential of MLPA analysis to establish p53 function was further evaluated by including the p53-responsive cell cycle inhibitor p21. We applied this modified MLPA procedure in 26 CLL samples subjected to IR. These results were compared to other read-outs of p53 function: western blotting for p53 and p21 and analysis of CD95 surface expression.

Chapter 3

61

Methods

Patient samples, cell culture and irradiationCLL PMBC fractions were obtained after informed consent and contained >90% of CD5+/CD19+ B cells as assessed by flow cytometry. Characteristics of the 26 patients in this study are listed in table 1. CLL cells and Burkitt lympho-ma line Ramos (p53 negative) were maintained in IMDM supplemented with 10% heat-inactivated FCS, 100 U ml-1 penicillin and 100 mg ml-1 streptomycin. Thawed CLL and Ramos cells (5x106 cells ml-1) were exposed to gamma-ra-diation (5 Gy), and incubated for 16 hours at 37°C. Cells were lysed to obtain protein and RNA content or stained with antibodies for flow cytometry.

RT-MLPA analysis, Western blotting, and CD95 expressionThe multiplex RT-MLPA procedure is able to quantify expression of 34 apopto-sis-related genes in small RNA samples (7). MLPA was executed and analyzed as described previously (9). Criteria for p53 functionality were defined by the average radiation-induced gene induction minus 2x the standard deviation of samples displaying normal 11q and 17p status, as well as normal p53/p21 sta-tus by Western blotting. Fifteen samples fulfilled these criteria; yielding values of 4.4±1.4 (average±STDEV) for Puma and 6.5±2.1 for p21 (table 1), and cut-off values of 1.4 and 2.3, respectively. Samples that did not meet both cut-off criteria were considered p53 dysfunctional.Western blotting was performed as described before (9). Blots were probed with antibodies against p53 (Calbiochem, La Jolla, CA), p21 (Santa Cruz Bio-technology, Santa Cruz, CA), Puma (Cell Signal Technology, Danvers, MA) and Bcl-2 (Kordia Life Sciences, Leiden, The Netherlands). Samples were clas-sified as p53 functional when both p53 and p21 were upregulated upon IR, as type A p53 dysfunction when only p53 was upregulated, and as type B p53 dysfunction when neither p53 nor p21 was upregulated, as previously described (11).Irradiated and non-irradiated CLL were analyzed by flow-cytometry for sur-face expression of CD95 (IQ Products, Groningen, The Netherlands) within the population of CD19-positive (Becton Dickinson, San Jose, CA) cells. The increase in mean fluorescent intensity (MFI) upon IR was calculated.

Statistical analysisPaired sample 2-sided students T-test was used for analysis of differences be-


62

tween irradiated and non-irradiated CLL. P values <0.05 were considered sta-tistically significant. Pearson correlation tests were performed to obtain r- and p-values from two independent parameters.

Results and discussion

Blood-derived CLL cells of 26 patients were analyzed for p53 function by three criteria. Patients 1-23 were randomly included from our cohort and tested for p53 function by WB analysis for p53 and p21 protein induction upon radia-tion, while patients 24-26 were included based upon clinical refractoriness to standard chemotherapy and subsequent FISH analysis which showed 17p- status (table 1). All samples were analysed for CD95 induction and by MLPA. In total, six patients had detectable 17p deletions and four had 11q deletions as assessed by FISH (table 1). Of the 34 apoptosis-related genes in the MLPA probe set, 3 are acknowledged targets of p53: Puma, p21 and Bax. In all but two patients without 11q- or 17p-, these genes were induced upon IR; 4.0±0.4 fold for Puma (p<0.0001), 6.4±2.8 fold for p21 (p<0.0001) and to a lesser extent also two splice variants of Bax (Bax1: 2.3±0.4 fold, p<0.0001 and Bax2: 2.1±0.4 fold, p<0.0001, respec-tively; figure 1B and table 1). Cut-off criteria were defined by considering first those samples with normal karyotype at 17p and 11q as well as clear p53 and p21 protein induction on Western blot. Fifteen CLL samples met these crite-ria, and their average gene induction minus 2x standard deviation was taken as cut-off value for normal p53 function. Next, p53 function in the patients with documented 11q- and 17p- were ana-lyzed. 11q- ranged from 21% to 95% (n=4). In none of these samples a p53 dysfunction could be detected by WB, MLPA or FACS method. This seems in line with recent data indicating that in the majority of patients with 11q-, the remaining ATM allele is intact and as a consequence p53 function is normal (1). In 3 chemotherapy-refractory patients with 17p- in the range of 64% to 94%, p53 dysfunction was apparent according to MLPA criteria, while CD95 FACS analysis was not discriminatory. In contrast, samples of three patients with a lower percentage of 17p- (11% to 39%) showed a normal p53 function by WB and MLPA. Interestingly, two patients without known 11q- or 17p- were p53 dysfunctional according to MLPA and WB analysis (patient 2 and 8; table 1). In case of patient 8 (type A dysfunction) additional cells for genetic

Chapter 3

63

analysis could unfortunately not be obtained because of rapid disease progres-sion and death. In summary, by applying threshold values for both Puma and p21 as detected by MLPA, p53 dysfunction could reliably detected in this co-hort of CLL patients, and was fully concordant among distinct methods. For CD95 FACS analysis (cut-off value of 0.8 for increase in MFI) and Bax gene induction, the criteria were not sufficient to discriminate p53 dysfunction (see table 1). These differences are illustrated in Fig 1C, where average fold induc-tion for Puma, p21 and CD95 is plotted for all samples assigned normal or dysfunctional p53 status. In Fig 1D, the correlation between CD95 and Puma is plotted (left panel) which yields a clear separation between the two groups, as opposed to the correlation plot between CD95 and Bax (right panel).To approximate the percentage of p53 dysfunctional cells in a sample that would be detectable by MLPA, RNA obtained from IR-treated p53 functional CLL cells was mixed with RNA from IR-treated (p53-negative) Ramos cells (figure 1E). Values for Bad and XIAP are used in this example as negative control because their expression is equal in CLL and Ramos cells and not influenced by IR treatment. In this experiment, the threshold for the Puma transcript above which p53 dysfunction could be detected is about 40% of p53 dysfunctional cells, in good agreement with data for the clinical samples (table 1). Taken together, our findings indicate that MLPA is a useful tool to detect p53 dysfunction in CLL samples, and possibly other malignant cells as well. Ad-vantages of the method are that it is amenable to standardisation and automa-tion and is relatively cheap. We conclude that this type of functional read-out for p53 could be a valuable contribution to diagnostic and therapeutic criteria, and should be further tested in prospective clinical trials.

Acknowledgements

Arnon P. Kater is supported by a ‘Veni’ grant from ZonMw (The Netherlands Organization for Health Research and Development). We thank Ronald van Ree for critical reading of the manuscript and comments.


64

A

p21

0 20 40 60 80 1000

2

4

6

8

% dysfunctional RNA

rela

tive

expr

essi

on

p53

p21

Bcl-2

- +

3

- +

4

- +

5

- +

6

- +

7

- +

8

- +

9

Puma

Puma

0 20 40 60 80 1000

2

4

6

8

% dysfunctional RNA

rela

tive

expr

essi

on

Bad

0 20 40 60 80 1000.0

0.5

1.0

% dysfunctional RNA

rela

tive

expr

essi

on

XIAP

0 20 40 60 80 1000.0

1.5

3.0

% dysfunctional RNA

rela

tive

expr

essi

on

E

DC

0 2 4 6 80.0

2.5

5.0

7.5

Puma (fold increase)

CD

95 (f

old

incr

ease

) r = 0.68p = 0.0003

0 1 2 3 4

0.0

2.5

5.0

7.5

Bax1 (fold increase)

CD

95 (f

old

incr

ease

) r = 0.69p= 0.0002

Puma p21 CD950

2

4

6

8

10WT n=20DYSF n=5

3.9E-05

3.8E-04

4.2E-03

fold

indu

ctio

n

Puma

- +

0

5

10

p= 0.0001

IR

rela

tive

expr

essi

on

p21

- +

0

5

10

15

20

p= 0.0001

IR

rela

tive

expr

essi

on

BBax1

- +

5

10

15

20

p= 0.0001

IR

rela

tive

expr

essi

on

Bcl-2

- +

0

2

4

6

8

10

IR

rela

tive

expr

essi

on

* * *

Chapter 3

65

Figure 1. A PBMC samples from CLL patients were treated with ionizing radiation (5Gy, indicated by + symbol). After 16 hours, cells were lysed and Western blotting for p53, p21 and Puma was performed; Bcl-2 serves as a loading control (not influenced by IR). CLL patient 8 was diagnosed with a type A p53 dysfunction. B RT-MLPA analysis performed on IR-treated PMBC samples from 26 CLL patients. Puma, p21 and two splice variants of Bax (Bax1 and Bax2) were differentially expressed upon IR. The filled triangles (▲) represent p53 dysfunctional samples. Expression of Bcl-2 (not influenced by IR) is shown as a control. Gene signals are expressed relative to normalized total signals set at 100%. Sample #10, which was excluded from further analysis due to high background values for Puma, p21 and Bax in untreated cells, is indicated with *. C Average gene induction and standard deviation for Puma, p21 or CD95 as established by MLPA or FACS. Values were calculated for all samples designated normal (WT, n=20) or dysfunctional (DYSF, n=5) for p53 function. P values for differences among the two groups are indicated.D Correlation between results of MLPA and CD95 surface expression. Data are plotted as fold increase compared to non-irradiated samples. The line in each plot is the best-fit obtained via linear regression of the plotted data; r- and p- values were calculated using Pearson correlation test; a circle is drawn around the p53 dysfunctional samples. This analysis illustrates that, although the correlation between CD95 and Bax induction is excellent, the quantitative analysis does not separate p53 functional and dysfunctional samples.E As input for MLPA, RNA from p53 functional CLL was mixed with RNA from (p53 negative) Ramos cells (both isolated 16 hours after IR treatment of the cells). The rela-tive expression of p53 responsive genes Puma and p21 is plotted against the percentage of p53 dysfunctional (Ramos) RNA present in the sample. Bad and XIAP (expression equal in CLL and Ramos and not influenced by IR) are used as a control.


66

Patie

nt ID

Rai

stag

eIg

VH

m

utat

ion

a

Chr

omos

omal

abno

rmal

ities

1I

+13

q- (2

3%)

2IV

-13

q- (8

0%)

30

-tr

isom

y 12

(60%

)4

IV-

triso

my

12 (7

4%)

5II

+t(1

4q32

) ; tr

isom

y 12

(63%

)6

0-

triso

my

127

II+

none

8II

-nt

9II

I-

13q-

10IV

+13

q-11

II+

nt12

0+

13q-

13I

+13

q-14

II+

13q-

150

nt16

II14

q- (5

3%)

17II

t(11

;14)

(13%

) ; 1

3q-

180

nt19

I+

11q-

(21%

)

p53

func

tion

(WB)

; p21

and

p5

3 up

regu

latio

n up

on IR

b

Pum

a(M

LPA

); fo

ld

indu

ctio

n up

on IR

c

p21

(MLP

A);

fold

in

duct

ion

upon

IR c

CD

95(F

AC

S); f

old

indu

ctio

n up

on IR

c

norm

al3.

010

.61.

8ty

pe B

1.2

1.5

0.9

norm

al6.

36.

74.

2no

rmal

5.0

4.7

3.6

norm

al7.

69.

86.

0no

rmal

6.4

3.5

2.3

norm

al4.

9nt

4.0

type

A1.

22.

51.

1no

rmal

2.7

4.6

3.3

norm

al

- d

- d3.

4

norm

al3.

07.

74.

1no

rmal

3.4

4.1

3.7

norm

al3.

16.

72.

7no

rmal

3.0

4.6

2.2

norm

al4.

45.

32.

0no

rmal

4.3

6.5

4.0

norm

al4.

18.

02.

0no

rmal

4.2

7.7

1.9

norm

al3.

97.

71.

5

Tabl

e 1.

Pat

ient

cha

ract

eris

tics a

nd re

sults

from

FIS

H a

naly

sis a

nd p

53 fu

nctio

n te

sts

Chapter 3

67

Patie

nt ID

Rai

stag

eIg

VH

m

utat

ion

a

Chr

omos

omal

abno

rmal

ities

20II

-11

q- (9

5%)

21II

-11

q- (1

5%);

17p-

(11%

); 13

q-22

II11

q- (2

5%);

17p-

(15%

); 13

q-

23I

+17

p- (3

9%);

13q-

24IV

+17

p- (5

3%) ;

13q

- (93

%)

25II

I-

17p-

(64%

)26

IV-

17p-

(94%

)e

Avg±

STD

evf

Cut-

offf

p53

func

tion

(WB)

; p21

and

p5

3 up

regu

latio

n up

on IR

b

Pum

a(M

LPA

); fo

ld

indu

ctio

n up

on IR

c

p21

(MLP

A);

fold

in

duct

ion

upon

IR c

CD

95(F

AC

S); f

old

indu

ctio

n up

on IR

c

norm

al3.

87.

92.

3no

rmal

3.9

5.5

1.8

norm

al3.

413

.62.

1no

rmal

2.2

5.9

2.6

nt0.

91.

41.

6nt

0.8

1.9

1.4

nt0.

81.

11.

04.

4±1.

56.

5±2.

13.

2±1.

21.

42.

30.

8

(a)I

gVH

mut

atio

n: m

utat

ion

stat

us o

f the

imm

unog

lobu

lin h

eavy

chai

n, o

btai

ned

via

sequ

ence

ana

lysis

; sam

ples

with

≥2%

diff

eren

ce

from

the

non-

mut

ated

gen

e se

quen

ce w

ere

cons

ider

ed m

utat

ed(b

) typ

e A

: dys

func

tion

of th

e p5

3 pa

thw

ay a

ssoc

iate

d w

ith p

53 m

utat

ion

(17p

-); t

ype

B: d

ysfu

nctio

n of

the

p53

path

way

ass

ocia

ted

with

ATM

mut

atio

n (1

1q-)

(11)

; nt:

not t

este

d(c

) fol

d in

duct

ion

com

pare

d to

non

-irra

diat

ed sa

mpl

e, fo

r cut

-off

valu

es se

e M

etho

ds se

ctio

n(d

) Sam

ple

#10

was

exc

lude

d fr

om a

naly

sis d

ue to

hig

h ba

ckgr

ound

val

ues f

or p

21, P

uma

and

Bax

in th

e un

trea

ted

cells

, whi

ch p

re-

clud

ed p

rope

r cal

cula

tion

of ra

diat

ion-

indu

ced

gene

exp

ress

ion.

(e) c

ompl

ex k

aryo

type

(f

) Cut

-off

valu

es w

ere

calc

ulat

ed a

s ave

rage

gen

e or

CD

95 in

duct

ion

in 1

5 re

liabl

e p5

3WT

sam

ples

(ind

icat

ed in

bol

d, u

nder

lined

), m

inus

2x

the

stan

dard

dev

iatio

n.IR

: ion

izin

g ra

diat

ion;

FA

CS:

fluo

resc

ence

-act

ivat

ed ce

ll so

rter


68

Reference List

1. Austen B, Skowronska A, Baker C, et al. Mutation status of the residual ATM allele is an important determinant of the cellular response to chemotherapy and survival in patients with chronic lymphocytic leukemia containing an 11q deletion. J Clin Oncol. 2007;25:5448-5457.

2. Best OG, Gardiner AC, Majid A, et al. A novel functional assay using etoposide plus nutlin-3a detects and distinguishes between ATM and TP53 mutations in CLL. Leukemia. 2008. 3. Byrd JC, Gribben JG, Peterson BL, et al. Select high-risk genetic features predict earlier progression following chemoimmunotherapy with fludarabine and rituximab in chronic lym-phocytic leukemia: justification for risk-adapted therapy. J Clin Oncol. 2006;24:437-443. 4. Carter A, Lin K, Sherrington PD, et al. Detection of p53 dysfunction by flow cytometry in chronic lymphocytic leukaemia. Br J Haematol. 2004;127:425-428. 5. Dicker F, Kater AP, Prada CE, et al. CD154 induces p73 to overcome the resistance to apop-tosis of chronic lymphocytic leukemia cells lacking functional p53. Blood. 2006;108:3450-3457. 6. Dicker F, Herholz H, Schnittger S, et al. Screening for TP53 Mutations Identifies Chronic Lymphocytic Leukemia Patients with Rapid Disease Progression. ASH Annual Meeting Ab-stracts. 2007;110:490. 7. Eldering E, Spek CA, Aberson HL, et al. Expression profiling via novel multiplex assay al-lows rapid assessment of gene regulation in defined signalling pathways. Nucleic Acids Res. 2003;31:e153. 8. Grever MR, Lucas DM, Dewald GW, et al. Comprehensive assessment of genetic and mo-lecular features predicting outcome in patients with chronic lymphocytic leukemia: results from the US Intergroup Phase III Trial E2997. J Clin Oncol. 2007;25:799-804. 9. Mackus WJ, Kater AP, Grummels A, et al. Chronic lymphocytic leukemia cells display p53-dependent drug-induced Puma upregulation. Leukemia. 2005;19:427-434. 10. Malcikova J, Smardova J, Pekova S, et al. Identification of somatic hypermutations in the TP53 gene in B-cell chronic lymphocytic leukemia. Mol Immunol. 2008;45:1525-1529. 11. Pettitt AR, Sherrington PD, Stewart G, et al. p53 dysfunction in B-cell chronic lymphocytic leukemia: inactivation of ATM as an alternative to TP53 mutation. Blood. 2001;98:814-822. 12. Zenz T, Trbusek M, Smardova J, et al. p53 Inactivation in CLL: Pattern of 110 TP53 Muta-tions. ASH Annual Meeting Abstracts. 2007;110:2064.

Chapter 3

4Redirection of CMV specific CTL towards

CLL via CD20 targeted HLA/CMV complexes

Rogier Mous1, Philip Savage3,4, Ester BM Remmerswaal2, René AW van Lier2, Eric Eldering2 and Marinus HJ van Oers1

Departments of 1Hematology and 2Experimental Immunology Academic Medical Center, Amsterdam;

3Alexis Biotech Ltd., London, UK; 4Department of Medical Oncology, Charing Cross Hospital, London, UK

Leukemia. 2006 Jun;20(6):1096-102

73

Abstract

Chronic lymphocytic leukemia (CLL) is a slowly progressing malignancy of CD5+ B cells, for which at present no curative treatment is available. In our current study, we apply a novel bridging reagent to redirect cytomegalovirus (CMV) specific cytotoxic T lymphocytes (CTL) to target CLL. A streptavidin fused anti-CD20 single chain variable fragment (scFv) is used in combina-tion with biotinylated MHC class I molecules containing CMV pp65 peptide (HLA/CMV). We demonstrate that CLL cells coated with this CD20-HLA/CMV complex can be lysed by autologous CMV specific CTL with similar efficiency as CLL cells directly loaded with CMV-peptide. Killing is HLA re-stricted and occurs at scFv CD20 concentrations of ≥100 ng ml-1 and HLA/CMV concentrations of ≥20 ng ml-1. Furthermore, complex coated CLL cells induce both proliferation and cytokine production (interferon γ, tumor ne-crosis factor α, and macrophage inflammatory protein-1 β) in CMV specific CD8+ T cells. Hereby, a necessary step towards possible application of CD20-HLA/CMV complexes for immunotherapy of B cell malignancies is consti-tuted.

Novel bridging reagent for active immunotherapy in CLL

74

Introduction

At present there is no curative treatment for B cell chronic lymphocytic leuke-mia (CLL). The slow progression and rather long median survival make CLL an attractive target for immunotherapy. For adequate T cell mediated cancer immunotherapy it is necessary to induce activation and differentiation of tu-mor-specific T cells. However, in CLL (like in many other types of cancer) this is severely hampered by the poor antigen presenting capacity of the malignant B cells, which express low levels of co-stimulatory molecules and therefore es-cape immune surveillance. Furthermore, there is evidence that CLL is accom-panied by T cell dysfunction(8;26). Nevertheless, CD40 stimulation of CLL cells was found to enhance costimulatory capacity of the tumor cells(31) and it has been demonstrated that not only allogenic(1;10) but also autologous(12) CLL directed T cells can be generated in vitro. Moreover, in a phase I clinical trial autologous, ex vivo CD40L transduced CLL cells induced a reduction of the tumor mass upon reinfusion(29). Although the specific antigen remains elusive, these studies show that active immunisation against CLL is possible. A disadvantage however of the discussed strategies is the need for complex ex vivo treatment of tumor cells. We have previously shown that in patients with CLL considerably expanded numbers of cytomegalovirus (CMV) reactive CD8+ T cells are present(15). CMV specific CD8+ T cells in a latent stage of the infection possess an effector phenotype, with high contents of perforin and granzymes(5). This contrasts with the CD8+ T cell pool against viruses which are cleared by the immune system like influenza and Respiratory Syncytial Virus (9). Recently we have demonstrated that these CMV-specific CD8+ cells are potent cytotoxic effec-tor cells when directed against CLL cells loaded with CMV peptide(11). Im-portantly, these CD8+ cells do not require ex vivo (re)stimulation but display their cytotoxicity directly after isolation from peripheral blood(11). These data demonstrate the potential to redirect autologous CMV specific CTL towards CLL cells for cancer immunotherapy.To develop this approach towards applications in vivo we investigated the pos-sibility to deliver CMV peptides specifically to CLL cells via antibody-targeted HLA class I/peptide complexes in an in vitro study. We have used a targeted complex (TC) consisting of a streptavidin (SA) fused anti-human CD20 single chain variable fragment (B9E9 scFv) coupled to CMV-peptide loaded bioti-nylated HLA class I(25).

Chapter 4

75

We compared CLL coated with these TC with CLL directly loaded with CMV peptide with respect to efficacy and specificity of autologous T cell mediated killing. Moreover, we analysed their capacity to induce proliferation and cyto-kine production by CMV-specific T cells.

Methods

Patient samplesAfter obtaining informed consent, 30 ml of blood was drawn from patients fulfilling the diagnostic criteria for CLL (18). All patients had Rai clinical stage 0-II and had not received prior treatment. Peripheral blood mononuclear cells (PBMC) obtained via density gradient centrifugation were directly fro-zen in fetal calf serum (FCS) containing 10% DMSO (Sigma Chemical Co.) and stored in liquid nitrogen before use(15). CLL PMBC fractions contained >90% of CD5+/CD19+ B cells as assessed by flowcytometry. On the day of use, cells were thawed and cultured in Iscove’s modified essential medium (IMDM) supplemented with 10% heat-inactivated FCS, 100 U ml-1 penicillin and 100 mg ml-1 streptomycin. B cells were obtained from a PBMC fraction of a healthy donor by depletion of T cells, monocytes and macrophages(13) us-ing anti-CD3, anti-CD14 and anti-CD16 immunomagnetic beads and a mag-netic particle concentrator (both Dynal A.S., Oslo, Norway).

ReagentsHLA-A2–binding NLVPMVATV CMVpp65 peptide and HLA-B7–binding TPRVTGGGAM CMVpp65 peptide (IHB-LUMC peptide synthesis library facility, Leiden, The Netherlands) were used. The B9E9 scFvSA fusion protein, consisting of the variable region of murine IgG2a anti-human-CD20 fused to streptavidin (scFv CD20) has been described before(25) and was used in combination with biotinylated human MHC class I molecules containing CMVpp65 peptide (HLA/CMV). In these experiments, CMVpp65 peptide containing MHC class I molecules HLA-A*0201/NLVPMVATV (HLA-A2) and HLA-B*0702/TPRVTGGGAM (HLA-B7), generated by ProImmune (Oxford, UK), were used. Complexes, formed by scFvSA and HLA/CMV, are hereafter referred to as targeted constructs (TC).


76

Expansion of virus-specific autologous cytotoxic T-lymphocytes (CTL)Thawed PBMC (>90% CLL cells) from CMV seropositive CLL patients were used for expansion of CMV specific CTL as described before(11). Briefly, PBMC (at a final concentration of 5x106 cells ml-1) were stimulated with CMVpp65 peptide (1.25 μg ml-1) loaded CLL cells and IL-2 (50 U ml-1, Biotest Ag, Dreieich, Germany). After one week, cells were restimulated on a weekly basis with CMV peptide loaded irradiated (30 Gy) EBV transformed cell-lines expressing either HLA-A2 or HLA-B7 (5x104 cells ml-1) in the presence of IL-2. Percentage of CMV specific CTL in culture was determined by FACS analy-sis staining cell culture samples with CD8-PE (Becton Dickinson, San Jose, CA) and APC-conjugated A2 or B7 CMV tetramers (Sanquin, Amsterdam, the Netherlands).

Cytotoxicity assayCTL activity was measured in a standard 51Cr release assay(28). As targets, freshly thawed CLL cells and B cells were used. Target cells were labelled with 50 µl 51Cr (30 μg ml-1, 500 μCi mg-1, Amersham, Buckinghamshire, England). After washing, cells were incubated at 37°C either with 0.1 μg ml-1 CMVpp65 for 1 hour or with scFv CD20 (50 μg ml-1) for 30 min and, subsequently, with HLA/CMV (at various concentrations) for 20 min. Autologous CMV-spe-cific effector cells (obtained as described above) were incubated with 3000 51Cr-labeled target cells at a 4:1 effector:target ratio. After 4 hours, released radioactivity was measured and specific lysis was calculated according to the following formula: percentage of specific release = ((experimental release - spontaneous release)/(maximum release - spontaneous release)) x 100 %. Re-sults are presented as specific lysis and represent the median of specific lysis in 6 replicate samples ± SEM.

In vitro stability assayHLA class I negative Daudi cells were targeted with TC as described before(23). Briefly, Daudi cells were incubated with scFv CD20 at 50 μg ml-1 for 30 min and, after washing, with HLA/CMV complexes at 2.5 μg ml-1 for another 20 min. After labelling, cells were washed and resuspended in medium and grown at 37°C in a 5% CO2 containing atmosphere. At various time points, samples of cells were taken, washed, stained with a FITC labelled anti-HLA class I an-tibody (PharMingen, San Diego, CA) and analysed by flowcytometry.Furthermore, HLA-A2 negative CLL cells were targeted with HLA-A2 TC.

Chapter 4

77

CLL cells were incubated with scFv CD20 at 50 μg ml-1 for 30 min and, after a washing step, with HLA/CMV complexes at 2.5 μg ml-1 for another 20 min. After labelling, cells were washed and resuspended in medium and grown at 37°C in a 5% CO2 containing atmosphere. Immediately after labelling and after 24 hours, samples were taken, washed, stained with a FITC labelled anti-HLA-A2 antibody (PharMingen, San Diego, CA) and analysed by flowcytom-etry. Furthermore, TC coated CLL (t=24h) and non-treated CLL cells from the same donor (t=24h) were used as targets in a standard 51Cr release assay with CMV specific HLA-A2 CTL as effector cells at a 1:1 effector:target ratio.

TC induced proliferation of CMV specific cytotoxic T-lymphocytesFreshly thawed PBMCs from CMV seropositive CLL patients were labelled with CFSE (Molecular Probes, Eugene, OR). After washing, cells were incu-bated with scFv CD20 (50 μg ml-1) and, subsequently, with HLA/CMV (500 ng ml-1). The labelled cells were then cultured in medium at 37°C. On day 7, samples were taken from the cell cultures and stained with CD8 PE and APC-conjugated CMV tetramers.

Cytokine production assaysCMV specific effector cells were stimulated for 6 h at 37°C with TC coated CLL(3;19). All stimulations were performed in medium containing 4 μg ml-1 anti-CD28 (15E8) and 2 μg ml-1 anti-CD29 (TS2/16). For the last 5 h of cul-ture, brefeldin A (Sigma-Aldrich) was added in a final concentration of 10 μg ml-1. After culture, the cells were fixed in FACS lysing solution (BD Biosci-ences) and permeabilized (BD Biosciences). On those cells, an intercellular staining was performed using anti-IFN-γ-FITC, anti-TNF-a-FITC and anti-MIP-1β- PE (all from BD Biosciences). To discriminate CTL from CLL, anti-CD8 PerCP Cy5.5 (BD Biosciences) was used, whereas anti-C69 APC (BD Biosciences) staining was applied to visualize activated cells.In order to analyse the cytotoxic capacity of cytokine producing CMV spe-cific CD8 cells, CMV specific effector cells were stimulated with TC coated or CMVpp65 treated CLL cells as described above.To isolate the IFN-γ secreting cells, they were labelled using an IFN-γ capture assay as described before(2). Briefly, the cells were washed in ice cold buffer (phosphate buffered saline containing 0.5% BSA and 2 mM EDTA) and resuspended in ice cold medium. Next, IFN-γ capture reagent was added (Miltenyi Biotech, Bergisch Gladbach, Germany) and the cells were put on ice for 5 minutes. Then warm culture


78

medium was added and the cells were incubated at 37°C for 45 minutes under slow continuous rotation. After incubation, ice cold buffer was added and the cells were spinned down. Subsequently, the pellet was resuspended in ice cold buffer. PE-conjugated IFN-γ detection antibody (Miltenyi Biotech) and FITC-conjugated antibodies against CD4, CD14, CD16 and CD19 (all BD Biosci-ences) were added and the cells were incubated on ice for another 10 minutes. After washing with ice cold buffer, the cells were resuspended in RPMI con-taining 5% FCS. CD4/14/16/19 negative IFN-γ positive cells were sorted by flowcytometry (FACS Aria, BD Biosciences) and collected. The sorted cells were cultured for 4 days (in the presence of IL-2) and subsequently used as ef-fector cells in a standard chromium release assay at a 1:1 effector:target ratio.

Results

Lysis of HLA/CMV targeted CLL by CMV specific CTLWe tested the efficacy of in vitro expanded CMV specific CTL in a standard 4h chromium release assay against autologous CLL cells either loaded with CMVpp65 or targeted with TC formed by scFv CD20 and HLA/CMV. Using a 50 μg ml-1 concentration of scFv CD20 and varying concentrations of HLA/CMV, targeted CLL cells were lysed. When diluting the HLA/CMV reagent, lysis of targeted CLL became suboptimal at concentrations ≥ 4 ng ml-1 (figure 1A), whereas lysis of CMVpp65 loaded CLL was not efficient at concentra-tions ≥ 100 pg ml-1 (data not shown). In view of the molecular weight of both constructs (HLA/CMV Mw=45kDa; CMVpp65 Mw=1kDa) at a molar base 100 pg ml-1 of CMVpp65 is comparable to 4.5 ng ml-1 HLA/CMV. Thus, lysis of TC targeted CLL appears as efficient as lysis of CLL presenting CMVpp65 via autologous HLA class I molecules. Pre-incubation of the effector cells with HLA/CMV molecules did not inhibit lysis of TC coated CLL (data not shown). Moreover, pre-incubation of the effector cells with complete (pre-coupled) TC induced efficient lysis of uncoated CLL (data not shown), indicating that free HLA/CMV molecules or TC do not interfere with cytotoxic T cell function. Next, scFv CD20 concentrations were varied (using a constant HLA/CMV concentration of 100 ng ml-1). Lysis was limited below concentrations of 100 ng ml-1 (figure 1B). Molecular weights of scFv CD20 and HLA/CMV are equal (=45kDa). scFv CD20 can bind a maximum number of 4 biotinylated HLA molecules thereby forming tetramers. Since lysis remains efficient at equal

Chapter 4

79

Figure 1. CTL me-diated lysis of CMV loaded CLL.CLL cells either di-rectly loaded with CMVpp65 or coated with CD20-HLA/CMV (TC) were used as targets for ex vivo expanded CMV spe-cific CTL in a stan-dard chromium re-lease assay (E:T ratio = 4:1). In A, lysis of CMVpp65 (100 μg ml-1) loaded CLL (HLA-A2) is com-pared to CLL coated with a fixed concen-tration of scFv CD20 (50 μg ml-1) and vary-ing concentrations of HLA/CMV. Results shown are the mean ± SEM of five indepen-dent experiments. B The effect of dilu-tion of scFv CD20 on CTL mediated lysis of TC coated CLL cells. In all conditions, the concentration of HLA/CMV was 100 ng ml-1. C CTL mediated lysis

of TC coated CLL cells is HLA-restricted and antigen specific. HLA-B7 CMV specific CTL were used as effector cells. Target cells were either CLL cells from a HLA-B7 donor coated with B7-TC, or CLL cells from a HLA-A2 donor coated with either A2-TC or B7-TC ([scFv CD20] = 50 μg ml-1; [HLA/CMV] = 500 ng ml-1). The right hand bar rep-resents activity of non-CMV specific autologous T cells against TC coated CLL.

A

2000 400 100 20 4

5

15

25

35

45

scFv CD20 (ng ml-1)

spec

ific

lysi

s (%

)

B

0.84201005002500---HLA/CMV (ng ml-1)

+++++++--scFv CD20 (50 µµµµg ml-1)

-------+-CMVpp65 (100 µµµµg ml-1)

-10

0

10

20

30

40

50

60

70

% s

peci

fic ly

sis

*

0

10

20

30

40

% s

peci

fic ly

sis

B7B7B7effector

B7A2B7TC

A2A2B7target

C


80

scFv CD20 and HLA/CMV concentrations (both 100 ng ml-1) this suggests that, even in a monomer form, TC facilitate killing of the targeted cells. To test whether the TC behave as genuine peptide presenting MHC molecules, CLL cells from a HLA-A2 patient were coated with either A2-TC or B7-TC. B7-TC coated HLA-A2 CLL were lysed by CMV specific HLA-B7 CTL to a comparable extent as B7-TC coated autologous (HLA-B7) CLL. In contrast, A2-TC coated HLA-A2 CLL were unaffected by CMV specific HLA-B7 CTL (Figure 1 C). Importantly, autologous, non-CMV specific CD8+ T cells (con-taining only 1.76% CMV specific CTL) did not induce lysis of TC coated CLL (Figure 1C). Therefore, it can be concluded that there is a restriction to lysis determined by the HLA type of the TC and the antigen specificity of the CTL, indicating MHC/peptide-TCR interactions trigger CTL activity similar to au-tologous MHC molecules.

The effect of CD20 surface expression on TC induced lysis CLL cells have a low surface expression of CD20 as compared to normal B cells(20). To test the effect of CD20 surface expression on susceptibility to

0.8420100500--HLA/CMV (ng ml-1)

++++++-scFv CD20 (50 µµµµg ml-1)

-10

0

10

20

30

40

50

60CD20 lowCD20 highB-cell

% s

peci

fic ly

sis

Figure 2. The effect of CD20 expression on CTL mediated lysis of CD20-HLA/CMV coated B cells. CMV specific CTL were used as effector cells in a 4 hour chromium release assay. As tar-gets (E:T ratio = 4:1) we used either CLL cells from a donor with (unusually) high CD20 surface expression (CD20high; MFI =538), “classical” CLL cells (CD20low; MFI=88) or normal B cells (MFI = 770). A fixed concentration of ScFvCD20 was used in combina-tion with varying concentrations of HLA/CMV.

Chapter 4

81

TC mediated lysis, CLL cells (CD20 expression: mean fluorescence intensity (MFI)=88), normal B cells (MFI=770) and CLL cells with (unusually) high CD20 surface expression (CD20high; MFI=538) were coated with TC using varying HLA/CMV concentrations. Only at HLA/CMV concentrations below 4 ng ml-1, the lysis of TC coated CLL by CMV specific CTL was impaired, whereas the lysis of normal B cells and CD20high CLL remained unaffected (figure 2). Therefore, the tenfold lower CD20 expression on CLL does not prevent efficient lysis by TC.

- +0

5

10

15

20

25

TC

% s

peci

fic ly

sis

A B

HLA

iso

HLA a,b,c

0 24h 48h

HLA

iso

HLA

iso

MFI

Figure 3. In vitro stability of the CD20-HLA/CMV targeted complex (TC). A HLA class I negative CD20 positive Daudi cells were coated with scFv CD20 (50 μg ml-1) and subsequently with HLA-A2/CMV (2.5 μg ml-1). The presence of the TC was detected with a FITC labelled anti-HLA class I antibody and analysed by flowcytom-etry, both immediately after labelling and 24 and 48 hours after labelling (iso = isotype control). B HLA-A2 negative CLL were coated with scFv CD20 (50 μg ml-1) and subse-quently with HLA-A2/CMV (2.5 μg ml-1). 24 hours after labelling, the TC coated CLL cells were used as targets in a standard 4h chromium release assay (E:T ratio = 1:1).

In vitro stability of TC To evaluate the stability of the cell bound TC in time, we targeted (HLA class I negative) Daudi cells with TC. The presence of the TC on the cells was de-tected using a broad reacting anti-HLA class I antibody and FACS analysis (figure 3A). TC appeared to be stable for at least 24 hours when targeted to Daudi cells. To analyse the stability of TC on CLL cells , we targeted HLA-A2 negative CLL cells with HLA-A2 TC, and used these cells afters 24 hours as targets in a stan-dard chromium release assay. The presence of TC on the cells was detected


82

with an anti-HLA-A2 antibody and FACS analysis. At 24 hours after target-ing, the presence of the TC on CLL as detected by FACS analysis was severely reduced (data not shown). However, these cells were still lysed effectively by CMV specific HLA-A2 CTL (figure 3B). Thus, apparently 24 hours after bind-ing of TC there still is a sufficient amount present on CLL to induce lysis.

TC induced proliferation and cytokine production in CMV specific CTLIn order to activate the immune system in vivo, CMV specific CTL must be capable of proliferation and production of cytokines upon antigen encounter. We tested both capacities in an in vitro setting. To track antigen specific prolif-eration, we labelled a PBMC fraction of a CLL patient which contained CMV specific CTL (0.32% of total lymphocytes) first with CFSE and, subsequently, with a combination of scFv CD20 (50 μg ml-1) and HLA/CMV (500 ng ml-1).

- TC pp65

CD8

CFSE

tetr

amer

0.1 6.0 11.6

MFI

Figure 4. Proliferation of CMV specific CTL.Expansion of CMV specific CD8+ T cells was assessed after stimulation of PBMC with CLL cells coated with CD20-HLA/CMV TC or directly loaded with CMVpp65, all in the presence of IL-2 (50 U ml-1) . At day 7 the percentage of CMV specific CTL in the to-tal lymphocyte population was determined by CD8 and CMV tetramer double staining. Before start of the assay, PBMC were labelled with CFSE to track cell division within the CMV specific CTL population (as depicted in the lower panel of the figure).

Chapter 4

83

After labelling, the cells were incubated for 7 days in the presence of IL-2 (50 U ml-1). Proliferation of CMV specific CTL was measured by analysing the percentage of CD8+/tetramer+ cells in the lymphocyte population. Within the CD8+/tetramer+ fraction, we measured the intensity of the CFSE signal (fig-ure 4). Whereas non-CMV labelled PBMC showed no increase in CMV spe-cific CTL, an increased percentage of CMV specific CTL was observed in the HLA/CMV labelled PMBC fraction (6.0% of total lymphocytes) which was accompanied by a decrease in CFSE signal intensity in most cells (indicating proliferation has occurred). Furthermore, we measured the production of several cytokines by CMV spe-cific CD8 cells upon 6 hours of stimulation with TC coated CLL. TC coated CLL induced activation of CMV specific CD8 cells (as indicated by CD69 up-regulation) resulting in production of IFN-γ, TNF-α and MIP-1β (figure 5A). Next, we wanted to confirm that these cytokine producing cells are indeed the same cells that can induce lysis of those TC coated cells. To this end we stimulated CMV specific CTL with either TC or CMVpp65 coated CLL. Sub-sequently, we labelled IFN-γ producing cells with an IFN-γ detection kit and sorted them (figure 5B). Four days later, we tested the capacity of those cells to lyse CMVpp65 loaded CLL. We noticed that both TC and CMVpp65 pre-stimulated IFN-γ secreting cells were capable to induce lysis of both CMVpp65 loaded or TC coated CLL (figure 5C). Thus, the TC are capable of inducing immune activation as assessed by proliferation and cytokine production by CMV specific CTL.

Discussion

In the present study we have shown that CLL cells, targeted with complexes consisting of a streptavidin(SA) fused anti-human CD20 single chain vari-able fragment coupled to CMV-peptide loaded biotinylated HLA class I mol-ecules (= targeted complexes: TC), can be lysed by autologous CMV specific CTL with similar efficiency as CLL cells directly loaded with CMV-peptide. Furthermore, we have demonstrated that TC coated CLL induce both prolif-eration and cytokine production (interferon γ, tumor necrosis factor α, and macrophage inflammatory protein-1β) in CMV specific CD8+ T cells. Most active immunotherapies require knowledge of the eliciting antigen and ex vivo manipulation of patient cells. The first aspect can be circumvented by


84

IFN-

TNF-αααα

MIP-1ββββ

CD

69

0.3 12.3

0.3

2.2 34.1

TC-

8.5

pp6520.2

57.3

22.3

A

Figure 5. CD20-HLA/CMV TC induced production of cytokines by CMV specific CTL. A Ex vivo expanded autologous CMV specific CTL were incubated with CLL cells that were coated with CD20-HLA/CMV TC or (as a positive control) directly loaded with CMVpp65. The negative control (-) consisted of uncoated CLL cells. After 1 hour of incubation, brefeldin A was added to prevent excretion of produced cytokines into the medium. After 6h of stimulation, the production of IFN-γ, TNF-α and MIP-1β by the CMV specific CTL (selected by CD8 staining) was determined by intracellular staining. For counterstaining, CD69 was used as a marker for cell activation. B CMV specific CTL were incubated with TC or CMVpp65 coated CLL for 4 hours. Subsequently, IFN-γ secreting cells were labelled using an IFN-γ detection kit. Double staining with a com-

bination of FITC labelled CD4/14/16/19. C CD4/14/16/19 nega-tive and IFN-γ positive cells were selected and incubated for four days in the presence of IL-2. Subsequently these cells were used as effectors in a standard chromium re-lease assay. In separate experiments (indicated by the interrupted line), ei-ther CMVpp65 loaded or TC coated CLL served as targets (E:T ratio = 1:1).

- pp65 - TC0

25

50

targets

stimulation beforepurification:

TCpp65

% s

peci

fic ly

sis

CB

- TC pp65

CD4/14/16/19

IFN

- γγ γγ

0.1 26.013.8

Chapter 4

85

applying the resident anti-CMV immunity. The second aspect was addressed in previous research by Savage et al., where the powerful effector mechanisms of virus specific cytotoxic T cells were successfully linked with the specificity of monoclonal antibodies. In vitro experiments with CD20 expressing tumor cells and influenza or EBV derived viral peptides, showed that CD20-targeted HLA class I/peptide complexes were expressed on the cultured tumor cells, and that these targeted tumor cells could be lysed very efficiently by PBMC from healthy donors, in vitro prestimulated with viral antigen(23). In vivo studies with SCID mice xenografted with the human B cell lymphoma Daudi line showed adequate protection from tumor growth when HLA-A2/influ-enza peptide targeted lymphoma cells were inoculated together with influenza specific CTL(22). Whether these findings can be extrapolated to a clinical situation is highly dependent on the penetration of the reagent in tissues. This aspect was addressed by Lev et al., who demonstrated the efficacy of a scFv (anti-CD25 or anti-mesothelin) / HLA-A2 chimerical fusion protein in nude mice bearing pre-established human tumor xenografts (14). It could be ar-gued that one step targeting by such a (~65 kDa) reagent enables higher tissue penetration compared to the two step approach described here. However, in a recent phase I study in B–cell non-Hodgkin lymphoma patients, the strepta-vidin CD20 scFv fragment in combination with biotinylated 99Yttrium showed adequate penetration in lymph nodes, subcutaneous localisations, joints and even testes (4). Since the molecular weight of biotinylated HLA/CMV is simi-lar (~45 kDa) it is expected that the penetration of HLA/CMV will be com-parable. Although a one step approach avoids the exposure of (potentially immunogenic) streptavidin, a two-step labelling procedure enables higher flexibility for patients with different HLA types. Our data on T cell proliferation and cytokine production induced by TC coat-ed CLL are in line with a recent study showing that human immunodeficiency virus (HIV) and Karposi sarcoma antigen containing HLA complexes induce expansion and activation of autologous HIV and Karposi sarcoma specific CTL (27). This emphasizes the potential of antibody targeted MHC complexes as a new approach for therapeutic vaccination. In our study intracellular stain-ing revealed antigen triggered production of IFN-γ, TNF-α and MIP-1β in CD8 positive T cells. IFN-γ is known to recruit T cells, activate macrophages and to induce upregulation of MHC class I and II(24). TNF-α also activates macrophages(21). Finally, MIP-1β is a cytokine which is thought to be in-volved in the activation of monocytes, macrophages and, importantly, den-


86

dritic cells (DC)(17). This activation of macrophages and DC may facilitate the uptake and presentation of tumor cell particles. In vitro it has been shown that offering apoptotic bodies of CLL cells to DC can induce autologous tu-mor specific CTL(12). Thus, in CLL the CD20-HLA/CMV complexes might be a double edged sword: direct induction of both T cell mediated killing and a secondary activation of CLL-specific T cell mediated immune response. CMV-specific CD8+ cells are excellent effector cells for possible use in clinical cellular immunotherapy studies. They are proven to be cytotoxic, not auto-reactive and capable of homing in a broad variety of tissues. The approach is widely applicable because 70–90% of healthy adults are CMV positive. In our recent study 72% of CLL patients were found to be CMV-seropositive(15). It has to be noted that the latter figure might be a slight underestimation of the real proportion of CMV positive patients because hypogammaglobulinaemia is frequently present in (advanced) CLL.Especially in CLL there are a number of advantages of the presented approach over the use of (radiolabeled) CD20 antibodies. Monotherapy with conven-tional doses of rituximab, a human/mouse chimeric CD20 monoclonal anti-body, has only limited efficacy(16). This is probably due to the low expression of CD20 in CLL precluding optimal complement activation(6). Moreover, in contrast to many cell lines, freshly isolated CLL cells are resistant to CD20 mediated antibody dependent cellular cytotoxicity (ADCC) by NK cells, an important second effector mechanism in vivo of CD20 monoclonal antibod-ies(7). Radiolabeled CD20 antibodies have considerable side effects, notably myelosuppression and increased risk for secondary myelodysplasia or acute myeloid leukaemia(30).However, unlike ADCC or complement dependent cytotoxicity, T cell acti-vation and killing requires only a few TCR to drive the process. Notably vi-rus specific CTL require only minute amounts of peptide presented on target cells. The potency of CMV specific CTL is emphasised by the findings that the tenfold lower CD20 surface expression on CLL compared to normal B cells hardly seems to affect the lysis of CLL (Figure 2).The findings of our study constitute a necessary step towards possible applica-tion of CD20-HLA/CMV complexes for immunotherapy of B cell malignan-cies. The next step will be to test the CD20-HLA/CMV targeting complexes in vivo in a combined mouse B cell tumor/CMV model. It is obvious that this re-cently recognized capacity to redirect existing antiviral immunity towards tu-mor cells has a utility in cancer immunotherapy far beyond CMV and CLL.

Chapter 4

87

Acknowledgments

We thank Berend Hooibrink for flowcytometric sorting of interferon gamma secreting cells.

Reference List

1. Arditti FD, Aviner S, Dekel B, et al. Eradication of B-CLL by autologous and allogeneic host nonreactive anti-third-party CTLs. Blood. 2005;105:3365-3371.

2. Bitmansour AD, Douek DC, Maino VC, et al. Direct ex vivo analysis of human CD4(+) mem-ory T cell activation requirements at the single clonotype level. J Immunol. 2002;169:1207-1218.

3. Dhein J, Walczak H, Baumler C, et al. Autocrine T-Cell Suicide Mediated by Apo-1/(Fas/Cd95). Nature. 1995;373:438-441.

4. Forero A, Weiden PL, Vose JM, et al. Phase 1 trial of a novel anti-CD20 fusion protein in pretargeted radioimmunotherapy for B-cell non-Hodgkin lymphoma. Blood. 2004;104:227-236.

5. Gamadia LE, Rentenaar RJ, Baars PA, et al. Differentiation of cytomegalovirus-specific CD8(+) T cells in healthy and immunosuppressed virus carriers. Blood. 2001;98:754-761.

6. Golay J, Lazzari M, Facchinetti V, et al. CD20 levels determine the in vitro susceptibility to rituximab and complement of B-cell chronic lymphocytic leukemia: further regulation by CD55 and CD59. Blood. 2001;98:3383-3389.

7. Golay J, Manganini M, Facchinmi V, et al. Rituximab-mediated antibody-dependent cel-lular cytotoxicity against neoplastic B cells is stimulated strongly by interleukin-2. Haemato-logica. 2003;88:1002-1012.

8. Gorgun G, Holderried TAW, Zahrieh D, et al. Chronic lymphocytic leukemia cells in-duce changes in gene expression of CD4 and CD8 T cells. Journal of Clinical Investigation. 2005;115:1797-1805.

9. Heidema J, de Bree GJ, de Graaff PMA, et al. Human CD8(+) T cell responses against five newly identified respiratory syncytial virus-derived epitopes. Journal of General Virology. 2004;85:2365-2374.


88

10. Hoogendoorn M, Wolbers JO, Smit WM, et al. Generation of B-cell chronic lymphocytic leukemia (B-CLL)-reactive T-cell lines and clones from HLA class I-matched donors using modified B-CLL cells as stimulators: implications for adoptive immunotherapy. Leukemia. 2004;18:1278-1287.

11. Kater AP, Remmerswaal EB, Nolte MA, et al. Autologous cytomegalovirus-specific T cells as effector cells in immunotherapy of B cell chronic lymphocytic leukaemia. Br J Haematol. 2004;126:512-516.

12. Kokhaei P, Choudhury A, Mahdian R, et al. Apoptotic tumor cells are superior to tumor cell lysate, and tumor cell RNA in induction of autologous T cell response in B-CLL. Leuke-mia. 2004;18:1810-1815. 13. Lens SMA, Drillenburg P, den Drijver BFA, et al. Aberrant expression and reverse signal-ling of CD70 on malignant B cells. British Journal of Haematology. 1999;106:491-503.

14. Lev A, Noy R, Oved K, et al. Tumor-specific Ab-mediated targeting of MHC-peptide com-plexes induces regression of human tumor xenografts in vivo. Proceedings of the National Academy of Sciences of the United States of America. 2004;101:9051-9056.

15. Mackus WJ, Frakking FN, Grummels A, et al. Expansion of CMV-specific CD8+CD45RA+CD27- T cells in B-cell chronic lymphocytic leukemia. Blood. 2003;102:1057-1063.

16. McLaughlin P, Grillo-Lopez AJ, Link BK, et al. Rituximab chimeric Anti-CD20 monoclo-nal antibody therapy for relapsed indolent lymphoma: Half of patients respond to a four-dose treatment program. Journal of Clinical Oncology. 1998;16:2825-2833.

17. Menten P, Wuyts A, Van Damme J. Macrophage inflammatory protein-1. Cytokine & Growth Factor Reviews. 2002;13:455-481. 18. Oscier D, Fegan C, Hillmen P, et al. Guidelines on the diagnosis and management of chronic lymphocytic leukaemia. British Journal of Haematology. 2004;125:294-317. 19. Picker LJ, Singh MK, Zdraveski Z, et al. Direct Demonstration of Cytokine Synthe-sis Heterogeneity Among Human Memory/Effector T-Cells by Flow-Cytometry. Blood. 1995;86:1408-1419. 20. Rossmann ED, Lundin J, Lenkei R, et al. Variability in B-cell antigen expression: impli-cations for the treatment of B-cell lymphomas and leukemias with monoclonal antibodies. Hematol J. 2001;2:300-306.

21. Ruddle NH. Tumor-Necrosis-Factor (Tnf-Alpha) and Lymphotoxin (Tnf-Beta). Current Opinion in Immunology. 1992;4:327-332.

Chapter 4

89

22. Savage P, Cowburn P, Clayton A, et al. Anti-viral cytotoxic T cells inhibit the growth of cancer cells with antibody targeted HLA class I/peptide complexes in SCID mice. Interna-tional Journal of Cancer. 2002;98:561-566.

23. Savage P, Cowburn P, Clayton A, et al. Induction of viral and tumour specific CTL re-sponses using antibody targeted HLA class I peptide complexes. Br J Cancer. 2002;86:1336-1342.

24. Schroder K, Hertzog PJ, Ravasi T, et al. Interferon-gamma: an overview of signals, mecha-nisms and functions. Journal of Leukocyte Biology. 2004;75:163-189. 25. Schultz J, Lin Y, Sanderson J, et al. A tetravalent single-chain antibody-streptavidin fusion protein for pretargeted lymphoma therapy. Cancer Res. 2000;60:6663-6669. 26. Scrivener S, Kaminski ER, Demaine A, et al. Analysis of the expression of critical activa-tion/interaction markers on peripheral blood T cells in B-cell chronic lymphocytic leukae-mia: evidence of immune dysregulation. British Journal of Haematology. 2001;112:959-964. 27. Stebbing J, Gazzard B, Patterson S, et al. Antibody-targeted MHC complex-directed ex-pansion of HIV-1- and KSHV-specific CD8(+) lymphocytes: a new approach to therapeutic vaccination. Blood. 2004;103:1791-1795. 28. van Leeuwen EM, Gamadia LE, Baars PA, et al. Proliferation requirements of cytomegalo-virus-specific, effector-type human CD8(+) T cells. Journal of Immunology. 2002;169:5838-5843. 29. Wierda WG, Cantwell MJ, Woods SJ, et al. CD40-ligand (CD154) gene therapy for chronic lymphocytic leukemia. Blood. 2000;96:2917-2924.

30. Witzig TE, White CA, Gordon LI, et al. Safety of yttrium-90 ibritumomab tiuxetan radio-immunotherapy for relapsed low-grade, follicular, or transformed non-Hodgkin’s lymphoma. Journal of Clinical Oncology. 2003;21:1263-1270.

31. Yellin MJ, Sinning J, Covey LR, et al. T-Lymphocyte T-Cell B-Cell Activating Molecule Cd40-l Molecules Induce Normal B-Cells Or Chronic Lymphocytic-Leukemia B-Cells to Ex-press Cd80 (B7/Bb-1) and Enhance Their Costimulatory Activity. Journal of Immunology. 1994;153:666-674.


5Adequate synapse formation between leukemic B cells and effector T cells following stimulation with artificial

TCR ligandsRogier Mous1, Philip Savage4,5, Eric Eldering2, Peter Teeling3,

Marinus HJ van Oers1 and René AW van Lier2

Departments of 1Hematology, 2Experimental Immunology and 3Pathology,Academic Medical Center, Amsterdam;

4Alexis Biotech Ltd., London, UK; 5Department of Medical Oncology, Charing Cross Hospital, London, UK

Leukemia & Lymphoma (in press)

93

Abstract

Artificial T cell receptor (TCR) ligands can be used to direct virus specific cytotoxic T lymphocytes (CTL) towards tumor cells. Because of their size, these constructs may differ from cognate peptides in their ability to induce T cell activation. We here analyzed signaling outcomes upon synapse formation between human cytomegalovirus (CMV)-specific CTL and chronic lympho-cytic leukemia (CLL) cells through targeted complexes (TC) containing anti-CD20 single-chain variable fragment and biotinylated major histocompatibil-ity complex (MHC) class I molecules presenting peptides from CMVpp65.TC coated CLL cells were effectively lysed by CMVpp65-specific CTL but in-duced less interferon gamma (IFN-γ) production than peptide loaded targets. Confocal microscopy revealed that TC induced a redistribution of TCR/CD3 but not CD2 towards the immunological synapse. Furthermore, morphologi-cal examination of immunological synapses showed smaller and “patchy” in-teractions between TC coated B cells and CTL as compared to peptide coated targets. Finally, pre-incubation of CTL with CD2 antibodies led to a Fc-de-pendent redistribution of CD2 into TC-induced synapses and restored IFN-γ production by CMV-specific CTL. Thus, redistribution of CD2 towards the immunological synapse appears to be essential for full T cell activation. However, CD2 triggering is not required for efficient lysis of tumor cells, demonstrating that CTL require only minimal stimulation to release their cytotoxic content.

T cell activation by artificial TCR ligands

94

Introduction

Like in other types of cancer, non-responsiveness to chemotherapy due to dysfunction of p53 is a serious problem in the treatment of CLL (1;4;10). Therefore, an urgent need exists for therapies that do not rely on functional p53. One possible way to circumvent p53 dysfunction may be passive immu-notherapy. In this respect, treatment of CLL with the CD52 monoclonal an-tibody alemtuzumab seems promising, since it showed to be effective against CLL with mutated p53 (32).Besides the use of monoclonal antibodies, active immunotherapy for CLL is also the topic of investigation in the search for p53 independent therapeutic strategies. In this perspective, several groups have described techniques to elicit CLL specific T cell responses in vitro, either via dendritic cells transfected with CLL mRNA (24) or by enhancing immunogenicity of CLL cells via CD40 liga-tion (16;34;37). However, the induction of T cell responses against CLL cells in vivo may be hampered by immuno-tolerance towards the tumor. To solve this problem, we and others have attempted to exploit the powerful cytotoxic capacities of antiviral CTL by redirecting them to tumor cells. Previously, we have observed that CLL patients have an increased number of CTL directed against CMV without any signs of viral reactivation (22). Furthermore, we have demonstrated that the cytolytic capacity of these CTL can be used to lyse autologous CMV peptide-loaded CLL cells (18). Recent studies indicate that this strategy may also be applicable in vivo, since we could show that CLL cells were effectively lysed by CMV-specific CTL when targeted with targeted complexes (TC) consisting of MHC class I molecules containing CMVpp65 peptide and a CD20 single chain variable antibody fragment (23).A potential danger of using a strong TCR agonist specific for expanded popu-lations of effector T cells is the massive release of cytokines that can lead to the cytokine release syndrome (2;7;33). To evaluate the efficacy and potential risk of CTL activation induced by TC we compared T cell activation parameters induced by targets loaded with either CMVpp65 peptide or TC. Furthermore, the immunological synapse formation between CMV-specific CTL and either TC–coated or CMV-peptide-loaded CLL cells were compared as to the role of accessory molecules as well as to the dimensions of the synapse.

Chapter 5

95

Methods

Patient samplesAfter obtaining informed consent, 30 ml of blood was drawn from patients fulfilling the diagnostic criteria for CLL (26). Freshly isolated PBMC were fro-zen in IMDM supplemented with 20% fetal calf serum (FCS) and 10% DMSO (Sigma Chemical Co.) and stored in liquid nitrogen before use. On the day of use, cells were thawed (viability after thawing > 90%) and cultured in RPMI 1640 containing 10% v/v heat-inactivated FCS, 100 U ml-1 penicillin and 100 μg ml-1 streptomycin. All procedures concerning recruitment of patients and collection of patient materials were approved by the Medical Ethical Commit-tee of the Academic Medical Center.

ReagentsHLA-A2-binding NLVPMVATV CMVpp65 peptide and HLA-B7-binding TPRVTGGGAM CMVpp65 peptide (IHB-LUMC peptide synthesis library facility, Leiden, The Netherlands) were used. The B9E9 scFvSA fusion protein, consisting of the variable region of murine IgG2a anti-human-CD20 fused to streptavidin (scFv CD20) (29) was used to target CD20 positive cells with biotinylated human MHC class I molecules containing CMVpp65 peptide (HLA/CMV). In our current studies, CMVpp65 peptide containing MHC class I molecules HLA-A*0201/NLVPMVATV (HLA-A2) and HLA-B*0702/TPRVTGGGAM (HLA-B7), generated by ProImmune (Oxford, UK), were used. Complexes, formed by scFv CD20 and HLA/CMV are hereafter referred to as targeted constructs (TC). Antibodies against CD2 (CLB-CD2.1, CLB-T11/1), LFA1 (CLB-54) and CD80 (DAL 1) were purified from hybridoma culture supernatants. F(ab’)2 fragments from CLB-CD2.1 and CLB-T11/1 were produced by pepsin diges-tion. All antibodies were used at a final concentration of 10 μg ml-1.

In vitro expansion of CMV specific cytotoxic T lymphocytes (CTL)Thawed PBMC from three HLA-A2 or HLA-B7 CMV seropositive CLL pa-tients (all Rai stage II and untreated) were used for expansion of CMV-specific CTL as described before (18). Briefly, PBMC (at a final concentration of 5-10x106 cells ml-1) were cultured in IMDM containing 10% v/v Human Pooled Serum (HPS; Cambrex, East Rutherford, NJ), CMVpp65 peptide (1.1 μM) and IL-2 (50 U ml-1 Biotest Ag, Dreieich, Germany). After one week of culture, cells


96

were restimulated every 7 days with CMV peptide loaded irradiated (30 Gy) Epstein Barr virus (EBV) transformed B cell-lines expressing either HLA-A2 or HLA-B7 (5x104 cells ml-1) in the presence of IL-2. The percentage of CMV-specific CTL in culture was determined by FACS analysis staining cell culture samples with CD8-PE (Becton Dickinson, San Jose, CA) and APC-conjugated A2 or B7 CMV tetramers (Sanquin, Amsterdam, the Netherlands). Cytotoxicity assayCTL activity was measured in a flow cytometry-based assay. As targets, freshly thawed CLL cells were used. Target cells were labeled with CellTrace Far Red DDAO-SE (Molecular Probes, Eugene, OR) to distinguish them from effector cells. After washing, the labeled CLL cells were incubated at 37°C either with CMVpp65 peptide for 1 hour or with scFv CD20 (1.1 μM) for 30 min and sub-sequently (after one washing step) with HLA/CMV for 20 min. Autologous CMV-specific effector cells (obtained as described above) were incubated with the labeled target cells at a 1:1 effector:target ratio. After 4 hours, target cell death was determined by 3,3’-dihexyloxacarbocyanine iodide (DiOC6(3)) staining. Gating on DDAO-SE positive cells, cell death was measured by de-termining the percentage of DiOC6(3)low cells within the target population. In some conditions, effector cells were pre-incubated for 2 hours with con-canamycin A (100nM; Sigma-Aldrich) or target cells were pre-treated with CD95 blocking antibody (Fas2; 10 μg ml-1) to determine the role of granule excretion and Fas ligation in target cell killing, respectively.

Calcium release assayEBV transformed B cells were either loaded with CMVpp65 peptide (250 nM) or coated with TC (HLA/CMV concentration 250 nM). Next, these cells were allowed to adhere to the bottom of poly-L-lysine coated 96 wells flat bottom culture plates for 10 minutes at 37°C. Subsequently, CMV-specific CTL loaded with fluo-4 AM (Molecular Probes, Eugene, OR; final concentration 2μM) were added to the wells and the plates were inserted into a Wallac 1420 Vic-tor3TM plate reader (PerkinElmer, Waltham, MA). Over time, calcium release was determined by measuring Fluo-4 AM fluorescence (Eexp) with intervals of 10 seconds (excitation filter 488nm, emission filter 535nm). At the end of each assay, the maximal calcium signal (Emax) was determined by adding 0.1% Triton-X to each well and subsequently the zero signal (E0) was determined by adding 4M EDTA to each well. The emission values were recalculated to per-

Chapter 5

97

centage of maximal experimental release using the formula: (Eexp-E0)/(Emax-E0) x 100%.

Cytokine production assaysCMV-specific effector cells were stimulated for 6 h at 37°C with CMVpp65 peptide loaded or TC coated CLL cells. All stimulations were performed in medium containing 4 μg ml-1 anti-CD28 (CLB-CD28/1) and 2 μg ml-1 anti-CD29 (TS2/16)(9;27). For the last 5 h of culture, brefeldin A (Sigma-Aldrich) was added in a final concentration of 10 μg ml-1. After stimulation, the cells were fixed in FIX&PERM Solution A (An der Grub, Kaumberg, Austria) and permeabilised in Solution B (An der Grub) in the presence of anti-IFN-γ FITC, anti-CD69 PE, anti-CD3 PerCP Cy5.5 and anti-CD8 APC (all BD Bio-sciences). IFN-γ production and CD69 expression in CMV-specific CTL (gate on CD3+/CD8+ cells) were evaluated via flow cytometry.

Transmission electron microscopy EBV transformed B cell lines were either loaded with CMVpp65 peptide (250 μM) or coated with TC (HLA/CMV concentration 250 μM). Next, the labeled EBV transformed B cells were added to CMV-specific CTL in a 96 wells round bottom plate. The plates were then centrifuged and incubated at 37°C/ 5% CO2 for 10 minutes to enable conjugate formation. Subsequently, preparation for electron microscopy was performed as previously described(9). Briefly, upon fixation in Karnovsky’s fixative (0.1 M cacodylate buffer pH 7.4 contain-ing 4% w/v formaldehyde and 2.5% w/v glutaraldehyde) samples were treated with 1% osmium tetroxide plus 0.5% uranyl acetate and embedded in Epon Resin (Hexion Specialty Chemicals, Hoogland, The Netherlands). Ultra-thin sections were stained with uranyl acetate and lead citrate, and subsequently examined under a Philips CM10 electron microscope; images were obtained and measured with iTEM software (Soft Imaging System GmbH). For deter-mination of intermembrane separation distance, several conjugates were ana-lyzed and > 100 measurements were performed (CMVpp65 peptide as well as TC) on tight junctions where both membranes had a trilaminar appearance.

Confocal fluorescence microscopyImaging of conjugates between target and effector cells was performed as pre-viously described (20). Peptide and TC loading of EBV transformed B cells


98

was performed as for electron microscopy. During the last 15 minutes of load-ing, CMTMR Orange (0.5 μM; Molecular Probes, Eugene, OR) was added to discriminate target cells from effector cells. Next, loaded targets were incu-bated with CMV-specific CTL to induce conjugate formation. In some con-ditions, antibodies against CD2 (CLB-CD2.1) or CLB-CD2.1-derived F(ab’)2 fragments were added to the CTL prior to conjugate formation. Subsequently, the cells were gently resuspended and dropped onto poly-L-lysine coated glass coverslips. After 5 minutes, the coverslips were washed two times with phos-phate-buffered saline (PBS) and subsequently the attached cells were fixed with 3% v/v paraformaldehyde. Cells were stained with CD3 FITC (Pharmingen, San Diego, CA) or FITC labeled CLB-CD2.1 antibody (CD2). Blocking anti-bodies and F(ab’)2 fragments against CD2 (CLB-CD2.1) were visualized using FITC labeled goat-anti-mouse F(ab’)2 fragments (Jackson ImmunoResearch Laboratories, West Grove, PA). Next, the coverslips were washed with PBS and briefly incubated in PBS with 0.1% w/v saponin (Calbiochem, San Diego, CA) and 3% v/v FCS. Finally the coverslips were embedded in Vectashield with DAPI (Vector Laboratories, Burlingame, CA). Imaging was performed with a Leica TCS-SP2 confocal microscope and Leica LCS software. Excitation/de-tection of the 3 fluorophores was performed as follows: FITC: 488nm/500-520nm, CMTMR orange: 594nm/600-630nm and DAPI 405nm/420-475nm. To avoid crosstalk between the fluorophores sequential line scan was used. In each microscopy slide, at least 30 conjugates between T cells and target cells were analyzed and the localization of CD3 and CD2 (randomly distributed or focused to the point of interaction between two cells) was assessed by eye in single plane scans. Additional Z-stack imaging was performed on most slides to confirm the position of CD3 and CD2 (focused or randomly distributed) observed in single plane scans.

Results

TC induce submaximal cytokine production by CMV-specific CTLIn previous studies we established that CD20-targeted HLA/CMVpp65 com-plexes (TC) were able to induce lysis of targeted CLL cells and to trigger cyto-kine production by autologous CMV-specific CTL (23). Here, we compared the limiting concentrations of CMV pp65 peptide and TC for lysis of CLL cells as well as target cell-induced cytokine production by CTL. In line with our

Chapter 5

99

0

TCpp65

102 103 104 105 106

25

50

75

pp65 (pM)

% C

D69

+ /(CD

8+ )

00

5

10

15

20

102 103 104 105 106

pp65 (pM)

% IF

Nγγ γγ+ /(C

D8+ )

*p<0.05

C

D

*

A

00

25

50

75

100

pp65TC

10 1 10 2 10 3 10 4 10 5 10 6

pp65 (pM)

% o

f max

imal

resp

onse

- Fas2 CMA0

25

50

75

100 pp65TC

% o

f max

imal

res

pons

e

B

Figure 1. Both CMVpp65 peptide and TC coated CLL cells induce granule-medi-ated lysis and IFN-γ production by CMV specific CTL. CLL cells were coated with either CMVpp65 peptide or CD20 targeted HLA/CMVpp65 complexes (TC). Next, the limiting peptide concentration for lysis of coated CLL cells and IFN-γ production by CMV-specific CTL were determined (E:T ratio = 1:1). A The effect of peptide concen-tration on lysis of coated CLL cells. The efficacy of lysis is reflected as % of the maxi-mal effect (maximal specific lysis for pp65: 45% and TC: 40%). The average of three independent experiments is shown. Error bars indicate SEM. B CLL cells were pre-incubated with a blocking antibody against CD95 (Fas2; 10 μg ml-1) or CMV-specific CTL were pre-treated with concanamycin A (CMA; 100 ng ml-1). Subsequently, lysis of either CMVpp65 peptide- or TC-coated CLL was determined (peptide concentration 250 nM). Results are from a single experiment; maximum specific lysis induced in this experiment was 47% for CMVpp65 peptide- and 39% for TC-coated CLL. Lysis of non-coated CLL cells by CMV specific CTL was <5% compared to control C The effect of peptide concentration on IFN-γ production by CMV-specific CTL triggered by CLL cells either coated with CMVpp65 peptide or TC. Mean of three independent experiments, error bars indicate SEM. D Activation (indicated by upregulation of CD69) of CMV-specific CTL by either CMVpp65 peptide or TC coated CLL. Mean of three independent experiments, error bars indicate SEM.


100

previous observations, CMVpp65 peptide and TC induced equally efficient lysis of targeted CLL cells. At concentrations above 1nM both CMVpp65 pep-tide and TC gave maximal lysis of coated CLL (45% vs 40%, respectively; fig-ure 1A). To explore the mechanism(s) of cytotoxicity by both agonists, either CMV-specific CTL were pre-treated with concanamycin A or the target cells

0

10

20

30

40

50

60

70

pp65TC-

18 36time (min)

% o

f max

imal

rele

ase

A

B *

* ns

*p<0.05

0

- pp65 TC0.0

0.5

1.0

1.5

∆∆ ∆∆ %

of m

axim

al r

elea

se m

in-1

Figure 2. CMVpp65 peptide and TC trigger calcium release in CMV specific CTL. CMV-specific CTL were loaded with fluo-4 AM (2μM) and subsequently incubated with plate-adhered EBV transformed B cells that were treated with either CMVpp65 peptide or TC (peptide con-centration 250 nM). Over time, changes in free calcium signal of the total CTL population were measured with via fluorometry (A488nm/E535nm). Non-CMV treated target cells (-) served as a control. A representative calcium release experiment. The results are presented as percentage of maximal experimental release (see materials and methods) B Quantification of the changes in free calcium signal per minute (Δ % of maximal release min-

1) within the CMV specific CTL population dur-ing the first 10 minutes after incubation with target cells (average of five independent experi-ments; error bars indicate SEM).

were pre-incubated with the CD95 blocking antibody Fas2. Subsequently, the effect on lysis of CMVpp65 or TC coated CLL cells was tested. As depicted in figure 1B, lysis of both CMVpp65 peptide and TC coated CLL could only be blocked by pre-treating CMV-specific CTL with concanamycin A, demon-strating that lysis of both CMVpp65- and TC-coated target cells is mediated by release of cytolytic granules by CMV-specific CTL. A quantitative difference between peptide and TC became apparent when cy-tokine production capacity was tested. CMV-specific CTL were triggered to produce IFN-γ when stimulated by CLL cells loaded with CMVpp65 peptide and TC concentrations of 10nM and higher. IFN-γ production induced by TC reached a plateau at 250 nM, whereas a similar concentration of CMVpp65

Chapter 5

101

peptide induced more IFN-γ production (6.55 ± 1.53 vs 16.36 ± 3.42 % (p=0.02) of CD8+ T cells, respectively; figure 1C). IFN-γ production induced by CMVpp65 peptide only reached a plateau at concentrations higher than 10 μM (data not shown). Notably, upregulation of early activation marker CD69 on CD8+ T cells was comparable after stimulation with either CMVpp65 pep-tide or TC coated CLL cells (figure 1D). Besides small inter-donor variance, no difference was observed between HLA-A2/CMV and HLA-B7/CMV in-duced CTL activity (data not shown).

TC induce calcium release in CMV-specific CTLThe release of calcium from intracellular storage compartments into the cy-toplasm is one of the earliest signaling events upon TCR triggering (25). To investigate the potency of TC to mobilize Ca++, cumulative changes in cyto-plasmatic calcium levels in populations of either CMVpp65 peptide or TC triggered CTL were measured (see materials and methods). Both CMVpp65 peptide and TC treated target cells induced a marked increase in free cal-cium levels in CMV-specific CTL populations compared to CTL populations that were exposed to non-CMV treated target cells (figure 2A). Notably, there was no quantitative difference observed between CMVpp65 peptide and TC induced calcium release (measured as Δ % of maximal release min-1; figure 2B).

CD2 is not fully triggered in TC stimulated CTLNext, a more detailed analysis of synapse formation around autologous MHC class I presenting CMVpp65 peptide or TC was performed. To this end, the lytic capacity of CMV-specific CTL was tested after pre-incubation with an-tibodies against CD2, CD80 and LFA1. Lysis of both CMVpp65 peptide and TC-coated CLL could only be inhibited by antibodies against LFA1 (figure 3A). Notably, in agreement with published data on influenza specific T cell clones (36), antibodies against CD2 or CD80 did not affect lysis by CMV-spe-cific CTL.Next, the influence of antibodies against CD2 and CD80 on IFN-γ production by CMV-specific CTL was tested. CLL cells were coated with CMVpp65 pep-tide or TC (peptide concentration 250 nM) and subsequently used as stimu-lus for CMV-specific CTL that were pre-incubated with blocking antibodies. Since CMV-specific CTL lack expression of CD28, antibodies against CD80, as expected, did not affect IFN-γ production (figure 3B). Furthermore, block-


102

*p<0.05**p<0.01

lysis

-

LFA

1

CLB

-CD

2.1

CD

80

0

1

2pp65TC

***

rela

tive

resp

onse

antibody:

A

B

antibody:

IFN-γγγγ production

-

CLB

-CD

2.1 2

CLB

-CD

2.1

F(ab

')

CLB

-T11

/1 2C

LB-T

11/1

F(a

b')

CD

80

0

1

2

3

rela

tive

resp

onse

* *

** ****

CD2

C no CMV pp65 TC

+CD2 mAb

-

IFN-

CD

69

6.33 7.96

2.89 15.33

0.37

Figure 3. The effect of antibod-ies against synapse molecules on CMVpp65 peptide or TC induced lysis (A) and IFN-γ production (B) by CMV spe-cific CTL. CLL cells were coated with ei-ther CMVpp65 peptide or TC (peptide concentration 250 nM) and incubated with CMV-spe-cific CTL (E:T ratio = 1:1) that were pre-treated with antibodies against CD2 (CLB-CD2.1 and CLB-T11/1), CD80 or LFA1. The effect of these antibodies on A lysis (4h) and B IFN-γ produc-tion (4h) is reflected as relative response ([response in presence of antibody]/[response without antibody]). In the absence of antibodies, TC coated targets induced IFN-γ production in 3.00% of CD8+ T cells compared to 9.87% of CD8+ T cells trig-gered with CMVpp65 loaded targets. Specific lysis observed in the absence of blocking anti-bodies was 22% for CMVpp65 peptide- and 14% for TC-coat-ed CLL cells. The bars in both graphs represent the average of three independent experiments ± SEM. C FACS plots of IFN-γ produc-

tion by CMV-specific CTL (gated on CD3+CD8+ cells). Plots are representative pictures for CTL triggered with non-coated CLL cells (no CMV), CMVpp65 peptide coated CLL cells (pp65) or TC coated CLL (TC), respectively. The lower plots show the effect of pre-incubation of the effector cells with an antibody against CD2 (CLB-CD2.1). The percentage of IFN-γ+/CD69+ cells is indicated in the upper right corner of each plot.

Chapter 5

103

ing CD2 with CLB-T11/1 monoclonal antibody seemed to have an (nonsig-nificant) inhibitory effect on CMVpp65 peptide induced IFN-γ production by CMV-specific CTL. Surprisingly, both CD2 specific antibodies CLB-CD2.1 and CLB-T11/1 induced increased production of IFN-γ by TC stimulated CMV-specific CTL (from 3.00% to 7.01% CD8+/IFN-γ+ cells), approaching the level of CMVpp65 induced IFN-γ production (9.87% CD8+/IFN-γ+ cells). Not only the percentage of IFN-γ positive CD8+ T cells increased in the pres-ence of anti-CD2 antibody (for example from 2.89% to 15.33 % in the experi-ment depicted in figure 3C), but the presence of CD2 antibodies also increased the amount of IFN-γ produced per cell. In contrast, when CMV-specific CTL were pre-incubated with F(ab’)2 fragments of CD2 antibodies CLB-T11/1 and CLB-CD2.1, IFN-γ production by both CMVpp65 peptide and TC stimulated CTL was strongly inhibited, suggesting that the stimulatory effect of the CD2 antibodies was Fc-mediated.

Transmission electron microscopy imaging of the immunological synapse Previously, Choudhuri et al. demonstrated that by introducing MHC class I molecules with an elongated ectodomain on antigen presenting cells, subse-quent T cell activation (measured by cytokine production) was diminished due to an increased intermembrane separation distance within the immuno-logical synapse(9). Since the size of the TC exceed that of natural occurring MHC-I molecules (~90 kDa vs ~45 kDa), we questioned whether this would affect the dimensions of the immunological synapses induced by these con-structs. To evaluate this, sections of embedded conjugates between CTL and either CMVpp65 peptide or TC coated EBV transformed B cell targets were made and the dimensions of the formed synapses were examined using trans-mission electron microscopy. In two independent experiments, the intermem-brane separation distance within the tight junctions formed between target and effector cells and the length of the tight junctions were determined. By measuring intermembrane separation distances in tight junctions formed between either CMVpp65 peptide or TC coated targets and effector cells, some variance in intermembrane separation distance within both CMVpp65 peptide and TC induced synapses was observed. Nevertheless, there was no difference observed in mean intermembrane separation distance between CMVpp65 peptide (mean ± SEM: 8.91 ± 1.85 nm) and TC (8.71 ± 1.85 nm) induced synapses (two-tailed non-parametric T-test: p= 0.48, figure 4B). In a subsequent experiment, the length of tight junctions in conjugates between


104

Figure 4. Transmission electron microscopy imaging of CMVpp65 peptide and TC induced synapses. Conjugates be-tween CMV-specific CTL and either CMVpp65 peptide or TC coated EBV transformed B cells were fixed and embed-ded in paraffin (see ma-terials and methods). Subsequently, sections of the embedded conjugates were stained with uranyl acetate and lead citrate and analyzed using trans-mission electron micros-copy. A images of conjugates between CMV-specific CTL and CMVpp65 pep-tide (upper images) or TC (lower images) coated targets. On the left hand overview images are dis-played (scale bar repre-sents 2 μm), on the right hand zoomed images are shown (scale bar repre-sents 0.5 μm). The tight junctions are indicated by the black arrows.

B intermembrane separation distance within tight junctions. Results of a single experi-ment are shown. Bars represent mean of all measurements, the error bars indicate SEM. C Percentage of tight junctions with a length shorter or longer than 300nm. For both CMVpp65 peptide and TC, 16 conjugates were analyzed and within these conjugates, the length of individual tight junctions was measured (total number of tight junctions measured: 40 CMVpp65 peptide- and 45 TC-induced tight junctions).

B

0

2

4

6

8

10

12

pp65 TC

inte

rmem

bran

e se

para

tion

dist

ance

(nm

)A

CTL

EBV

CTLEBV

TC

pp65

zoom

C

< 300 > 3000

25

50

75 pp65 (n=40)TC (n=45)

length of tight junctions (nm)

% o

f tig

ht ju

nctio

ns

p = 0.24

p = 0.48

Chapter 5

105

target and effector cells was measured. We observed a broad variance in length of tight junctions and also in the number of tight junctions per conjugate. It appeared as if synapses between EBV cells loaded with CMVpp65 peptide and CMV-specific CTL were formed by less but longer tight junctions than syn-apses between TC coated EBV cells and CMV-specific CTL. Thus, TC induced synapses had a “patchy” appearance compared to CMVpp65 peptide induced counterparts (figure 4A and C). However, despite visible qualitative differenc-es in length between TC and CMVpp65 peptide induced tight junctions, this difference was not quantifiable in a statistically significant manner (two-tailed non-parametric T-test: p=0.24).

CD2 is not recruited to TC induced synapsesThe results of the CTL activation experiments in the presence of blocking an-tibodies suggested incomplete CD2 triggering in TC induced T cell activation. Taken together with the observed tight junction morphology of TC induced synapses this raised the question if and where CD2 is located in TC-induced synapses compared to MHC-I/peptide induced synapses. To investigate this, conjugate formation between CMV-specific CTL and either CMVpp65 peptide or TC-coated target cells was allowed and subsequently the position of CD2 in synapses was visualized through confocal fluorescence microscopy. Conjugate formation between T cells and target cells was antigen specific, since no conju-gate formation was observed when CMV-specific T cells were incubated with non-coated target cells (data not shown). Furthermore, in about 50% of all CMVpp65 peptide and TC-induced conjugates, CD3 was concentrated in the synapse between target and effector cell (figure 5). CD2 was also concentrated in CMVpp65 peptide induced synapses, albeit to a lower extent (29.7 ± 0.4 % of all conjugates). In contrast, in TC-induced synapses CD2 remained randomly distributed on the surface of CMV-specific CTL. This could be manipulated by incubating CMV-specific CTL with an antibody against CD2 (CLB-CD2.1) prior to conjugate formation. As a result, CD2 was now recruited towards TC-induced synapses (21.0 ± 8.7%), while CD2 remained randomly distributed on CTL forming conjugates with CMVpp65 peptide loaded targets. This re-cruitment appeared to be Fc-mediated since pre-incubation of CMV-specific CTL with F(ab’)2 fragments of CLB-CD2.1 resulted in an even distribution of CD2 on CTL forming conjugates with TC coated targets (data not shown).


106

Figure 5 Confocal flu-orescence microscopy of conjugates between CMVpp65 peptide or TC coated targets and CMV-specific CTL. EBV transformed B cells were coated with either CMVpp65 peptide or TC and loaded with CMTMR Orange to dis-criminate them from ef-fector cells. A Representative pic-tures of conjugates be-tween target and effec-tor cells demonstrate the position of CD3 (upper panels) and CD2 (mid-dle panels) in the immu-nological synapse (IS; magnification 40x). The bottom pictures were taken from conjugates of CMVpp65 peptide or TC coated targets and CMV-specific CTL that were pre-incubated with an antibody against CD2 (CLB-CD2.1). The positioning of CD2 in these conjugates was subsequently visualized by a FITC labeled goat anti-mouse antibody.

(Cells that do not stain for CD3, CD2 or CMTMR Orange are CLL cells that remain within the effector cell population after in vitro expansion of CMV-specific CTL) B Per-centage of conjugates in which either CD3 or CD2 is located in the IS (bars represent the mean of three experiments). In each experiment, at least 30 conjugates were evaluated per condition; error bars indicate SEM.

pp65 TCA

B

CD3

CMTMR

DAPI

CD2

CMTMR

DAPI

CD2mAb/GααααM

CMTMR

DAPI

CD3 CD2 CD2 (+ blocking Ab)0

25

50

75pp65TC

% c

lust

erin

g

ns

p<0.005 p<0.05

Chapter 5

107

Discussion

In previous studies, we demonstrated that CMV-specific CTL can be redirected towards CLL cells that are targeted with anti-CD20 scFv-MHC-I/CMVpp65 complexes (TC) (23). In the present study, we explored the mechanism behind TC induced CTL activation. We show that TC can efficiently trigger CTL me-diated lysis. However, compared to CMVpp65 activated CTL, IFN-γ produc-tion by TC activated CMV-specific CTL is less efficient. We provide evidence that the reduced capacity to induce IFN-γ production by TC can be explained by incomplete CD2 triggering because:1) CD2 did not translocate towards TC induced synapses and 2) CD2 antibodies restored TC-induced IFN-γ produc-tion to the level of CMVpp65 peptide stimulated CTL and induced the redis-tribution of CD2 towards the immunological synapse. The latter suggests that not only CD2 ligation, but importantly, clustering of CD2 in the immunologi-cal synapse is mandatory to give full CTL activation. Various studies have addressed the role of CD2 in CTL activation (8;15;30;31), but the exact role of CD2 in human effector T cell activation is still not com-pletely understood. A recent publication by Espagnolle et al. suggests that in helper T cells CD2 ligation is essential for sustained calcium signaling and subsequent helper cell activation (11). Our results suggest that in CTL the in-duction of both lytic activity and calcium mobilization appear to be relatively independent from CD2 ligation. These findings are in line with studies from other groups demonstrating that interfering with CD2-CD58 interaction does not affect TCR polarization towards the APC (38) nor the lysis of target cells (19). The latter shows that TCR signaling still occurs in the absence of CD2 ligation. Therefore, it seems that CD2 has a role in modulating certain aspects of TCR signaling in CTL rather than being essential for the initiation of proxi-mal signaling events. The TC used in our studies has a higher molecular weight than autologous MHC-I (± 90 kDa vs ± 45 kDa). Since the dimensions of CD2-CD58 conju-gates seem to be similar to the dimensions of a (normal) TCR-MHC-I conju-gate (35), the most likely position for CD2-CD58 interactions is in the centre of the immunological synapse (close to TCR-MHC-I). Therefore, it can be speculated that the size of TC changes the dimensions of the immunologi-cal synapse and thereby lead to inefficient CD2 recruitment. Choudhuri et al. have demonstrated that altering the dimensions of MHC-I indeed affects the dimensions of the immunological synapse resulting in impaired T cell activa-


108

tion(9). In our current study, transmission electron microscopy analysis of CTL-target conjugates indicated that the synapse formation with TC coated B-cells is different from immunological synapse formation with CMVpp65 coated targets. The apparent predominance of short tight junctions in TC-in-duced conjugates compared to CMVpp65 peptide-induced conjugates (where there is a tendency towards long tight junctions) may reflect deviations from the normal immunological synapse structure in TC-induced synapses. If cor-rect, this would explain both the inability of CD2 to fully redistribute towards the synapse and could also account for CD2 antibodies entering TC-induced synapses through the “gaps” between these tight junctions and engage with Fc receptors on the target cell, inducing an artificial clustering of CD2 and thereby full activation of the CTL (21). This would then also explain the effect of CD2 antibodies on CMVpp65 peptide-induced T cell activation, because according to the above-proposed model the predominant long synapses be-tween CMVpp65 peptide-loaded target cells and CTL will not allow antibody-CD2 complexes to cluster in the vicinity of TCR-MHC-I complexes, a process which apparently seems to be essential for full CTL activation. Alternatively, it can be argued that the reduced IFN-γ production by CMV-specific CTL when triggered with TC is caused by a suboptimal amount of peptide present on the targeted CLL cell (12) as a result of the low surface expression of CD20 (3). However, since surface staining of CTL for the early activation marker CD69 showed a similar concentration-dependent upregulation for pp65 peptide and TC, whereas at the same time the curves for IFN-γ production diverge, this scenario seems unlikely. Thus, we favor the explanation that the observed structural differences between CMVpp65 peptide and TC-induced synapses account for the difference in IFN-γ production by CMV-specific CTL.In the presence of an antibody against CD80, CMV-specific CTL exhibit both lytic activity and IFN-γ production. This was expected, as CMV-specific CTL have an effector-memory phenotype which allows them to become activated without receiving costimulatory signals(5). It is this capacity that makes them ideal effectors for a T cell based immunotherapy for CLL, since CLL is known to express only low levels of costimulatory ligands on its surface (14). Further-more, the effector-memory phenotype suggests that only minute amounts of peptide presented on the CLL cells by means of TC may be sufficient to trigger CMV-specific CTL to lyse the targeted cancer cells. Indeed, our experiments indicate that lysis of target cells is achieved at much lower peptide concentra-tions than required for the induction of cytokine production by CMV-specific

Chapter 5

109

CTL, thereby confirming the observations of Faroudi et al. (12) . Finally, lysis of peptide loaded targets could only be blocked by antibodies against LFA1, demonstrating that CMV-specific CTL require only minimal stimulation to exert powerful lytic activity. Recently, it has been described that CLL cells induce impaired actin polym-erization in T cells, resulting in defective immunological synapse formation with subsequent impairment of T cells function(28). Apparently this is not relevant in our model of redirected virus specific T cells, because these CTL induce efficient lysis of CLL target cells. Our study raises the question whether the reduced induction of cytokine production by TC as compared to the di-rectly CMV peptide loaded CLL target cells, is good or bad news for the in vivo application of TC for anti-tumor therapy. Since it has been suggested that the immune system in CLL patients is skewed towards Th2 (13), the induction of Th1 cytokines may be crucial to support a proper anti-tumor response. On the other hand, it has also been described that one of the most important Th1 cytokines, IFN-γ, may serve as a survival factor for CLL cells (6). More im-portantly, full activation of the total pool of CMVpp65-reactive CTL at once might even result in a cytokine storm as a side-effect (17;33). In this respect, the relative inability of TC to induce the production and release of cytokines may in fact be favorable and broaden their clinical applicability. Whether the latter holds true is still to be tested in upcoming in vivo studies, where the TC will be injected into transgenic mice that express human CD20 on all B cells. These studies should point out to what degree activation of autologous virus-specific cells via these B cell-directed TC indeed induces lysis of the targeted B cells and release of cytokines in vivo. Nevertheless, we expect that redirect-ing viral immune responses in vivo through TC will be a feasible and highly tumor cell-specific immunotherapy.

Acknowledgements

The authors would like to thank Jan van Marle for his contribution in the con-focal fluorescence microscopy imaging.


110

Reference List

1. Aas T, Borresen AL, Geisler S, et al. Specific P53 mutations are associated with de novo resistance to doxorubicin in breast cancer patients. Nat Med. 1996;2:811-814. 2. Abramowicz D, Schandene L, Goldman M, et al. Release of tumor necrosis factor, inter-leukin-2, and gamma-interferon in serum after injection of OKT3 monoclonal antibody in kidney transplant recipients. Transplantation. 1989;47:606-608. 3. Almasri NM, Duque RE, Iturraspe J, et al. Reduced expression of CD20 antigen as a charac-teristic marker for chronic lymphocytic leukemia. Am J Hematol. 1992;40:259-263. 4. Barratt PL, Seymour MT, Stenning SP, et al. DNA markers predicting benefit from adjuvant fluorouracil in patients with colon cancer: a molecular study. Lancet. 2002;360:1381-1391. 5. Bertram EM, Dawicki W, Sedgmen B, et al. A switch in costimulation from CD28 to 4-1BB during primary versus secondary CD8 T cell response to influenza in vivo. J Immunol. 2004;172:981-988.

6. Buschle M, Campana D, Carding SR, et al. Interferon gamma inhibits apoptotic cell death in B cell chronic lymphocytic leukemia. J Exp Med. 1993;177:213-218.

7. Chatenoud L, Ferran C, Reuter A, et al. Systemic reaction to the anti-T-cell monoclonal antibody OKT3 in relation to serum levels of tumor necrosis factor and interferon-gamma [corrected]. N Engl J Med. 1989;320:1420-1421.

8. Chavin KD, Qin L, Yon R, et al. Anti-CD2 mAbs suppress cytotoxic lymphocyte activity by the generation of Th2 suppressor cells and receptor blockade. J Immunol. 1994;152:3729-3739. 9. Choudhuri K, Wiseman D, Brown MH, et al. T-cell receptor triggering is critically depen-dent on the dimensions of its peptide-MHC ligand. Nature. 2005;436:578-582. 10. el Rouby RS, Thomas A, Costin D, et al. p53 gene mutation in B-cell chronic lymphocytic leukemia is associated with drug resistance and is independent of MDR1/MDR3 gene expres-sion. Blood. 1993;82:3452-3459. 11. Espagnolle N, Depoil D, Zaru R, et al. CD2 and TCR synergize for the activation of phospholipase Cgamma1/calcium pathway at the immunological synapse. Int Immunol. 2007;19:239-248. 12. Faroudi M, Utzny C, Salio M, et al. Lytic versus stimulatory synapse in cytotoxic T lym-phocyte/target cell interaction: manifestation of a dual activation threshold. Proc Natl Acad Sci U S A. 2003;100:14145-14150.

Chapter 5

111

13. Fayad L, Keating MJ, Reuben JM, et al. Interleukin-6 and interleukin-10 levels in chron-ic lymphocytic leukemia: correlation with phenotypic characteristics and outcome. Blood. 2001;97:256-263. 14. Freedman AS. Immunobiology of chronic lymphocytic leukemia. Hematol Oncol Clin North Am. 1990;4:405-429. 15. Gorog G, Batory G, Laskay T, et al. Effect of anti-human pan-T monoclonal antibodies on lymphocyte proliferative and cytotoxic functions. Cell Immunol. 1985;96:184-198. 16. Hoogendoorn M, Wolbers JO, Smit WM, et al. Generation of B-cell chronic lymphocytic leukemia (B-CLL)-reactive T-cell lines and clones from HLA class I-matched donors using modified B-CLL cells as stimulators: implications for adoptive immunotherapy. Leukemia. 2004;18:1278-1287. 17. Jahrling PB, Hensley LE, Martinez MJ, et al. Exploring the potential of variola virus in-fection of cynomolgus macaques as a model for human smallpox. Proc Natl Acad Sci U S A. 2004;101:15196-15200. 18. Kater AP, Remmerswaal EB, Nolte MA, et al. Autologous cytomegalovirus-specific T cells as effector cells in immunotherapy of B cell chronic lymphocytic leukaemia. Br J Haematol. 2004;126:512-516. 19. Khanna R, Burrows SR, Suhrbier A, et al. EBV peptide epitope sensitization restores hu-man cytotoxic T cell recognition of Burkitt’s lymphoma cells. Evidence for a critical role for ICAM-2. J Immunol. 1993;150:5154-5162.

20. Leupin O, Zaru R, Laroche T, et al. Exclusion of CD45 from the T-cell receptor signaling area in antigen-stimulated T lymphocytes. Curr Biol. 2000;10:277-280. 21. Li J, Smolyar A, Sunder-Plassmann R, et al. Ligand-induced conformational change with-in the CD2 ectodomain accompanies receptor clustering: implication for molecular lattice formation. J Mol Biol. 1996;263:209-226. 22. Mackus WJ, Frakking FN, Grummels A, et al. Expansion of CMV-specific CD8+CD45RA+. Blood. 2003;102:1057-1063. 23. Mous R, Savage P, Remmerswaal EB, et al. Redirection of CMV-specific CTL towards B-CLL via CD20-targeted HLA/CMV complexes. Leukemia. 2006;20:1096-1102. 24. Muller MR, Tsakou G, Grunebach F, et al. Induction of chronic lymphocytic leukemia (CLL)-specific CD4- and CD8-mediated T-cell responses using RNA-transfected dendritic cells. Blood. 2004;103:1763-1769.


112

25. Oettgen HC, Terhorst C, Cantley LC, et al. Stimulation of the T3-T cell receptor complex induces a membrane-potential-sensitive calcium influx. Cell. 1985;40:583-590. 26. Oscier D, Fegan C, Hillmen P, et al. Guidelines on the diagnosis and management of chronic lymphocytic leukaemia. Br J Haematol. 2004;125:294-317. 27. Picker LJ, Singh MK, Zdraveski Z, et al. Direct demonstration of cytokine synthesis het-erogeneity among human memory/effector T cells by flow cytometry. Blood. 1995;86:1408-1419. 28. Ramsay AG, Lee AM, Gribben JG. Impaired Actin Polymerization Results in Defective Immunological Synapse Formation in T Cells in Chronic Lymphocytic Leukemia. ASH An-nual Meeting. 2007. 29. Schultz J, Lin Y, Sanderson J, et al. A tetravalent single-chain antibody-streptavidin fusion protein for pretargeted lymphoma therapy. Cancer Res. 2000;60:6663-6669. 30. Selvaraj P, Plunkett ML, Dustin M, et al. The T lymphocyte glycoprotein CD2 binds the cell surface ligand LFA-3. Nature. 1987;326:400-403. 31. Shaw S, Luce GE, Quinones R, et al. Two antigen-independent adhesion pathways used by human cytotoxic T-cell clones. Nature. 1986;323:262-264. 32. Stilgenbauer S, Dohner H. Campath-1H-induced complete remission of chronic lympho-cytic leukemia despite p53 gene mutation and resistance to chemotherapy. N Engl J Med. 2002;347:452-453. 33. Suntharalingam G, Perry MR, Ward S, et al. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med. 2006;355:1018-1028. 34. Tolba KA, Bowers WJ, Hilchey SP, et al. Development of herpes simplex virus-1 amplicon-based immunotherapy for chronic lymphocytic leukemia. Blood. 2001;98:287-295. 35. van der Merwe PA, McNamee PN, Davies EA, et al. Topology of the CD2-CD48 cell-adhe-sion molecule complex: implications for antigen recognition by T cells. Curr Biol. 1995;5:74-84. 36. van Kemenade FJ, Kuijpers KC, de Waal-Malefijt R, et al. Skewing to the LFA-3 adhesion pathway by influenza infection of antigen-presenting cells. Eur J Immunol. 1993;23:635-639. 37. Wierda WG, Cantwell MJ, Woods SJ, et al. CD40-ligand (CD154) gene therapy for chronic lymphocytic leukemia. Blood. 2000;96:2917-2924.

Chapter 5

113

38. Zaru R, Cameron TO, Stern LJ, et al. Cutting edge: TCR engagement and triggering in the absence of large-scale molecular segregation at the T cell-APC contact site. J Immunol. 2002;168:4287-4291.


6Granzyme B induced apoptosis is

enhanced by small molecule inhibitors of XIAP

Rogier Mous1, Michael Bots3, Margot Jak1, Mohamed Ahdi1, René AW van Lier2, Eric Eldering2, Marinus HJ van Oers1

and Arnon P Kater1

Departments of 1Hematology and 2Experimental Immunology and 3Laboratory of Experimental Oncology and Radiotherapy,

Academic Medical Center, Amsterdam, the Netherlands

Manuscript in preparation

117

Abstract

Recent studies point out that tumor cells may be protected against cytotoxic T cell attack via increased expression of X-linked inhibitor of apoptosis protein (XIAP). Small molecule inhibitors of XIAP, which have demonstrated their potential as single agents or as enhancers of existing treatment strategies for various types of cancer, may therefore augment immunotherapeutic treat-ment regimens. To test this hypothesis, we monitored the effect of a sublethal dose of these inhibitors on apoptosis induced by cytotoxic T cells (CTL). First, we observed that XIAP inhibitor 1396-11 enhanced DNA fragmentation in-duced by recombinant human Granzyme B in Jurkat cells. However, when this inhibitor was used in cytotoxicity and redirected killing assays using P815 or CLL cells as target cells we did not observe synergy between effector cell-induced target cell death and 1396-11. Therefore, our data do not provide a rationale for using small molecule inhibitors of XIAP in CTL-mediated im-munotherapeutic approaches.

XIAP inhibitors and cytotoxicity

118

Introduction

Chronic lymphocytic leukemia (CLL) is a disease that is characterized by a slow accumulation of mature B lymphocytes, which is considered the result of defects in apoptosis. Both changes within death receptor pathways (such as resistance to Fas-mediated apoptosis(23;31)) as well as mitochondrial path-ways (predominantly via overexpression of Bcl-2(8;16)) have been implied with decreased sensitivity to apoptosis of CLL cells. Over the past few years, modern drug design has produced agents that specifically target the molecules involved, such as Bcl-2 antisense oligonucleotide oblimersen(30) and various BH-3 mimetics(28;29). These agents all act by activating upstream apopto-sis pathways. Nevertheless, actual apoptosis induction can still be prevented downstream by inhibitor of apoptosis proteins (IAPs(21)) that are capable of direct binding to caspase molecules(10). Strikingly, various tumor types have increased expression of IAPs(35) and therefore the threshold for apoptosis induction is raised in these tumor cells. Within the IAP family, especially X-linked inhibitor of apoptosis protein (XIAP) is a crucial player(12) because it directly binds to and thereby prevents activation of caspases-3, -7 and -9(11). It is therefore no surprise that various therapeutic strategies have been de-veloped for inhibiting the effect of XIAP(13;22;27). Some of these strategies, such as small molecule inhibitors of XIAP(32), might soon be applicable in a clinical setting. These small molecule drugs have been designed to target and inhibit the BIR2 domain of XIAP. Binding of these drugs to XIAP results in the dissociation of caspase-3 from XIAP and as a consequence restoration of caspase-3 activity(36). Importantly, small molecule XIAP inhibitors have already been demonstrated to induce tumor cell death(6;7), to augment the effect of various cytotoxic drugs(32) and also the sensitivity to Fas ligation of tumor cells(18). Interestingly, recent reports also suggest benefit of inhibiting XIAP on CTL mediated cell death(14;17). This finding may have clinical rel-evance for CLL because CTL-based strategies are the topic of investigation for the development of novel therapies(20) and also the success of allogeneic stem cell transplantation in CLL depends on host-reactive donor-derived CTL(15). Importantly, from the fact that allogeneic stem cell transplantation is also ef-fective in CLL with 17p-(5), it can be deducted that CTL mediated cell death may be at least partially independent of p53. Given that p53 dysfunction is strongly correlated to therapy resistance in CLL(4), these observations war-rant the development of active immunotherapeutic strategies against therapy-

Chapter 6

119

resistant CLL. In this perspective, it would be interesting to examine the effect of small molecule XIAP inhibitors on CTL attack. In the present study, we found that apoptosis of tumor cells induced by re-combinant granzyme B (in the presence of recombinant perforin) could be enhanced by pre-incubation with a sub-lethal dose of XIAP inhibitor. We subsequently tested the sensitivity of various tumor cell types to apoptosis-induction by small molecule inhibitors of XIAP and finally tested whether the observed synergy between granzyme B and small molecule inhibitors of XIAP would also result in increased tumor cell death upon CLT attack.

Methods

CellsMurine mastocytoma cell line P815, human T cell lymphoma derived J16 cell line (Jurkat) and B cell lymphoma cell line Ramos and Daudi were maintained in Iscove’s Modified Dulbecco’s Medium (IMDM; Gibco) supplemented with 10% v/v Fetal Calf Serum (FCS) and antibiotics. After obtaining informed consent, peripheral blood was drawn from CLL patients. From this material CLL cells were isolated, checked for purity (>90% CD5+/CD19+ cells) via flow cytometry, frozen in 10% DMSO and stored in liquid nitrogen.

Small molecule inhibitors of XIAPSmall molecule inhibitors of XIAP were kindly provided by the Torrey Pines Institute for Molecular Studies, San Diego, CA and have been described previ-ously(32). For the experiments, inhibitors 1396-11 (MW 611.8) and 1540-14 (MW 506.4) were used. As a control, the chemically related but inactive com-pound 1540-20 (MW 515.4) was used.

In vitro expansion of CMV specific cytotoxic T lymphocytes (CTL)PBMC from CMV seropositive CLL patients were used for expansion of CMV-specific CTL as described before(19). Briefly, PBMC (at a final concentration of 5-10x106 cells ml-1) were cultured in IMDM containing 10% v/v Human Pooled Serum (HPS; Cambrex, East Rutherford, NJ), CMVpp65 peptide (1.1 μM) and IL-2 (50 U ml-1 Biotest Ag, Dreieich, Germany). After one week of culture, cells were re-stimulated every 7 days with CMV peptide loaded ir-radiated (30 Gy) Epstein Barr virus (EBV) transformed B cell-lines express-


120

ing either HLA-A2 or HLA-B7 (5x104 cells ml-1) in the presence of IL-2. The percentage of CMV-specific CTL in culture was determined by FACS analysis staining cell culture samples with CD8-PE (Becton Dickinson, San Jose, CA) and APC-conjugated A2 or B7 CMV tetramers (Sanquin, Amsterdam, the Netherlands).

Apoptosis assaysTumor cell lines or freshly thawed CLL cells were incubated in culture medi-um with various concentrations of small molecule inhibitors of XIAP. After 6 or 24 hours, the percentage of apoptotic cells was assessed via flow cytometry upon staining with 3,3’-dihexyloxacarbocyanine iodide (DiOC6(3)) while ne-crotic cells were detected using propidium iodide (Sigma-Aldrich). All cells that were DiOC6(3)bright/propidium iodide- were considered viable cells, and the percentage of apoptotic cells was calculated as follows: apoptotic cells (%) = 100 – viable cells. In some assays, pan-caspase inhibitor zVAD was added to the tumor cells 30 minutes prior to incubation with XIAP inhibitors. In these assays, besides DiOC6(3) staining, cleavage of caspase-3 was assessed via in-tracellular staining (BD Pharmingen, San Jose, CA).

Granzyme B/perforin killing assayJ16 cells were washed 3 times in Hank’s balanced salt solution (HBSS, Invi-trogen, Breda, the Netherlands) supplemented with 0.4% bovine serum al-bumin (BSA) and resuspended in HBSS containing 20mM Hepes buffer (pH 7.2), 2mM CaCl2 and 0.4% BSA. The cells were then pre-incubated with small molecule inhibitors of XIAP for 30 minutes (final concentration 10μM). Next, sub-lethal doses of recombinant human perforin in combination with different concentrations of recombinant human granzyme B (Sigma, Zwijn-drecht, the Netherlands) were added. The cells were then incubated at 37°C for 4 hours. Hereafter, DNA fragmentation was measured by lysing the cells in Nicoletti buffer (0.1% w/v sodium citrate, pH 7.4/0.1% v/v Triton X-100/50μg ml-1 propidium iodide). Via flow cytometry, DNA content was measured in the nuclei that remained after lysis (sub-G0 /G1 signal was considered apop-totic).

Redirected killing assayFreshly isolated PBMC from a healthy donor were stimulated with plate-coat-ed anti-CD3 antibody (16A9T) for 72 hours. These cells were subsequently

Chapter 6

121

used as effector cells in a redirected killing assay. As targets, P815 cells were labeled with CellTrace Far Red DDAO-SE (Molecular Probes, Eugene, OR) to distinguish them from effector cells. The labeled target cells then pre-in-cubated with small molecule inhibitors of XIAP (final concentration 10μM) for 30 minutes. Next, effector cells were added in a 1:1 ratio and finally, CD3 antibody (1x1) was added in various concentrations. After 4 hours, target cell death was measured in a flow cytometry-based assay by DiOC6(3) staining. Gating on DDAO-SE positive cells, cell death was measured by determining the percentage of DiOC6(3)low cells within the target population.

Cytotoxicity assayCTL activity was measured in a flow cytometry-based assay. As targets, freshly thawed CLL cells were used. CLL cells were first labeled with CellTrace Far Red DDAO-SE. After washing, the labeled CLL cells were incubated at 37°C with CMVpp65 peptide for 1 hour. The labeled target cells were subsequently pre-incubated with small molecule inhibitors of XIAP (final concentration 5μM) for 30 minutes. Autologous CMV-specific effector cells (obtained as described above) were incubated with the labeled target cells at a various effector:target ratios. After 4 hours, target cell death was determined by DiOC6(3) staining (gating on DDAO-SE positive cells).

ADCC killing assayJ16 cells (stably transfected with PI-9 or mock expression vector)(2) were la-beled with DDAO-SE and opsonized with CD45 mAb 15D9 (ascites; dilution 1:300). These cells were then incubated in duplicates in culture medium with effector cells (freshly thawed PBMC from healthy donors) at various effector:target (E:T) ratios for 4 h at 37°C, 5% CO2. After 4 hours, target cell death was determined by DiOC6(3) staining (gating on DDAO-SE positive cells). Spe-cific lysis was subsequently calculated as follows: [(% viability of the cells in the control group - % viability of the cells test group)/% viability of the cells in the control group] x 100%.


122

Results

XIAP inhibitors synergize with granzyme B induced apoptosisIt has been described that granzyme B induced apoptosis can be inhibited by overexpression of XIAP(14). To test whether this protective effect can be over-come by treating tumor cells (which frequently overexpress XIAP) with small molecule inhibitors of XIAP, we incubated J16 cells with sub-lethal doses of perforin and various concentrations of granzyme B and evaluated the effect of small molecule inhibitors of XIAP on tumor cell apoptosis (by measur-ing DNA fragmentation). First, we observed that target cell death induced by granzyme B was dose-dependent (figure 1). However, in the presence of XIAP inhibitor 1396-11, target cell death could be achieved at much lower concen-trations of granzyme B compared to conditions in which no XIAP inhibitor or the inactive compound 1540-20 was added. This effect was found to be syner-gistic, because incubation of target cells with the inhibitors in the absence of granzyme and perforin did not affect cell viability.

- PFN 2 1 0.5 0.25 0.125 0.063 0.0310

10

20

30

40

50

60medium1396-111540-20

human GrB (ug/ml)

% D

NA

frag

men

tatio

n

*

*

*p < 0.05

Figure 1. Small molecule inhibitors of XIAP synergize with Granzyme B induced apoptosis. J16 cells were treated with sublethal doses of recombinant human perforin (PFN) in combination with various concentrations of recombinant human granzyme B in the presence of XIAP inhibitors. After 4 hours, the cells were lysed and remaining nuclei were stained with propidium iodide to detect fragmented DNA content via flow cytometry (fragmented DNA appears as a sub-G0 /G1 peak, see methods). The bars rep-resent the mean of three independent experiments, error bars indicate SEM.

Chapter 6

123

XIAP inhibitors induce apoptosis in various cell typesSmall molecule inhibitors of XIAP can induce apoptosis in various tumor cell types(32). Because we wanted to test whether there was a synergistic effect of XIAP inhibitors on cytotoxic T cell (CTL) induced tumor cell death, we as-sessed the effect of XIAP inhibitors on tumor cell death.

P815

0 10 200

25

50

75

100

XIAP inhibitor (µµµµM)

% a

popt

otic

cel

ls

Daudi

0 10 200

25

50

75

100


% a

popt

otic

cel

ls

J16

0 5 100

25

50

75

100


% a

popt

otic

cel

ls

Ramos

0 5 100

25

50

75

100

1396-111540-141540-20


% a

popt

otic

cel

ls

A

CLL

0 5 100

25

50

75

100 1396-111540-141540-20


% a

popt

otic

cel

ls

B DRamos

0 5 100

25

50

75

100 WT 1540-14

WT 1540-20

DN2 1540-14

DN21540-20

DN9 1540-14

DN9 1540-20


% a

popt

otic

cel

ls

DiOC6(3)

nt 1396-11 1396-11 + zVADC

11% 45% 16%

8% 48% 2%

cleaved caspase-3

coun

ts

Figure 2. Small mol-ecule inhibitors of XIAP induce apop-tosis in tumor cell lines and primary CLL cells. Tumor cell lines and PBMC from a CLL patient were incubated with small molecule inhibitors of XIAP for 24 hours. Subsequently, the per-centage of apoptotic cells was assessed via DiOC6(3)/propidium iodide staining. A apoptosis induced by XIAP inhibitors in various cell lines. B the effect of XIAP inhibitors on Ra-mos cells that either express a dominant negative form of cas-pase-2 or caspase-9. C the effect of pre-incubation of Ramos cells with pan-cas-pase inhibitor zVAD (100μM) on apopto-sis induced by XIAP

inhibitor 1396-11 (5μM). D the effect of XIAP inhibitors on PBMC from a CLL patient (95% CD5+/CD19+ cells).


124

We observed that both active components 1396-11 and 1540-14 induced tu-mor cell apoptosis overnight in a dose-dependent fashion in various tumor cell lines (figure 2A), while equivalent concentrations of the inactive com-pound 1540-20 did not induce tumor cell death. Similar concentrations of the inhibitors did not induce any apoptosis within 4 hours of incubation (data not shown); only a toxic effect was observed at very high concentrations (20μM and higher). Furthermore, the effect of the small molecule inhibitors of XIAP appeared to be independent of caspase-2 or caspase-9 activity, since apoptosis of Ramos cells induced by XIAP inhibitors was not affected by overexpres-sion of dominant negative forms of caspase-2 (DN-2) and caspase-9 (DN-9), respectively (figure 2B). Nevertheless, caspase activity (other than caspase-2 and –9) was found to be essential for XIAP inhibitor induced cell death, be-cause pre-incubation of Ramos cells with pan-caspase inhibitor ZVAD-fmk prevented both mitochondrial depolarization and cleavage of caspase-3 upon incubation with XIAP inhibitors (figure 2C). Finally, primary CLL cells also underwent apoptosis when incubated with XIAP inhibitors (figure 2D).

XIAP inhibitors do not synergize with CTL mediated cell deathWe tested whether the observed synergy between small molecule inhibitors of XIAP and granzyme B mediated apoptosis might also enhance CTL mediated tumor cell death in an immunotherapy setting. First, we used polyclonally stimulated T cells from a healthy donor as effector cells in a redirected killing assay (see methods). These effector cells were subsequently redirected towards murine mastocytoma P815 cells in the presence of XIAP inhibitors (final con-centration 5μM). We observed a small additive effect of the treatment with XIAP inhibitors compared to non-treated tumor cells, but no synergy was ob-served (figure 3A). Alternatively, we pre-treated CLL cells with XIAP inhibi-tors (final concentration 5μM), loaded them with CMV immunodominant peptide CMVpp65 and subsequently used these cells as targets for ex vivo ex-panded autologous CMV specific CTL (see methods). Varying both effector:target ratio and CMVpp65 concentration on the target cells, we did not ob-serve a synergy between CTL mediated tumor cell death and XIAP inhibitors but only a mild additive effect (figure 3B). Since we observed that CLL cells have high expression of PI-9 (data not shown), we tested whether this might be responsible for the absence of synergy between XIAP inhibitors and cyto-toxicity. We assessed the effect of PI-9 overexpression on granule-mediated cell death as measured in our assay using PI-9 or mock transduced Jurkat cells

Chapter 6

125

as targets in an ADCC killing assay. As depicted in figure 3C, overexpression of PI-9 did not impair granule-mediated cell death.

102 103 104 105 106 1070

25

50

75

100

pp65 (pM)

% a

popt

otic

targ

et c

ells

-E1:1

6 1:8 1:4 1:2 1:1 2:1 4:10

25

50

75

100

-1396-11 2.5µM1396-11 5µM

E:T ratio

% a

popt

otic

targ

et c

ells

A

B

P815

10-9 10-8 10-7 10-6 10-5 10-4 10-30

25

50

75

100 -1396-111540-141540-20

CD3 mAb dilution

% a

popt

otic

targ

et c

ells

0.625

:11.2

5:1 2.5:1 5:1 10

:120

:140

:1

0

10

20

30mockPI-9

E:T ratio

spec

ific

lysi

s

C

CLL

J16

Figure 3. XIAP inhibitors do not enhance CTL-induced target cell death. Cells were pre-incubated for 30 minutes with small molecule inhibitors of XIAP. Subsequently, these cells were used as targets in a CTL killing assay. A The effect of the addition of XIAP inhibitors in a 4 hour redirected killing assay (see methods). The mean of five independent experiments is shown, error bars indicate SEM. B CLL cells (pre-treated with XIAP inhibitors) were loaded with CMVpp65 peptide and subsequently used as targets for in vitro expanded autologous CMV-specific CTL. In this assay, both effector:target ratio and CMVpp65 peptide concentration were varied. C Jurkat cells transfected with a PI-9 or mock expression vector were used as targets in an ADCC killing assay (see methods). The lines represent the average of two duplicate measurements within one experiment; error bars indicate range.

Discussion

In the current study, we tested whether small molecule inhibitors of XIAP enhanced tumor cell death induced by CTL. We observed that, in a sublethal dose, XIAP inhibitor 1396-11 enhanced DNA fragmentation induced by re-combinant Granzyme B. However, we only observed a mild additive effect of


126

this inhibitor on CTL induced target cell death (assessed via measurement of mitochondrial potential loss).XIAP overexpression by tumor cells may be a potent mechanism to prevent caspase activation(35). This is underlined by the results from our study, which demonstrate that treatment with small molecule inhibitors of XIAP 1396-11 or 1540-14 alone induces caspase-dependent cell death. The latter suggests that tumor cells may use overexpression of XIAP as a final threshold to coun-terbalance the continuous drive to undergo apoptosis(24). Interestingly, the fact that small molecule inhibitors of XIAP have been designed to target the BIR2 domain of XIAP (which is known to selectively bind and block the activ-ity of caspase-3 and –7(34)) implies that these inhibitors target the intersec-tion of the intrinsic and extrinsic apoptosis route. It is therefore not surpris-ing that these components synergize with various kinds of apoptosis-inducing agents(18;32). It has been demonstrated that Granzyme B induced target cell death can be an-tagonized by overexpression of XIAP(14;17). In line with this finding, we here demonstrate that XIAP inhibitor 1396-11 potentates Granzyme B induced DNA fragmentation in target cells, one of the final steps in apoptosis. Never-theless, when used in a CTL killing assay, it did not enhance target cell death, which may be explained in several ways. First of all, due to technical limita-tions we were unable to measure effector caspase activity in the target cells but only assessed target cell death by monitoring the mitochondrial potential loss, a relatively early event in cell death. Alternatively, the synergy between Granzyme B and XIAP inhibitor 1396-11 was established in J16 cells, which express low levels of PI-9, the natural inhibitor of Granzyme B. In contrast, we observed that CLL cells, as well as many other tumor cells(26), express very high levels of PI-9. Therefore, all Granzyme B that enters the CLL cells may be neutralized by PI-9 before exerting its caspase-activating activity. If true, this would also imply that Granzyme B cannot be responsible for CTL induced cell death in CLL. This assumption is supported by our observation that PI-9 overexpressing J16 cells were lysed in a similar fashion as PI-9 low J16 cells (figure 3C). Indeed, there is limited evidence that CTL can induce tar-get cell death independent of Granzyme B in a non-caspase dependent fash-ion(37). The latter is supported by the observation that target cells treated with pan-caspase inhibitor zVAD-FMK are still effectively lysed by CTL(9;25). All above-mentioned observations raise the question what mediator(s) se-creted by CTL other than Granzyme B might then induce lysis of CLL cells.

Chapter 6

127

Since previous studies from our group demonstrated that CMV specific CTL depend on degranulation to exert their lytic effect(19), perforin seems to be a good candidate. Moreover, perforin is capable of inducing target cell death in the absence of Granzyme A and B(37). Alternatively, other members from the Granzyme family like Granzyme K or H may also fulfill this role(1;3) (al-though the latter is mainly found in NK cells(33)). Importantly, all above-mentioned molecules exert their cytolytic function in a caspase-independent fashion, so that overexpression of IAPs will not protect tumor cells from the lytic activity of these molecules. In summary, our study demonstrates that small molecule inhibitors of XIAP enhance Granzyme B induced apoptosis but paradoxically do not induce in-creased target cell death when used in in vitro killing assays. It therefore raises the question whether Granzyme B can be held responsible for killing of CLL cells (which in general have high expression of PI-9) by CMV specific CTL. Nevertheless, our study also confirms the strong cytotoxic potential of these CTL as mediators of active immunotherapy for CLL.


128

Reference List

1. Bailey NC, Kelly CJ. Nephritogenic T cells use granzyme C as a cytotoxic mediator. Eur J Immunol. 1997;27:2302-2309.

2. Bots M, Van Bostelen L, Rademaker MT, et al. Serpins prevent granzyme-induced death in a species-specific manner. Immunol Cell Biol. 2006;84:79-86. 3. Bratke K, Kuepper M, Bade B, et al. Differential expression of human granzymes A, B, and K in natural killer cells and during CD8+ T cell differentiation in peripheral blood. Eur J Im-munol. 2005;35:2608-2616. 4. Byrd JC, Gribben JG, Peterson BL, et al. Select high-risk genetic features predict earlier progression following chemoimmunotherapy with fludarabine and rituximab in chronic lym-phocytic leukemia: justification for risk-adapted therapy. J Clin Oncol. 2006;24:437-443.

5. Caballero D, Garcia-Marco JA, Martino R, et al. Allogeneic transplant with reduced inten-sity conditioning regimens may overcome the poor prognosis of B-cell chronic lymphocytic leukemia with unmutated immunoglobulin variable heavy-chain gene and chromosomal ab-normalities (11q- and 17p-). Clin Cancer Res. 2005;11:7757-7763. 6. Carter BZ, Gronda M, Wang Z, et al. Small-molecule XIAP inhibitors derepress downstream effector caspases and induce apoptosis of acute myeloid leukemia cells. Blood. 2005;105:4043-4050.

7. Cillessen SA, Reed JC, Welsh K, et al. Small-molecule XIAP antagonist restores caspase-9-mediated apoptosis in XIAP-positive diffuse large B-cell lymphoma cells. Blood. 2007. 8. Cimmino A, Calin GA, Fabbri M, et al. miR-15 and miR-16 induce apoptosis by targeting BCL2. Proc Natl Acad Sci U S A. 2005;102:13944-13949. 9. Derby E, Reddy V, Kopp W, et al. Three-color flow cytometric assay for the study of the mechanisms of cell-mediated cytotoxicity. Immunol Lett. 2001;78:35-39. 10. Deveraux QL, Reed JC. IAP family proteins--suppressors of apoptosis. Genes Dev. 1999;13:239-252.

11. Deveraux QL, Roy N, Stennicke HR, et al. IAPs block apoptotic events induced by caspase-8 and cytochrome c by direct inhibition of distinct caspases. EMBO J. 1998;17:2215-2223.

12. Deveraux QL, Takahashi R, Salvesen GS, et al. X-linked IAP is a direct inhibitor of cell-death proteases. Nature. 1997;388:300-304.

Chapter 6

129

13. Fulda S, Wick W, Weller M, et al. Smac agonists sensitize for Apo2L/T. Nat Med. 2002;8:808-815. 14. Goping IS, Barry M, Liston P, et al. Granzyme B-induced apoptosis requires both direct caspase activation and relief of caspase inhibition. Immunity. 2003;18:355-365. 15. Gribben JG, Zahrieh D, Stephans K, et al. Autologous and allogeneic stem cell transplanta-tions for poor-risk chronic lymphocytic leukemia. Blood. 2005;106:4389-4396. 16. Hanada M, Delia D, Aiello A, et al. bcl-2 gene hypomethylation and high-level expression in B-cell chronic lymphocytic leukemia. Blood. 1993;82:1820-1828. 17. Kashkar H, Seeger JM, Hombach A, et al. XIAP targeting sensitizes Hodgkin lymphoma cells for cytolytic T-cell attack. Blood. 2006;108:3434-3440. 18. Kater AP, Dicker F, Mangiola M, et al. Inhibitors of XIAP sensitize CD40-activated chronic lymphocytic leukemia cells to CD95-mediated apoptosis. Blood. 2005;106:1742-1748. 19. Kater AP, Remmerswaal EB, Nolte MA, et al. Autologous cytomegalovirus-specific T cells as effector cells in immunotherapy of B cell chronic lymphocytic leukaemia. Br J Haematol. 2004;126:512-516. 20. Kater AP, van Oers MH, Kipps TJ. Cellular immune therapy for chronic lymphocytic leu-kemia. Blood. 2007;110:2811-2818. 21. LaCasse EC, Baird S, Korneluk RG, et al. The inhibitors of apoptosis (IAPs) and their emerging role in cancer. Oncogene. 1998;17:3247-3259. 22. LaCasse EC, Kandimalla ER, Winocour P, et al. Application of XIAP antisense to cancer and other proliferative disorders: development of AEG35156/ GEM640. Ann N Y Acad Sci. 2005;1058:215-234. 23. Laytragoon-Lewin N, Duhony E, Bai XF, et al. Downregulation of the CD95 receptor and defect CD40-mediated signal transduction in B-chronic lymphocytic leukemia cells. Eur J Haematol. 1998;61:266-271. 24. Lowe SW, Cepero E, Evan G. Intrinsic tumour suppression. Nature. 2004;432:307-315.

25. Malyguine A, Derby E, Brooks A, et al. Study of diverse mechanisms of cell-mediated cytotoxicity in gene-targeted mice using flow cytometric cytotoxicity assay. Immunol Lett. 2002;83:55-59.


130

26. Medema JP, de JJ, Peltenburg LT, et al. Blockade of the granzyme B/perforin pathway through overexpression of the serine protease inhibitor PI-9/SPI-6 constitutes a mechanism for immune escape by tumors. Proc Natl Acad Sci U S A. 2001;98:11515-11520. 27. Naumann U, Bahr O, Wolburg H, et al. Adenoviral expression of XIAP antisense RNA induces apoptosis in glioma cells and suppresses the growth of xenografts in nude mice. Gene Ther. 2007;14:147-161. 28. Oltersdorf T, Elmore SW, Shoemaker AR, et al. An inhibitor of Bcl-2 family proteins in-duces regression of solid tumours. Nature. 2005;435:677-681. 29. Paoluzzi L, Gonen M, Gardner JR, et al. Targeting Bcl-2 family members with the BH3 mimetic AT-101 markedly enhances the therapeutic effects of chemotherapeutic agents in in vitro and in vivo models of B-cell lymphoma. Blood. 2008;111:5350-5358. 30. Raynaud FI, Orr RM, Goddard PM, et al. Pharmacokinetics of G3139, a phosphorothio-ate oligodeoxynucleotide antisense to bcl-2, after intravenous administration or continuous subcutaneous infusion to mice. J Pharmacol Exp Ther. 1997;281:420-427. 31. Roue G, Lancry L, Duquesne F, et al. Upstream mediators of the Fas apoptotic transduc-tion pathway are defective in B-chronic lymphocytic leukemia. Leuk Res. 2001;25:967-980. 32. Schimmer AD, Welsh K, Pinilla C, et al. Small-molecule antagonists of apoptosis suppres-sor XIAP exhibit broad antitumor activity. Cancer Cell. 2004;5:25-35. 33. Sedelies KA, Sayers TJ, Edwards KM, et al. Discordant regulation of granzyme H and granzyme B expression in human lymphocytes. J Biol Chem. 2004;279:26581-26587. 34. Takahashi R, Deveraux Q, Tamm I, et al. A single BIR domain of XIAP sufficient for inhib-iting caspases. J Biol Chem. 1998;273:7787-7790. 35. Tamm I, Kornblau SM, Segall H, et al. Expression and prognostic significance of IAP-fam-ily genes in human cancers and myeloid leukemias. Clin Cancer Res. 2000;6:1796-1803. 36. Wang Z, Cuddy M, Samuel T, et al. Cellular, biochemical, and genetic analysis of mecha-nism of small molecule IAP inhibitors. J Biol Chem. 2004;279:48168-48176. 37. Waterhouse NJ, Sutton VR, Sedelies KA, et al. Cytotoxic T lymphocyte-induced killing in the absence of granzymes A and B is unique and distinct from both apoptosis and perforin-dependent lysis. J Cell Biol. 2006;173:133-144.

Chapter 6

7Summary and discussion

135

Drug-resistance due to p53 dysfunction

Chronic lymphocytic leukemia (CLL) is frequently regarded as an indolent disease that often does not require therapy. Nevertheless, when progression does occur and treatment is required, a considerable subset of all CLL pa-tients has a poor outcome due to frequent relapses and resistance against most chemotherapeutic agents. Drug-resistance has therefore become the most im-portant issue in the treatment of CLL (54). Especially CLL patients with pri-mary or acquired resistance to fludarabine treatment have a marked decreased overall survival. According to a study by Keating et al., those patients have a median survival of ten months (41). The most important genetic abnormality associated with fludarabine resistance is deletion of the short arm of chromo-some 17 (17p-), which contains the p53 locus. Because an intact p53 tumor suppressor pathway is essential for many “classical” chemotherapeutic drugs, CLL patients with 17p- usually respond poorly to chemotherapy (20;82). Un-fortunately, fluorescence in situ hybridization (FISH), which is currently used as standard technique to identify chromosomal abnormalities in CLL seems to underestimate the percentage of patients that will become resistant to che-motherapy (6;8). This might be because not only 17p- but also point muta-tions of p53 (not detectable via FISH) lead to p53 dysfunction and thereby drug resistance (19;90). Therefore, in addition to or even prior to screening for chromosomal abnormalities, it might be useful to assess p53 function in all patients with active disease to identify the patients that should receive p53-in-dependent regimens. Obviously, in order to be applicable in clinical practice, a p53 function test should be easy, reproducible, amenable for standardiza-tion and preferably inexpensive. At present, the most frequently used tech-nique to assess p53 function in tumor samples is Western blotting for p53 and p21 (upregulated upon p53 activation (21)) using lysates of irradiated cells. Recently however, several alternatives to this technique have been presented (4;11;18;50). In this thesis, we demonstrate that measuring transcript levels of the p53-responsive genes Puma, p21 and Bax in a reverse transcriptase PCR-based multiplex ligation probe assay (RT-MLPA) may be a reliable technique to test for p53 function in tumor samples (chapter 3). Using RT-MLPA, we also identified p53 dysfunction in samples from CLL patients without chro-mosomal abnormalities. Our data suggest that upregulation of Puma upon ionizing radiation (IR) treatment is a highly representative parameter for p53 function. In fact, recent observations suggest that Puma is also the most pre-

Summary and discussion

136

dominantly upregulated gene in vivo upon chemotherapeutic treatment of lo-cally advanced breast cancer (53). Therefore, RT-MLPA might not only serve to assess p53 function before onset of therapy but might also be used to moni-tor the effect of drug treatment in vivo in CLL.Since p53 dysfunction seems to be the most important cause of drug resis-tance in CLL, there is a definite need for p53-independent therapeutic agents. At this moment, several of these agents are available. Methylprednisolone for example induces p53-independent cell death (14) and indeed seems to be ef-fective in CLL patients with 17p- (80). Also monoclonal antibodies might act p53-independent. This can be deduced from the fact that tumor cells with mutated p53 can be killed in vitro both via antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC)(44). Notably, ADCC and CDC are thought to be the key mechanisms via which monoclonal antibodies exert their effect in vivo (27;67). For that reason, it is not surprising that treatment regimens containing alemtuzumab (Mabcam-path), a CD52-specific humanized antibody, are effective against CLL with 17p- (47;73). Furthermore, first line treatment with alemtuzumab leads to significantly improved progression-free survival as compared to treatment with chlorambucil (34). Importantly, various clinical studies point out that alemtuzumab also has additive value in the treatment of refractory CLL ei-ther as a single agent (55;65), added to fludarabine treatment (23) or used for consolidation (84). Unfortunately, patients treated with alemtuzumab often experience CMV infections due to immune suppression (46;84), probably be-cause alemtuzumab targets all lymphocytes and therefore will affect adaptive immunity. The mechanism of action of the other monoclonal antibody used in CLL, rituximab (Mabthera), is less clear. Rituximab as a single agent at the usual dosage has only moderate effects against refractory CLL (52). Therefore, in most recent studies it is administered in combination with fludarabine and cyclophosphamide (FCR)(86). The results from a phase II trial point out that FCR treatment results in a longer progression free survival when compared to historical controls (chemotherapy alone)(85). Nevertheless, the response of 17p- CLL to rituximab (in combination with fludarabine) seems to be inferior to that of CLL without 17p- (8). Therefore, this monoclonal antibody may be less effective against chemotherapy-resistant CLL than alemtuzumab.Modern drug development has lead to the discovery of other drugs that in-duce apoptosis of tumor cells independent of p53. Some of these drugs, like

Chapter 7

137

seliciclib (R-roscovitine) or bortezomib have already been tested in clinical trials for CLL (24;61) (seliciclib: phase II trial, results unpublished). Another very interesting group of drugs that might be used for the treatment of p53 dysfunctional CLL are p53 activating small molecules (7;26;36;57;83). It is claimed that these small molecules restore biological activity of inactive and some even of mutated p53, thereby inducing apoptosis in a broad range of tu-mor types. Results from in vivo studies demonstrate that small molecules CP-31398 and PRIMA-1 used as single agent strongly inhibit tumor growth while not affecting healthy tissues (36;77). Although still to be tested in clinical tri-als, these drugs seem to have strong potential for the treatment of tumors in which p53 plays a role in the origin or progression of the disease.The successful application of reduced intensity conditioning regimen in al-logeneic stem cell transplantation has recently brought attention to this ap-proach for the treatment of therapy-resistant CLL. Indeed, long-term remis-sions have been achieved with allogeneic stem cell transplantation in refractory CLL (37;68). The success of this approach is attributed to a graft-versus-leu-kemia effect mediated via donor-derived cytotoxic T lymphocytes (CTL)(31). Importantly, CTL have been demonstrated to lyse target cells in (an at least partially) p53-independent fashion (79). In line with this observation, prom-ising results have been obtained with allogeneic stem cell transplantation in patients with 11q- or 17p- (9). Therefore, CTL mediated therapies may be an alternative towards therapy for refractory CLL. This is discussed in the follow-ing paragraph.

Active immunotherapy for CLL

It has been shown that CLL is accompanied by T cell dysfunction (29;70), which may prevent the formation of efficient CLL-specific T cell responses. Nevertheless, several groups have described successful approaches to over-come T cell tolerance against tumor cells. One approach has been ex vivo transduction of CLL cells with CD40 ligand (CD40L), resulting in increased surface expression of costimulatory molecules on CLL cells. These CD40L-transduced CLL cells could subsequently be used to elicit CLL specific T cell responses in vitro (40). Moreover, CLL-specific T cell clones emerged in pa-tients upon administration of autologous CD40L-transduced CLL cells (87). The latter was accompanied by reduction of tumor mass in these patients,


138

suggesting that infusion of CD40L-transduced CLL cells even broke immu-notolerance against non-transduced tumor cells. Alternative approaches to-wards active immunotherapy for CLL, such as ex vivo generation of allogeneic or autologous CLL-specific CTL using either CLL cells (3;35) or DC pulsed with CLL lysates (43), have demonstrated promising results in vitro but have not been successfully applied in vivo as yet. A possible explanation for the latter might be that all methods described above require either ex vivo ma-nipulation of tumor cells or the use of allogeneic T cells. As an alternative, we therefore studied the possibility of redirecting virus-specific cytotoxic T cells (CTL) towards autologous CLL cells. We hypothesized that cytomegalovirus (CMV)-specific CTL would be proper candidates for this task, because they are unaffected by the general T cell dysfunction in CLL and, importantly, are not heavily dependent on co-stimulation to become activated. Indeed, previ-ous studies from our group confirmed the power of CMV-specific CTL since CLL cells loaded with the CMV immunodominant peptide pp65 were lysed by autologous CMV-specific CTL, even without in vitro pre-stimulation (39). Because the technique described above still requires ex vivo manipulation of tumor cells, we modified this technique by replacing the CMVpp65 peptide by targeted complexes (TC) consisting of HLA-I molecules containing CMVpp65 peptide coupled to a CD20 antibody fragment (chapter 4). We demonstrate that these TC coated CLL cells are lysed by autologous CMV-specific CTL as efficiently as CMVpp65 loaded CLL cells, but that cytokine production by TC stimulated CTL is less than cytokine production triggered by CMVp65 (chap-ter 5). We demonstrate that the latter is probably due to inefficient immu-nological synapse formation between CTL and TC-coated CLL cells. Never-theless, the reduced cytokine production by TC triggered CMV-specific CTL might be an advantage for the future clinical application of the TC because the reduced production of cytokines by CMV-specific CTL will possibly reduce the risk of the induction of a cytokine storm (74) while the targeted tumor cells still are effectively lysed. The results from a clinical trial with the CD20 antibody fragment suggest that TC should have excellent tissue penetration in vivo (25). In this study, even tumor cells in immune-privileged sites were efficiently targeted upon intrave-nous infusion of the CD20 antibody fragment. At present, animal experiments are being performed to see whether this also holds true for the complete CMV peptide-containing TC. Subsequently, additional experiments should also ad-dress whether CTL against CMV can be successfully activated to lyse all tu-

Chapter 7

139

mor cells in vivo. If the concept of redirecting virus-specific T cell responses against tumor cells without ex vivo manipulation indeed proves to be valid, it might be applicable for the treatment of other malignancies as well (13;63;64) because of the flexibility that the dual-component targeting construct offers. In chapter 6, we investigated the possibility to enhance tumor cell death in-duced by CMV-specific CTL. It is known that many tumor types (including CLL) express high levels of X-linked inhibitor of apoptosis protein (XIAP)(76). This may have consequences for active immunotherapy in CLL, since cell death induced by Granzyme B (an important mediator of CTL-induced cell death) can be antagonized by XIAP (28;38). Therefore, we tested whether CTL mediated lysis could be optimized by pre-treating tumor cells with small mol-ecule inhibitors of XIAP. We observed that small molecule XIAP inhibitor 1396-11 indeed synergized with Granzyme B induced DNA fragmentation (which is caspase-mediated) in Jurkat cells but surprisingly did not enhance CTL-induced cell death of CLL cells. In contrast to Jurkat cells, CLL cells ex-press high levels of proteinase inhibitor 9 (PI-9), an inhibitor of Granzyme B. This suggests that Granzyme B may not play an important role in CLL cell death induced by CMV-specific CTL. Alternatively, it also confirms that CTL-induced cell death is at least partially caspase-independent (17;51). It would therefore be interesting to further explore the mechanism behind CTL-medi-ated cell death in CLL cells, thereby possibly identifying new targets for drugs that enhance active immunotherapy in CLL.

Regulatory T cells: target for anticancer therapy?

Cancer patients (including CLL patients) frequently have increased numbers of regulatory T cells (Treg) in the peripheral blood (5;88). The data presented in chapter 2 of this thesis suggest that in CLL, tumor cells might facilitate the formation of Treg in a lymph node environment via CD70 ligation. This is in line with recent observations in non-Hodgkin’s lymphoma, where the tumor cells also seem to play a crucial role in the formation of Treg via CD70 ligation (89). It is tempting to speculate that increased formation of regulatory T cells through CD70 ligation by tumor cells might be a general mechanism behind immunotolerance in B cell malignancies. Although at present it is unclear whether these Treg indeed play a role in the disease, it has been described that eradicating Treg in vivo leads to enhanced tumor cell clearance (30;33;69;71).


140

Moreover, it has been suggested that Treg play an essential role in the initial phase of tumor formation and that eradication of Treg at this stage of the dis-ease even prevents tumorgenesis (22). Strikingly, there are reports stating that active immunotherapy results in increased numbers of tumor-specific Treg which diminish the effect of tumor vaccination (12;92). If Treg indeed play such a crucial role in tumorgenesis and may also hamper active immuno-therapy, it would be very interesting to see what effect eradicating Treg in can-cer patients will have. Because Treg, in contrast to naïve and memory T cells, have high surface expression of CD25 (beta subunit of the IL-2 receptor), this molecule would be a putative target for anti-Treg therapy. In this perspective, one of the agents that seems very interesting is denileukin diftitox or ONTAK (91). This drug is composed of recombinant human IL-2 coupled to diphthe-ria toxin. It has been approved by the FDA for the treatment of cutaneous T cell lymphoma (62), a tumor that has high expression of CD25. However, in experimental therapeutic settings for melanoma and renal cell carcinoma this agent also reduced frequencies of Treg, which was accompanied formation of tumor-specific T cell responses (16;49). Thus, the antitumor effect of denileu-kin diftitox in the treatment of cutaneous T cell lymphoma may partially be explained by its effect on regulatory T cells. A comparable phenomenon may also account for the potent antitumor effect of fludarabine treatment in CLL. A recent study by Beyer et al. showed that fludarabine treatment does not only eradicate tumor cells but also results in reduced frequencies and activity of regulatory T cells (5). Fludarabine probably has this effect on Treg because Treg are highly sensitive to apoptosis induction (75). Studies from our group demonstrate that Treg (in general) have a highly apoptosis-prone phenotype characterized by low expression of anti-apoptotic Bcl-2 and high expression of pro-apoptotic Noxa and high surface expression of Fas (CD95) (chapter 2). Surprisingly, we found an altered apoptotic gene expression profile in Treg of CLL patients which highly resembles the apoptotic gene expression profile of CLL cells (48). Therefore, some new drugs that have been designed to over-come apoptosis resistance in CLL might also be very effective against Treg in CLL, especially the ones targeting cells with high expression of Noxa (like seliciclib (32) and bortezomib (58)) and Bcl-2 (like oblimersen, a Bcl-2 anti-sense oligonucleotide (66)).

Chapter 7

141

Targeting Noxa and Bcl-2: killing two birds with one stone

It is generally assumed that CLL cells do not accumulate in vivo by increased proliferation but due to disturbed apoptosis. This is most likely due to in-creased expression of anti-apoptotic Bcl-2 (42). Strikingly, our data suggest that Treg from CLL patients also have increased expression of Bcl-2 compared to Treg of healthy donors (chapter 2). In contrast to Bcl-2 expression in CLL cells, the increased Bcl-2 expression in Treg from CLL patients is probably not caused by deletions involving Bcl-2-regulating microRNA (10), but more like-ly via other pathways, such as cytokine stimulation (15;78) or T cell receptor triggering (1). Because of its increased expression in both CLL cells and Treg of CLL patients, Bcl-2 seems an attractive target for the development of novel therapeutic strategies for CLL. Various agents that target Bcl-2 have been de-veloped (66;72;81), of which oblimersen (an antisense oligonucleotide target-ing Bcl-2) seems the most promising. This drug is now used in phase II and III trials, either as single agent or in combination with other drugs. Although less effective as a single agent due to severe side effects at therapeutic levels (60), at lower dosages it seems to have additive effect to treatment with classical cyto-toxic drugs (59). Nevertheless, considering the apoptotic gene expression pro-file of CLL (48), oblimersen would be even more effective in combination with drugs that preferably target cells with high expression of pro-apoptotic Noxa. Conveniently, Treg from CLL patients (as well as Treg from healthy individu-als) also express high levels of Noxa. Therefore, it is not surprising that both CLL cells and Treg are very sensitive to apoptosis induction via R-roscovitine (2)(chapter 2), a cyclin-dependent kinase inhibitor that acts via the Noxa/Mcl-1 axis (32). Unfortunately, R-roscovitine is less potent against cells that have increased expression of Bcl-2 (32). In this perspective, it would be very interesting to explore the effect of combined treatment with R-roscovitine and oblimersen in CLL. In addition, combining these therapeutic agents may also be very effective against therapy-resistant CLL because R-roscovitine induces apoptosis in a p53-independent fashion (2) while oblimersen also enhances (radiation-induced) cell death in p53-deficient cells (56).


142

Concluding remarks

From the data discussed above, it is clear that extensive knowledge of tumor biology and drug resistance mechanisms is essential for the development of treatment strategies for CLL patients that are resistant to “classical” chemo-therapy. However, care should be taken when extrapolating in vitro and even in vivo effects (in animals) to the clinic (45;74). Nevertheless, with our current knowledge of tumor biology, it should be possible to design therapeutic strate-gies that target tumor cells in a highly specific way with minimal side effects.

Reference List

1. Alves NL, Derks IA, Berk E, et al. The Noxa/Mcl-1 axis regulates susceptibility to apoptosis under glucose limitation in dividing T cells. Immunity. 2006;24:703-716. 2. Alvi AJ, Austen B, Weston VJ, et al. A novel CDK inhibitor, CYC202 (R-roscovitine), over-comes the defect in p53-dependent apoptosis in B-CLL by down-regulation of genes involved in transcription regulation and survival. Blood. 2005;105:4484-4491. 3. Arditti FD, Aviner S, Dekel B, et al. Eradication of B-CLL by autologous and allogeneic host nonreactive anti-third-party CTLs. Blood. 2005;105:3365-3371. 4. Best OG, Gardiner AC, Majid A, et al. A novel functional assay using etoposide plus nutlin-3a detects and distinguishes between ATM and TP53 mutations in CLL. Leukemia. 2008. 5. Beyer M, Kochanek M, Darabi K, et al. Reduced frequencies and suppressive function of CD4+CD25hi regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood. 2005;106:2018-2025. 6. Buhmann R, Kurzeder C, Rehklau J, et al. CD40L stimulation enhances the ability of con-ventional metaphase cytogenetics to detect chromosome aberrations in B-cell chronic lym-phocytic leukaemia cells. Br J Haematol. 2002;118:968-975. 7. Bykov VJ, Issaeva N, Shilov A, et al. Restoration of the tumor suppressor function to mutant p53 by a low-molecular-weight compound. Nat Med. 2002;8:282-288. 8. Byrd JC, Gribben JG, Peterson BL, et al. Select high-risk genetic features predict earlier progression following chemoimmunotherapy with fludarabine and rituximab in chronic lym-phocytic leukemia: justification for risk-adapted therapy. J Clin Oncol. 2006;24:437-443.

Chapter 7

143

9. Caballero D, Garcia-Marco JA, Martino R, et al. Allogeneic transplant with reduced inten-sity conditioning regimens may overcome the poor prognosis of B-cell chronic lymphocytic leukemia with unmutated immunoglobulin variable heavy-chain gene and chromosomal ab-normalities (11q- and 17p-). Clin Cancer Res. 2005;11:7757-7763. 10. Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A. 2002;99:15524-15529. 11. Carter A, Lin K, Sherrington PD, et al. Detection of p53 dysfunction by flow cytometry in chronic lymphocytic leukaemia. Br J Haematol. 2004;127:425-428. 12. Chakraborty NG, Chattopadhyay S, Mehrotra S, et al. Regulatory T-cell response and tumor vaccine-induced cytotoxic T lymphocytes in human melanoma. Hum Immunol. 2004;65:794-802.

13. Cheung NK, Modak S, Lin Y, et al. Single-chain Fv-streptavidin substantially improved therapeutic index in multistep targeting directed at disialoganglioside GD2. J Nucl Med. 2004;45:867-877. 14. Clarke AR, Purdie CA, Harrison DJ, et al. Thymocyte apoptosis induced by p53-depen-dent and independent pathways. Nature. 1993;362:849-852. 15. Cohen SB, Crawley JB, Kahan MC, et al. Interleukin-10 rescues T cells from apoptotic cell death: association with an upregulation of Bcl-2. Immunology. 1997;92:1-5. 16. Dannull J, Su Z, Rizzieri D, et al. Enhancement of vaccine-mediated antitumor immunity in cancer patients after depletion of regulatory T cells. J Clin Invest. 2005;115:3623-3633. 17. Derby E, Reddy V, Kopp W, et al. Three-color flow cytometric assay for the study of the mechanisms of cell-mediated cytotoxicity. Immunol Lett. 2001;78:35-39. 18. Dicker F, Kater AP, Prada CE, et al. CD154 induces p73 to overcome the resistance to apop-tosis of chronic lymphocytic leukemia cells lacking functional p53. Blood. 2006;108:3450-3457. 19. Dicker F, Herholz H, Schnittger S, et al. Screening for TP53 Mutations Identifies Chronic Lymphocytic Leukemia Patients with Rapid Disease Progression. ASH Annual Meeting Ab-stracts. 2007;110:490. 20. el Rouby S, Thomas A, Costin D, et al. p53 gene mutation in B-cell chronic lymphocytic leukemia is associated with drug resistance and is independent of MDR1/MDR3 gene expres-sion. Blood. 1993;82:3452-3459.


144

21. el-Deiry WS, Tokino T, Velculescu VE, et al. WAF1, a potential mediator of p53 tumor suppression. Cell. 1993;75:817-825. 22. Elpek KG, Lacelle C, Singh NP, et al. CD4+CD25+ T regulatory cells dominate multiple immune evasion mechanisms in early but not late phases of tumor development in a B cell lymphoma model. J Immunol. 2007;178:6840-6848. 23. Elter T, Borchmann P, Schulz H, et al. Fludarabine in combination with alemtuzumab is effective and feasible in patients with relapsed or refractory B-cell chronic lymphocytic leuke-mia: results of a phase II trial. J Clin Oncol. 2005;23:7024-7031. 24. Faderl S, Rai K, Gribben J, et al. Phase II study of single-agent bortezomib for the treat-ment of patients with fludarabine-refractory B-cell chronic lymphocytic leukemia. Cancer. 2006;107:916-924. 25. Forero A, Weiden PL, Vose JM, et al. Phase 1 trial of a novel anti-CD20 fusion protein in pretargeted radioimmunotherapy for B-cell non-Hodgkin lymphoma. Blood. 2004;104:227-236. 26. Foster BA, Coffey HA, Morin MJ, et al. Pharmacological rescue of mutant p53 conforma-tion and function. Science. 1999;286:2507-2510. 27. Golay J, Lazzari M, Facchinetti V, et al. CD20 levels determine the in vitro susceptibility to rituximab and complement of B-cell chronic lymphocytic leukemia: further regulation by CD55 and CD59. Blood. 2001;98:3383-3389. 28. Goping IS, Barry M, Liston P, et al. Granzyme B-induced apoptosis requires both direct caspase activation and relief of caspase inhibition. Immunity. 2003;18:355-365. 29. Gorgun G, Holderried TA, Zahrieh D, et al. Chronic lymphocytic leukemia cells induce changes in gene expression of CD4 and CD8 T cells. J Clin Invest. 2005;115:1797-1805. 30. Grauer OM, Sutmuller RP, van MW, et al. Elimination of regulatory T cells is essential for an effective vaccination with tumor lysate-pulsed dendritic cells in a murine glioma model. Int J Cancer. 2008;122:1794-1802. 31. Gribben JG, Zahrieh D, Stephans K, et al. Autologous and allogeneic stem cell transplanta-tions for poor-risk chronic lymphocytic leukemia. Blood. 2005;106:4389-4396. 32. Hallaert DY, Spijker R, Jak M, et al. Crosstalk among Bcl-2 family members in B-CLL: seliciclib acts via the Mcl-1/Noxa axis and gradual exhaustion of Bcl-2 protection. Cell Death Differ. 2007;14:1958-1967.

Chapter 7

145

33. Heier I, Hofgaard PO, Brandtaeg P, et al. Depletion of CD4+ CD25+ regulatory T cells inhibits local tumour growth in a mouse model of B cell lymphoma. Clin Exp Immunol. 2008;152:381-387. 34. Hillmen P, Skotnicki AB, Robak T, et al. Alemtuzumab compared with chlorambucil as first-line therapy for chronic lymphocytic leukemia. J Clin Oncol. 2007;25:5616-5623. 35. Hoogendoorn M, Wolbers JO, Smit WM, et al. Generation of B-cell chronic lymphocytic leukemia (B-CLL)-reactive T-cell lines and clones from HLA class I-matched donors using modified B-CLL cells as stimulators: implications for adoptive immunotherapy. Leukemia. 2004;18:1278-1287. 36. Issaeva N, Bozko P, Enge M, et al. Small molecule RITA binds to p53, blocks p53-HDM-2 interaction and activates p53 function in tumors. Nat Med. 2004;10:1321-1328. 37. Kahl C, Storer BE, Sandmaier BM, et al. Relapse risk in patients with malignant diseases given allogeneic hematopoietic cell transplantation after nonmyeloablative conditioning. Blood. 2007;110:2744-2748. 38. Kashkar H, Seeger JM, Hombach A, et al. XIAP targeting sensitizes Hodgkin lymphoma cells for cytolytic T-cell attack. Blood. 2006;108:3434-3440. 39. Kater AP, Remmerswaal EB, Nolte MA, et al. Autologous cytomegalovirus-specific T cells as effector cells in immunotherapy of B cell chronic lymphocytic leukaemia. Br J Haematol. 2004;126:512-516. 40. Kato K, Cantwell MJ, Sharma S, et al. Gene transfer of CD40-ligand induces autologous immune recognition of chronic lymphocytic leukemia B cells. J Clin Invest. 1998;101:1133-1141. 41. Keating MJ, O’Brien S, Kontoyiannis D, et al. Results of first salvage therapy for patients refractory to a fludarabine regimen in chronic lymphocytic leukemia. Leuk Lymphoma. 2002;43:1755-1762. 42. Kitada S, Andersen J, Akar S, et al. Expression of apoptosis-regulating proteins in chron-ic lymphocytic leukemia: correlations with In vitro and In vivo chemoresponses. Blood. 1998;91:3379-3389. 43. Kokhaei P, Choudhury A, Mahdian R, et al. Apoptotic tumor cells are superior to tumor cell lysate, and tumor cell RNA in induction of autologous T cell response in B-CLL. Leuke-mia. 2004;18:1810-1815.


146

44. Lapalombella R, Zhao X, Triantafillou G, et al. A novel Raji-Burkitt’s lymphoma model for preclinical and mechanistic evaluation of CD52-targeted immunotherapeutic agents. Clin Cancer Res. 2008;14:569-578. 45. Legrand N, Cupedo T, van Lent AU, et al. Transient accumulation of human mature thymocytes and regulatory T cells with CD28 superagonist in “human immune system” Rag2(-/-)gammac(-/-) mice. Blood. 2006;108:238-245. 46. Lin TS, Flinn IW, Lucas MS, et al. Filgrastim and alemtuzumab (Campath-1H) for refrac-tory chronic lymphocytic leukemia. Leukemia. 2005;19:1207-1210. 47. Lozanski G, Heerema NA, Flinn IW, et al. Alemtuzumab is an effective therapy for chronic lymphocytic leukemia with p53 mutations and deletions. Blood. 2004;103:3278-3281. 48. Mackus WJ, Kater AP, Grummels A, et al. Chronic lymphocytic leukemia cells display p53-dependent drug-induced Puma upregulation. Leukemia. 2005;19:427-434. 49. Mahnke K, Schonfeld K, Fondel S, et al. Depletion of CD4+CD25+ human regulatory T cells in vivo: kinetics of Treg depletion and alterations in immune functions in vivo and in vitro. Int J Cancer. 2007;120:2723-2733. 50. Malcikova J, Smardova J, Pekova S, et al. Identification of somatic hypermutations in the TP53 gene in B-cell chronic lymphocytic leukemia. Mol Immunol. 2008;45:1525-1529. 51. Malyguine A, Derby E, Brooks A, et al. Study of diverse mechanisms of cell-mediated cytotoxicity in gene-targeted mice using flow cytometric cytotoxicity assay. Immunol Lett. 2002;83:55-59. 52. McLaughlin P, Grillo-Lopez AJ, Link BK, et al. Rituximab chimeric anti-CD20 monoclo-nal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol. 1998;16:2825-2833. 53. Middelburg R, de Haas RR, Dekker H, et al. Induction of p53 up-regulated modulator of apoptosis messenger RNA by chemotherapeutic treatment of locally advanced breast cancer. Clin Cancer Res. 2005;11:1863-1869.

54. Montserrat E, Moreno C, Esteve J, et al. How I treat refractory CLL. Blood. 2006;107:1276-1283. 55. Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy is associated with prolonged sur-vival. J Clin Oncol. 2005;23:2971-2979.

Chapter 7

147

56. Mu Z, Hachem P, Pollack A. Antisense Bcl-2 sensitizes prostate cancer cells to radiation. Prostate. 2005;65:331-340. 57. Nahi H, Lehmann S, Mollgard L, et al. Effects of PRIMA-1 on chronic lymphocytic leukae-mia cells with and without hemizygous p53 deletion. Br J Haematol. 2004;127:285-291. 58. Nikiforov MA, Riblett M, Tang WH, et al. Tumor cell-selective regulation of NOXA by c-MYC in response to proteasome inhibition. Proc Natl Acad Sci U S A. 2007;104:19488-19493. 59. O’Brien S, Moore JO, Boyd TE, et al. Randomized phase III trial of fludarabine plus cyclo-phosphamide with or without oblimersen sodium (Bcl-2 antisense) in patients with relapsed or refractory chronic lymphocytic leukemia. J Clin Oncol. 2007;25:1114-1120. 60. O’Brien SM, Cunningham CC, Golenkov AK, et al. Phase I to II multicenter study of oblimersen sodium, a Bcl-2 antisense oligonucleotide, in patients with advanced chronic lym-phocytic leukemia. J Clin Oncol. 2005;23:7697-7702. 61. O’Connor OA, Wright J, Moskowitz C, et al. Phase II clinical experience with the novel proteasome inhibitor bortezomib in patients with indolent non-Hodgkin’s lymphoma and mantle cell lymphoma. J Clin Oncol. 2005;23:676-684. 62. Olsen E, Duvic M, Frankel A, et al. Pivotal phase III trial of two dose levels of denileukin diftitox for the treatment of cutaneous T-cell lymphoma. J Clin Oncol. 2001;19:376-388. 63. Osbourn JK, Field A, Wilton J, et al. Generation of a panel of related human scFv antibod-ies with high affinities for human CEA. Immunotechnology. 1996;2:181-196. 64. Pavlinkova G, Batra SK, Colcher D, et al. Constructs of biotin mimetic peptide with CC49 single-chain Fv designed for tumor pretargeting. Peptides. 2003;24:353-362. 65. Rai KR, Freter CE, Mercier RJ, et al. Alemtuzumab in previously treated chronic lympho-cytic leukemia patients who also had received fludarabine. J Clin Oncol. 2002;20:3891-3897. 66. Raynaud FI, Orr RM, Goddard PM, et al. Pharmacokinetics of G3139, a phosphorothio-ate oligodeoxynucleotide antisense to bcl-2, after intravenous administration or continuous subcutaneous infusion to mice. J Pharmacol Exp Ther. 1997;281:420-427. 67. Reff ME, Carner K, Chambers KS, et al. Depletion of B cells in vivo by a chimeric mouse human monoclonal antibody to CD20. Blood. 1994;83:435-445.

68. Ritgen M, Bottcher S, Stilgenbauer S, et al. Quantitative MRD monitoring identifies dis-tinct GVL response patterns after allogeneic stem cell transplantation for chronic lympho-cytic leukemia: results from the GCLLSG CLL3X trial. Leukemia. 2008.


148

69. Rudge G, Barrett SP, Scott B, et al. Infiltration of a mesothelioma by IFN-gamma-produc-ing cells and tumor rejection after depletion of regulatory T cells. J Immunol. 2007;178:4089-4096. 70. Scrivener S, Kaminski ER, Demaine A, et al. Analysis of the expression of critical activa-tion/interaction markers on peripheral blood T cells in B-cell chronic lymphocytic leukae-mia: evidence of immune dysregulation. Br J Haematol. 2001;112:959-964. 71. Sharma S, Dominguez AL, Lustgarten J. High accumulation of T regulatory cells prevents the activation of immune responses in aged animals. J Immunol. 2006;177:8348-8355. 72. Sinicrope FA, Penington RC, Tang XM. Tumor necrosis factor-related apoptosis-induc-ing ligand-induced apoptosis is inhibited by Bcl-2 but restored by the small molecule Bcl-2 inhibitor, HA 14-1, in human colon cancer cells. Clin Cancer Res. 2004;10:8284-8292. 73. Stilgenbauer S, Dohner H. Campath-1H-induced complete remission of chronic lympho-cytic leukemia despite p53 gene mutation and resistance to chemotherapy. N Engl J Med. 2002;347:452-453. 74. Suntharalingam G, Perry MR, Ward S, et al. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med. 2006;355:1018-1028. 75. Taams LS, Smith J, Rustin MH, et al. Human anergic/suppressive CD4(+)CD25(+) T cells: a highly differentiated and apoptosis-prone population. Eur J Immunol. 2001;31:1122-1131. 76. Tamm I, Kornblau SM, Segall H, et al. Expression and prognostic significance of IAP-fam-ily genes in human cancers and myeloid leukemias. Clin Cancer Res. 2000;6:1796-1803. 77. Tang X, Zhu Y, Han L, et al. CP-31398 restores mutant p53 tumor suppressor function and inhibits UVB-induced skin carcinogenesis in mice. J Clin Invest. 2007;117:3753-3764. 78. Teague TK, Marrack P, Kappler JW, et al. IL-6 rescues resting mouse T cells from apopto-sis. J Immunol. 1997;158:5791-5796. 79. Thiery J, Dorothee G, Haddada H, et al. Potentiation of a tumor cell susceptibility to autol-ogous CTL killing by restoration of wild-type p53 function. J Immunol. 2003;170:5919-5926. 80. Thornton PD, Matutes E, Bosanquet AG, et al. High dose methylprednisolone can induce remissions in CLL patients with p53 abnormalities. Ann Hematol. 2003;82:759-765. 81. Wang JL, Zhang ZJ, Choksi S, et al. Cell permeable Bcl-2 binding peptides: a chemical ap-proach to apoptosis induction in tumor cells. Cancer Res. 2000;60:1498-1502.

Chapter 7

149

82. Wattel E, Preudhomme C, Hecquet B, et al. p53 mutations are associated with resistance to chemotherapy and short survival in hematologic malignancies. Blood. 1994;84:3148-3157. 83. Weinmann L, Wischhusen J, Demma MJ, et al. A novel p53 rescue compound induces p53-dependent growth arrest and sensitises glioma cells to Apo2L/TRAIL-induced apoptosis. Cell Death Differ. 2008;15:718-729. 84. Wendtner CM, Ritgen M, Schweighofer CD, et al. Consolidation with alemtuzumab in patients with chronic lymphocytic leukemia (CLL) in first remission--experience on safety and efficacy within a randomized multicenter phase III trial of the German CLL Study Group (GCLLSG). Leukemia. 2004;18:1093-1101. 85. Wierda W, O’Brien S, Faderl S, et al. A retrospective comparison of three sequential groups of patients with Recurrent/Refractory chronic lymphocytic leukemia treated with fludarabi-ne-based regimens. Cancer. 2006;106:337-345. 86. Wierda W, O’Brien S, Wen S, et al. Chemoimmunotherapy with fludarabine, cyclophos-phamide, and rituximab for relapsed and refractory chronic lymphocytic leukemia. J Clin Oncol. 2005;23:4070-4078. 87. Wierda WG, Cantwell MJ, Woods SJ, et al. CD40-ligand (CD154) gene therapy for chronic lymphocytic leukemia. Blood. 2000;96:2917-2924. 88. Wolf AM, Wolf D, Steurer M, et al. Increase of regulatory T cells in the peripheral blood of cancer patients. Clin Cancer Res. 2003;9:606-612. 89. Yang ZZ, Novak AJ, Ziesmer SC, et al. CD70+ non-Hodgkin lymphoma B cells induce Foxp3 expression and regulatory function in intratumoral CD4+CD25 T cells. Blood. 2007;110:2537-2544. 90. Zenz T, Trbusek M, Smardova J, et al. p53 Inactivation in CLL: Pattern of 110 TP53 Muta-tions. ASH Annual Meeting Abstracts. 2007;110:2064. 91. Zhang L, Waters C, Nichols J, et al. Inhibition of HIV-1 RNA production by the diphtheria toxin-related IL-2 fusion proteins DAB486IL-2 and DAB389IL-2. J Acquir Immune Defic Syndr. 1992;5:1181-1187. 92. Zhou G, Drake CG, Levitsky HI. Amplification of tumor-specific regulatory T cells follow-ing therapeutic cancer vaccines. Blood. 2006;107:628-636.


Samenvatting voor niet-ingewijden

153

Chronische lymfatische leukemie (CLL)CLL is de meest voorkomende leukemie onder oudere mensen in Nederland. De ziekte wordt gekenmerkt door een woekering van kwaadaardige cellen die het meest lijken op B cellen, een bepaalde soort witte bloedcel. Deze cellen hopen zich op in de lymfeklieren, in de milt en in het beenmerg van CLL patiënten en zijn ook in groten getale aanwezig in het bloed (leukemie bete-kent letterlijk “wit bloed”). Tot een bepaald stadium van de ziekte hebben CLL patiënten hier echter meestal geen last van en hoeven zij ook niet behandeld te worden. Bij voortgang van de ziekte kan echter de aanmaak van gezonde witte bloedcellen (afweercellen), rode bloedcellen (nodig voor zuurstoftrans-port) en bloedplaatjes (bloedstolling) ernstig belemmerd worden. Doordat het beenmerg in latere fasen van de ziekte overspoeld wordt met CLL cel-len is het beenmerg dan niet meer in staat tot productie van gezonde bloed-cellen. Hierdoor zijn deze patiënten vatbaarder voor infecties, lijden zij aan bloedarmoede of hebben gemakkelijk bloedingen. Zij moeten in dit sta-dium van de ziekte dan ook behandeld worden. Helaas is er tot op heden geen medicijn waarmee CLL te genezen is. Dit komt doordat CLL cellen allerlei slimme “trucs” hebben die ze beschermen tegen de dood. Hierdoor zijn CLL cellen slecht gevoelig voor chemotherapie. Er is daarom behoefte aan nieuwe vormen van therapie die deze “trucs” weten te omzeilen.

Het afweersysteem van CLL patiëntenHet afweersysteem beschermt ons tegen invloeden van buitenaf en van bin-nenuit die schadelijk zijn voor het lichaam. Dit doet het afweersysteem door onderscheid te maken tussen lichaamseigen en lichaamsvreemde structuren. Als het afweersysteem iets lichaamsvreemds herkent komen vervolgens speci-fieke afweercellen in actie en vallen zij de lichaamsvreemde structuur (zoals bacteriën, virussen, schimmels of parasieten) aan. Naast lichaamsvreemde structuren kan het afweersysteem ook kwaadaardige lichaamseigen cellen (kankercellen) herkennen. Dit kan doordat kankercellen ten opzichte van ge-zonde cellen door fouten in het DNA (mutaties) genetisch veranderd zijn en zich daarom anders gedragen. Dit leidt er aan de ene kant toe dat ze ongelimi-teerd delen en vervolgens door hun ongeremde uitgroei de gezonde cellen verdringen. Aan de andere kant maakt dit de kankercellen ook kwetsbaar, om-dat door de mutaties ook allerlei veiligheidsmechanismen die ingebouwd zijn in iedere lichaamscel actief worden. Deze mechanismen kunnen er toe leiden dat de kankercellen dood gaan doordat er een zelfvernietigingprogramma

Samenvatting

154

(apoptose) in de cellen in gang wordt gezet of doordat het afweersysteem een signaal krijgt om de kankercellen aan te vallen. Hiernaast is het gevolg van de genetische veranderingen dat tumorcellen er aan de buitenkant anders uitzien dan gezonde lichaamscellen, wat ze herkenbaar maakt voor het afweersys-teem. In veel gevallen blijken deze twee mechanismen echter niet (goed genoeg) te werken en kunnen de kankercellen ongestoord doorgroeien. Dit laatste is ook het geval in CLL. Om niet geheel duidelijke reden hebben CLL cellen vaak een erg hoog gehalte aan Bcl-2, een eiwit dat apoptose sterk remt. Hierdoor gaan CLL cellen maar moeilijk dood. Hiernaast hebben CLL cellen soms nog andere mutaties die ze bestand maken tegen apoptose, waarvan de meest be-kende mutaties van p53 zijn. p53 is een eiwit dat een coördinerende rol speelt in apoptose die optreedt als gevolg van beschadigingen aan het DNA. Ook speelt p53 een essentiële rol in het afsterven van kankercellen na bestraling of chemotherapie. Onderzoek wijst uit dat bij patiënten bij wie een p53 mutatie wordt vastgesteld chemotherapie vaak weinig effect heeft. Naast bovengenoemde mutaties hebben CLL cellen ook een aantal “trucs” om het afweersysteem ervan te weerhouden de CLL cellen aan te vallen. CLL cellen zijn de kwaadaardige evenknie van B cellen, witte bloedcellen die ge-specialiseerd zijn in het activeren van andere afweercellen door eiwitten van lichaamsvreemde of potentieel gevaarlijke eiwitten in stukjes te breken en die vervolgens aan andere afweercellen te presenteren. Dit doen zij op een soort presenteerblaadjes die MHC-II heten. Vervolgens kunnen zij deze afweer-cellen instrueren alles wat een vergelijkbare structuur heeft aan te vallen. Om verschillende redenen zijn CLL cellen, in tegenstelling tot B cellen, zeer slecht in het activeren van afweercellen. Zo scheiden CLL cellen bepaalde stofjes uit, zogenaamde cytokines, die het afweersysteem juist tot rust brengen. Ook zijn op het oppervlak van CLL cellen een aantal eiwitten in zeer hoge of sterk verlaagde concentratie aanwezig in vergelijking met normale B cellen. Zo is er een lage concentratie van eiwitten die helpen bij activatie van afweercellen, de zogenaamde costimulatoire moleculen. Dit leidt ertoe dat CLL cellen, zelfs wanneer ze als kwaadaardig worden herkend, niet worden aangevallen door afweercellen omdat die niet voldoende geactiveerd worden in afwezigheid van voldoende costimulatie. Anderzijds is het oppervlak van CLL cellen zeer rijk aan CD200, een eiwit dat afremming van de afweercellen induceert. Tenslotte is het afweersysteem van CLL patiënten (net zoals dat van veel andere kanker-patiënten) rijk aan regulatoire T cellen, een soort ordebewaarders van het af-

Samenvatting

155

weersysteem. Deze cellen zijn een subgroep van T cellen, afweercellen die ge-traind zijn in het herkennen van één specifiek stukje van een eiwit (antigeen) van een virus of een kankercel. In tegenstelling tot andere T cellen gaan regu-latoire T cellen bij het herkennen van hun specifieke antigeen niet over tot de aanval of tot hulp inroepen van andere afweercellen maar geven juist signalen af die het afweersysteem afremmen. In het gezonde afweersysteem bestaat de taak van regulatoire T cellen eruit het lichaam te beschermen tegen (foutieve) afweerreacties tegen goedaardige lichaamscellen en hiernaast uit het gecon-troleerd laten verlopen van afweerreacties tegen virussen (vergelijkbaar met het afremmen van een raceauto met een parachute aan het einde van de race). In CLL zou de grote hoeveelheid regulatoire T cellen echter nadelig kunnen zijn, omdat hierdoor CLL-specifieke afweercellen worden afgeremd waardoor CLL cellen beschermd zijn tegen aanvallen van het afweersysteem.Alle hierboven genoemde veranderingen leiden ertoe dat het afweersysteem van CLL patiënten in een soort slaaptoestand verkeert, waarbij naast het aan-vallen van CLL cellen ook het aanvallen van bacteriën en virussen sterk belem-merd wordt. Uitzondering hierop vormen zeer speciale afweercellen tegen het cytomegalovirus (CMV). Deze zogenaamde cytotoxische T cellen herkennen specifiek lichaamscellen die geïnfecteerd zijn met CMV om deze vervolgens zonder hulp van andere afweercellen en zonder costimulatie te doden. Om-dat het virus de cellen van de geïnfecteerde persoon nodig heeft om zich-zelf te vermenigvuldigen wordt hierdoor de verspreiding van het virus in het lichaam geremd. In tegenstelling tot de meeste virussen wordt CMV echter nadat het het lichaam infecteert niet volledig uitgeroeid maar ontstaat er een soort evenwicht tussen het virus en het afweersysteem. Hierbij houdt het virus zich schuil op bepaalde plaatsen in het lichaam en maakt het lichaam continu cytotoxische T cellen tegen CMV aan om het lichaam te beschermen tegen eventuele wederopstanding (reactivatie) van het virus. Ondanks de sterke afremming van het afweersysteem in CLL patiënten zijn deze afweercellen in staat het virus onder de duim te houden. Verbazingwekkend genoeg zijn deze CMV-specifieke cytotoxische T cellen om onduidelijke reden zelfs in grote getale aanwezig in het bloed van CLL patiënten.

Focus van dit proefschriftIn dit proefschrift werd een aantal van de veranderingen in CLL cellen en het afweersysteem van CLL patiënten nader bestudeerd. Enerzijds werd hierbij onderzoek gedaan naar een tweetal mechanismen , dat mogelijk bescherming

Samenvatting

156

biedt voor CLL cellen, namelijk mutaties van p53 en het nauw gerelateerde p21 en de overmaat aan regulatoire T cellen. Anderzijds werd nader ingegaan op de in groten getale aanwezige cytotoxische T cellen tegen CMV die, zoals blijkt uit hoofdstuk 4 en 5, mogelijk een therapeutische rol in CLL kunnen spelen.

De bevindingen In hoofdstuk 2 werd onderzocht waarom CLL patiënten meer regulatoire T cellen hebben dan gezonde personen. Een belangrijk kenmerk van T cellen in het algemeen is dat ze “opgeleid” zijn om één specifiek antigeen te herkennen. Als een T cel dit antigeen tegenkomt leidt dit ertoe dat hij actief wordt en gaat delen. Wij onderzochten daarom als eerste of er in CLL patiënten aanwijzing is voor een dominant antigeen (bijvoorbeeld een antigeen op CLL cellen) dat de grote hoeveelheid regulatoire T cellen kan verklaren. Hiervoor isoleerden wij regulatoire T cellen uit bloed van CLL patiënten. Wij onderzochten vervolgens het DNA van deze regulatoire T cellen om te kijken of ze tegen één bepaald antigeen gericht waren. Hiernaast onderzochten wij de uiterlijke kenmerken van deze regulatoire T cellen. Beide methoden leverden geen aanwijzingen op dat er een specifiek antigeen betrokken is bij de vorming van regulatoire T cel-len van CLL patiënten. Wel vonden wij aanwijzingen dat CLL cellen afkomstig uit de lymfeklieren van CLL patiënten de aanmaak van regulatoire T cellen zouden kunnen stimuleren omdat zij hoge hoeveelheden van het eiwit CD70 op hun oppervlak hebben. Hiermee kunnen zij het celdelingsproces van re-gulatoire T cellen extra stimuleren. Tenslotte bleek uit verder onderzoek dat regulatoire T cellen bijzonder gevoelig zijn voor celdood. Dit lijkt met name te komen doordat regulatoire T cellen (zowel van gezonde personen als van CLL patiënten) grote hoeveelheden eiwitten bevatten die celdood bevorderen (Noxa en Fas) terwijl Bcl-2, een eiwit dat celdood remt, sterk verlaagd aan-wezig is. In regulatoire T cellen van CLL patiënten bleek de hoeveelheid Bcl-2 echter significant hoger te zijn dan bij gezonde personen, waardoor ze beter beschermd zijn tegen celdood.Samengevat wees onze studie dus uit dat CLL patiënten meer regulatoire T cellen hebben dan gezonde personen omdat ze 1. meer aangemaakt worden in de lymfeklieren door stimulatie van CD70 op CLL cellen aldaar en 2. be-schermd zijn tegen celdood door een hogere hoeveelheid Bcl-2 dan regula-toire T cellen van gezonde personen.

Samenvatting

157

Momenteel is het niet altijd even makkelijk om in te schatten hoe goed CLL patiënten gaan reageren op chemotherapie. Met chromosoomonderzoek lukt het om een groot deel van de patiënten aan te wijzen die ongevoelig worden voor chemotherapie. Bij een flink aantal patiënten blijkt dit echter niet goed te voorspellen omdat niet alle mutaties op te sporen zijn met chromosoomonder-zoek. In hoofdstuk 3 onderzochten wij daarom of een nieuwe techniek ons kan helpen CLL patiënten op te sporen die niet goed te behandelen zijn met chemotherapie. Wij gebruikten in dit hoofdstuk een test (MLPA) waarmee kan worden gekeken hoe actief bepaalde genen in een cel zijn. Wij ontdekten met behulp van deze MLPA dat in CLL cellen die blootgesteld worden aan radioactieve straling en hier vervolgens aan sterven bepaalde genen erg actief werden. Deze genen heten Puma, p21 en Bax. Deze genen werden echter niet actief bij CLL patiënten die mutaties hadden waardoor p53 niet werkt. De test bleek hierin zelfs veel gevoeliger dan chromosoomonderzoek, omdat zelfs een aantal patiënten werd gevonden zonder afwijkingen bij chromosoomonder-zoek maar met duidelijk verminderde gen-activiteit na bestraling. Dit is een belangrijke bevinding, omdat uit recente studies is gebleken dat patiënten met een niet werkend p53 slecht reageren op de meest gangbare chemotherapie voor CLL. Daarom zou de MLPA net als chromosoomonderzoek dus als test kunnen dienen om slecht behandelbare CLL patiënten te identificeren.

In hoofdstuk 4 en 5 wordt een techniek geïntroduceerd die in de toekomst mogelijk kan dienen om CLL patiënten te behandelen. Deze techniek maakt gebruik van een tweedelig eiwit. Het ene deel is een antilichaam dat gericht is tegen een eiwit dat op CLL cellen maar ook op gezonde B cellen voorkomt (CD20). Het uiteinde van dit antilichaam past als een sleutel in een slot op het tweede deel van het eiwit, een “presenteerblaadje” dat MHC-I heet waarop een deel van het CMV virus wordt gepresenteerd. Dit MHC-I dient er bij ge-zonde cellen onder andere voor het afweersysteem te waarschuwen dat de cel geïnfecteerd is met een virus door delen van dit virus op de buitenkant van de cel te presenteren op MHC-I moleculen. Dit geeft vervolgens cytotoxische T cellen die dit virus-deeltje herkennen de kans de cel te doden door giftige stoffen uit te scheiden. Door nu CLL cellen te “bedekken” met een jasje van dit tweedelige eiwit lijkt het voor CMV specifieke cytotoxische T cellen alsof die CLL cellen geïnfecteerd zijn en gaan zij tot de aanval over. Hierbij doden zij de CLL cellen en scheiden hiernaast cytokines uit waarmee andere delen van het afweersysteem kunnen worden geactiveerd. Dit laatste lijkt echter deels be-

Samenvatting

158

lemmerd te worden doordat het contact tussen CLL cellen en de cytotoxische T cellen niet optimaal verloopt. Hierdoor wordt CD2, een eiwit op de cyto-toxische T cel, niet afdoende gestimuleerd. Dit kan verholpen worden door een antilichaam tegen CD2 toe te voegen, waardoor wel de maximale activatie van de cytotoxische T cellen wordt bereikt. Op korte termijn zullen proefdierexperimenten worden uitgevoerd met het tweedelige eiwit. Deze experimenten zullen meer duidelijkheid verschaffen over de eventuele klinische toepasbaarheid van de hierboven beschreven methode.

In hoofdstuk 6 tenslotte werd onderzocht of het mogelijk is het doden van CLL cellen door CMV specifieke cytotoxische T cellen te verbeteren door de CLL cellen gelijktijdig te behandelen met een (toekomstig) medicijn. Dit medicijn is een remmer van XIAP, een eiwit dat celdood remt. Uit onze proeven blijkt dat deze XIAP remmer een samenwerking (synergie) kan aangaan met een van de belangrijke stoffen die uitgescheiden worden door de cytotoxische T cel, granzyme B. Toch heeft toevoeging van de XIAP remmer niet tot gevolg dat CLL cellen die aangevallen worden door cytotoxische T cellen beter ge-dood worden. Dit laatste suggereert dat het niet zinvol is CLL patiënten die in de toekomst behandeld zullen worden met het tweedelige eiwit (hoofdstuk 4 en 5) tevens te behandelen met de XIAP remmer.

ConclusieDit proefschrift laat zien dat aan de veranderingen zoals die worden waargenomen in het afweersysteem van CLL patiënten zowel negatieve als positieve aspecten zitten. Enerzijds zorgen deze veranderingen ervoor dat CLL cellen beschermd zijn tegen het afweersysteem en chemotherapie. Anderzijds levert bestudering van deze veranderingen belangrijke aanknopingspunten op voor het ontwikkelen van nieuwe behandelstrategieën voor CLL. Mogelijk dat hierin met name CMV specifieke cytotoxische T cellen een belangrijke rol kunnen spelen.

Samenvatting

Dankwoord

163

Mijn ervaringen tijdens de afgelopen jaren promotieonderzoek kan ik mis-schien het beste vergelijken met het rijden van de Tour de France: een opeen-volging van beproevingen, successen maar ook tegenvallende resultaten en na de slotetappe een voldaan gevoel dat de finish heelhuids bereikt is. Ik heb mij in deze periode bijzonder thuis gevoeld binnen het “team Hematologie/Ex-perimentele Immunologie” en wil daarom van de gelegenheid gebruik maken de mensen uit en rond dit team bedanken.

Als eerste de ploegleiders. Rien, ondanks jouw drukke agenda en de in-drukwekkende stapeltjes-collectie op je bureau was je altijd makkelijk benaderbaar en leerde je mij goed zaken te plannen door het overzicht te houden. Hiernaast wil ik je bedanken voor je steun tijdens het slopende sollicitatietraject naar een opleidingsplaats tot internist. René, ook de drempel naar jouw kantoor was (wellicht onbedoeld) altijd laag. Je leerde me doelgericht onderzoek te doen en daarna de resultaten kritisch te be-kijken. Ik denk dat de woorden “Vijf figuren is een artikel, vier figuren is een Nature-artikel” genoeg zeggen. Eric, ondanks onze strubbelingen half-koers is na de reorganisatie onze samenwerking steeds beter geworden. Jouw stempel op dit proefschrift is duidelijk te zien in een aantal hoofdstuk-ken. Ik ben dan ook blij dat je co-promotor op mijn proefschrift wilt zijn.

De teamleden: Margot, zonder jouw hulp was het proefschrift nooit afgeko-men. Hopelijk wordt het CMV immunotherapie-concept ook in vivo een groot succes! Arnon, onze samenwerking (soms over lange afstand) heeft tot een aantal mooie proeven geleid. Delfine, onze eerste dag in het lab samen en de spraakverwarring over de trekkast/zuurkast zal ik niet snel vergeten. Succes de 15e! De mechaniciens: Annelieke, René en Ester: zoals jullie kunnen zien is een aantal prachtexperimenten in dit proefschrift van jullie hand. Dank voor jullie hulp en gezelligheid. Ingrid, Dieuwertje, en SiLa: dank voor jullie geduld en het in goede banen leiden van mijn (vaak op het laatste moment “geplande”) proeven. Jan, Peter, Berend en Marc: jullie apparaatkennis heeft tot een aantal mooie plaatjes en resultaten geleid.Tenslotte mijn studenten, Mohamed en Adriana, die ieder een flinke aanzet voor twee van de hoofdstukken in dit proefschrift hebben gegeven: dank jullie wel/muchas gracias!

164

Ook speciale aandacht voor twee niet-teamleden. Philip, thank you for intro-ducing the CD20-HLA targeted complexes into our lab. Hopefully, the col-laboration between our groups will lead to many successes in the near future. Michael, dank voor de XIAP experimenten die je voor deze “domme dokter” hebt uitgevoerd.

Ik ben tijdens mijn promotie-onderzoek in de luxepositie geweest twee thuis-bases te hebben. Dit had als voordeel dat ik altijd veel aanspraak en gezelligheid had (en op 2 plaatsen verjaardagstractaties!). Daarom aan alle collega’s van de laboratoria Speciële Hematologie en Experimentele Immunologie/SKIL: dank jullie wel! In het bijzonder mijn kamergenoten: Marc, Dick, Kirsten, Bernard, Tania, Paul, Cristina, Laurian, Jaap, Felix en Ronald. Clandestiene virusexpe-rimenten, de kamer-opruimactie op G1-144 (“er wordt niets weggegooid!”), altijd een reserve-voorraad dorito’s, steeds wisselende desktopafbeeldingen en screensavers, mega-sterke koffie: need I say more? Alek: thanks for introdu-cing the Polish vodka parties to our lab. Na zdrowie!

In juni 2007 maakte ik een transfer naar “team RKZ”. Toen ik na mijn eerste werkdag ontredderd thuiskwam had ik het gevoel totaal de verkeerde keuze te hebben gemaakt. Gelukkig ben ik daar snel anders over gaan denken. Ik vermoed dat vooral de laagdrempeligheid, de vriendelijke sfeer, het fietsen naar Beverwijk en de fantastische collega’s daar een belangrijke rol in hebben gespeeld. Ook de nachtdiensten (zingende patiënten, filmhuis B4, politie en dronken mensen op de eerste hulp) maken deze periode tot een bijzondere ervaring. Niek, Gerrit, Ron en Emile: bedankt voor jullie hulp en steun bij het verwerven van de opleidingsplaats in het AMC. Roostermakers, dank jul-lie wel voor het inwilligen van mijn onmogelijke roosterwensen rondom de kerstdagen!

Bas, het (wedstrijd)mountainbiken (die wedstrijd was toch op zondag?) en het gebeuren eromheen (NK Spaarnwoude, Houffalize, Decathlon) heeft lang-zaam plaatsgemaakt voor wijn, Korsakoff en snowboarden. Onze vriendschap is gelukkig onveranderd. Michiel en Volker (Duitser): thanks for great nights out in the Winston, carnival in Cologne, beach/nightlife in Gran Canaria and many squash/sauna evenings. Brian, le Roc is inmiddels ieder jaar een vast agendapunt. Op naar 10 oktober! Vivianda, Jantien en Rachèl: sorry voor alle gemiste feestjes. Gelukkig heb ik goede herinneringen aan de (te schaarse) wel-

165

gelukte afspraken (The march of the penguins, Parade). Ik beloof beterschap! Francesca and Edo: thanks for many nice evenings and great Italian food and wine! Os amigos portugueses: Nuno (Vizinho/Baguette), Nuno (Santarém) e Tai, Hugo, Cristiana e Massi (semi-português), Tiago e Luciana, João (Povoas), Ana, Catarina, Ana Catarina: muito obrigado pela vossa hospitalidade e in-trodução à cultura e aos hábitos portugueses (bacalhau, rissóis, bagaço). A minha família portuguesa: Espírito Santo, fico muito contente que possas es-tar presente na defesa da minha tese. Muito obrigado por fazeres de Braga a minha segunda cidade natal. Lila e Nuno: vocês são os melhores cunhados que eu poderia desejar. Zusje: als kinderen uit een artsloze familie allebei dokter worden, hoe is het mogelijk? Hopelijk zullen we nog vele natrium/kalium vs. VT/VO duels uit-vechten. Papa en mama, ik herinner me de regenachtige dag in de zomer van 1995 waarop ik mij ging inschrijven aan de Anna’s Hoeve nog als de dag van gisteren. Het mag een wonder heten dat ik ondanks dit droevige tafereel onderzoek doen toch nog leuk ben gaan vinden. Dank jullie wel voor jullie onvoorwaardelijke steun en een altijd open deur.

Fofinha, obrigado pelo teu apoio e compreensão durante este período difícil. Sabes que eu farei o mesmo por ti. Eu amo-te.

List of publications

Mous R*, Jak M*, Remmerswaal EBM, Spijker R, Jaspers A, Yagüe A, Eldering E, van Lier RAW, van Oers MHJ. Enhanced survival and increased formation of regula-tory T cells in CLL. Submitted

Mous R, Jaspers A, van Oers MHJ, Kater AP, Eldering E. Detection of p53 dysfunction in chronic lymphocytic leukemia cells via multiplex quantification of p53 target gene induction. Submitted

Mous R, Savage P, Eldering E, Teeling P, van Oers MHJ, van Lier RAW. Adequate synapse formation between leukemic B cells and effector T cells following stimulation with artificial TCR ligands. Leukemia & Lymphoma (in press)

Mous R, Savage P, Remmerswaal EBM, van Lier RAW, Eldering E, van Oers MHJ. Redirection of CMV-specific CTL towards B-CLL via CD20-targeted HLA/CMV complexes. Leukemia 2006 Jun;20(6):1096-102

van Wetering S, van den Berk N, van Buul JD, Mul FP, Lommerse I, Mous R, ten Klooster JP, Zwaginga JJ, Hordijk PL. VCAM-1-mediated Rac signaling controls endothelial cell-cell contacts and leukocyte transmigration. Am J Physiol Cell Physiol 2003 Aug;285(2):C343-C352

*equal contribution of both authors

Curriculum vitae

De schrijver van dit proefschrift werd geboren in Amsterdam op 21 novem-ber 1977. Na het behalen van het Gymnasium diploma aan het Alberdingk Thijm College te Hilversum begon hij in 1995 aan de opleiding Medische Biologie aan de faculteit der Natuurwetenschappen, Wiskunde en Informatica van de Universiteit van Amsterdam. In 1997 startte hij hiernaast met de studie Geneeskunde aan de Universiteit van Amsterdam. In het kader van beide studies liep hij achtereenvolgens stage aan de afdeling Allergie van Sanquin te Amsterdam onder begeleiding van Prof. Dr. R.C. Aalberse (The association between IgG to D. pteronyssinus at age one and allergy at age five) en de afde-ling Experimentele ImmunoHematologie van Sanquin te Amsterdam onder begeleiding van Dr. S. van Wetering en Dr. P.L. Hordijk (The role of ICAM-1 (CD54) and VCAM-1 (CD106) crosslinking on the function of VE-cadherin in primary endothelial cells). In 2001 behaalde hij zijn doctoraal diploma van de opleiding Medische Biologie (afstudeerrichting Immunologie); kort hierna (in 2002) behaalde hij tevens zijn doctoraal diploma van de opleiding Geneeskunde. In aansluiting hierop begon hij met zijn co-schappen, welke hij in 2004 afrondde door het artsdiploma te behalen. Hij werd in 2004 aangesteld als arts-onderzoeker aan de faculteit Geneeskunde van de Universiteit van Amsterdam. Aldaar verrichte hij, verbonden aan de afdeling Hematologie, onderzoek naar het immuunsysteem van CLL patiënten onder supervisie van Prof Dr. M.H.J van Oers (Hematologie) en Prof. Dr. R.A.W. van Lier (Experimentele Immunologie). De resultaten van het promotieonderzoek staan beschreven in dit proefschrift. Momenteel is hij werkzaam als arts-assistent interne geneeskunde in het Rode Kruis Ziekenhuis te Beverwijk (vanaf april 2008 in het kader van de opleiding tot internist aan het Academisch Medisch Centrum te Amsterdam).

uva-dare (digital academic repository) studies on the ... · uva-dare is a service provided by the...

Documents