Interleukin 21 in Cancer and Immunotherapy

Interleukin 21 in cancer and immunotherapy

PH.D. THESIS

Henrik Søndergaard, M.Sc.

2009

Faculty of Science Novo Nordisk A/S University of Copenhagen Denmark Denmark

Interleukin 21 in cancer and immunotherapy Ph.D. Thesis 2009 © Henrik Søndergaard

ISBN 978-87-7611-388-4 Printed by SL grafik, Frederiksberg C, Denmark (www.slgrafik.dk)

Contents

Contents ...... 1

Preface and acknowledgements ...... 3

Abbreviations ...... 4

Summary ...... 5

Sammendrag (Danish summary) ...... 7

Chapter 1 – Introduction and objectives ...... 9

Specific objectives in part 1 ...... 10 Specific objectives in part 2 ...... 10 Chapter 2 – Background ...... 11

Cancer and immunotherapy ...... 11 The mouse as experimental system ...... 14 Chapter 3 – Manuscripts ...... 17

Paper I: ...... 19 Interleukin 21: roles in immunopathology and cancer therapy (Review)

Paper II: ...... 35 Interleukin 21 therapy increases the density of tumor infiltrating CD8+ T cells and inhibits syngeneic tumor growth

Paper III: ...... 49 Intratumoral interleukin 21 increases anti-tumor immunity, tumor-infiltrating CD8+ T cell density and activity, and enlarges draining lymph nodes

Paper IV: ...... 77 Endogenous interleukin 21 restricts CD8+ T cell expansion and is not required for tumor immunity

Chapter 4 – Discussion ...... 91

Chapter 5 – Conclusion ...... 99

Chapter 6 – Future perspectives ...... 101

References ...... 103

Front page illustration depicts IL-21 protein structure and was provided by Kent Bondensgaard, Department of Protein Structure and Biophysics, Novo Nordisk A/S, Denmark.

Preface and acknowledgements

The work in this thesis has been funded by an Industrial Ph.D. fellowship from the Danish Ministry of Science, Technology and Innovation in collaboration with Novo Nordisk A/S (NN). The experimental work was performed at NN facilities in Måløv, Denmark, in departments of Cancer Pharmacology, Histology, and Immunopharmacology from September 2006 to September 2009. During this period, 6 month research was performed in the department of Cellular Immunology at the Peter MacCallum Cancer Centre, Melbourne, Australia.

I am deeply indebted to my former and present supervisors Michael Kragh and Kresten Skak who made this project possible and guided me into the world of science; Michael for his invaluable help to kick-start this project with tremendous dedication, and Kresten for happily taking over as main supervisor with his endless source of positive energy and exceptional support throughout this entire project, I particularly enjoyed our many runs full of discussion until our breaths ran out. I am very thankful to Niels Ødum for accepting the task as university supervisor, safely guiding me through the university maze, and to Per Thor Straten for his commitment as co-supervisor and for our many fruitful discussions. I sincerely acknowledge the collaboration with my great colleagues at NN, especially, Elisabeth Douglas Galsgaard and Birte Jørgensen for introducing me to immunohistochemistry with their catching enthusiasm, Klaus Steensgaard Frederiksen for sharing his quantitative PCR expertise, Peter Thygesen for his great help with pharmacokinetics, Heidi Winther and Ken Heding who were always ready with helping hands, and to everyone in departments 903 and 479 for creating a great, fun and inspiring work atmosphere. A deep gratitude goes to Mark J. Smyth for giving me the opportunity to visit his laboratory at the Peter MacCallum Cancer Centre. I sincerely wish to thank Adam Uldrich for sharing his brilliant mind and for teaching me all I know about NKT cells and fishing in Port Philip Bay. A special thanks to Nicole Mclaughlin for her tireless help with experiments, delivered with the same enthusiasm as when we were mountain biking. John Stagg, thank you for your great friendship and for teaching me that surfing joyfully substitutes for science. And, thanks heaps to all the other amazing people in the cancer immunology program who made my stay down under unforgettable. Finally, I wish to thank my family and friends for their great mental support and for always reminding me of what is important, especially Kristina, for your love and support, and for enduring our long periods of separation. Måløv, Denmark, September 2009

Henrik Søndergaard

Abbreviations

αGC Alpha-galactosylceramide MM Metastatic melanoma ACT Adoptive cell transfer MMP Matrix metalloproteinase Ag Antigen MOG Myelin oligodendrocyte ADCC Antibody-dependent ceullular glycoprotein cytotoxicity MS Multiple sclerosis AOI Area of interest NK Natural killer cell BCR B cell receptor NKT Natural killer T cell CCL C-C motif ligand NOD Non-obese diabetic CD Crohn’s disease OVA Ovalbumin CIA Collagen-induced arthritis PBL Peripheral blood lymphocytes CTL CD8+ cytotoxic T cells PBMCs Peripheral blood mononuclear CTLA-4 Cytotoxic T lymphocyte antigen 4 cells CXCL Chemokine CXC motif ligand PD-L1 Programmed death receptor DC Dendritic cell ligand-1 EAE Experimental autoimmune PLP Proteolipid protein encephalomyelitis PTI Post tumour injection FasL Fas ligand RA Rheumatoid arthritis

γc Common IL-2 receptor γ chain RCC Renal cell carcinoma GC Germinal center rDC Regulatory dendritic cells IBD Inflammatory Bowel disease RIP Rat insulin promoter IBF IRF-4-binding protein S.c. Subcutaneous iDC Immature dendritic cells SLE Systemic lupus erythematosus IDO Indoleamine 2,3-dioxygenase T1D Type 1 diabetes IFN Interferon TCR T cell receptor IL Interleukin TGFβ Transforming growth factor β.

IL-21 Interleukin 21 Tfh Follicular helper T cells

IL-21R Interleukin 21 receptor Th Helper T cells

I.p. Intraperitoneal Th17 IL-17-producing helper T cells I.t. Intratumoral TILs Tumor infiltrating lymphocytes LN Lymph node TRAIL Tumour necrosis factor–related LPM Lamina propria monocytes apoptosis-inducing ligand

MDSC Myeloid-derived suppressor cells Tregs Regulatory T cells mAb Monoclonal antibody UC Ulcerative colitis mIL-21 Murine IL-21 WT Wild type MCA 3’-methylcholanthrene

Summary Interleukin (IL)-21 is a recently discovered cytokine with pleiotropic immunomodulatory effects and putative anti-tumor activity. This Ph.D. thesis examines the functions of IL-21 protein as cancer immunotherapy and the role of endogenous IL-21 in tumor immunity in preclinical mouse models. In Chapter 1, the specific objectives for the experimental work are introduced. Chapter 2 presents the theoretical background to cancer immunology and immunotherapy, including specific barriers to immunotherapy and current therapeutic advances. This is followed by a brief review of the mouse as a model organism for drug testing in cancer and immunology with specific considerations for IL-21. Chapter 3 contains four original manuscripts; a review manuscript introduces IL-21 and IL-21 receptor (IL-21R) immunobiology and reviews the current knowledge concerning IL-21 in cancer therapy and immunopathology, and the following 3 manuscripts present the experimental work of this thesis:

Paper I: Søndergaard H. and Skak K. Interleukin 21: roles in immunopathology and cancer therapy. Tissue Antigens. 2009 Oct. 21, (Epub ahead of print)

Paper II: Søndergaard H., Frederiksen K.S., Thygesen P.,Galsgaard E.D., Skak K. Kristjansen P.E.G. and Kragh M. Interleukin 21 therapy increases the density of tumor infiltrating CD8+ T cells and inhibits syngeneic tumor growth. Cancer Immunol. Immunother. 2007, Sep;56(9):1417-28. Epub. 2007 Feb. 7

Paper III: Søndergaard H., Galsgaard E.D., Bartholomæussen M., Ødum N. and Skak K. Intratumoral interleukin 21 increases anti-tumor immunity, tumor-infiltrating CD8+ T cell density and activity, and enlarges draining lymph nodes. J. Immunother. in press

Paper IV: Søndergaard H., Coquet J.M., Uldrich A.P., McLaughlin N, Godfrey D.I., Sivakumar P.V. Skak K. and Smyth M.J. Endogenous interleukin 21 restricts CD8+ T cell expansion and is not required for tumor immunity. J. Immunol. 2009 Dec 1;183(11):7326-36. Epub 2009 Nov 13

Paper II and III focus on the anti-tumor effect of IL-21 protein therapy following intraperitoneal, subcutaneous and intratumoral administration in two preclinical mouse cancer models - B16 melanoma and RenCa renal cell carcinoma. Herein, the responsible effector cells for IL-21 anti-tumor activity are determined and the effects of IL-21 are evaluated on the density and activity of tumor infiltrating T cells and on tumor draining lymph nodes. Paper IV

investigates the role of endogenous IL-21 in immunosurveillance, and primary and secondary tumor immunity using various experimental tumor models in IL-21- and IL-21R-deficient mice with focus on NK, NKT and CD8+ T cell responses. The results obtained in Paper II-IV are discussed in chapter 4. Chapter 5 summarizes the main conclusions obtained in this thesis and chapter 6 outlines the perspectives for future research concerning IL-21 in cancer immunotherapy. A list of references is given at the end of the thesis (excluding those in Paper I-IV).

Sammendrag (Danish summary)

Interleukin (IL)-21 er et fornyeligt opdaget cytokin med omfattende immunmodulerende effekter og med formodet anti-tumor aktivitet. Denne Ph.d. afhandling undersøger funktionen af IL-21 protein som cancer immunterapi og rollen af endogent IL-21 i tumorimmunitet ved hjælp af prækliniske musemodeller. I kapitel 1 introduceres de specifikke mål for det eksperimentelle arbejde. I kapitel 2 gives en introduktion til cancerimmunologi og immunterapi med fokus på specifikke barrierer overfor immunterapi og nuværende terapeutiske fremskridt. Dette efterfølges af et kort overblik over musen som modelorganisme for afprøvning af lægemidler indenfor cancer og immunologi med særlige betragtninger for IL-21. Kapitel 3 indeholder fire originale manuskripter; en oversigtsartikel der introducerer immunbiologien bag IL-21 og IL-21 receptoren (IL-21R) samt giver et overblik over den aktuelle viden indenfor IL- 21 som cancerterapi og i immunpatologi, og de efterfølgende 3 manuskripter præsenterer det eksperimentelle arbejde i denne afhandling:

Paper I: Søndergaard H. and Skak K. Interleukin 21: roles in immunopathology and cancer therapy. Tissue Antigens. 2009 Oct. 21, (Epub ahead of print)

Paper II og III fokuserer på anti-tumor effekten af IL-21 proteinterapi givet som intraperitoneal, subkutan og intratumoral administration i to prækliniske musemodeller for cancer - B16 melanom and RenCa renalcelle carcinom. Heri bestemmes de celler der er nødvendige for anti-tumor aktiviteten af IL-21 og effekten af IL-21 evalueres på densiteten og

aktiviteten af tumor infiltrerende T celler samt på tumor drænende lymfeknuder. Paper IV undersøger rollen af endogent IL-21 i immunovervågning, samt primær og sekundær tumorimmunitet, ved hjælp af forskellige eksperimentelle tumormodeller i IL-21- og IL-21R- deficiente mus med fokus på NK, NKT and CD8+ T celle responser. De opnåede resultater i Paper II-IV diskuteres i kapitel 4. Kapitel 5 opsummerer hovedkonklusionerne der er opnået i denne afhandling og kapitel 6 udstikker perspektiver for fremtidig forskning omkring IL-21 i cancerimmunterapi. Til slut er givet en liste over referencer (eksklusive referencerne i Paper I-IV).

Chapter 1 – Introduction and objectives

Conventional therapy is rarely curative against disseminated cancers, manifesting the need to explore new treatment strategies. Recognition of the immune system as an intricate player in cancer development has prompted the concept of immunotherapy, where modulation of the immune system is proposed as a novel approach to fight cancer. Cytokines are signaling molecules secreted by the immune system and represent one way to systemically modulate immune responses with clinical proof of concept for the treatment of cancer (Rosenberg, 2001; Smyth et al., 2004). Interleukin (IL)-21 is a recently discovered cytokine produced by CD4+ T cells and NKT cells (Parrish-Novak et al., 2000; Coquet et al., 2007), and targets a broad range of immune cells within both the lymphoid and myeloid lineage (Spolski and Leonard, 2008). IL-21 has shown encouraging anti-tumor activity in various mouse models, but so far most studies have been performed in models with little clinical relevance, like artificial cytokine secreting tumors, models immunogenically enhanced with foreign antigens, or by the use of cytokine producing plasmids (Leonard and Spolski, 2005). However, it remains unknown if these results are reproducible using recombinant IL-21 protein therapy in more clinically relevant models. Also, there is limited in vivo data describing the anti-tumor mechanisms of IL-21, and it remains elusive whether the anti-tumor effect of IL-21 is caused by an increase in the number or function of effector cells, improved homing, better survival or perhaps other secondary effects. Finally, there is a general lack of knowledge concerning feasible routes of administration of IL-21 and potential biomarkers of IL-21 anti- tumor activity. Altogether, it will be essential to clarify these issues for the understanding and development of IL-21 as a cancer immunotherapy.

Extensive work in experimental animals have recently associated endogenous IL-21 signaling with the development of several major autoimmune diseases including, systemic lupus erythematosus (SLE) (Bubier et al., 2009), type 1 diabetes (T1D) (Spolski et al., 2008; Sutherland et al., 2009), rheumatoid arthritis (RA) (Young et al., 2007), inflammatory bowel disease (IBD) (Fina et al., 2008), and possibly multiple sclerosis (MS) (Nurieva et al., 2007). Together, these data highlight that IL-21 is mainly a proinflammatory cytokine and for this reason it has been suggested that neutralization of IL-21 could be of potential benefit to patients suffering from these diseases (Ettinger et al., 2008). However, the role of endogenous IL-21 during cancer development, growth and metastasis remains unknown and has potential ramifications for such approaches.

Therefore, the objectives of this thesis is divided into two parts; part 1 is the main part (Paper II and III) and will focus on the evaluation of IL-21 protein as a cancer immunotherapy and part 2 (Paper IV) will focus on the role of endogenous IL-21 in tumor immunity.

Specific objectives in part 1

Here, two clinically relevant murine cancer models will be established to investigate the therapeutic anti-tumor activity of IL-21 protein using the murine orthologue protein (mIL-21) and the following questions will be addressed: 1. Can IL-21 protein therapy given by various routes of administration inhibit established subcutaneous tumor-growth in the B16 melanoma and RenCa renal cell carcinoma model? 2. How does IL-21 anti-tumor effect of local administration compare to systemic administration? 3. What immune cells are responsible for the anti-tumor activity of IL-21? 4. How does IL-21 therapy affect the density and activity tumor infiltrating T cells and the number and proliferation of T cells in tumor draining lymph nodes?

Specific objectives in part 2

Here, IL-21- and IL-21R-deficient mice will be subjected to various experimental tumors and the following question will be addressed: 1. How does IL-21 signaling-deficiency affect tumor immunosurveillance, primary tumor immunity conducted by NK, NKT and CD8+ T cells and secondary CD8+ T cell memory?

Chapter 2 – Background

Cancer and immunotherapy Cancer is a leading cause of death world-wide and accounted for a total of 7.4 million deaths in 2004 (~13% of all deaths); a number which is expected to rise rapidly with the increase in global ageing1. Cancer is a generic term for a large group of diseases that can arise from all nucleated cells in our body. The hallmarks of cancer is a transformation of normal cells by a series of inherited and acquired genetic mutations, which provide growth and survival advantages, and eventually generate malignant neoplasms able to invade adjacent tissues and spread to distant organs (Hanahan and Weinberg, 2000). The spreading of cancer cells known as metastasis is a defining feature of the disease and is the major cause of death from cancer.

Chemotherapy is still the first line treatment of most disseminated cancers, but despite the arrival of more than 20 new compounds in the last decade, chemotherapy is only curative in very few cases (Savage et al., 2009). Daily, patients and physicians are faced with the shortcomings of these conventional treatments and clearly new approaches are needed. Cancer immunotherapy is a novel approach aiming to harness our immune system to combat cancer, and has the potential to specifically target cancer cells with limited systemic toxicity.

Our immune system is a tremendously potent defense system, which protects us from a large and versatile array of microbial intruders, and the idea of using its inherent strengths to fight cancer is appealing. In 1970, the concept of cancer immunosurveillance was conceived, which proposed the existence of immunological mechanisms that eliminate potentially dangerous mutant cells (Burnet, 1970). Since then, this concept has been substantiated by evidence that both the innate and adaptive parts of the immune system indeed recognize, shape and partly inhibit cancer development (van der et al., 1991; Shankaran et al., 2001; Smyth et al., 2001; Dunn et al., 2002). Still, cancers clearly develop in the presence of a competent immune system, showing that the immune system alone is not equipped to protect against all cancers.

Immune recognition of transformed cells during early cancer formation is suggested to shape an emerging lesion by deleting particularly abnormal and immunogenic cell-clones in a process called immunoediting. Since cancers are genetically instable and consist of a heterogenic population of cells, this process eventually promotes the outgrowth of immune-escape variants (Dunn et al., 2002; Dunn et al., 2004).

1 According to the World Health Organization, Fact sheet N°297, February 2009, www.who.int/cancer

The existence of immunoediting implies a period of equilibrium where immunoediting outbalances tumor-growth, and evidence of this was recently shown in experimental animals (Koebel et al., 2007). Whether emerging human cancers occasionally are eliminated during periods of immunoediting is difficult to show, but clinically detectable tumors that have escaped show signs of an immunological selection process; loss or downregulation of molecules in the antigen-processing machinery is a typical finding in human cancers (Cabrera et al., 1996; Hicklin et al., 1998; Garcia-Lora et al., 2003), and loss of immunogenicity has been found in trials with specific antigen targeting (Khong and Restifo, 2002; Yee et al., 2002). Similar shaping and generation of treatment-resistant clones within established tumors is often the reason that chemotherapy fails to control cancers (Mellor and Callaghan, 2008).

These inherent problems outline the challenge in modern cancer therapy, where inadequate host responses and poor therapeutic measures result in treatment-refractory tumors that eventually progress. Because cancer immunotherapy targets the host response, it represents an entirely new frontline to combine with conventional therapies, and if properly applied, it might facilitate additional selection pressure to eradicate cancers.

However, in addition to the generation of immune-tolerated variants, cancer cells also exploit a panel of immunosuppressive mechanisms, generally disabling immune reactivity (Gajewski et al., 2006; Rabinovich et al., 2007). These include alterations in the antigen presentation machinery, poor immune cell chemoattraction, lack of activating co-stimulatory signals, secretion of immunosuppressive factors, activation of negative regulatory pathways, and specific recruitment of immunosuppressive cell populations. Thus, successful cancer immunotherapy needs to circumvent these many evasive mechanisms, which are illustrated in Figure 1.

The defining goal in cancer immunotherapy is to increase the interaction and reactivity between the immune system and cancers. Tumor infiltrating lymphocytes (TILs) are a common term for immune-effector cells in the defence against cancer and several reports show that TILs are beneficial in human cancers. Particularly, the number of CD8+ cytotoxic T cells and an + + increased ratio of CD8 cytotoxic T cells/CD4 regulatory T cells (Tregs) are independent prognostic factors for improved overall survival (Clemente et al., 1996; Galon et al., 2006; Pages et al., 2005; Gao et al., 2007; Naito et al., 1998; Piersma et al., 2007; Sato et al., 2005; Sharma et al., 2007; Schumacher et al., 2001; Zhang et al., 2003)

Figure 1. Tumor-intrinsic barriers for successful cancer immunotherapy. Tumor cells exploit several immunosuppressive mechanisms to evade immune responses. Generally, tumors do not induce significant inflammation, which is needed for proper immune cell chemotaxis. Secretion of soluble factors from tumor cells or from resident regulatory cells i.e. natural killer T (NKT) cells, works to maintain immature dendritic cells (iDC), and promote development and recruitment of regulatory dendritic cells (rDC), regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSC), which further contribute to the suppressive environment. Increased activity of catabolic enzymes i.e. indoleamine 2,3-dioxygenase (IDO) increases depletion of essential amino acids required for effector cell activity. Defects in the antigen presentation machinery lead to low levels of tumor antigen presentation, restricting immune recognition. Increased expression of negative co-stimulatory receptors such as programmed death receptor ligand-1 (PD-L1) on tumors and cytotoxic T lymphocyte antigen 4 (CTLA-4) on CD8+ T cells limit effector cell funtions and expression of apoptosis-inducing ligands by tumors can terminate effector cells. TRAIL, tumour necrosis factor– related apoptosis-inducing ligand; FasL, Fas ligand; TGFβ, transforming growth factor β. (Drawn with inspiration from (Gajewski et al., 2006; Rabinovich et al., 2007))

For that reason, many cancer immunotherapeutic approaches focus on boosting the amount and functionality of TILs (reviewed in Blattman and Greenberg, 2004; Dougan and Dranoff, 2009). The passive infusion of monoclonal antibodies (mAb) is the new paradigm in cancer therapy; several are approved for clinical use and are either targeting tumor-associated surface proteins or growth factor receptors overexpressed by tumors. The anti-tumor function of these antibodies is not completely understood, but includes steric inhibition of target receptors, complement activation, and activation of cell-mediated cytotoxicity. The next generation of mAb mainly focuses on re-generating immune co-stimulation or blocking of regulatory pathways as outlined in figure 1. Tumor-specific vaccines based on attenuated tumor cells, viral vectors expressing tumor-associated antigens or modified dendritic cells (DC) have all been explored to boost the pool of tumor reactive cells, but so far only prophylactic vaccines against virally-induced cancer has been approved. Adoptive cell transfer (ACT) is another approach that relies on ex vivo expansion of cancer-specific T cells isolated from

patient TIL populations. While this type of therapy is restricted to highly specialized institutions, it has shown response rates of ~50% in traditionally unmanageable cancers e.g. metastatic melanoma, emphasizing the potential of approaches that modulate TILs (Rosenberg et al., 2008).

Cytokines are secreted proteins able to systemically modulate immune responses. Interleukin (IL)-2 and interferon (IFN)-α are approved cytokine-therapies for metastatic melanoma (MM) and advanced renal cell carcinoma (RCC) (Dougan and Dranoff, 2009). IFN-α has important anti-viral activities and IL-2 is a potent T cell growth factor, and while both therapies show evidence of immune-mediated anti-tumor activity their mechanism of action remains unclear. Although these therapies offer moderate response rates, they still represent the only effective and potentially curative treatment for patients with MM and RCC. However, the side effects of these therapies are considerable and resemble symptoms of systemic infections with hypotension, fever and malaise. IL-2 can induce a potentially lethal vascular leak syndrome, which requires intensive care, and this has limited its general use. In addition to their therapeutic effects, cytokines are important tools for many of the new approaches outlined above; cytokines are used as adjuvants boosting anti-tumor vaccines and are critical for the ex vivo expansion of TILs for ACT. Clearly, cytokines are attractive candidates for cancer immunotherapy and the exploration of new cytokine-therapies is warranted. Interleukin (IL)-21 is a novel cytokine related to IL-2, with profound immunomodulatory effects and putative anti-tumor activity, which is reviewed in Paper I of this thesis.

The mouse as experimental system The mouse was used as the model organism in all experiments throughout this thesis due to its biological similarity, practicality and pedigree in studies of cancer, immunology and IL-21. Generally, the mouse reflects human immunobiology remarkably well and although the mouse and human immune systems clearly are not identical, the mouse is the model of choice in experimental immunology (Mestas and Hughes, 2004). In IL-21 research, the mouse is almost exclusively used as the experimental animal and for several reasons. Mouse and human IL-21 and IL-21R show great overall sequence homology with specific conservation in areas of cytokine-receptor interactions (Parrish-Novak et al., 2000). Furthermore, the tissue distribution of the IL-21R seems to be analogous between mice and humans, and although some discrepancies have been described, most of the cellular responses to IL-21 are similar in mice and humans (Spolski and Leonard, 2008). Thus, it is reasonable to assume that results from mice using the murine orthologue protein mIL-21 will be guidelines for what to expect in the human organism.

The mouse has a long history in models of cancer (Frese and Tuveson, 2007) and for many reasons. Mice are easy to handle, require low amounts of drug for pharmacologic activity and are practical for larger group sizes desired for statistical comparisons. The use of transplantable tumors where cancer cells are injected e.g. subcutaneously or intravenously gives a very reproducible onset of disease and homogenous disease development, which is critical for controlled pharmacological drug testing. Measureable endpoints are also a necessity; subcutaneous injection of tumor cells allows objective monitoring of disease progression through direct measurement of tumor size and intravenous injection results in quantifiable lung metastasis development. Furthermore, the existence of numerous mutant mouse strains with specific immunodeficiencies and the wide use of mice for gene-targeting are both very helpful tools to clarify the mechanism of action of immunotherapies and also to determine the contribution of endogenous proteins in disease development and control, which are both objectives in this thesis. However, translation of results from mice to humans should always be done with careful considerations for the different size, metabolism, pharmacokinetics and varying biology that exist between these species and importantly how well the model depicts the development and course of the human disease. But generally, the mouse seems to be a suitable animal model to study the anti-tumor effects of IL-21 and evaluate the role of IL-21-deficiency in tumor immunity.

Chapter 3 – Manuscripts

The work in this thesis is covered by 3 original manuscripts preceded by a review manuscript which will add to the introduction of the thesis. Papers II and III will cover the objectives of part 1 and Paper IV will cover the objectives of part 2.

Paper I. Interleukin 21: roles in immunopathology and cancer therapy (Review) This review will introduce the biology of IL-21 and IL-21R, including the intracellular signaling properties, identified target genes and expression pattern of the IL-21R. The cellular immunobiology of IL-21 is reviewed describing the cell types that produce IL-21 and the effects of IL-21 on the major immune cell subsets. Finally, the review covers the most recent advances of IL-21 in cancer immunotherapy and the emerging role of IL-21 in various immunopathologies.

Paper II. Interleukin 21 therapy increases the density of tumor infiltrating CD8+ T cells and inhibits syngeneic tumor growth This manuscript examines the anti-tumor effect of IL-21 protein by intraperitoneal and subcutaneous administration in B16 melanomas and RenCa renal cell carcinomas. The anti- tumor effect of early and delayed administration of IL-21 is explored along with the pharmacokinetics and biodistribution of IL-21. The responsible effector cells for the anti-tumor effect of IL-21 are determined and the effect of IL-21 on the density of tumor infiltrating T cells is evaluated.

Paper III. Intratumoral interleukin 21 increases anti-tumor immunity, tumor- infiltrating CD8+ T cell density and activity, and enlarges draining lymph nodes This manuscript compares the anti-tumor effect of intratumoral and subcutaneous administration of IL-21 in B16 melanomas and RenCa renal cell carcinomas. The effect of intratumoral administration of IL-21 is examined on the density and activity of tumor infiltrating T cells, on local versus systemic tumor-growth, and on the number and proliferation of T cells in tumor draining lymph nodes.

Paper IV. Endogenous IL-21 restricts CD8+ T cell expansion and is not required for tumor immunity This manuscript examines the role of endogenous IL-21 signaling during natural tumor immunosurveillance, and primary and secondary tumor immunity using IL-21- and IL-21R- deficient mice subjected to various experimental tumors controlled by NK, NKT or CD8+ T cells or with responsiveness to IL-21 therapy.

During the course of this work, contributions have also been made to two separate original manuscripts, which are not included in this thesis and will only be referenced in the text:

1. Eriksen, K.W., Søndergaard H, Woetmann A, Krejsgaard T, Skak K, Geisler C, Wasik M.A., Ødum N., The combination of IL-21 and IFN-α boosts STAT3 activation, cytotoxicity and experimental tumor therapy. Mol Immunol. 2009 Feb;46(5):812-20. Epub 2008 Oct 22.

2. Skak K, Søndergaard H, Frederiksen K.S., Ehrnrooth E, In vivo antitumor efficacy of interleukin-21 in combination with chemotherapeutics. Cytokine. 2009 Dec;48(3):231- 8. Epub 2009 Aug 25.

Paper I:

Interleukin 21: roles in immunopathology and cancer therapy (Review)

Søndergaard H. and Skak K., Tissue Antigens. 2009 Oct. 21, (Epub ahead of print)

Tissue Antigens ISSN 0001-2815

REVIEWARTICLE IL-21: roles in immunopathology and cancer therapy H. Søndergaard & K. Skak

Immunopharmacology, Novo Nordisk A/S, Maløv,˚ Denmark

Key words Abstract autoimmunity; cancer; cytokine; interleukin-21 Cytokines are secreted signalling molecules with decisive effects on haematopoiesis, innate and adaptive immunity, and immunopathology. Interleukin (IL)-21 is a novel Correspondence cytokine produced by activated CD4+ T cells and natural killer T (NKT) cells. IL-21 Henrik Søndergaard is part of a family of cytokines which include IL-2, -4, -7, -9 and -15 that all Novo Nordisk A/S share the common IL-2 receptor γ chain (γ ) in their individual receptor complexes. Department of Immunopharmacology c Novo Nordisk Park F6.2.30 IL-21 receptor (IL-21R) is widely expressed on both myeloid and lymphoid cell DK-2760 Maløv˚ lineages and IL-21 actions include co-stimulation of B cell differentiation and Denmark immunoglobulin (Ig) production, co-mitogen of T cells, and stimulation of NK and Tel: 45 4443 1376 + CD8+ T cell cytotoxic function. Initially, IL-21 was recognized for its anti-tumour Fax: 45 4443 4537 + effects in several preclinical tumour models, warranting its currently ongoing clinical e-mail: [email protected] development as a cancer immunotherapeutic. More recently, IL-21 has been associated Received 21 August 2009; with the development of a panel of autoimmune and inﬂammatory diseases, where accepted 21 August 2009 neutralization of IL-21 has been suggested as a potential new therapy. In this review, we will cover the latest discoveries of IL-21 as a cancer therapy and its implications doi: 10.1111/j.1399-0039.2009.01382.x in immunopathologies.

Introduction the immunobiology of IL-21 and cover the advances of this interesting cytokine in cancer therapy and its emerging role Cytokines are secreted polypeptides used primarily by the in autoimmune and inflammatory diseases. immune system for intercellular communication. Cytokines are vital in all aspects of immunology from early haematopoiesis to the generation, maintenance and contraction of both IL-21 and IL-21 receptor biology, signalling innate and adaptive immune responses. Therefore, the timely and expression introduction of cytokines or blockade of their signalling In 2000, a new type 1 cytokine receptor was discovered and pathway can have decisive effects on immune processes. denoted novel orphan interleukin receptor (NILR) (1). Soon Historically, identification of novel cytokines, and under- after, a functional cloning approach identified a four-helix- standing of their production and function have been funda- bundle cytokine with structural relation to IL-2, IL-4 and mental for the development of new clinical strategies. The IL-15 as the natural ligand to NILR, naming the new enti- use of recombinant cytokines have pioneered in the field of ties IL-21 and IL-21 receptor (IL-21R) (2). The IL-21R gene cancer immunotherapy where interleukin (IL)-2 and inter- is located directly downstream of IL-4R-α on human chro- feron (IFN)-α have been used for more than two decades mosome 16, and IL-21R has an amino acid sequence with and still represent effective treatments for certain cancers, e.g. greatest homology to the IL-2 receptor beta-chain (IL-2R-β). melanoma and renal cell carcinoma (RCC). The blockade of The IL-21 gene is located on human chromosome 4q26-q27 endogenous cytokines and their signalling pathways represent approximately 180 kb from the IL-2 gene and the mature novel approaches in the fight against many autoimmune dis- IL-21 polypeptide consists of 131 amino acid residues most eases; particularly the neutralization of tumour necrosis factor similar to IL-15 (2). (TNF)-α and its signalling have revolutionized the treatment Human and murine IL-21 and IL-21R show approximately of rheumatoid arthritis. 60% overall amino acid sequence homology with significant The discovery of IL-21 adds to a still growing panel of conservation in regions of cytokine–receptor interaction (2), human cytokines, and this cytokine has not only shown poten- and to a large extent the function of IL-21 has been shown to tial as cancer therapy, but also as an attractive target for be similar in mouse and human, although several discrepancies several autoimmune diseases. Here, we will briefly review have also been described.

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak

Figure 1 The common gamma chain (γc) cytokine family and interleukin (IL)-21 intracellular signalling. IL-2, IL-4, IL-7, IL-9, IL-15 and IL-21 all share the common IL-2 receptor (IL-2R)γc in their individual receptor complexes. IL-21 signals through its unique IL-21 receptor (IL-21R) in a heterodimeric complex with γc. Upon ligand interaction, the IL-21R/γc complex recruits and phosphorylates janus kinases (JAK)1 and JAK3, which downstream activates signal transducers and activators of transcription (STAT). IL-21 primarily activates STAT3, but also STAT1 and more transiently STAT5a and 5b. The phosphatidylinositol-3-kinase (PI-3K)-AKT and mitogen-activated protein kinase (MAPK) pathways are also involved in IL-21 signalling. Target genes positively regulated by IL-21 are indicated.

The IL-21R signals as a heterodimeric complex with the Studies in IL-21R-deficient mice show that IL-21 is not crit- common IL-2 receptor gamma chain (γc, CD132) (3), mak- ical for normal haematopoiesis (11, 12). In the T cell lineage, ing IL-21 the newest member of the γc cytokine family, which IL-21R is not expressed until thymocytes differentiate into also includes IL-2, IL-4, IL-7, IL-9 and IL-15 (Figure 1). CD4 and CD8 double positive cells. Low levels of IL-21R are Like other type 1 cytokine receptors, the IL-21R/γc com- found on mature CD4+ and CD8+ T cells which increase in plex is a receptor tyrosine kinase, which upon ligand inter- response to T cell receptor (TCR) stimulation (13). In contrast, action recruits and phosphorylates janus kinases (JAK) that NKT cells express IL-21R ex vivo without prior activation. In subsequently activates signal transducers and activators of the B cell lineage the earliest detection of IL-21R is on pre- transcription (STAT). Similar to its other family members B cells, and mature follicular B cells express higher levels IL-21R recruits and activates JAK1 and JAK3 (3). Down- compared with T cells, and this is further increased upon B stream, IL-21 signalling primarily activates STAT3, but also cell activation. Mature marginal zone B cells have lower IL- STAT1 and more transiently STAT5a and STAT5b (3, 4). 21R expression compared with follicular B cells, and plasma The phosphatidylinositol-3-kinase (PI-3K)-AKT and mitogen- cells express very low IL-21R levels (13, 14). Overall, IL-21R activated protein kinase (MAPK) pathways have also been shows the highest expression on activated lymphocytes, indi- suggested to transmit IL-21 signals (4). The target genes for cating that IL-21 mainly acts as a co-stimulant of activated IL-21 signalling still remains to be fully elucidated, but tar- lymphocytes. get genes identified so far include cyclin A/B/E, granzyme A/B, IFN -γ, CXCR3, CXCR6, Bcl-3, JAK3, IL-21 and IL- 21R (5–7), indicating that IL-21 plays a role in cell cycle IL-21 in immunobiology progression, cellular activation, trafficking, cell survival and Humans deficient of the γc receptor suffer from X-linked positively regulates its own expression (Figure 1). severe combined immunodeficiency (XSCID) syndrome, a IL-21 production is restricted to activated CD4+ T cells (2), condition where T cells and NK cells are completely absent and activated natural killer T (NKT) cells (8), whereas and B cells are present but functionally impaired. This shows the IL-21R is much more widely expressed, including the sum of actions and impact that the γc-dependent cytokines B cells, T cells, NK cells, NKT cells, dendritic cells (DC), have on normal immunobiology. IL-21 is no exception and its macrophages, keratinocytes and intestinal fibroblasts (9, 10), pleiotropic effects correspond to the cellular distribution of its indicating a broad range of actions of IL-21. receptor. These are summarized in Figure 2.

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology

Figure 2 The major actions of IL-21. IL-21 is produced by CD4+ T cells and natural killer (NK) T cells in response to T cell receptor (TCR) activation, and modulates a broad range of both myeloid and lymphoid immune cells. The effects of IL-21 are generally context dependent and various forms of co-stimulation determine the cellular response: B cells proliferate and differentiate in response to B cell receptor (BCR) and CD40 co-stimulation, but undergo apoptosis without; CD8+ T cells primarily expand and increase their cytotoxicity together with IL-15 or TCR co-stimulation; resting NK cells express very little IL-21R, but in concert with IL-2 or IL-15, IL-21 drives terminal NK cell differentiation. IL-21 shows autocrine stimulation of its own production, activation of NKT cells and regulation of CD4+ T cell differentiation and proliferation. IL-21 does not directly modulate regulatory T cells (Tregs), but it has been suggested to reduce Treg differentiation. Dendritic cells (DC) remain immature in response to IL-21, whereas macrophages become activated.

B cells IL-21R knockout mice had reduced serum levels of IgG, B cells express the highest levels of IL-21R, making B cells but increased levels of IgE (12), and IL-21 given at the time prime responders to IL-21. Initially, IL-21 was found to of immunization reduces switching to IgE, but not IgG by spe- augment anti-CD40-induced B cell proliferation but inhibit ciﬁc inhibition of germline C(ε) transcription in B cells (16). proliferation induced by anti-IgM and IL-4 (2). IL-21 also However, the regulation of Ig class switching by IL-21 is inhibited B cell proliferation and induced apoptosis when also context dependent, because IL-21 inhibits IgE switch- B cells were activated via innate toll-like receptors (TLR) ing induced by IL-4 and PHA stimulation, whereas it pro- recognizing the bacterial component lipopolysaccaride (LPS) motes IgE switching induced by IL-4 and anti-CD40 stimula- or viral DNA cytosine-guanine motifs (CpG) (13). Further tion (17). studies have clariﬁed that IL-21 mainly inhibits B cell responses IL-21 co-stimulation is a strong inducer of B lymphocyte- and induces apoptosis of resting B cells in the absence of induced maturation protein-1 (BLIMP-1), a transcriptional proper co-stimulation, whereas cross-linking of both the B cell master switch controlling the terminal differentiation of receptor (BCR) and CD40 induces extensive proliferation and B cells into plasma cells, and IL-21 can directly induce plasma differentiation into immunoglobulin (Ig) producing plasma cell differentiation of anti-IgM-stimulated B cells. In addi- cells with enhanced IgG production (15). tion, IL-21 induces Bcl-6, which has been hypothesized to be

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak involved in the differentiation of germinal center (GC) B cells of the IL-21R in response to TCR stimulation, showing that into memory B cells (15), suggesting a more complex role of IL-21 primarily affects activated CD8+ T cells. Alone, IL-21 IL-21 in B cell differentiation. does not induce significant proliferation of CD8+ T cells, Taken together, IL-21 is a critical regulator of B cell but in response to antigen-independent stimulation IL-21 co- responses; B cells faced with IL-21 in the context of antigen- stimulates proliferation and expansion together with IL-7 or specific BCR stimulation and T cell co-stimulation will IL-15 of both na¨ıve and activated CD8+ T cells (7, 26). IL- undergo class switch recombination and differentiate into anti- 21 increases IFN-γ and IL-2 production and sustains the body producing plasma cells. In contrast, B cells encountering expression of CD62L and CD28 on IL-15-stimulated CD8+ IL-21 during unspecific TLR stimulation or without proper T cells. CD28 is an important co-receptor engaged by DC T cell help will undergo apoptosis. ligands during TCR stimulation, which is lost on senescent T cells (26).

CD4+ T cells IL-21R-deficient mice showed reduced antigen-specific CD8 T cell expansion and cytotoxicity compared with WT in CD4 T cells stimulated via their TCR are main producers of + + response to viral stimulation (7), highlighting a role for IL-21 IL-21, and TCR stimulation increases CD4 T cell expression + also in antigen-specific CD8 T cell expansion and func- of IL-21R, giving IL-21 an autocrine role in CD4 T cell + + tion. Similarly, in vitro expansion of antigen-specific CD8 responses. + T cells in blood from melanoma patients was substantially IL-21 is not a classical helper T cell (Th)1 or Th2 cytokine, increased by IL-21 in concert with autologous DCs pulsed but can be produced by CD4+ T cells during both highly Th1 with a melanoma-specific antigen (27). This effect was caused and Th2 skewed immune responses in vivo (18). CD4+ fol- by increased proliferation and survival, and could not be licular helper T cells (Tfh), identified by their expression of the chemokine receptor CXCR5, express high levels of IL- mimicked by IL-2, IL-7 or IL-15. Consistent with antigen- independent stimulation, IL-21-co-stimulated antigen-specific 21 required for Tfh cell development, cognate B cell help and GC formation (19). The recently characterized IL-17- CD8+ T cells showed high CD28 expression, increased IL- 2 production, high-avidity TCRs and increased cytotoxic producing CD4+ helper T cells (Th17) also produce IL-21 activity. and are distinct from the classical Th1 and Th2 cells. Th17 cells are involved in the clearance of certain pathogens and Taken together, IL-21 co-stimulates antigen-dependent and -independent proliferation, expansion, survival, and cytotoxi- in autoimmune pathogenesis (20). Regulatory CD4+ T cells city of CD8 T cells. Furthermore, IL-21 maintains CD8 T (Tregs), known for their immunosuppressive effects and char- + + acterized by high expression of IL-2Rα (CD25) and the tran- cell expression of CD28 and increases their IFN-γ and IL-2 scription factor forkhead box P3 (FoxP3), have not yet shown production, creating a more robust and independent CD8+ T evidence of IL-21 production. cell response. In response to anti-CD3/CD28 stimulation, IL-21 potently co-stimulates CD4+ T cell proliferation and IFN-γ production, which counteracts Treg suppression (21, 22), but IL-21 NK cells has not shown any direct effects on Tregs (21, 23). Although, Mature resting NK cells show very low expression of IL- during CD4+ T cell differentiation IL-21 has been suggested 21R, but IL-21R expression is up-regulated on both mature to inhibit FoxP3 expression and in turn increase T 17 cell h NK cells and on NK cell precursors upon activation. IL-21 development (24). IL-21 can drive the differentiation of T 17 h enhances NK cell differentiation from bone-marrow-derived cells in concert with transforming growth factor (TGF)β, but precursors (2), but IL-21R knockout mice have normal devel- in the presence of other proinflammatory cytokines such as opment and activity of mature NK cells (11), showing that IL-6, IL-21 is dispensable for T 17 differentiation and rather h IL-21 is redundant for normal NK cell development and serves as an auto-amplification loop for T 17 cell expan- h maturation. This is because NK precursor cells do not sion (25). In addition, IL-21 regulates its own production in express IL-21R, but IL-15, which is critical for normal an autocrine fashion through STAT3 activation (6). NK cell development, induces IL-21R expression (28) and Overall, IL-21 is produced by several different CD4 T cell + subsequently IL-21 can accelerate the NK cell maturation subtypes, it is essential in T cell development and GC fh process (29). formation, and serves as an autocrine amplification loop for Stimulation of NK cells with viral particles, IL-2 or IL-15 its own production and for CD4 T cell proliferation and + greatly enhances responsiveness to IL-21, which then induces survival. a large granular phenotype with increased cytolytic activity, IFN-γ production and perforin expression (11, 30). However, CD8+ T cells despite the co-stimulation of NK cell maturation and function, In addition to B cells, CD8+ T cells are the primary respon- IL-21 inhibits proliferation of NK cells induced by IL-2 and ders to IL-21. CD8+ T cells also increase their expression IL-15 and increases NK cell apoptosis (11, 30).

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology

NKT cells concomitant immunity to parental cell rechallenge (36). Since then, several groups have studied IL-21 anti-tumour effects NKT cells express inhibitory and activating NK cell receptors together with a restricted repertoire of TCRs. In con- by this approach, using genetically modified tumour cells trast to conventional T cells, the NKT cell TCR recognizes from both mouse and humans (see Table 1). In general, these lipid antigens such as the agonist alpha-galactosylceramide (α- studies show very potent anti-tumour activity of IL-21 with GC), presented by the MHC-like molecule CD1d expressed reports of complete tumour rejection in most studies. In these on DCs. Resting NKT cells express the IL-21R and produce studies, NK cells and CD8+ T cells were responsible for the IL-21 in response to anti-CD3 or α-GC stimulation (8). Fol- anti-tumour activity, with involvement of both IFN-γ and lowing anti-CD3 stimulation, the level of IL-21 production perforin, but with no role for CD4+ T cells (see Table 1). Although these data reflect the potential of IL-21 therapy and is even higher in NKT cells compared with splenic CD4+ T cells, suggesting that NKT cells may represent an important point out possible effector mechanisms of IL-21, tumour cells source of IL-21. NKT cells also respond to IL-21 stimulation; secreting IL-21 are not clinically translatable. IL-21 increases survival and proliferation of NKT cells and A more relevant approach came with therapeutic admin- enhances α-GC-stimulated proliferation and expansion. Fur- istration of IL-21-expressing plasmids. Using this approach, thermore, IL-21 co-stimulation of NKT cells increases their Wang et al. showed significant anti-tumour activity against granular morphology, granzyme B expression and cytokine B16 melanomas and MCA205 fibrosarcomas mediated by NK production (8). These data suggest that NKT cells represent an cells, with a minor role for CD8+ T cells but not CD4+ important source of IL-21 and that IL-21 generally increases T cells (39). Brady et al. showed that IL-21-expressing plas- NKT cell activation. mids injected at the time of tumour cell inoculation mediated NK cell- and perforin-dependent reduction in lung and liver metastasis, whereas IFN-γ, Fas ligand or TNF-related Dendritic cells apoptosis-inducing ligand (TRAIL) did not contribute (30). IL-21 has mainly shown inhibitory effects on DCs. IL-21 Nakano et al. showed significant tumour growth inhibition of maintains bone-marrow-derived DCs in a more immature subcutaneous head and neck squamous cell carcinomas using state characterized by increased phagocytotic activity and IL-21 plasmids, dependent on T cells, NK cells and with gen- decreased antigen presentation, thereby limiting the activation eration of tumour-specific Ig (42). Together, these data show of antigen-specific T cell responses (31, 32). This indicates a that IL-21 administration by a more therapeutic approach has potential role for IL-21 also in the restriction or termination a similar mechanism of anti-tumour activity even adding a of immune responses. However, DCs pretreated with IL- possible role for B cells, but perhaps with less dramatic anti- 21 and α-GC increases stimulation of NKT cell IFN-γ tumour effects compared with the IL-21-secreting tumours. production (33), indicating that the role of IL-21 on DCs is Moroz et al. were the first to use recombinant IL-21 pro- more complex and needs further investigation. tein and showed significant tumour growth inhibition and prolonged survival in an ovalbumin (OVA)-expressing lym- Macrophages phoma model when IL-21 was injected intraperitoneally (i.p.) early (from day 2 to day 12 post-tumour inoculation) or late IL-21R is expressed on both mouse and human macrophages (from day 12 to day 22 post-tumour inoculation) (40). This and in contrast to DCs, there is limited evidence that IL-21 was a CD8 T cell-dependent effect where IL-21 increased has proinflammatory effects on macrophages. IL-21 signalling + the expansion and cytotoxicity of OVA-specific CD8 T cells, is not required for macrophage development, but increases + and generated a more durable response compared with IL-2 macrophage production of the neutrophil chemoattractant and IL-15 leading to concomitant immunity towards tumour CXCL8/IL-8, increases phagocytosis and protease activity, rechallenge. Interestingly, pretreatment with IL-21 (day 4 and also the ability to stimulate antigen-specific CD4+ T cell − to day 6 post-tumour inoculation) completely abrogated the proliferation (34, 35). anti-tumour effect of IL-21, indicating that the timing of IL-21 is critical for the outcome. In a fully syngeneic sys- IL-21 as cancer therapy tem, we have shown significant tumour growth inhibition of B16 melanomas and RenCa RCCs in response to therapeu- Preclinical data tic administration of IL-21 protein, particularly by subcuta-

The stimulatory effects of IL-21 on NK cells and CD8+ neous administration (43). The anti-tumour activity was CD8+ T cells suggest that IL-21 may possess anti-tumour activity. T cell-dependent, and IL-21 increased the density of tumour In 2003, the first evidence of IL-21 anti-tumour activity infiltrating CD8+ T cells, a finding associated with improved was shown: murine colon carcinoma cells transduced with prognosis in human cancers. Using stereotactical injections of the IL-21 gene were completely rejected in syngeneic mice. IL-21 protein, Daga et al. showed significant tumour rejection Rejection was dependent on both NK and T cells and with of intracranially implanted gliomas, mediated by NK cells and

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak

Table 1 Preclinical anti-tumour activity and mechanisms of IL-21 monotherapy

Tumour model Treatment Anti-tumour effect Mechanism of action References

Colon 26 carcinoma (syngeneic) IL-21-secreting tumour Complete tumour rejection NK- and T cells, IFN-γ (36) ↑ production (spleen) B16F1 melanoma, MethA IL-21-secreting tumour Complete tumour rejection CTL, NK-, not CD4+ T cells (37) fibrosacoma (syngeneic) partly perforin mediated, not IFN-γ dependent AsPC1 pancreatic carcinoma IL-21-secreting tumour Significant tumour growth Partly NK cells, IFN-γ (38) ↑ (human xenograft) inhibition production (spleen cells) B16 melanoma, MCA205 IL-21-expressing plasmids Significant tumour growth NK cells, minor CTL role not (39) fibrosarcoma (syngeneic) (d5 and 12 PTI) inhibition and increased CD4 T cells, NK cell + ↑ survival cytotoxicity, no IFN-γ increase (serum) B16F10 melanoma, RenCa renal IL-21-expressing plasmids Reduced number of lung and NK cells, perforin dependent, (30) cell carcinoma, DA3 mammary (d -2 or 1 PTI) liver metastasis not IFN-γ, Fas ligand or TRAIL carcinoma (syngeneic) OVA-expressing, E.G7 thymoma IL-21 protein 20–100 μg i.p. Significant tumour growth CTL, early moderate role for NK (40) (semi-syngeneic) early (d2–12 PTI) 1 /day inhibition and increased cells, not CD4 T cells, and × + ↑ late (d12–22 PTI) 1 /day survival compared with IL-2 sustained Ag-specific CTL × and IL-15 response, cytotoxicity ↑ Several carcinomas with IL-21 protein 50 μg i.p. Reduced number of lung NK cells, NKG2D and perforin (41) endogenous or transfected i.v. metastasis: (d0–3 PTI) metastasis dependent, not IFN-γ and Fas NKG2D ligands (syngeneic) spontaneous: (d10–12 PTI) ligand SCCVII head and neck IL-21-expressing plasmids Significant tumour growth T cells, NK cells and (42) squamous cell carcinoma (d5, 12, 19 and 26 PTI) inhibition and increased tumour-specific (syngeneic) survival immunoglobulins B16F0 melanoma, RenCa renal IL-21 protein 50 μg i.p. or s.c. Significant tumour growth CTL, not NK cells (43) cell carcinoma (syngeneic) d3 or 8 PTI, 1 /day, B16 inhibition density of tumour infiltrating × ↑ d7 or 12 PTI, 3 /week, RenCa CD8 T cells × + GL261 glioma (syngeneic) IL-21-secreting tumour or IL-21 Significant tumour rejection B cells and NK cells (44) protein 1 μg i.t. tumour-specific IgG, ↑ serum-induced ADCC and ↑ complement-mediated lysis

PTI, post tumour injection; NK, natural killer; IFN, interferon; IL, interleukin; CTL, CD8+ cytotoxic T cells; OVA, ovalbumin; TRAIL, tumour necrosis factor-related apoptosis-inducing ligand; ADCC, antibody-dependent ceullular cytotoxicity.

B cells, with increases in tumour-specific antibodies, serum- whether this is secondary to increases in CD8+ T cell induced antibody-dependent cellular cytotoxicity (ADCC) and expansion or whether IL-21 also has direct effects on T cell complement-dependent tumour cell lysis (44). Taken together, trafficking. Generally, it remains to be clarified at what stage these results show that therapeutic use of recombinant IL- during a tumour immune response IL-21 is beneficial; does 21 protein is effective in the treatment of various preclinical IL-21 mainly shape the initiation and expansion processes in tumours. tumour draining lymph nodes or does it primarily modulate Taken as a whole, IL-21 anti-tumour activity depends tumour infiltrating lymphocytes during the effector phase. on NK cells, CD8 T cells or both, with a possible role + In addition to its immune-mediated anti-tumour mecha- for B cell-derived tumour-specific Ig. CD4 T cells seem + nisms, IL-21 has direct cytotoxic effects on certain B cell lym- less important, but so far total CD4 T cell depletions + phomas (45), and potential anti-angiogenic effects of IL-21 have been made, and more specific depletion of CD4+ T has been suggested (46). Table 1 summarizes the main pre- cell subsets, e.g. Tregs and Th17 cells is needed to fully clarify their role. Perforin, which is critical for both NK clinical in vivo anti-tumour data of IL-21 monotherapy. In addition to having anti-tumour effects when given as and CD8+ T cell cytotoxicity, is up-regulated by IL-21 and in contrast to Fas ligand or TRAIL, perforin seems monotherapy, IL-21 also shows significant additive effects to be important for IL-21 anti-tumour activity. IL-21 also together with a range of other biologicals modifying both increases IFN-γ production, but the need for IFN-γ is more innate and adaptive responses. IL-21 protein alone enhanced varied than perforin, probably depending on specific tumour the anti-tumour effect against B16 melanomas treated with cell susceptibility to this molecule. IL-21 generally increases adoptively transferred transgenic CD8+ T cells recognizing tumour infiltrating CD8+ T cells, but it remains to be shown a melanoma antigen and an antigen-specific vaccine (7). But

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology in concert with IL-15, IL-21 significantly boosted the anti- partial responses (RCC) were observed according to response tumour effect by increasing the circulating level of adop- evaluation criteria in solid tumours (RECIST). IL-21 also tively transferred CD8+ T cells and augmenting their IFN-γ showed evidence of immune activation, with increases in production. The sequential stimulation of NKT cells with granzyme B, perforin, IFN-γ and chemokine C-X-C motif the agonist α-GC followed by IL-21 stimulation synergis- receptor (CXCR)3 mRNA expression in peripheral blood tically inhibited B16 melanoma lung metastasis formation NK cells and CD8+ T cells (5, 53), suggestive of increased by boosting the concomitant NK cell activation induced by effector cell function. Overall, phase I results endorsed NKT cells (47). In combination with a potent triple anti- progress to phase II trials. body cocktail (Trimab), IL-21 helped to completely eradi- To date, two phase II trials have been completed, where IL- cate established tumours (48). Trimab consists of anti-DR5, 21 efficacy was evaluated in patients with stage IV MM (55) anti-CD40 and anti-CD137 (4-1BB), which together induce and in combination with a novel tyrosine kinase inhibitor a powerful T cell-dependent anti-tumour response through (sorafinib) in RCC [Bhatia et al., J Clin Oncol 27:15s, 2009 DR5-mediated tumour cell apoptosis and generation of anti- (suppl; abstr 3023), ASCO Annual Meeting]. Out of 24 MM gen, co-stimulation of DCs via CD40, and T cell activation patients treated by i.v. administration of IL-21, one partial through CD137 stimulation. and one complete response were observed per RECIST, and Several other interesting combination partners for IL-21 adverse effects were similar to those observed in the phase have recently been reviewed (49), expanding the potential of I trails. In the combination of IL-21 and sorafinib, 14 out IL-21 in cancer therapy even further. Still, many of the com- of 29 previously treated RCC patients completed 21 weeks bination partners so far tested with IL-21 are only in early treatment with stable disease or better and with an acceptable clinical testing or otherwise difficult to clinically translate. safety profile. However, we have recently published that IL-21 has additive Mild lymphocytopenia is a frequently observed adverse anti-tumour effects in combination with certain chemothera- effect of IL-21, but interestingly IL-21 increased the frequency pies, provided that IL-21 treatment is delayed compared to of lymphocytes expressing CD62L and CCR7 (55), indicat- chemotherapy (50). These data indicate that IL-21 is feasible ing that IL-21-induced lymphocytopenia might be caused by a in combination with conventional therapies. redistribution of lymphocytes to secondary lymphoid compart- Another interesting aspect of IL-21 in cancer therapy is its ments. Consistent with the phase I trials, NK cells and CD8+ use in the ex vivo generation of antigen-specific CD8 T cells + T cells in phase II trials showed significant increases in per- for adoptive cell therapy. IL-21 conditioning in vitro induces forin, granzyme B and IFN-γ expression following IL-21 (55). a unique differentiation program in CD8 T cells yielding a + Also, CD8 T cells and NK cells increased their expression of highly effective anti-tumour response upon adoptive transfer, + the activation markers CD25 and CD69 in response to IL-21, which cannot be mimicked with IL-2 or IL-15 (51). whereas CD4 T cells did not, indicating biologically dif- In conclusion, IL-21 alone or in combination with other + ferent effects of IL-21 on these lymphocytes. These results biological response modifiers stimulates both the innate and confirm that IL-21 is a well-tolerated drug that has objective adaptive arm of the immune system which shows significant anti-tumour activity in human cancer patients associated with anti-tumour activity in several different preclinical tumour signs of relevant immune activation. models. These findings have paved the way for IL-21 clinical Further trials are currently ongoing, evaluating IL-21 alone trials, which are currently ongoing. and in combination with other compounds. Preliminary reports from a trial of IL-21 in combination with rituximab (anti- Clinical data CD20 antibody) for the treatment of non-hodgkin’s lymphoma IL-2, which is closely related to IL-21, is approved for the (ClinicalTrials.gov identifier: NCT00347971) is encourag- treatment of metastatic melanoma (MM) and RCC. Although ing (55). The final results of these and future clinical trials IL-2 shows encouraging responses in groups of patients with will be very interesting to follow. these generally unmanageable diseases, its use is limited because of severe toxicities, e.g. vascular leak syndrome, requiring intensive care (52). IL-21 in immunopathology In phase I clinical trials IL-21 safety was tested in patients with MM and RCC (53, 54). IL-21 was well tolerated; Potent stimulation of immune responses inherently risks the the most common dose-related adverse effects were flu-like development of autoimmunity. Given that IL-21 stimulates symptoms such as fever, fatigue, chills and myalgia, and NK and T cell-mediated tumour immunity, plays a central dose-limiting toxicities included lymphopenia, neutropenia, role in B cell differentiation and antibody production, and thrombocytopenia and hepatotoxicity with increases in liver amplifies the expansion of proinflammatory Th17 cells, it is enzymes. Forty-seven MM and 19 RCC patients were treated not difficult to appreciate that host-derived IL-21 could also with IL-21. Of these, two complete responses (MM) and four be a key player in immunopathologies (see Table 2).

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak

Table 2 The role of IL-21 in autoimmune diseases

Disease Preclinical data References Human data References

Systemic lupus erythematosus Lupus-prone BXSB-Yaa, Sanroque and (56–58) Polymorphisms in IL-21 and IL-21R genes (60, 61) MRL-Faslpr mice overexpress IL-21 associate with SLE IL-21R.Fc ameliorates lupus and IL-21R (58, 59) Low IL-21R expression on B cells in SLE (62) deficiency protects from lupus patients correlates with disease activity Type 1 diabetes NOD mice overexpress IL-21, IL-21R (63, 66, 67) Polymorphisms in the IL-2/IL-21 locus and (68, 69) deficiency protects from diabetes, and in the IL-21 and IL-21R genes associate transgenic expression of IL-21 induces with T1D diabetes C57BL/6 mice Ag-specific CD4+ T cells in RIP-OVA mice (65) overexpress IL-21 counteracting Treg suppression of diabetes Rheumatoid arthritis IL-21R.Fc ameliorates CIA in DBA/1 mice (70) Inflamed synovial tissue, synovial (73, 74) lymphocytes and PBL from RA patients overexpress IL-21R IBF-deficient mice overexpress IL-21 and (71) IL-21R.Fc blocks inflammatory (75) develop spontaneous arthritis cytokine-release in RA synovial cell cultures IL-21R-deficient K/BxN mice are protected (72) Polymorphisms in IL-2/IL-21 locus (76) from arthritis associate with RA Inflammatory bowel disease IL-21 co-stimulation of TGFβ-treated (77) Polymorphisms in IL-2/IL-21 locus (78, 79, 81)

CD4+CD25− T cells inhibits colitis associate with UC and CD, and inﬂamed suppression in SCID mice receiving gut mucosa from UC and CD patients

untreated CD4+CD25− T cells overexpress IL-21 IL-21 is overexpressed in gut mucosa from (78) IL-21 from LPM cells increases MMP (10) DSS- and TNBS-induced colitis and IL-21 release from intestinal fibroblasts in CD deficiency protects from colitis patients IL-21R.Fc inhibits CCL20 secretion and T (80) cell chemotaxis by IBD mucosa Multiple sclerosis IL-21 administration prior to, but not post (82) IL-21 drives secondary autoimmunity in MS (85) MOG immunization increases EAE patients treated with severity lymphocyte-depleting Ab (alemtuzumab) IL-21R-deficient mice are protected from (24) MOG-EAE IL-21/IL-21R-deficient mice are not (83) protected from MOG-EAE and IL-21R.Fc increases severity of PLP-EAE

Ag, antigen; SLE, systemic lupus erythematosus; NOD, non-obese diabetic; RIP, rat insulin promoter; T1D, type 1 diabetes; OVA, ovalbumin; RA, rheumatoid arthritis; CIA, collagen induced arthritis; PBL, peripheral blood lymphocytes; IBF, IRF-4-binding protein; TGF, transforming growth factor; UC, ulcerative colitis; CD, Crohn’s disease; LPM, lamina propria monocytes; TNBS, trinitrobenzene sulfonic acid; MMP, matrix metalloproteinase; DSS, dextran sulfate sodium; MOG, myelin oligodendrocyte glycoprotein; PLP, myelin proteolipid protein; EAE, experimental autoimmune encephalomyelitis

Systemic lupus erythematosus of IL-21 (57). The lupus-prone MRL-Faslpr mouse showed Systemic lupus erythematosus (SLE) is a chronic autoimmune increased IL-21 production from CD4+ T cells and IL-21R.Fc disease that can affect connective tissues throughout the body. administration ameliorated disease severity (58). And, under- Hallmarks of SLE are high levels of circulating autoantibodies lining the role of IL-21 in experimental SLE, homozygous / thought to arise from aberrant apoptosis, characteristic rashes, IL-21R− −BXSB-Yaa mice failed to develop renal disease and glomerulonephritis with associated renal dysfunction. and mortality (59). These studies outline a potential beneﬁt BXSB-Yaa mutant mice, which spontaneously develop a of IL-21 neutralization in SLE patients, particularly given the condition similar to SLE, with lymphadenopathy, hyper- critical role for IL-21 in Tfh cell and GC development, plasma gammaglobulinemia, and severe immune-complex-mediated cell differentiation and antibody production. glomerulonephritis, have elevated serum levels of IL-21 (56). In humans, polymorphisms in the human IL-21 and IL- In another mutant mouse strain, the sanroque mouse, a muta- 21R gene show association with SLE (60, 61) and peripheral tion in the roquin protein negatively regulates Tfh cell devel- B cells from SLE patients have decreased IL-21R expres- opment. This mouse developed a similar lupus-like syndrome sion correlating with increased disease activity (62). Taken associated with increased Tfh cell levels and overproduction together, these data indicate a putative role for IL-21 in the

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology pathogenesis of SLE, but future studies are required to better principally involves autoimmune attacks of the joints produc- clarify its role in human disease. ing inflammatory synovitis that often progresses to destruction of the articular cartilage and joint deformities. Type 1 diabetes In animal models of RA, IL-21R.Fc reduced the histological and clinical signs of collagen-induced arthritis (CIA) Type 1 diabetes (T1D) is a condition caused by specific in DBA/1 mice immunized with bovine collagen (70). In autoimmune destruction of insulin-producing β-cells in the another study, interferon regulatory factor 4 (IRF-4)-binding pancreas, resulting in loss of blood glucose control and is protein (IBF)-deficient mice showed increased IL-21 and fatal without insulin-replacement therapy. IL-17 production leading to spontaneous development of a In non-obese diabetic (NOD) mice, a commonly used model RA-like condition (71). K/BxN mice spontaneously develop of T1D, the insulin-dependent diabetes susceptibility locus autoantibody-dependent arthritis, but IL-21R-deficient K/BxN Idd3 on chromosome 3 has major impact on the spontaneous mice were completely refractory to disease (72). Here, IL-21 disease development and spans the region for the IL-2 and mainly had a pathogenic role in Tfh development and autoan- IL-21 genes. It was initially shown that NOD mice over- tibody production, whereas T 17 cells and T seemed to be express IL-21 and it was postulated that IL-21 could drive h regs uninvolved (72). homeostatic expansion in NOD mice leading to the genera- In RA patients, IL-21R expression was increased in tion of autoreactive T cells (63). This theory was questioned inflamed synovial tissue and lymphocytes, and in periph- by findings of genetic variations in the IL-2 gene, leading eral blood lymphocytes compared with osteoarthritis (OA) to impaired T suppressive functions that associated with reg patients (73, 74). Also, IL-21 production was detected in RA diabetes development, and this was instead proposed as the synovial tissue cultures, and IL-21R.Fc inhibited secretion of causal link of Idd3 (64). T can protect against diabetes regs proinflammatory cytokines from the synovial tissue (75). In development as shown in another diabetes model; however, a recent study, it was proposed that genetic polymorphisms here, diabetic mice had normal T suppressive capacity, but reg in the IL-2/IL-21 locus were associated with the development instead showed overexpression of IL-21 which counteracted of RA (76). Overall, increasing evidence suggests that IL-21 T -mediated suppression (65). In support, two recent studies reg could be involved in the pathogenesis of RA. independently show that IL-21R-deficient NOD mice completely fails to develop diabetes, correlating with reduced insulitis, decreased infiltration of CD4+ and CD8+ T cells Inflammatory bowel disease in the pancreas, and no increases in T levels (66, 67). Fur- reg Inflammatory bowel disease (IBD) mainly comprises ulcer- thermore, in contrast to IL-21R / NOD splenocytes, adoptive + + ative colitis (UC) and Crohn’s disease (CD), where both transfer of IL-21R / NOD splenocytes does not induce dia- − − involve idiopathic autoimmune destruction of the gut mucosa, betes in NOD/scid mice and transgenic expression of IL-21 but differ in the areas of the gut they affect. induces spontaneous diabetes in diabetes-resistant C57BL/6 In experimental colitis, CD4 CD25 T cell transfer to mice (67). These data strongly advocate for a central role of + − SCID mice results in colitis development, co-transfer of IL-21 in diabetes development in NOD mice and suggest that TGFβ-treated CD4 CD25 T cells ameliorates disease, IL-21 is a relevant susceptibility gene in the Idd3 locus. Much + − whereas co-transfer of TGFβ and IL-21-treated CD4 CD25 effort has been put into identifying the one gene that confers + − T cells fails to suppress colitis development (77). This is the diabetes association of the NOD Idd3 locus, but it appears thought to occur because IL-21 inhibits TGFβ-induced FoxP3 that IL-2 and IL-21 are not mutually exclusive diabetes sus- expression and T differentiation of CD4 CD25 T cells ceptibility genes. reg + − and instead promotes differentiation of T 17 cells produc- In a recent large genome-wide association study of human h ing high levels of IL-17 as well as IL-21 (77). In experi- T1D, the chromosome region 4q27 was identified as one mental colitis induced by dextran sulfate sodium (DSS) or of six new robust disease-associated loci (68). This locus is trinitrobenzene sulfonic acid (TNBS), IL-21 expression was the human orthologue of the Idd3 containing both the IL-2 and IL-21 gene. In another analysis of IL-21 and IL-21R increased in the gut mucosa, and IL-21-deficient mice were gene-sequence variants, polymorphisms in both genes were highly protected from disease correlating with reduced Th17 association with diabetes (69). These data indicate that IL-21 cell activity (78). could have a role in human T1D development, but the strong In humans, IL-21 expression is increased in inflamed, but set of experimental data still lack supportive human data to not unaffected gut mucosa from both UC and CD patients (78, confirm this hypothesis. 79). Isolated lamina propria T cells from CD patients showed reduced IL-17 production when stimulated in the presence of anti-IL-21 antibody (78), indicating that IL-21 is present in Rheumatoid arthritis IBD and has potential impact on IL-17 responses, which are Rheumatoid arthritis (RA) is a chronic, inflammatory disorder believed to be pathogenic in CD. Intestinal fibroblasts isolated of idiopathic etiology that affects many tissues and organs. It from colitis patients and healthy controls express IL-21R, and

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak

IL-21aloneorinconcertwithTNF-α significantly induced their IL-21- and IL-21R-deficient mice to chronic lymphocytic secretion of matrix metalloproteinases (MMP), thought to play choriomeningitis virus (LCMV) (86–88). In these studies, IL- a key role in tissue degradation (10). Also, IL-21R.Fc treatment 21 was needed neither during the acute phase of viral infec- reduced the MMP production from intestinal fibroblasts tions, nor for the maintenance of memory CD8+ T cells after induced by supernatants from CD lamina propria mononuclear resolved infections, but without IL-21 chronic presence of cells (10). Moreover, IL-21 stimulation induced secretion of virus antigen resulted in exhaustion of CD8+ T cell responses the T cell attractant, CCL20/MIP-3α from colon epithelial and poor viral control. These studies highlight yet another cells, and anti-IL-21 blocked T cell chemotaxis induced by important aspect of IL-21 with potential ramifications for supernatants from IBD mucosal explants (80). In genome-wide novel therapies. Here it is also worth noticing that the role association studies single nucleotide polymorphisms (SNPs) of endogenous IL-21 in cancer control, which also includes within the IL-2/IL-21 locus 4q27 associated with both UC chronic persistence of antigen, remains to be determined. and CD (81). Overall, evidence is accumulating that IL-21 is a very relevant cytokine in the pathogenesis of IBD with a role in both tissue remodelling and T cell trafficking. Concluding remarks In less than a decade, IL-21 has advanced from being a Multiple sclerosis novel member of the γc receptor family to a cytokine in Multiple sclerosis (MS) is an autoimmune disease of unknown clinical trials for the treatment of cancer, with implica- etiology, which attacks the central nervous system (CNS), tions in several autoimmune pathogeneses. As cancer therapy, leading to demyelination and loss of physical and cognitive IL-21 monotherapy shows moderate clinical responses, but so function. far clinical trials have been limited to pretreated, end-stage Experimental autoimmune encephalomyelitis (EAE) is the patients and it would be very interesting to evaluate IL-21 in classical animal model of MS, where immunization with patients with less advanced disease. The manageable toxicity myelin-derived peptides or proteins induces CNS inflam- of IL-21 encourages combination with other drugs, and such mation mimicking human disease. In EAE, IL-21 protein options ought to be pursued in future trials. Continued research administration prior to myelin oligodendrocyte glycoprotein into IL-21 anti-cancer biology will be essential to further clar- (MOG)-peptide immunization increased the severity of dis- ify its immunotherapeutic effects, support future clinical trials ease, whereas IL-21 administration during disease progression and perhaps identify novel interesting combination partners. did not increase severity (82). Initially, MOG immunization The data reviewed here clearly suggest that neutralization of IL-21-deficient mice showed reduced EAE disease activity of IL-21 could hold therapeutic value in several major ascribed to a lack of Th17 cell differentiation (24). However, immunopathologies, where particularly IBD, RA and SLE more recent findings have challenged this, showing no reduc- seem to be relevant candidates. However, disclosing the role tion or even exacerbated EAE in IL-21- and IL-21R-deficient of IL-21 in human immunopathologies will be essential to mice and competent Th17 cell differentiation (83). This dis- warrant the development of new IL-21-blocking compounds. crepancy may arise from genetic variations in the mice used Hopefully, the next few years of IL-21 research will clarify in the different experiments (83). Also, in EAE induced by its full potential in cancer therapy and its intriguing role in myelin proteolipid protein (PLP), administration of IL-21R.Fc immunopathologies. before and after peptide immunisation increased the severity of disease associated with decreased Treg numbers and Foxp3 expression (84). References Altogether, the role of IL-21 in EAE is controversial, and a direct link between IL-21 and human MS remains to be 1. Ozaki K, Kikly K, Michalovich D et al. Cloning of a type I shown. However, a recent publication shows that MS patients cytokine receptor most related to the IL-2 receptor beta chain. Proc Natl Acad Sci U S A 2000: 97: 11439–44. who developed secondary autoimmunity resulting from treat- 2. Parrish-Novak J, Dillon SR, Nelson A et al. Interleukin 21 and ment with a lymphocyte-depleting monoclonal antibody, its receptor are involved in NK cell expansion and regulation alemtuzumab, had greatly elevated serum IL-21, which was of lymphocyte function. Nature 2000: 408: 57–63. genetically predetermined. It is possible that IL-21 drives 3. Asao H, Okuyama C, Kumaki S et al. Cutting edge: the excess homeostatic T cell expansion and apoptosis leading common gamma-chain is an indispensable subunit of the IL-21 to secondary autoimmunity (85). Clearly, these data warrant receptor complex. J Immunol 2001: 167: 1–5. future studies of IL-21 in the pathogenesis of MS. 4. Zeng R, Spolski R, Casas E et al. The molecular basis of IL-21-mediated proliferation. Blood 2007: 109: 4135–42. 5. Frederiksen KS, Lundsgaard D, Freeman JA et al. IL-21 Chronic viral infections induces in vivo immune activation of NK cells and CD8(+) Very recently, a trio of papers showed that host IL-21 was crit- T cells in patients with metastatic melanoma and renal cell ical for the control of chronic viral infections by subjecting carcinoma. Cancer Immunol Immunother 2008: 57: 1439–49.

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology

6. Caprioli F, Sarra M, Caruso R et al. Autocrine regulation of 25. Korn T, Bettelli E, Oukka M et al. IL-17 and Th17 Cells. IL-21 production in human T lymphocytes. J Immunol 2008: Annu Rev Immunol 2009: 27: 485–517. 180: 1800–7. 26. Alves NL, Arosa FA, van Lier RA. IL-21 sustains CD28 7. Zeng R, Spolski R, Finkelstein SE et al. Synergy of IL-21 and expression on IL-15-activated human naive CD8+ T cells. IL-15 in regulating CD8+ T cell expansion and function. J Exp J Immunol 2005: 175: 755–62. Med 2005: 201: 139–48. 27. Li Y, Bleakley M, Yee C. IL-21 influences the frequency, 8. Coquet JM, Kyparissoudis K, Pellicci DG et al. IL-21 is phenotype, and affinity of the antigen-specific CD8 T cell produced by NKT cells and modulates NKT cell activation and response. J Immunol 2005: 175: 2261–9. cytokine production. J Immunol 2007: 178: 2827–34. 28. Perez SA, Mahaira LG, Sotiropoulou PA et al. Effect of IL-21 9. Spolski R, Leonard WJ. Interleukin-21: basic biology and on NK cells derived from different umbilical cord blood implications for cancer and autoimmunity. Annu Rev Immunol populations. Int Immunol 2006: 18: 49–58. 2008: 26: 57–79. 29. Sivori S, Cantoni C, Parolini S et al. IL-21 induces both rapid 10. Monteleone G, Caruso R, Fina D et al. Control of matrix maturation of human CD34+ cell precursors towards NK cells metalloproteinase production in human intestinal fibroblasts by and acquisition of surface killer Ig-like receptors. Eur J interleukin 21. Gut 2006: 55: 1774–80. Immunol 2003: 33: 3439–47. 11. Kasaian MT, Whitters MJ, Carter LL et al. IL-21 limits NK 30. Brady J, Hayakawa Y, Smyth MJ et al. IL-21 induces the cell responses and promotes antigen-specific T cell activation: functional maturation of murine NK cells. J Immunol 2004: a mediator of the transition from innate to adaptive immunity. 172: 2048–58. Immunity 2002: 16: 559–69. 31. Brandt K, Bulfone-Paus S, Jenckel A et al. Interleukin-21 12. Ozaki K, Spolski R, Feng CG et al. A critical role for IL-21 in inhibits dendritic cell-mediated T cell activation and induction regulating immunoglobulin production. Science 2002: 298: of contact hypersensitivity in vivo. J Invest Dermatol 2003: 1630–4. 121: 1379–82. 13. Jin H, Carrio R, Yu A et al. Distinct activation signals 32. Brandt K, Bulfone-Paus S, Foster DC et al. Interleukin-21 determine whether IL-21 induces B cell costimulation, growth inhibits dendritic cell activation and maturation. Blood 2003: arrest, or Bim-dependent apoptosis. J Immunol 2004: 173: 102: 4090–8. 657–65. 33. Maeda M, Yanagawa Y, Iwabuchi K et al. IL-21 enhances 14. Good KL, Bryant VL, Tangye SG. Kinetics of human B cell dendritic cell ability to induce interferon-gamma production by behavior and amplification of proliferative responses following natural killer T cells. Immunobiology 2007: 212: 537–47. stimulation with IL-21. J Immunol 2006: 177: 5236–47. 34. Pelletier M, Bouchard A, Girard D. In vivo and in vitro roles 15. Konforte D, Simard N, Paige CJ. IL-21: an executor of B cell of IL-21 in inflammation. J Immunol 2004: 173: 7521–30. fate. J Immunol 2009: 182: 1781–7. 35. Ruckert R, Bulfone-Paus S, Brandt K. Interleukin-21 16. Suto A, Nakajima H, Hirose K et al. Interleukin 21 prevents stimulates antigen uptake, protease activity, survival and antigen-induced IgE production by inhibiting germ line induction of CD4+ T cell proliferation by murine C(epsilon) transcription of IL-4-stimulated B cells. Blood macrophages. Clin Exp Immunol 2008: 151: 487–95. 2002: 100: 4565–73. 36. Ugai S, Shimozato O, Kawamura K et al. Expression of the 17. Wood N, Bourque K, Donaldson DD et al. IL-21 effects on interleukin-21 gene in murine colon carcinoma cells generates human IgE production in response to IL-4 or IL-13. Cell systemic immunity in the inoculated hosts. Cancer Gene Ther Immunol 2004: 231: 133–45. 2003: 10: 187–92. 18. Pesce J, Kaviratne M, Ramalingam TR et al. The IL-21 37. Ma HL, Whitters MJ, Konz RF et al. IL-21 activates both receptor augments Th2 effector function and alternative innate and adaptive immunity to generate potent antitumor macrophage activation. J Clin Invest 2006: 116: 2044–55. responses that require perforin but are independent of 19. Silver JS, Hunter CA. With a little help from their friends: IFN-gamma. J Immunol 2003: 171: 608–15. interleukin-21, T cells, and B cells. Immunity 2008: 29: 7–9. 38. Ugai S, Shimozato O, Yu L et al. Transduction of the IL-21 20. Wei L, Laurence A, Elias KM et al. IL-21 is produced by and IL-23 genes in human pancreatic carcinoma cells produces Th17 cells and drives IL-17 production in a STAT3-dependent natural killer cell-dependent and -independent antitumor manner. J Biol Chem 2007: 282: 34605–10. effects. Cancer Gene Ther 2003: 10: 771–8. 21. Peluso I, Fantini MC, Fina D et al. IL-21 counteracts the 39. Wang G, Tschoi M, Spolski R et al. In vivo antitumor activity regulatory T cell-mediated suppression of human CD4+ T of interleukin 21 mediated by natural killer cells. Cancer Res lymphocytes. J Immunol 2007: 178: 732–9. 2003: 63: 9016–22. 22. Ferrari-Lacraz S, Chicheportiche R, Schneiter G et al. IL-21 40. Moroz A, Eppolito C, Li Q et al. IL-21 enhances and sustains promotes survival and maintains a naive phenotype in human CD8+ T cell responses to achieve durable tumor immunity: CD4+ T lymphocytes. Int Immunol 2008: 20: 1009–18. comparative evaluation of IL-2, IL-15, and IL-21. J Immunol 23. Wuest TY, Willette-Brown J, Durum SK et al. The influence of 2004: 173: 900–9. IL-2 family cytokines on activation and function of naturally 41. Takaki R, Hayakawa Y, Nelson A et al. IL-21 enhances tumor occurring regulatory T cells. J Leukoc Biol 2008: 84: 973–80. rejection through a NKG2D-dependent mechanism. J Immunol 24. Nurieva R, Yang XO, Martinez G et al. Essential autocrine 2005: 175: 2167–73. regulation by IL-21 in the generation of inflammatory T cells. 42. Nakano H, Kishida T, Asada H et al. Interleukin-21 triggers Nature 2007: 448: 480–3. both cellular and humoral immune responses leading to

IL-21 in cancer and immunopathology H. Søndergaard & K. Skak

therapeutic antitumor effects against head and neck squamous 60. Sawalha AH, Kaufman KM, Kelly JA et al. Genetic cell carcinoma. J Gene Med 2006: 8: 90–9. association of interleukin-21 polymorphisms with systemic 43. Sondergaard H, Frederiksen KS, Thygesen P et al. Interleukin lupus erythematosus. Ann Rheum Dis 2008: 67: 458–61. 21 therapy increases the density of tumor infiltrating CD8(+) 61. Webb R, Merrill JT, Kelly JA et al. A polymorphism within T cells and inhibits the growth of syngeneic tumors. Cancer IL21R confers risk for systemic lupus erythematosus. Arthritis Immunol Immunother 2007. Rheum 2009: 60: 2402–7. 44. Daga A, Orengo AM, Gangemi RM et al. Glioma 62. Mitoma H, Horiuchi T, Kimoto Y et al. Decreased expression immunotherapy by IL-21 gene-modified cells or by of interleukin-21 receptor on peripheral B lymphocytes in recombinant IL-21 involves antibody responses. Int J Cancer systemic lupus erythematosus. Int J Mol Med 2005: 16: 2007: 121: 1756–63. 609–15. 45. Gelebart P, Zak Z, Anand M et al. Interleukin-21 effectively 63. King C, Ilic A, Koelsch K et al. Homeostatic expansion of induces apoptosis in mantle cell lymphoma through a T cells during immune insufficiency generates autoimmunity. STAT1-dependent mechanism. Leukemia 2009. Cell 2004: 117: 265–77. 46. Castermans K, Tabruyn SP, Zeng R et al. Angiostatic activity 64. Yamanouchi J, Rainbow D, Serra P et al. Interleukin-2 gene of the antitumor cytokine interleukin-21. Blood 2008: 112: variation impairs regulatory T cell function and causes 4940–7. autoimmunity. Nat Genet 2007: 39: 329–37. 47. Smyth MJ, Wallace ME, Nutt SL et al. Sequential activation of 65. Clough LE, Wang CJ, Schmidt EM et al. Release from NKT cells and NK cells provides effective innate regulatory T cell-mediated suppression during the onset of immunotherapy of cancer. J Exp Med 2005: 201: 1973–85. tissue-specific autoimmunity is associated with elevated IL-21. 48. Smyth MJ, Teng MW, Sharkey J et al. Interleukin 21 enhances J Immunol 2008: 180: 5393–401. antibody-mediated tumor rejection. Cancer Res 2008: 68: 66. Spolski R, Kashyap M, Robinson C et al. IL-21 signaling is 3019–25. critical for the development of type I diabetes in the NOD 49. Skak K, Kragh M, Hausman D et al. Interleukin 21: mouse. Proc Natl Acad Sci U S A 2008: 105: 14028–33. combination strategies for cancer therapy. Nat Rev Drug 67. Sutherland AP, Van BT, Wurster AL et al. Interleukin-21 is Discov 2008: 7: 231–40. required for the development of type 1 diabetes in NOD mice. 50. Skak K, Søndergaard H, Frederiksen KS, Ehrnrooth E. In vivo Diabetes 2009: 58: 1144–55. antitumor efficacy of interleukin-21 in combination with 68. Todd JA, Walker NM, Cooper JD et al. Robust associations of chemotherapeutics. Cytokine 2009: [Epub ahead of print]. four new chromosome regions from genome-wide analyses of 51. Hinrichs CS, Spolski R, Paulos CM et al. IL-2 and IL-21 type 1 diabetes. Nat Genet 2007: 39: 857–64. confer opposing differentiation programs to CD8+ T cells for 69. Asano K, Ikegami H, Fujisawa T et al. Molecular scanning of adoptive immunotherapy. Blood 2008: 111: 5326–33. interleukin-21 gene and genetic susceptibility to type 1 52. Atkins MB. Interleukin-2: clinical applications. Semin Oncol diabetes. Hum Immunol 2007: 68: 384–91. 2002: 29: 12–7. 70. Young DA, Hegen M, Ma HL et al. Blockade of the 53. Davis ID, Skrumsager BK, Cebon J et al. An open-label, interleukin-21/interleukin-21 receptor pathway ameliorates two-arm, phase I trial of recombinant human interleukin-21 in disease in animal models of rheumatoid arthritis. Arthritis patients with metastatic melanoma. Clin Cancer Res 2007: 13: Rheum 2007: 56: 1152–63. 3630–6. 71. Chen Q, Yang W, Gupta S et al. IRF-4-binding protein inhibits 54. Thompson JA, Curti BD, Redman BG et al. Phase I study of interleukin-17 and interleukin-21 production by controlling the recombinant interleukin-21 in patients with metastatic activity of IRF-4 transcription factor. Immunity 2008: 29: melanoma and renal cell carcinoma. J Clin Oncol 2008: 26: 899–911. 2034–9. 72. Jang E, Cho SH, Park H et al. A positive feedback loop of 55. Davis ID, Brady B, Kefford RF et al. Clinical and biological IL-21 signaling provoked by homeostatic CD4+. J Immunol efficacy of recombinant human interleukin-21 in patients with 2009: 182: 4649–56. stage IV malignant melanoma without prior treatment: a phase 73. Jungel A, Distler JH, Kurowska-Stolarska M et al. Expression IIa trial. Clin Cancer Res 2009: 15: 2123–9. of interleukin-21 receptor, but not interleukin-21, in synovial 56. Ozaki K, Spolski R, Ettinger R et al. Regulation of B cell fibroblasts and synovial macrophages of patients with differentiation and plasma cell generation by IL-21, a novel rheumatoid arthritis. Arthritis Rheum 2004: 50: 1468–76. inducer of Blimp-1 and Bcl-6. J Immunol 2004: 173: 5361–71. 74. Li J, Shen W, Kong K et al. Interleukin-21 induces T-cell 57. Vinuesa CG, Cook MC, Angelucci C et al. A RING-type activation and proinflammatory cytokine secretion in ubiquitin ligase family member required to repress follicular rheumatoid arthritis. Scand J Immunol 2006: 64: 515–22. helper T cells and autoimmunity. Nature 2005: 435: 452–8. 75. Andersson AK, Feldmann M, Brennan FM. Neutralizing IL-21 58. Herber D, Brown TP, Liang S et al. IL-21 has a pathogenic and IL-15 inhibits pro-inflammatory cytokine production in role in a lupus-prone mouse model and its blockade with rheumatoid arthritis. Scand J Immunol 2008: 68: 103–11. IL-21R.Fc reduces disease progression. J Immunol 2007: 178: 76. Daha NA, Kurreeman FA, Marques RB et al. Confirmation of 3822–30. STAT4, IL2/IL21, and CTLA4 polymorphisms in rheumatoid 59. Bubier JA, Sproule TJ, Foreman O et al. A critical role for arthritis. Arthritis Rheum 2009: 60: 1255–60. IL-21 receptor signaling in the pathogenesis of systemic lupus 77. Fantini MC, Rizzo A, Fina D et al. IL-21 regulates erythematosus in BXSB-Yaa mice. Proc Natl Acad Sci U S A experimental colitis by modulating the balance between Treg 2009: 106: 1518–1523. and Th17 cells. Eur J Immunol 2007: 37: 3155–63.

H. Søndergaard & K. Skak IL-21 in cancer and immunopathology

78. Fina D, Sarra M, Fantini MC et al. Regulation of gut autoimmune encephalomyelitis. J Immunol 2008: 180: inﬂammation and th17 cell response by interleukin-21. 7097–101. Gastroenterology 2008: 134: 1038–48. 84. Piao WH, Jee YH, Liu RL et al. IL-21 modulates CD4+ 79. Monteleone G, Monteleone I, Fina D et al. Interleukin-21 CD25+ regulatory T-cell homeostasis in experimental enhances T-helper cell type I signaling and interferon-gamma autoimmune encephalomyelitis. Scand J Immunol 2008: 67: production in Crohn’s disease. Gastroenterology 2005: 128: 37–46. 687–94. 85. Jones JL, Phuah CL, Cox AL et al. IL-21 drives secondary 80. Caruso R, Fina D, Peluso I et al. A functional role for autoimmunity in patients with multiple sclerosis, following interleukin-21 in promoting the synthesis of the T-cell therapeutic lymphocyte depletion with alemtuzumab chemoattractant, MIP-3alpha, by gut epithelial cells. (Campath-1H). J Clin Invest 2009: 119: 2052–61. Gastroenterology 2007: 132: 166–75. 86. Elsaesser H, Sauer K, Brooks DG. IL-21 is required to control 81. Festen EA, Goyette P, Scott R et al. Genetic variants in the region harbouring IL2/IL21 associated with ulcerative colitis. chronic viral infection. Science 2009: 324: 1569–72. Gut 2009: 58: 799–804. 87. Yi JS, Du M, Zajac AJ. A vital role for interleukin-21 in the 82. Vollmer TL, Liu R, Price M et al. Differential effects of IL-21 control of a chronic viral infection. Science 2009: 324: during initiation and progression of autoimmunity against 1572–6. neuroantigen. J Immunol 2005: 174: 2696–701. 88. Frohlich A, Kisielow J, Schmitz I et al. IL-21R on T cells 83. Coquet JM, Chakravarti S, Smyth MJ et al. Cutting edge: is critical for sustained functionality and control of chronic IL-21 is not essential for Th17 differentiation or experimental viral infection. Science 2009: 324: 1576–80.

Paper II:

Interleukin 21 therapy increases the density of tumor infiltrating CD8+ T cells and inhibits syngeneic tumor growth

Søndergaard H., Frederiksen K.S., Thygesen P.,Galsgaard E.D., Skak K. Kristjansen P.E.G. and Kragh M., Cancer Immunol Immunother. 2007, Sep;56(9):1417-28. Epub. 2007 Feb.

Cancer Immunol Immunother (2007) 56:1417–1428 DOI 10.1007/s00262-007-0285-4

ORIGINALARTICLE

Interleukin 21 therapy increases the density of tumor inﬁltrating CD8+ T cells and inhibits the growth of syngeneic tumors

Henrik Søndergaard Æ Klaus S. Frederiksen Æ Peter Thygesen Æ Elisabeth D. Galsgaard Æ Kresten Skak Æ Paul E. G. Kristjansen Æ Niels Ødum Æ Michael Kragh

Received: 13 October 2006 / Accepted: 24 December 2006 / Published online: 7 February 2007 Ó Springer-Verlag 2007

Abstract Interleukin (IL)-21 is a recently discovered cells by immunohistochemistry. Whereas both routes cytokine in early clinical development, which has of IL-21 administration significantly inhibited growth shown anti-tumor activity in various animal models. In of small, established RenCa and B16 tumors, only s.c. the present study, we examine the anti-tumor activity therapy significantly inhibited the growth of large, of IL-21 protein therapy in two syngeneic tumor established tumors. We found a greater bioavailability models and its effect on the density of tumor infiltrat- and significant drainage of IL-21 to regional lymph ing T cells. We treated mice bearing established sub- nodes following s.c. administration, which could ac- cutaneous B16 melanomas or RenCa renal cell count for the apparent increase in anti-tumor activity. carcinomas with intraperitoneal (i.p.) or subcutaneous Specific depletion of CD8+ T cells with monoclonal (s.c.) IL-21 protein therapy and subsequently scored antibodies completely abrogated the anti-tumor activ- the densities of tumor infiltrating CD4+ and CD8+ T ity, whereas NK1.1+ cell depletion did not affect tumor growth. In accordance, both routes of IL-21 administration significantly increased the density of tumor + H. Søndergaard (&) E. D. Galsgaard infiltrating CD8 T cells in both B16 and RenCa tu- K. Skak M. Kragh Á Á Á mors; and in the RenCa model s.c. administration of Department of Cancer Pharmacology, IL-21 led to a significantly higher density of tumor Biopharmaceuticals Research Unit, Novo Nordisk A/S, + Novo Nordisk Park F6.2.30, DK 2760 Ma˚løv, Denmark infiltrating CD8 T cells compared to i.p. administra- + e-mail: [email protected] tion. The densities of CD4 T cells were unchanged following IL-21 treatments. Taken together, these data K. S. Frederiksen demonstrate that IL-21 protein has anti-tumor activity Department of Molecular Genetics, Biopharmaceuticals Research Unit, Novo Nordisk A/S, in established syngeneic tumors, and we show that Novo Alle, DK 2880 Bagsværd, Denmark IL-21 therapy markedly increases the density of tumor infiltrating CD8+ T cells. P. Thygesen Department of Exploratory ADME, Keywords Interleukin-21 Biopharmaceuticals Research Unit, Novo Nordisk A/S, Á Novo Nordisk Park, DK 2760 Ma˚løv, Denmark Tumor infiltrating lymphocytes Cancer Á Á Melanoma Renal cell carcinoma P. E. G. Kristjansen Á Department of Development Projects 05, Novo Nordisk A/S, Novo Alle, Abbreviations DK 2880 Bagsværd, Denmark IL-21 Interleukin 21 i.v. Intravenous N. Ødum i.p. Intraperitoneal Department of Molecular Biology and Physiology and Department of Medical Microbiology and Immunology, s.c. Subcutaneous University of Copenhagen, Copenhagen, Denmark TILs Tumor infiltrating lymphocytes

123

1418 Cancer Immunol Immunother (2007) 56:1417–1428

+ NK cells Natural killer cells sion of antigen stimulated CD8 cytotoxic T cells as CTLs Cytotoxic T lymphocytes well as enhance their cytolytic activity in vitro AOI Area of interest [14,17,39]. In vivo, IL-21 has been shown to enhance AUC Area under the curve the expansion, activity, and long-term survival of + WT Wild type ovalbumin-specific CD8 T cells detected in lymph LN Lymph node nodes (LNs) in mice bearing ovalbumin-expressing IP-10 Interferon-inducible protein 10 E.G7 lymphomas [20]. Based on these results we MIG Monokine induced by interferon gamma anticipate that IL-21 in vivo is able to increase the I-TAC Interferon-inducible T cell alpha number and/or reactivity of TILs. chemoattractant In previous studies of IL-21-mediated anti-tumor activity, the cytokine was primarily administered via plasmid gene delivery [38], tumor cell secretion Introduction [7,8,19,37], or used in systems where the immune response was enhanced by introduction of foreign anti- Successful immune-based cancer therapy needs to en- gens or adoptive transfer of tumor specific lymphocytes hance the interaction and/or reactivity between the [13,20,26,39]. Clearly the effects of IL-21 protein immune system and the cancer cells. In tumor immu- therapy also need to be investigated in simple nity, tumor infiltrating lymphocytes (TILs) are re- syngeneic models using more conventional routes of garded as the primary effector cells, and the number of administration, which are more clinically relevant. The TILs has previously been correlated with prolonged use of IL-21 protein therapy in a native syngeneic survival in cancer patients [6,18]. The number of tumor model could help to determine whether IL-21 is able to infiltrating CD8+ T cells and the CD8+/CD4+ T cell modulate TILs in vivo. The evaluation of s.c. admin- ratio have been shown to be independent prognostic istration of IL-21 is of major interest for the clinical factors for improved survival in several different hu- application of IL-21 because of patient convenience man cancers [21,22,28,30]. Thus, novel treatments able and since increased tolerability and sustained efficacy to increase the number of tumor-specific, tumor-infil- has been shown in clinical trials with IL-2 by this route trating immune effector cells could hold a promising of administration [11]. clinical future. In this study, we demonstrate the anti-tumor effects IL-21 is the latest member of the common c-chain- of s.c. and i.p. IL-21 protein therapy in two syngeneic dependent cytokine family and is currently in early tumor models. Our data indicate that IL-21 protein via clinical development for the treatment of cancer. IL-21 both routes of administration can inhibit established was discovered as a product of activated CD4+ T tumor growth and that s.c. administration could be helper cells, and its unique receptor IL-21R has been applicable in the clinic. Furthermore, we show that IL- identified on a broad range of immune cells including 21 therapy strongly increases the density of CD8+ TILs B, T, NK, and dendritic cells [3,23]. IL-21 has pleio- without changing the CD4+ TILs, and that the CD8+ T tropic immune modulatory activity, which has shown cells are essential for the IL-21-induced anti-tumor encouraging anti-tumor effects in several different activity. animal models [16]. CD8+ cytotoxic T lymphocytes (CTLs), NK cells, or both have been identified as the main mediators of IL-21 anti-tumor activity [16], and in Materials and methods this respect IL-21 has been suggested to play a role in the transition from innate to adaptive immunity [14]. Mice However, it remains to be shown whether its anti-tumor activity in vivo is mediated by modulation of the Wild type (WT) female C57BL/6 and BALB/c mice tumor infiltrating effector cell populations. In vitro, IL- were purchased from Taconic Europe A/S, Lille 21 stimulation induced the differentiation and matu- Skensved, Denmark, whereas female C57BL/6 nude ration of human NK cells from progenitor cells [23,31] mice (B6.Cg/Ntac-Foxn1nu N9) were acquired from and also increased the cytolytic activity of mature Taconic, Hudson, NY, USA. The animals were 6 weeks activated human NK cells [23]. Murine NK cells have old on arrival and were allowed to acclimatize for at shown more biphasic responses to IL-21 stimulation least one week before start of experiments. WT mice in vitro, depending on IL-21 concentration, co-stimuli, were housed in a standard animal facility whereas nude and cell maturation stage [14,24,36]. In mice and mice were isolated in an immunodeficient facility sep- humans, IL-21 has been shown to increase the expan- arated by a barrier. Light was controlled on a 12-h

123

Cancer Immunol Immunother (2007) 56:1417–1428 1419 light–dark cycle, and the animals were given free access Treatment with IL-21 was either initiated early with to food and drinking water. The animals were observed ~5 mm3 mean tumor volume or late with ~50 mm3 daily for clinical signs and their body weights were re- mean tumor volume. Fifty lg of IL-21 or PBS in a corded regularly. All experiments were conducted in dosing volume of 200 lL was administered either accordance with corporate and governmental policies. intraperitoneally (i.p.) or subcutaneously (s.c.) in the contralateral flank 1x/daily in the C57BL/6-B16 model Cell lines and 3x/week in the BALB/c-RenCa model. The 50 lg dose was chosen on the basis of dose-titration experi- C57BL/6 derived B16 (F0) melanoma cells (American ments prior to this work (data not shown). Termination Type Culture Collection (ATCC), CRL-6322) and criteria were a tumor volume of 1,000 mm3 or more BALB/c derived RenCa renal cell carcinoma cells than 20% weight loss from time of cell inoculation. (kindly provided by Dr. Robert H. Wiltrout, NCI at Frederick, MD, USA) were cultured in RPMI 1640 with In vivo immune cell depletion GlutaMAXTM supplemented with 10% heat-inactivated FCS, sodium pyruvate (RenCa only), non-essential ami- In vivo CD8+ and NK1.1+ cell depletion was performed no acids (RenCa only), and 5% penicillin-streptomycin by using anti-mouse CD8 (clone 2.43) and anti-mouse (all from GIBCO Cell Culture, Invitrogen, Denmark). NK1.1 (clone PK136) monoclonal antibodies, respectively, as previously reported [33]. C57BL/6 mice Reagents and antibodies inoculated s.c. in the right flank with 105 B16 melanoma cells were injected with antibodies (100 lg/ Recombinant murine IL-21 (IL-21) protein was pro- mouse) i.p. on day –1, 0, 6, and 12 in relation to the vided by Novo Nordisk A/S, Denmark and Zymoge- onset of treatment. Treatment with 50 lg IL-21 netics, Inc., WA, USA and used in all experiments. The administered s.c. was initiated when mean tumor stock solutions contained IL-21 in a concentration of volume had reached ~5 mm3 corresponding to early 5.5–10 mg/ml and working preparations were diluted in treatment. Prior to the experiment, the dose and PBS. Radioactive 125I conjugated IL-21 was produced schedule of the depleting antibodies were verified by at Novo Nordisk A/S, Denmark. The depleting anti- flow cytometry analysis showing complete depletion of mouse CD8 (clone 2.43, TIB-210 from ATCC) and CD8+ and NK1.1+ cells throughout the course of the anti-mouse NK1.1 (clone PK136, HB191 from ATCC) experiment (data not shown). monoclonal antibodies were obtained from supernatants of hybridomas cultured at Novo Nordisk A/S, In vitro tumor cell proliferation assay Denmark. The antibodies were purified in-house by affinity chromatography. For the histological exami- Potential effects of IL-21 on the growth of B16 mela- nation we used rat anti-mouse CD4 (L3T4, clone noma cells and RenCa carcinoma cells in vitro were RM4–5) (BD Pharmingen, CA, USA), rat anti-mouse examined in a colorimetric assay using the prolifera- CD8 (Ly-2, clone 53–6.7) (BD Pharmingen, CA, tion reagent 4-[3-(4-Iodophenyl)-2-(4-nitrophenyl)-2H- USA), rat serum IgG2a (Serotec, UK), and biotin- 5-tetrazolio]-1,3 benzene disulfonate (WST-1) (Roche conjugated donkey anti-rat IgG (Jackson ImmunoRe- diagnostics GmbH, Germany). Briefly, 3000 B16 or search Laboratories, Inc., PA, USA). RenCa cells/well were incubated in 96-well plates in their appropriate media and stimulated with increasing In vivo tumor models concentrations of IL-21 protein. After 48 h cell growth, cells were incubated with WST-1 for 2 h at 37�C. On day 0, C57BL/6 or BALB/c mice were inoculated Absorbance at 450 nm was used to determine the rel- s.c. in the right flank with 105 B16 melanoma or RenCa ative proliferation of cells. renal cell carcinoma cells, respectively. All mice were randomized and ear-tagged prior to treatment. The IL-21 pharmacokinetics tumor volume was measured as two perpendicular diameters approximately three times per week, and IL-21 was administered intravenous (i.v.), i.p. and s.c. calculated by the following formula: (50 lg/animal) to BALB/c mice. Mice were anesthe- tized with FORENE� Isoflurane (Abbot Scandinavia p 2 AB, Sweden) after 5 min, 15 min, 30 min, 1 h, 2 h, 4 h, Volume d d2 if d1\d2; ¼ 6 � 1 � and 8 h post injections, blood was collected from the where d represents the two diameters: retro-orbital sinus, and a DuoSet sandwich ELISA kit

123

1420 Cancer Immunol Immunother (2007) 56:1417–1428 was used to detect IL-21 protein in the serum (R&D used as secondary antibody. Sections were fixed in 4% systems, Inc., MN, USA). Briefly, goat anti-mouse IL- paraformaldehyde at 4�C and endogenous biotin 21 antibodies were coated overnight onto a 96-well activity was blocked using Biotin blocking system from plate at RT. Serum samples were diluted 1:10 or 1:100 Dako A/S, Denmark. Prior to the antibody stainings in PBS with 1% BSA and IL-21 standards were like- non-specific binding was blocked by incubation in TBS wise supplemented with normal mouse serum. A bio- with 3% skim milk, 3% BSA, and 7% donkey serum. tin-labeled goat anti-mouse IL-21 antibody was used as Incubation with the primary antibodies was made at detection antibody, and for the color reaction strepta- 4�C over night followed by 1 h incubation at RT with vidin-HRP with tetramethylbenzidine (TMB) was used the secondary antibody both diluted in TBS with 0.5% as substrate (Sigma–Aldrich Chemie GmbH, Ger- skim milk, 3% BSA, and 7% donkey serum. Strepta- many). The result was measured as absorbance at vidin conjugated alkaline phosphatase and Liquid 450 nm. The lower limit of detection was ~0.06 ng/ml Permanent Red Chromogen (Dako A/S, Denmark) with a dynamic range of 0.06–100 ng/ml. The serum was used to visualize positive cells and sections were concentration-time data were analyzed by a non-com- counterstained with Mayer’s hematoxylin to reveal partmental pharmacokinetic analysis using WinNonlin nuclei morphology. Professional (Pharsight, Inc., CA, USA) based on a sparse blood sampling schedule, where the mean serum Quantification of tumor infiltrating T cells concentration-time profiles of blood samples from three animals per time point were performed. The area A stereological method was established to quantify under the curve (AUC), a measure of drug exposure the density of tumor infiltrating T cells. Images from and the bioavailability are reported. immunohistochemically stained tumor sections were analyzed via online light microscopy at 20· magnifi- IL-21 biodistribution cation using C.A.S.T. grid software ver. 2.3.1.3 from Olympus Denmark A/S, Denmark. In all tumor sec- Mice were injected i.p. with 0.05 ml of 125I-IL-21 tions (one from each tumor) the density of tumor (~2 lCi/animal) or Na125I as control in the lower right infiltrating lymphocytes was blindly scored counting all abdominal quadrant or subcutaneously in the right foot positive cells intratumorally and relating them to an pad in order to use the popliteal LN as an exclusive area of interest (AOI), representing the total tumor lymph drainage site. After 15 min, 30 min, 1 h, 2 h, area measured stereologically, excluding necrotic tu- 4 h, 6 h, and 8 h animals were sacrificed and the fol- mor tissue and non-tumor tissue such as peritumoral lowing tissues/organs were collected and analyzed: connective tissue. The C.A.S.T. grid software and a heart, lungs, liver, spleen, kidneys, thyroidal gland, motorized stage system enabled side by side imaging to small and large intestines, mesenteric LNs, right and ensure that no area was evaluated twice or omitted. left popliteal LNs, and skin at injection site after s.c. administration, i.e. the foot. Na125I solution was used Statistics to monitor how free 125I would behave contra protein bound 125I in order to detect if the 125I from the protein Student’s t-test (two-tailed, assuming equal variance) was detached in vivo (data not shown). The c-radiation was used for statistical evaluations of differences be- was measured in all samples by a Cobra Auto-Gamma tween treated and control groups. Data are shown as gamma-counter (PerkinElmer, Inc., MA, USA) and mean ± SEM and a P value less than 0.05 was con- related to the total c-radiation of the initial injected sidered statistically significant. dose of 125I-IL-21 in %.

Immunohistochemistry of tumor inﬁltrating T cells Results

Six lm cryo-sections were made from tumor biopsies IL-21 does not inhibit B16 or RenCa tumor cell taken out at termination of the therapeutic studies. growth in vitro Sections were immunohistochemically stained with rat anti-mouse CD4 (clone RM4-5) or rat anti-mouse CD8 To determine whether IL-21 protein had any direct (clone 53-6.7) antibodies (5 lg/ml), whereas rat serum inhibitory effects on B16 or RenCa tumor cells we IgG2a was used as matching isotype control. Biotin- performed a tumor cell proliferation assay with increa- conjugated donkey anti-rat IgG (diluted 1:3,000) was sing concentrations of IL-21 (0–5,000 ng/ml). We found

123

Cancer Immunol Immunother (2007) 56:1417–1428 1421 no effects on B16 or RenCa cell growth after 48 h Subcutaneous IL-21 protein therapy significantly incubation (Fig. 1). Consistent with these results, no IL- inhibits B16 and RenCa tumor growth 21 receptor mRNA expression in either tumor cell line examined by quantitative RT-PCR analysis was found Subcutaneous syngeneic B16 and RenCa tumors were (data not shown). established in their respective hosts, C57BL/6 and BALB/c mice, by injection of 105 cells/animal. In both models IL-21 treatment was initiated either early, when tumors were just palpable (~5 mm3 mean tumor volume), or late, when tumors were more established 3 3 (~50 mm mean tumor volume). B16 Early treatment with IL-21 using either route of 2.5 RenCa administration showed significant (P < 0.001) growth 2 inhibition of B16 melanoma tumors (Fig. 2a). In the early treatment of RenCa carcinomas only s.c. admin- 1.5 istration of IL-21 showed significant growth inhibition +/- SEM 1 (P < 0.001), whereas i.p. administration showed a

Absorbance 450 nm 450 Absorbance borderline significant growth inhibition (P = 0.06) 0.5 (Fig. 2b). The difference between s.c. and i.p. admin- 0 istration in the early treatment of RenCa carcinomas 0 1 10 100 1000 5000 IL-21 concentration (ng/mL) was likewise close to statistical significance (P = 0.09). In the late treatment regimens s.c. administration of Fig. 1 IL-21 does not inhibit proliferation of B16 and RenCa IL-21 showed significant (P < 0.05) growth inhibition cells in vitro. B16 cells (filled square) or RenCa cells (square) in both models, whereas i.p. administration was (3,000/well) were grown in appropriate media and stimulated with the indicated concentrations of IL-21 for 48 h. Absorbance unable to significantly inhibit tumor growth (Fig 2c, d). at 450 nm was measured after 2 h incubation with WST-1 Generally, the late treatment regimen showed less proliferation reagent. Mean ± SEM of duplicates growth inhibition than the early treatment, and s.c.

a 1100 b 600 PBS i.p. PBS i.p. )

1000 ) 3 IL-21 i.p. 3 m 900 m 500 IL-21 i.p. m m ( (

800 IL-21 s.c. IL-21 s.c. e e 400 m 700 m u M u l l M E o o

600 E v S v

S 300 - r - r / 500 / o o + + m 400 m u u

t 200 t

n 300 n a

a Treatment start

e Treatment start 200 e 100 M M 100 0 0 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 78 9 10 11 12 13 14 15 16 17 18 19 20 21 Days post B16 tumor inoculation Days post RenCa tumor inoculation

900 c 1000 PBSi.p. d PBS i.p. )

) 900 800 3 3 IL-21i.p. IL-21 i.p. m m 800 700 m m

( IL-21 s.c. ( IL-21 s.c.

700 e e 600 m m u M u 600 l M l

E 500 o E o S v S v 500 r r 400 o o +-/ +-/ 400 m m u

u 300 t Treatment start t 300 Treatment start n n a

a 200

200 e e M M 100 100 0 0 7 8 9 10 11 12 13 14 15 16 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Days post B16 tumor inoculation Days post RenCa tumor inoculation

Fig. 2 IL-21 therapy inhibits growth of syngeneic B16 melano- mean tumor volume. Fifty lg IL-21 or PBS was injected i.p. or s.c. mas and RenCa carcinomas. All animals were injected with (contralateral to the tumor site) daily in the B16 model (a and c) 105 cells s.c. in the right ﬂank and randomized prior to treatment and 3·/week in the RenCa model (b and d). Mean ± SEM, a start as indicated. Treatment was started either early (a and b) n = 12, b n = 15, c and d n = 10, P = 0.06, *P < 0.05, **P < 0.01, with ~5 mm3 mean tumor volume or late (c and d) with ~50 mm3 ***P < 0.001, compared to PBS controls by Student’s t-test

123

1422 Cancer Immunol Immunother (2007) 56:1417–1428 administration appeared to be more efficient than i.p. mice are fully NK cell competent and T cell deficient, administration, although this difference did not reach verified by flow cytometry (data not shown). This result statistical significance (Fig. 2). suggests that T cells and not NK cells are essential for Alongside these results, we monitored animal the anti-tumor activity of IL-21 in this model. To fur- weight and general health in response to IL-21 ther examine the specific cell types involved in the anti- administration and found no treatment associated ef- tumor activity, we depleted B16 tumor-bearing C57BL/ fects on animal health (data not shown), suggesting 6 WT mice with anti-CD8 and/or anti-NK1.1 antibod- that the treatment was well tolerated. ies prior to treatment start with s.c. IL-21 again initiated early. As shown in Fig. 3b, significant anti-tumor CD8+ T cells are essential for the anti-tumor activity was still exhibited in NK.1.1+ cell-depleted activity of IL-21 animals (P < 0.01) compared to vehicle controls, whereas CD8+ T cell depletion completely abrogated In order to determine which immune effector cells the tumor growth inhibition of IL-21. Together, these were important for the anti-tumor activity of IL-21 results show that CD8+ T cells are the essential cells for protein in our models, C57BL/6 nude mice and wild the anti-tumor activity of IL-21 in our model. type (WT) mice specifically depleted of CD8+ T cells or NK1.1+ cells were used. As shown in Fig. 3a, early IL-21 significantly increases the density of tumor s.c. IL-21 protein therapy was unable to inhibit B16 infiltrating CD8+ T cells tumor growth in C57BL/6 nude mice. C57BL/6 nude Based on the finding that CD8+ T cells were responsible for the anti-tumor activity in our models we examined whether IL-21 increased the number of tu- a 1600 + Vehicle s.c. mor infiltrating CD8 T cells as a possible mechanism )

3 1400 IL-21 s.c. of action. Immunohistochemistry was used to stain m

m + +

( 1200 CD4 and CD8 T cells in tumor biopsies from our e

m 1000

u in vivo therapeutic studies. One section from each tu- M l o E v S

800

- mor biopsy obtained at the end of the late treatment r / o +

m 600 experiments was stained (Fig. 2c, d). The density of u t

n 400 TILs was scored by counting all positive cells in a a Treatment start e

M 200 stereologically deﬁned area of interest (AOI) in which

0 necrotic areas and peritumoral connective tissue were 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 excluded. Thus, our results reﬂect the density of in- Days post B16 tumor inoculation tratumoral inﬁltrating lymphocytes, which are in direct

b 1000 Vehicle s.c. contact with tumor cells and have the potential of ) 3 900 IL-21 s.c. + ∆NK1.1 + ∆CD8 performing cytotoxic effects. Representative pictures m 800 m IL-21 s.c. + ∆CD8 (

of anti-CD8 stained B16 melanomas and RenCa car-

e 700 IL-21 s.c. + ∆NK1.1 m cinomas with and without IL-21 treatments are shown u IL-21 s.c.

M 600 l o E v S 500 in Fig. 4. In B16 melanomas the density of tumor - r / o + 400 +

m inﬁltrating CD4 T cells was unchanged following IL- u

t 300 +

n 21 treatments (Fig. 5a), whereas the density of CD8 T a 200 Treatment start e

M 100 cells showed a significant 7–10-fold increase (P < 0.05) 0 following i.p. and s.c. administration of IL-21 (Fig. 5b). 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 In RenCa carcinomas the density of CD4+ T cells was Days post B16 tumor inoculation likewise unchange after IL-21 treatments (Fig. 5c) and Fig. 3 The anti-tumor effect of IL-21 is CD8+ T cell dependent. the density of CD8+ T cells was increased 3-fold after C57BL/6 nude (a) or C57BL/6 WT mice (b) were injected with 5 i.p. administration (P < 0.05) and 8-fold after s.c. 10 B16 melanoma cells s.c. in the right flank and randomized administration (P < 0.01) of IL-21 (Fig 5d). Moreover, prior to treatment start as indicated. WT animals were depleted of NK1.1+ cells, CD8+ cells or both using monoclonal Ab the difference between i.p. and s.c. IL-21 administra- administrated i.p. on day –1, 0, 6 and 12 compared to treatment tion was significant (P < 0.05) reflecting the improved start. Fifty lg IL-21 or vehicle was injected s.c. (contralateral to efficacy obtained in the RenCa carcinomas with s.c. the tumor site) daily from day 3 after tumor inoculation. administration (Fig. 2d). Together these data show that Mean ± SEM, n = 10, *P < 0.05, **P < 0.01, compared to vehicle control, CD8-depleted and NK1.1 + CD8-depleted groups by IL-21 strongly increases the density of tumor infiltrat- Student’s t-test ing CD8+ T cells without affecting the CD4+ T cell

123

Cancer Immunol Immunother (2007) 56:1417–1428 1423

Fig. 4 Pictures of tumor infiltrating CD8+ T cells in B16 melanomas and RenCa carcinomas. Representative pictures of cryo-sections at ·20 magnification showing tumor infiltrating CD8+ T cells in B16 melanomas: PBS (a), i.p. IL-21 (b), s.c. IL-21 (c) and RenCa carcinomas: PBS (d), i.p. IL-21 (e), s.c. IL- 21 (f). Tumor biopsies were obtained at the end of the experiments shown in Fig. 2c, d. CD8+ T cells were stained with liquid permanent red by immunohistochemistry according to ‘‘Materials and methods’’ and they are indicated by arrows. Sections have been counterstained with Mayer’s hematoxylin yielding a blue nucleus stain

CD4+ T cells CD8+ T cells 16 a 16 b 14 (AOI) (AOI) 14 2 2 12 12 µm µm 6 6 10 10 8 8

+/- SEM 6 6 +/- SEM 4 4 2 2 Mean no. of cells/10 0 Mean no. of cells/10 0

160 c 160 d 140 (AOI) 140 2 (AOI) 120 2 120 µm µm 6 100 6 100 80 80

+/- SEM 60 +/- SEM 60 40 40 20 20

Mean no. of cells/10 0 0 Mean no. of cells/10 PBS i.p. IL-21 i.p. IL-21 s.c PBS i.p. IL-21 i.p. IL-21s.c.

Fig. 5 IL-21 increases the density of tumor inﬁltrating CD8+ T istry. All positive cells located intratumorally were counted in cells. The bar plots show the density of tumor inﬁltrating CD4+ T one section from each biopsy and related to a stereologically cells (a and c) and CD8+ T cells (b and d) in B16 melanomas (a measured area of interest (AOI) excl. necrotic areas and non- and b) and the RenCa carcinomas (c and d). Tumor biopsies tumor tissue, such as connective tissue. Bars represent mean ± were obtained at the end of the experiments shown in Fig. 2c, d SEM, n = 7–10, *P < 0.05, **P < 0.01, Student’s t-test. AOI area and stained for CD8+ and CD4+ T cells by immunohistochem- of interest

123

1424 Cancer Immunol Immunother (2007) 56:1417–1428 density, consequently increasing the CD8+/CD4+ T cell a 90 125I-IL-21 s.c. ratio in both tumor models. 80 70 Subcutaneous administration of IL-21 results in a slow release from the injection site 60 and signiﬁcant draining to regional lymph nodes 50 40 In order to examine whether differences in the 30 biodistribution of IL-21 could account for the observed Radioactivity/Dose % 20 efﬁcacy after s.c. versus i.p administration, the kinetics 10 of the two different routes of administration was 0 compared. Particularly, we wanted to investigate the 0 1 2 3 4 5 6 7 8 degree of regional LN drainage of IL-21 after s.c. Hours post injection administration, since IL-21 in this compartment could be relevant during the generation of immune b 6 125I-IL-21 s.c., right node (injection side) responses. 125I labeled IL-21 was used to measure the 5 125 kinetics of IL-21 distribution in several major organs I-IL-21 s.c., left node 125I-IL-21 i.p., right node and tissues as listed in Materials and methods. Figure 6 4 shows the distribution over time at the s.c. injection site, i.e the right foot (Fig. 6a), and in the popliteal LNs 3 (Fig. 6b). The results show that upon s.c. injection 125I-IL-21 is released slowly from the injection site and 2

a considerable amount is drained through the regional Radioactivity/Dose % 1 LN (Cmax ~ 4.7% of injected dose). In contrast, very low levels was detected in the popliteal LN node after 0 i.p. injection and in the contralateral LN after s.c. 0 1 2 3 4 5 6 7 8 injection (Cmax < 0.1% of injected dose). We found no Hours post injection specific retention or major differences in the biodis- 125 tribution between i.p. and s.c. over time in the major Fig. 6 I-IL-21 shows slow release and significant lymph drainage following s.c. administration. 125I-IL-21 was injected organs: heart, lungs, liver, spleen, kidneys, and small either i.p. or s.c. in the right footpad in BALB/c mice, and and large intestines (data not shown). The thyroid animals were sacrificed at indicated time points and c-radiation gland generally showed an increased radioactivity over at the subcutaneous injection site (a) and in popliteal LNs (b) time after both routes of administration due to its was measured. Each data point represents the mean ± SEM of triplicate mice retention of free 125I.

IL-21 has a greater bioavailability and higher peak 10000 IL-21 i.p. serum concentration after subcutaneous IL-21 s.c. administration 1000 IL-21 i.v.

Next, we examined the serum concentration-time 100 proﬁles of IL-21 after i.v., i.p. and s.c. administration to examine whether the pharmacokinetics of IL-21 con- +/-SEM 10 tributes to the observed efﬁcacy differences. IL-21 was 1

detectable in serum after 5 min with all administration Serum IL- 21 conc. (ng/mL) routes. A peak serum concentration of 113 ng/ml of 0.1 IL-21 was reached 1 h post i.p administration, whereas 0 1 2 3 4 5 6 7 8 s.c. administration provided a higher peak IL-21 con- Hours post injection centration of 230 ng/ml 2 h post injection (Fig. 7). S.c. Fig. 7 Greater bioavailability and peak serum conc. of IL-21 administration resulted in a more sustained concen- after s.c. administration. Fifty lg IL-21 was injected i.p. in the tration of IL-21 in serum and 2.5-fold higher AUC lower right abdominal quadrant, s.c. in the right ﬂank or i.v. in compared to i.p administration (665 h ng/ml vs. the tail vein of BALB/c mice. Blood samples were drawn at · indicated time points and the serum concentration of IL-21 was 242 h·ng/ml after s.c. and i.p. administration, respec- measured by ELISA. Each data point represents mean ± SEM tively). The AUC after i.v. administration was even of triplicate mice

123

Cancer Immunol Immunother (2007) 56:1417–1428 1425 higher (1,010 h·ng/ml) due to the very high initial which may also have contributed to an improved anti- IL-21 concentration that within 1 h decreased to a level tumor immune response. IL-21 has been shown to ex- approximately 10-fold lower than both i.p. and s.c pand antigen stimulated CD8+ T cells [17,39] and this administrations (Fig. 7). The bioavailability was 24% effect might be further improved by the direct stimu- and 66% for i.p. and s.c. administration, respectively. lation of CD8+ T cells in LNs during an immune re- These results indicate that IL-21 has a prolonged sponse. This notion is supported by our finding that pharmacokinetic profile following s.c. administration CD8+ T cells were indispensable for the reduced tumor with a higher peak serum concentration and greater growth observed after IL-21 treatment. Interestingly, it bioavailability compared to i.p. administration. has also been shown that the site of immunization produces a subsequent site-specific homing of T cells and accompanying anti-tumor activity [5]. Thus, s.c. Discussion administration of IL-21 might also mount an immune response targeted more specifically against the s.c. Previous studies of IL-21 anti-tumor activity have been compartment, where the tumors in this study were conducted with IL-21 secreting tumors [7,8,19,37], IL- located. In the B16 model the efficacy difference 21 expression plasmids [38], in model systems immu- between s.c. and i.p. was less pronounced compared to nogenically enhanced with foreign antigens, or by the the more immunogenic RenCa model. It is not clear at use of adoptive transfer of antigen specific lymphocytes this point whether this difference is caused by differ- [13,20,26,39]. Although these studies support the con- ences between the two tumor cell lines or differences cept that IL-21 aids in the destruction of cancer, these between their hosts (BALB/c and C57BL/6 mice for methods of IL-21 administration are not clinically RenCa and B16 tumors, respectively). It is possible applicable. More recently, a number of studies have that different drug delivery methods may affect tumors shown anti-tumor effects of intraperitoneal adminis- with different immunogenicity differently, as recently tration of recombinant IL-21 protein [13,32,35]. In described [4]. Altogether, our results suggest that s.c. these studies IL-21 therapy was given either in com- administration of IL-21 could be advantageous and bination with other therapies, as prophylactic therapy deserves clinical evaluation. at or before tumor inoculation, and only for very few In our experiments we found that depletion of days duration. Here, we have used intraperitoneal and NK1.1+ cells did not reduce the anti-tumor effect of subcutaneous administration of IL-21 protein to mice IL-21, whereas depletion of CD8+ cells or growth bearing established syngeneic tumors in order to study of tumors in athymic mice completely abrogated the the effects of IL-21 alone under more therapeutically effect of IL-21. Taken together, these data show that relevant conditions, and IL-21 was administered CD8+ T cells are required for the anti-tumor activity in throughout the experiments to maximize the treatment our model, whereas NK cells played an insignificant effect. The use of s.c. administration of cytokines for role under these settings. In the literature either CD8+ cancer therapy has previously been applied success- T cells, NK cells, or both cell types have been described fully, as s.c. administration of IL-2 resulted in less ad- as required for the IL-21 anti-tumor activity depending verse events yet maintaining efficacy and enabling on the models used [16]. Most of the studies demon- outpatient treatments [11]. In addition, s.c. adminis- strating NK cell-mediated anti-tumor activity of IL-21 tration is believed to be more convenient for the have been performed in i.v. metastasis models where patients. We report that IL-21 protein therapy signifi- IL-21 was administered at the time of tumor establish- cantly inhibits tumor growth in two syngeneic tumor ment or in models using IL-21-expressing tumors models. The effect was not due to a direct inhibition of [2,19,34,35,37]. Two major differences might explain tumor cell proliferation, as also shown previously with these different effector mechanisms. First, we have B16 tumor cells [38]. Our data indicate that s.c. treated subcutaneous tumors, which are less accessible administration of IL-21 in the RenCa model is at least to NK cells compared to intravenous tumors. Second, in as effective as or perhaps more effective than i.p. our experiments treatment was initiated only after the administration with early as well as late therapy initi- tumors became established. The larger tumor burden in ation. This might be due to prolonged availability of this setting might overwhelm the effect of NK cells as IL-21, since s.c. administration of IL-21 resulted in a they are unable to clonally expand. In the i.v. metastasis more prolonged presence of the protein in serum and model where IL-21 therapy is given at the time of tumor higher bioavailability compared to i.p. administration. inoculation, NK-mediated killing of the tumor cells Furthermore s.c. administration also resulted in a sig- may eradicate the tumor cells almost completely before nificant passage of IL-21 through regional lymphatics, an adaptive immune response can evolve, consistent

123

1426 Cancer Immunol Immunother (2007) 56:1417–1428 with the notion that the efficacy in this model was of antigen-stimulated CD8+ T cells [17,20,39], which in sustained in Rag-1-/- mice [2]. In a study by Ma et al. turn could yield a greater pool of tumor specific T cells [19] rejection of IL-21 secreting B16F1 tumor cells infiltrating the tumor. Another possibility is that IL-21 inoculated subcutaneously required both NK cells and sustains the survival of CD8+ T cells infiltrating the CD8+ T cells. In this model, NK cell-depletion resulted tumor, which has been supported by both in vitro and in rapid growth of the tumors, suggesting an early role in vivo data [1,20]. Particularly, it has been shown that of NK cells, whereas CD8+ T cell-depletion resulted IL-21 is able to sustain CD28 expression on CD8+ T in delayed tumor growth, suggesting that adaptive cells and increase their IL-2 production [1], both of immunity might have played a role later on to kill which could enhance their survival in a challenging remaining tumor cells, as suggested by the authors [19]. tumor environment [25]. In this context, it should be By contrast Wang et al. showed that depletion of NK noted that our results represent the conditions about cells completely abrogated the anti-tumor effect of 8–10 days after the first IL-21 dose, and one day after IL-21 expressing plasmids administered day 5 and 12 the last dose, thus the effects of IL-21 have been sus- post tumor inoculation with MCA205 fibrosarcomas tained for more than a week from the initial stimula- [38]. Although these results apparently contradict the tion. Finally, it is possible that IL-21 increases the hypothesis suggested by us and others [4,19] that NK specific homing of CD8+ T cells into tumors. IL-21 has cells primarily play a role early in the anti-tumor been shown to increase CXC chemokines such as IP-10, response, it is possible that the much slower growth rate MIG and I-TAC in IL-21-secreting tumors [8], which of MCA205 cells compared to B16 cells allows NK cells could work as T cell attractants [27], but to date there to kill MCA205 cells before the tumor burden becomes are no data demonstrating that systemic IL-21 delivery too large, even though treatment is initiated day 5. improves chemokine-mediated homing of CD8+ T cells Also, differences in immunogenicity and expression of to tumors. Presently, the mechanism of IL-21 anti- ligands for activating and inhibitory NK cell receptors tumor activity remains to be fully elucidated, but the between the two tumor lines might contribute to these finding that IL-21 increased tumor infiltration of CD8+ different results. T cells is a step forward that encourages the use of IL-21 Several studies have shown that IL-21 can stimulate in oncology. CD8+ T cells to proliferate, increase their cytotoxicity, In conclusion, we have shown that IL-21 protein and sustain survival [1,14,17,20,39]. In order to elicit mono-therapy inhibited established syngeneic tumor tumor cytotoxicity, CD8+ T cells must infiltrate the tu- growth in two preclinical models of melanoma and mor to come in close contact with the tumor cells. The renal cell carcinoma. Furthermore, we have shown that benefit of TILs and specifically CD8+ TILs and the IL-21 markedly increased the density of tumor-infil- CD8+/CD4+ TIL ratio have also been shown in several trating CD8+ T cells, which are essential for the anti- different human cancers where they were predictive of tumor effect. These findings support the use of IL-21 as improved survival [9,12,21,22,28,30]. In this study, we a promising new anti-cancer drug. demonstrate that IL-21 protein given therapeutically significantly increased the density of tumor infiltrating Acknowledgement We would like to thank Heidi Winther, CD8+ T cells without changing the CD4+ T cell density, Bodil Andreasen, Birte Jørgensen and Kirsten Meeske for + + technical assistance with the experiments, and Mark Smyth for consequently increasing the CD8 /CD4 T cell ratio in valuable discussion of the manuscript. both models. Interestingly, the density of CD8+ TILs was significantly higher after s.c. administration of IL-21 compared to i.p. administration in the RenCa carcino- References mas, supporting our notion that s.c. administration might be favorable. Generally, the RenCa carcinomas 1. Alves NL, Arosa FA, van Lier RA (2005) IL-21 sustains CD28 expression on IL-15-activated human naive CD8+ T also showed approximately 10-fold higher density of cells. J Immunol 175:755–762 infiltrating T cells compared to B16 tumors reflecting 2. Brady J, Hayakawa Y, Smyth MJ, Nutt SL (2004) IL-21 the higher inherent immunogenicity of RenCa as pre- induces the functional maturation of murine NK cells. viously reported [15,29]. In a recently published article J Immunol 172:2048–2058 + 3. Brandt K, Bulfone-Paus S, Jenckel A, Foster DC, Paus R, a similar increase in CD8 T cell infiltration was ob- Ruckert R (2003) Interleukin-21 inhibits dendritic cell- served in a small number of tumors following challenge mediated T cell activation and induction of contact hyper- with an IL-21 expressing mouse bladder cancer [10]. sensitivity in vivo. J Invest Dermatol 121:1379–1382 There are several possible explanations for the 4. Cappuccio A, Elishmereni M, Agur Z (2006) Cancer + immunotherapy by interleukin-21: potential treatment increased density of CD8 TILs after IL-21 stimulation. strategies evaluated in a mathematical model. Cancer Res First, IL-21 has been shown to increase the proliferation 66:7293–7300

123

Cancer Immunol Immunother (2007) 56:1417–1428 1427

5. Chang CJ, Tai KF, Roffler S, Hwang LH (2004) The IL-21 activates both innate and adaptive immunity to gen- immunization site of cytokine-secreting tumor cell vaccines erate potent antitumor responses that require perforin but influences the trafficking of tumor-specific T lymphocytes are independent of IFN-gamma. J Immunol 171:608–615 and antitumor efficacy against regional tumors. J Immunol 20. Moroz A, Eppolito C, Li Q, Tao J, Clegg CH, Shrikant PA 173:6025–6032 (2004) IL-21 enhances and sustains CD8+ T cell responses to 6. Clemente CG, Mihm MC Jr, Bufalino R, Zurrida S, Collini achieve durable tumor immunity: comparative evaluation of P, Cascinelli N (1996) Prognostic value of tumor infiltrating IL-2, IL-15, and IL-21. J Immunol 173:900–999 lymphocytes in the vertical growth phase of primary cuta- 21. Naito Y, Saito K, Shiiba K, Ohuchi A, Saigenji K, Nagura H, neous melanoma. Cancer 77:1303–1310 Ohtani H (1998) CD8+ T cells infiltrated within cancer cell 7. Comes A, Rosso O, Orengo AM, Di Carlo E, Sorrentino C, nests as a prognostic factor in human colorectal cancer. Meazza R, Piazza T, Valzasina B, Nanni P, Colombo MP, Cancer Res 58:3491–3494 Ferrini S (2006) CD25+ regulatory T cell depletion augments 22. Pages F, Berger A, Camus M, Sanchez-Cabo F, Costes A, immunotherapy of micrometastases by an IL-21-secreting Molidor R, Mlecnik B, Kirilovsky A, Nilsson M, Damotte D, cellular vaccine. J Immunol 176:1750–1758 Meatchi T, Bruneval P, Cugnenc PH, Trajanoski Z, Fridman 8. Di Carlo E, Comes A, Orengo AM, Rosso O, Meazza R, WH, Galon J (2005) Effector memory T cells, early metas- Musiani P, Colombo MP, Ferrini S (2004) IL-21 induces tasis, and survival in colorectal cancer. N Engl J Med tumor rejection by specific CTL and IFN-gamma-dependent 353:2654–2666 CXC chemokines in syngeneic mice. J Immunol 172:1540– 23. Parrish-Novak J, Dillon SR, Nelson A, Hammond A, Spre- 1547 cher C, Gross JA, Johnston J, Madden K, Xu W, West J, 9. Diederichsen AC, Hjelmborg JB, Christensen PB, Zeuthen Schrader S, Burkhead S, Heipel M, Brandt C, Kuijper JL, J, Fenger C (2003) Prognostic value of the CD4+/CD8+ ratio Kramer J, Conklin D, Presnell SR, Berry J, Shiota F, Bort S, of tumour infiltrating lymphocytes in colorectal cancer and Hambly K, Mudri S, Clegg C, Moore M, Grant FJ, Lofton- HLA-DR expression on tumour cells. Cancer Immunol Im- Day C, Gilbert T, Rayond F, Ching A, Yao L, Smith D, munother 52:423–428 Webster P, Whitmore T, Maurer M, Kaushansky K, Holly 10. Furukawa J, Hara I, Nagai H, Yao A, Oniki S, Fujisawa M RD, Foster D (2000) Interleukin 21 and its receptor are in- (2006) Interleukin-21 gene transfection into mouse bladder volved in NK cell expansion and regulation of lymphocyte cancer cells results in tumor rejection through the cytotoxic function. Nature 408:57–63 T lymphocyte response. J Urol 176:1198–1203 24. Parrish-Novak J, Foster DC, Holly RD, Clegg CH (2002) 11. Geertsen PF, Gore ME, Negrier S, Tourani JM, von der MH Interleukin-21 and the IL-21 receptor: novel effectors of NK (2004) Safety and efficacy of subcutaneous and continuous and T cell responses. J Leukoc Biol 72:856–863 intravenous infusion rIL-2 in patients with metastatic renal 25. Petrulio CA, Kim-Schulze S, Kaufman HL (2006) The tu- cell carcinoma. Br J Cancer 90:1156–1162 mour microenvironment and implications for cancer immu- 12. Haanen JB, Baars A, Gomez R, Weder P, Smits M, de Gruijl notherapy. Expert Opin Biol Ther 6:671–684 TD, von Blomberg BM, Bloemena E, Scheper RJ, van Ham 26. Roda JM, Parihar R, Lehman A, Mani A, Tridandapani S, SM, Pinedo HM, van den Eertwegh AJ (2006) Melanoma- Carson WE III (2006) Interleukin-21 enhances NK cell specific tumor-infiltrating lymphocytes but not circulating activation in response to antibody-coated targets. J Immunol melanoma-specific T cells may predict survival in resected 177:120–129 advanced-stage melanoma patients. Cancer Immunol Im- 27. Romagnani P, Annunziato F, Lazzeri E, Cosmi L, Beltrame munother 55:451–458 C, Lasagni L, Galli G, Francalanci M, Manetti R, Marra F, 13. He H, Wisner P, Yang G, Hu HM, Haley D, Miller W, O’hara Vanini V, Maggi E, Romagnani S (2001) Interferon-induc- A, Alvord WG, Clegg CH, Fox BA, Urba WJ, Walker EB ible protein 10, monokine induced by interferon gamma, and (2006) Combined IL-21 and Low-Dose IL-2 therapy induces interferon-inducible T-cell alpha chemoattractant are pro- anti-tumor immunity and long-term curative effects in a duced by thymic epithelial cells and attract T-cell receptor murine melanoma tumor model. J Transl Med 4:24 (TCR) alphabeta+ CD8+ single-positive T cells, TCRgam- 14. Kasaian MT, Whitters MJ, Carter LL, Lowe LD, Jussif JM, madelta+ T cells, and natural killer-type cells in human Deng B, Johnson KA, Witek JS, Senices M, Konz RF, thymus. Blood 97:601–607 Wurster AL, Donaldson DD, Collins M, Young DA, Grusby 28. Sato E, Olson SH, Ahn J, Bundy B, Nishikawa H, Qian F, MJ (2002) IL-21 limits NK cell responses and promotes Jungbluth AA, Frosina D, Gnjatic S, Ambrosone C, Kepner antigen-specific T cell activation: a mediator of the transition J, Odunsi T, Ritter G, Lele S, Chen YT, Ohtani H, Old LJ, from innate to adaptive immunity. Immunity 16:559–569 Odunsi K (2005) Intraepithelial CD8+ tumor-infiltrating 15. Krup OC, Kroll I, Bose G, Falkenberg FW (1999) Cytokine lymphocytes and a high CD8+/regulatory T cell ratio are depot formulations as adjuvants for tumor vaccines. I. associated with favorable prognosis in ovarian cancer. Proc Liposome-encapsulated IL-2 as a depot formulation. J Im- Natl Acad Sci USA 102:18538–18543 munother 22:525–538 29. Scheffer SR, Nave H, Korangy F, Schlote K, Pabst R, Jaffee 16. Leonard WJ, Spolski R (2005) Interleukin-21: a modulator of EM, Manns MP, Greten TF (2003) Apoptotic, but not ne- lymphoid proliferation, apoptosis and differentiation. Nat crotic, tumor cell vaccines induce a potent immune response Rev Immunol 5:688–698 in vivo. Int J Cancer 103:205–211 17. Li Y, Bleakley M, Yee C (2005) IL-21 influences the fre- 30. Schumacher K, Haensch W, Roefzaad C, Schlag PM (2001) quency, phenotype, and affinity of the antigen-specific CD8 Prognostic significance of activated CD8(+) T cell infiltra- T cell response. J Immunol 175:2261–2269 tions within esophageal carcinomas. Cancer Res 61:3932– 18. Lipponen PK, Eskelinen MJ, Jauhiainen K, Harju E, Terho 3936 R (1992) Tumour infiltrating lymphocytes as an independent 31. Sivori S, Cantoni C, Parolini S, Marcenaro E, Conte R, prognostic factor in transitional cell bladder cancer. Eur J Moretta L, Moretta A (2003) IL-21 induces both rapid Cancer 29A:69–75 maturation of human CD34+ cell precursors towards NK 19. Ma HL, Whitters MJ, Konz RF, Senices M, Young DA, cells and acquisition of surface killer Ig-like receptors. Eur J Grusby MJ, Collins M, Dunussi-Joannopoulos K (2003) Immunol 33:3439–3447

123

1428 Cancer Immunol Immunother (2007) 56:1417–1428

32. Smyth MJ, Crowe NY, Godfrey DI (2001) NK cells and NKT 37. Ugai S, Shimozato O, Kawamura K, Wang YQ, Yamaguchi cells collaborate in host protection from methylcholan- T, Saisho H, Sakiyama S, Tagawa M (2003) Expression of the threne-induced ﬁbrosarcoma. Int Immunol 13:459–463 interleukin-21 gene in murine colon carcinoma cells gener- 33. Smyth MJ, Kelly JM, Baxter AG, Korner H, Sedgwick JD ates systemic immunity in the inoculated hosts. Cancer Gene (1998) An essential role for tumor necrosis factor in natural Ther 10:187–192 killer cell-mediated tumor rejection in the peritoneum. J Exp 38. Wang G, Tschoi M, Spolski R, Lou Y, Ozaki K, Feng C, Kim Med 188:1611–1619 G, Leonard WJ, Hwu P (2003) In vivo antitumor activity of 34. Smyth MJ, Wallace ME, Nutt SL, Yagita H, Godfrey DI, interleukin 21 mediated by natural killer cells. Cancer Res Hayakawa Y (2005) Sequential activation of NKT cells and 63:9016–9022 NK cells provides effective innate immunotherapy of cancer. 39. Zeng R, Spolski R, Finkelstein SE, Oh S, Kovanen PE, J Exp Med 201:1973–1985 Hinrichs CS, Pise-Masison CA, Radonovich MF, Brady JN, 35. Takaki R, Hayakawa Y, Nelson A, Sivakumar PV, Hughes S, Restifo NP, Berzofsky JA, Leonard WJ (2005) Synergy of Smyth MJ, Lanier LL (2005) IL-21 enhances tumor rejection IL-21 and IL-15 in regulating CD8+ T cell expansion and through a NKG2D-dependent mechanism. J Immunol function. J Exp Med 201:139–148 175:2167–2173 36. Toomey JA, Gays F, Foster D, Brooks CG (2003) Cytokine requirements for the growth and development of mouse NK cells in vitro. J Leukoc Biol 74:233–242

123

Paper III:

Intratumoral interleukin 21 increases anti-tumor immunity, tumor-infiltrating CD8+ T cell density and activity, and enlarges draining lymph nodes

Søndergaard H., Galsgaard E.D., Bartholomæussen M., Ødum N. and Skak K., J Immunother in press

Intratumoral interleukin 21 increases anti-tumor immunity, tumor-infiltrating CD8+ T cell density and activity, and enlarges draining lymph nodes*

Henrik Søndergaard1, Elisabeth D. Galsgaard1, Monica Bartholomæussen1, Per Thor Straten2, Niels Ødum3, Kresten Skak1 1Biopharmaceuticals Research Unit, Novo Nordisk A/S, Måløv Denmark, 2Center for Cancer 3 Immunotherapy, Department of Hematology, Herlev University Hospital, Denmark, Institute of Biology, University of Copenhagen, Denmark

Corresponding author: Henrik Søndergaard, M.Sc. Novo Nordisk A/S Department of Immunopharmacology Novo Nordisk Park F6.2.30 DK-2760, Måløv, Denmark Phone (direct): +45 44431376 Fax: +45 44434537 E-mail: [email protected]

*This work was funded by Novo Nordisk A/S.

Short running title: Intratumoral IL-21 increases anti-tumor immunity, TIL activity and density, and enlarges LNs

Key words: intratumoral, interleukin-21, tumor-infiltrating lymphocytes, T cells, draining lymph nodes, cancer, melanoma, renal cell carcinoma

Abbreviations: IL-21, interleukin 21; SC, subcutaneous; IT, intratumoral; TILs, tumor-infiltrating lymphocytes; NK cells, natural killer cells; Tregs, regulatory T cells; AOI, area of interest; WT, wild type; LN, lymph node; CXCL, chemokine CXC motif ligand; CCL, C-C motif ligand

Abstract

Interleukin (IL)-21 is a novel cytokine in clinical development for the treatment of cancer. In this study, we have compared the efficacy of subcutaneous (SC) and intratumoral (IT) administration of IL-21 protein in two syngeneic mouse tumor models, RenCa renal cell carcinoma and B16 melanoma, and investigated the mechanisms by which IL-21 enhances CD8+ T cell-mediated anti- tumor immunity. We found that in comparison to SC administration, IT administration of IL-21 more potently inhibited tumor growth and increased survival. This correlated with increased densities of tumor- infiltrating CD8+ and CD4+CD25- T cells, but not CD4+CD25+FoxP3+ T cells. Furthermore, IT administration of IL-21 increased degranulation, and expression of IFNγ and granzyme B in tumor- infiltrating CD8+ T cells. Tumors injected with IL-21 grew slower than contralateral tumors, suggesting that the increased efficacy of IT administration of IL-21 was due to a local rather than systemic effect. IT administration of IL-21 led to enlarged tumor-draining lymph nodes (LNs), with increased naïve lymphocyte numbers and proliferation of activated lymphocytes, suggesting that local administration of IL-21 generally benefits the tumor microenvironment and activates tumor- draining LNs. Overall, our data suggest that IL-21 augments CD8+ T cell-mediated anti-tumor immunity through increased proliferation and effector function and acts both on tumor-infiltrating CD8+ T cells as well as on the draining lymph nodes. IT administration led to superior CD8+ T cell proliferation, effector function and anti-tumor efficacy, suggesting that IT administration of IL-21 may be clinically useful in patients with unresectable tumors.

Introduction as Tregs, the release of immune suppressive soluble mediators e.g. IL-10 and TGFβ, and The finding of tumor-infiltrating lymphocytes generation of antigen-loss variants. Clearly, (TILs) in human cancers shows that the TILs are faced with many hurdles in the immune system to some degree recognizes tumor microenvironment that hinder cancers. Several reports have shown the successful immunotherapy. Therefore, novel benefit of TILs in human cancers, particularly measures are needed that can boost both the the number of CD8+ T cells and an increased number and reactivity of TILs and at the same ratio of CD8+/regulatory T cells (Tregs) are time avoid infiltration of suppressive immune associated with improved patient prognosis cells. (1-10). Adoptive cell therapy (ACT) with ex Interleukin 21 (IL-21) is the latest member of vivo expanded autologous TILs has shown the common γ-chain cytokine family and is encouraging response rates, indicating the currently in clinical development for the potential of trying to enhance TIL numbers treatment of cancer. IL-21 is produced by (11). However, far from all cancers contain activated CD4+ T cells and is particularly adequate amounts of TILs, and the anti-tumor found in follicular helper T cells and Th17 activity of TILs are rarely sufficient to control cells, and activated NKT cells (13-19). The the disease. This is mainly due to a variety of IL-21 receptor (IL-21R) is widely expressed inherent mechanisms that facilitate tumor throughout the immune system including immune escape (12): Insufficient T cell macrophages, B, T, NK, NKT and dendritic chemotaxis, lack of co-stimulation and cells (20). Hence, IL-21 has pleiotropic increased inhibitory ligand expression, co- actions on the immune system which include infiltration of suppressive immune cells such co-stimulation of B cell maturation and Ig

production, increased NK cell cytotoxicity increased density and activation of tumor- and bone marrow recruitment, co-stimulation infiltrating CD8+ T cells, suggesting that local of T cell proliferation as well as CD8+ T cell administration of IL-21 could improve patient expansion and cytotoxicity (20). In vivo, IL- outcome and is warranted for clinical 21 has shown encouraging anti-tumor activity investigation. in several different animal models, where the anti-tumor activity was mainly dependent on + Materials and Methods NK cells, CD8 T cells or both (21). Currently, IL-21 is in phase II clinical trials Mice for the treatment of stage IV melanoma and Wild type (WT) C57BL/6 and BALB/c renal cell carcinoma (22). So far, clinical female mice were purchased from Taconic results have shown that the drug is generally Europe A/S, Lille Skensved, Denmark, well tolerated (23), and immune activation whereas female C57BL/6nu/nu (nude) mice and signs of clinical activity have also been (B6.Cg/Ntac-Foxn1nu N9) and female observed (22;24). To support the further β -/- tm1Jae development of IL-21 as an oncology drug it C57BL/6 2m (B6.129-B2m N12) were acquired from Taconic, Hudson, NY, USA. will be important to fully understand the The animals were 6-8 weeks old on arrival pleiotropy of this cytokine to yield the best and were allowed to acclimatize for at least clinical outcome, hereby understand the one week before start of experiments. WT therapeutic functions of the cytokine in vivo, mice were housed in a standard animal establish the most optimal route of β -/- administration and possibly identify potential facility whereas nude and 2m mice were biological markers of efficacy. isolated in a facility for immunodeficient Previously, we have shown that subcutaneous animals. Light was controlled on a 12 hr light- administration of IL-21 protein was more dark cycle, and the animals were given free effective than intraperitoneal administration access to food and drinking water. The and led to increased densities of tumor- animals were observed daily for clinical signs infiltrating CD8+ T cells in syngeneic mouse and their body weights were recorded tumor models (25). However, it still remains regularly. All experiments were approved by to be shown whether a local administration of the relevant ethical committees and conducted IL-21 can further augment the anti-tumor in accordance with corporate policies on activity compared to systemic treatment, and animal welfare. whether it would be even more potent in modulating the density and activity of tumor- Cell lines infiltrating T cells. IL-2, which is closely C57BL/6-derived B16 (F0) melanoma cells related to IL-21 and approved as therapy for (American Type Culture Collection (ATCC), stage IV melanoma and renal cell carcinoma, CRL-6322) and BALB/c-derived RenCa renal has previously shown improved patient cell carcinoma cells (kindly provided by Dr. outcome and less severe adverse effects after Robert H. Wiltrout, NCI at Frederick, MD, USA) were cultured in RPMI 1640 with local administration compared to systemic TM treatment (26). GlutaMAX supplemented with 10% heat- In this study, we have compared intratumoral inactivated FCS, sodium pyruvate (RenCa and subcutaneous administration of IL-21 in only), nonessential amino acids (RenCa only), two syngeneic subcutaneous mouse tumor and 5% penicillin-streptomycin (all from models. Our data show potent anti-tumor GIBCO Cell Culture, Invitrogen, Denmark). efficacy and increased survival after intratumoral administration, enlargement of Reagents and antibodies tumor-draining lymph nodes (LNs) and Recombinant murine IL-21 (IL-21) protein provided by Novo Nordisk A/S, Denmark was

used in all experiments. The stock solutions subcutaneously in the right flank with 105 contained IL-21 in a concentration of 10.5 B16 melanoma cells. BALB/c mice were mg/mL and working preparations were inoculated in the right or in both flanks with diluted in PBS, IL-21-vehicle diluted 2x105 RenCa renal cell carcinoma cells. All similarly was used as control. For the mice were randomized and ear-tagged prior to histological examination we used rat anti- treatment. The tumor volume was measured mouse CD4 (clone RM4-5), rat anti-mouse as two perpendicular diameters with a digital CD8a (clone 53-6.7) (both from BD caliper approximately three times per week, Pharmingen, CA, USA), rat serum IgG2a and calculated by the following formula: (Serotec, UK), and biotin-conjugated donkey 2 anti-rat IgG (Jackson ImmunoResearch = × × < Volume 0.5 d1 d 2 , if d1 d 2 , where d represents Laboratories Inc., PA, USA). For flow cytometry analysis the following antibodies the two diameters. were used: Rat anti-mouse CD45 Allophycocyanin (APC)-Cy7 (clone 30-F11), Treatment with IL-21 was initiated when tumors reached a mean volume between 40 rat anti-mouse CD8a APC (clone 53-6.7), rat 3 3 anti-mouse CD4 Peridinin Chlorophyll mm to above 100 mm . Fifty µg IL-21 or Protein Complex (PerCP) (clone RM4-5), rat vehicle in a dosing volume of 30 µL using a anti-mouse CD19 PE-Cy7/PerCPCy5.5 (clone 0.3 ml BD Micro-Fine+ insulin syringe with a 1D3), rat anti-mouse CD25 Fluorescein 8 mm 30G needle (BD Pharmingen, CA, isothiocyanate (FITC) (clone 7D4), rat anti- USA) was administered intratumorally (IT) in mouse CD62L PE (clone MEL-14), rat anti- the center of the tumor nodule or in a volume mouse IFNγ APC (clone XMG1.2), rat IgG1, of 200 µL subcutaneously (SC) 3x/week in κ (isotype), rat anti-mouse CD107a FITC the BALB/c-RenCa model and 1x/daily in the C57BL/6-B16 model. This dose was chosen (clone 1D4B), rat IgG2a, κ FITC (isotype), 7- on the basis of experiments in our previous amino-actinomycin D (7-AAD) staining work (25). Termination criteria were a tumor solution (all from BD Pharmingen, CA, volume of 1000 mm3, which was used as a USA), hamster anti-mouse TCRβ FITC (clone surrogate survival endpoint in Kaplan-Meyer β H57-597), hamster anti-mouse TCR Alexa analysis, or more than 20% weight loss from Flour 750 (clone H57-597), rat anti-mouse time of cell inoculation. CD8 Pacific Blue (clone 53-6.7), rat anti- mouse CD44 FITC/APC (clone IM7), rat anti- Immunohistochemistry of tumor- mouse FoxP3 PE (clone FJK-16s), rat anti- infiltrating T cells mouse granzyme B PE (clone 16G6), rat Six µm cryo-sections were made from tumor- IgG2b, κ PE (isotype) (all from eBioscience, biopsies taken out at termination of the CA, USA), rat anti-mouse CD62L APC therapeutic studies. Sections were (clone MEL-14) (Beckman Coulter, CA, immunohistochemically stained with rat anti- USA), rat anti-mouse CD4 Pacific Orange mouse CD4 or rat anti-mouse CD8 antibodies (clone R2a30), rat anti-mouse CD45 Pacific (5 µg/ml), whereas rat serum IgG2a was used Orange (clone 30-F11) (both from CALTAG as matching isotype control. Biotin- laboratories, Invitrogen, CA, USA), rabbit conjugated donkey anti-rat IgG (diluted anti-rat/human KI67 FITC (clone SP6) and 1:3000) was used as secondary antibody. rabbit IgG FITC (isotype) (both from Abcam Sections were fixed in 4% paraformaldehyde plc, UK). at 4°C and endogenous biotin activity was blocked using Biotin blocking system from In vivo tumor models nu/nu Dako A/S, Denmark. Prior to the antibody On day 0, C57BL/6 WT, C57BL/6 and staining non-specific binding was blocked by C57BL/6 β2m-/- were inoculated

incubation in TBS with 3% skim milk, 3% stained with appropriate antibodies for 30-40 BSA, and 7% donkey serum. Incubation with min and washed. A FoxP3 staining kit the primary antibodies was made at 4°C over (eBioscience, CA, USA) was used for night followed by 1 h incubation at RT with intracellular staining of cells according to the the secondary antibody, both diluted in TBS manufactures instructions. Briefly, cells were with 0.5% skim milk, 3% BSA, and 7% fixed for 60 min using the Fix/Perm buffer, donkey serum. Streptavidin conjugated washed once and stained for 30 min in alkaline phosphatase and Liquid Permanent permeabilization buffer. The stained cells Red Chromogen (Dako A/S, Denmark) was were transferred to BD TruCOUNT tubes used to visualize positive cells and sections (BD Biosciences, CA, USA) for were counterstained with Mayer’s quantification of cells. An LSR II flow hematoxylin to reveal nuclei morphology. cytometer (BD Biosciences, CA, USA) was used for acquisition and data were analyzed Stereological quantification of tumor- using BD FACSDiva software (BD infiltrating T cells Biosciences, CA, USA). Live lymphocytes A stereological method was used to quantify were indentified based on FSC/SSC the density of tumor-infiltrating T cells. properties and gated as 7AAD-CD45+. The Images from immunohistochemically stained absolute number of TCRβ+CD8+, tumor sections were analyzed via online light TCRβ+CD4+, TCRβ+CD4+CD25-, microscopy at 20× magnification using TCRβ+CD4+CD25+, C.A.S.T. grid software ver. 2.3.1.3 from TCRβ+CD4+CD25+FoxP3+ and TCRβ-CD19+ Olympus Denmark A/S, Denmark. In all cells were calculated based on the number of tumor sections (one from each tumor) the acquired TruCOUNT beads related to the density of tumor-infiltrating lymphocytes was total number of beads per tube as provided by blindly scored counting all positive cells the manufacturer. The absolute number of the intratumorally and relating them to an area of different cell populations in each biopsy was interest (AOI), representing the total tumor related to the biopsy weight. CD8+ T cell area measured stereologically, excluding activation was analyzed directly ex vivo necrotic tumor tissue and non-tumor tissue detecting the percentage of CD107a+ and such as peritumoral connective tissue. The IFNγ+ TCRβ+CD8+ T cells as well as the C.A.S.T. grid software and a motorized stage median fluorescence intensity of granzyme B system enabled side by side imaging to ensure in TCRβ+CD8+ T cells, all compared to that no area was omitted or evaluated twice. matching isotype controls.

Flow cytometric analysis of tumor- Quantification of tumor-draining lymph infiltrating lymphocytes node lymphocytes Tumor-biopsies were taken out at the On day 0, BALB/c mice were inoculated termination of experiments or on day 1, 5, and subcutaneously in the right hind foot pad with 10 post treatment-start and weighed. 1x106 RenCa renal cell carcinoma cells. Mice Mechanical dissection with scalpel blades were randomized and ear-tagged prior to followed by enzymatic digestion for 1 h at treatment and tumor volume was measured 37°C with collagenase type IV (Sigma-aldrich three times per week as the perpendicular GmbH, Germany) and DNase I (Roche diameters and calculated as described above. Diagnostics GmbH, Germany) were used to Intratumoral treatment with 50 µg IL-21 was dissociate the tumor tissue. Single cell initiated approximately on day 12 post tumor suspensions were made by passage through a inoculation when the mean tumor volume > 40µm BD Falcon cell strainer (BD 40 mm3 and given 5x/week for 8 days. One Biosciences, CA, USA). Cells were surface day after the last IL-21 injection the popliteal

lymph nodes (LNs) in the tumor-bearing and Results contralateral legs were removed. LNs were forced through a 40µm BD Falcon cell Intratumoral administration of IL-21 strainer (BD Biosciences, CA, USA) using a augments anti-tumor efficacy pestle, and total viable cell numbers were Previously, we have shown superior anti- counted on a ViCell automated cell viability tumor efficacy of subcutaneous (SC) counter (Beckman Coulter, CA, USA). Cells administration of IL-21 compared to were stained with appropriate antibodies for intraperitoneal therapy in mice (25). In this 30 min. and washed. Intracellular staining study we wanted to investigate whether was done using a FoxP3 staining kit equivalent doses of IL-21 given locally could (eBioscience, CA, USA) according to further augment the anti-tumor effect of this manufactures instructions described above. cytokine. Initially, we inoculated RenCa renal Acquisition was performed on a LSR II flow cell carcinoma cells subcutaneously in cytometer (BD Biosciences, CA, USA) and syngeneic BALB/c mice. Intratumoral (IT) data were analyzed using BD FACSDiva IL-21 treatment was initiated when tumors software (BD Biosciences, CA, USA). Cell were established with a mean tumor volume frequencies obtained by flow cytometry were above 40 mm3. Our results showed that 50 µg used to calculate the absolute number of cell IL-21 administered intratumorally potently subpopulations based on the total cell count. inhibited growth of RenCa tumors compared The following gating strategy was used: FSC- to vehicle treatment (p<0.001) (Figure 1a). H/FSC-A was used to eliminate doublets and Tumor growth in mice receiving intratumoral FSC-A/SSC-A to gate live lymphocytes. administration of IL-21 was slower than in TCRβ, CD4, CD25, CD8, and CD19 were mice receiving subcutaneous IL-21, although used to discriminate main T cell populations the difference did not reach statistical and B cells. CD62L and CD44 were used to significance (Figure 1b). Titration of IL-21 further discriminate between activated and showed that 10 µg administered naive T cells. KI67 expression was used to intratumorally was equipotent to identify proliferating cells compared to a subcutaneous administration of 50 µg (Figure matching isotype control.. 1b). In this experiment, the mean tumor volume at the onset of treatment was above Statistics 100 mm3, thus the subcutaneous Student’s t-test (two-tailed, assuming equal administration of IL-21 only showed variance), two-tailed Mann-Witney U-test or borderline significant growth inhibition (p= One-way ANOVA with Tukey’s post test was 0.09), but we know from previous used for statistical evaluations of differences experiments that the starting tumor size is between treated and control groups as important for the observed effect of IL-21 due indicated in figure legends. Mantel Cox Log to shortening of the therapeutic window as all Rank test was used to evaluate statistical mice are terminated at a fixed tumor size differences in Kaplan-Meyer analyses and (1000 mm3) (25). To further investigate correlation data were compared by the whether the effect of intratumoral Spearman non-parametric correlation administration of IL-21 was due to effects in coefficient. Data are generally shown as the local tumor environment we also treated mean±SEM unless otherwise noted. animals with subcutaneous RenCa tumors Bonferroni correction was used to correct for inoculated in both flanks. The intratumorally mass significance and a P value less than 0.05 treated tumors grew significantly slower than was considered statistically significant. the contralateral tumors (Figure 1c), suggesting that a high concentration of IL-21

in the local tumor environment provides the findings in the RenCa model, tumors increased anti-tumor activity. treated with intratumoral IL-21 grew slower compared to subcutaneous treatment, and Intratumoral IL-21 increases long-term again 10 µg IL-21 administered anti-tumor effect compared to intratumorally showed equal potency to 50 µg subcutaneous administration subcutaneously (Figure 3b). We also In light of the increased efficacy on tumor- evaluated the long-term survival effects of growth inhibition of intratumoral subcutaneous and intratumoral administration administration of IL-21 compared to of IL-21 in the B16 model with similar results subcutaneous administration we next compared to the RenCa experiments, although investigated whether this effect would slightly less significant (data not shown). translate into improved long-term survival. These results indicate that RenCa tumors are Domestic ethical guidelines on animal more responsive to IL-21 treatment, and experiments prohibit use of death as an expression of MHC class I was increased on endpoint, thus Kaplan-Meyer survival RenCa cells compared to B16 cells analysis was conducted with the surrogate (Supplementary figure 1). survival endpoint ‘time to tumor size >1000 To confirm our previous findings that CD8+ T mm3’. BALB/c mice bearing subcutaneous cells were important for the anti-tumor effect RenCa tumors were treated with subcutaneous of IL-21 now using intratumoral or intratumoral administration of 50 µg IL-21. administration, we treated B16 melanoma Here, significantly increased survival time bearing C57BL/6nu/nu mice which are T cell was observed in mice injected intratumorally deficient but NK cell competent and β2M with IL-21 compared to both vehicle knockout mice which lack CD8+ T cells and treatment (p<0.01) and to subcutaneous NKT cells (Figure 3c and d). These administration (p<0.05) (Figure 2), suggesting experiments showed no effect of intratumoral that the additional anti-tumor effect of administration of IL-21 in neither mouse intratumoral administration of IL-21 improves model, suggesting that CD8+ T cells are long-term survival. Furthermore, our data required for the anti-tumor effect of IL-21 showed that once the treatment was when using intratumoral administration of IL- discontinued on day 29, regressing tumors 21. Similarly, RenCa tumors in BALB/cnu/nu started to grow again, indicating that the mice were refractory to subcutaneous IL-21 continuous presence of IL-21 was needed to treatment, as were RenCa tumors in WT sustain the anti-tumor effect (data not shown). BALB/c mice depleted of CD8+ T cells, whereas significant tumor-growth inhibition Intratumoral IL-21 increases CD8+ T cell- was maintained in mice depleted of NK cells dependent anti-tumor effect in B16 (data not shown). melanomas Following the experiments in the RenCa Intratumoral IL-21 increases the density of model we also explored whether these tumor-infiltrating T cells findings were of more general relevance and We previously found that the density of could be reproduced in the B16 melanoma tumor-infiltrating CD8+ but not CD4+ T cells model. C57BL/6 mice bearing established was significantly increased after subcutaneous subcutaneous B16 melanomas were treated IL-21 treatment (25). Here, we evaluated the with subcutaneous or intratumoral effects of intratumoral IL-21 administration administration of 50 µg IL-21. The results on the density of tumor-infiltrating CD4+ and showed that intratumoral administration of CD8+ T cells. Immunohistochemistry was IL-21 significantly inhibited B16 tumor used to stain B16 tumor-biopsies taken out at growth (p<0.001) (Figure 3a). Also, similar to the end of the experiment shown in figure 3a.

One section from each tumor-biopsy was was not significant, possibly due to large stained and the density of tumor-infiltrating T variation in the intratumoral IL-21 group. The cells was scored by counting positively density of CD4+CD25- but not CD4+CD25+ T stained cells in a stereologically measured cells also showed a significant increase after area of interest (AOI), which excluded intratumoral IL-21 treatment (data not peritumoral connective tissue and necrotic shown). To evaluate the importance of these tissue. Representative pictures are shown of TILs on disease outcome we correlated the CD8+ (Figure 4a and b) and CD4+ (Figure 4c density of the different TILs to final tumor and d) stained cells in IL-21 and vehicle volume. The density of CD8+ T cells showed treated tumors. The density of CD8+ T cells the strongest inverse correlation with final was increased 9.3 fold following intratumoral tumor volume (p=0.0005) (Figure 5b); the IL-21 treatment compared to controls density of CD4+CD25- T cells also correlated (p<0.001) (Figure 4e). Furthermore, in inversely with final tumor volume, but less contrast to our previous findings with significant than CD8+ T cells, whereas the subcutaneous IL-21, intratumoral IL-21 also density of CD4+CD25+ T cells showed no significantly increased the density of tumor- correlation (data not shown). To further infiltrating CD4+ cells (Figure 4f), although analyze the tumor-infiltration of T cells after the increase was less pronounced (2.7 fold, IL-21 treatment we took out RenCa tumor- p<0.01) compared to the increase in CD8+ T biopsies on day 1, 5 and 10 post treatment- cells. In comparison to our previous data start in a separate but similar experiment to using subcutaneous administration (25) these the one above. On day 1, there was no data show comparable levels of CD8+ and difference in the density of T cells between CD4+ T cells in vehicle treated animals and treatments (Figure 5c). On day 5 (equal to 3 indicate that intratumoral administration of doses with 3x/week treatment schedule), the IL-21 was more potent in increasing both density of CD8+ T cells was strongly CD8+ and CD4+ tumor-infiltrating T cells in increased after intratumoral IL-21 treatment B16 melanomas. compared to both vehicle and subcutaneous IL-21 treatment (p<0.01). Also, the density of The density of tumor-infiltrating CD8+ and CD4+CD25- T cells was significantly CD4+CD25- T cells, but not Tregs increases increased compared to vehicle treatment after intratumoral administration of IL-21 (p<0.05). Subcutaneous administration of IL- and correlates with tumor growth 21 did not show any significant increase in inhibition the tumor-infiltrating T cells at this time To directly compare the effects of point, suggesting a more potent and rapid intratumoral and subcutaneous administration effect on the density of tumor-infiltrating T of IL-21 on the density of TILs we quantified cells by intratumoral administration. On day TILs in RenCa tumors by flow cytometry at 10, the density of CD8+ T cells after the end of an experiment similar to figure 1b. intratumoral IL-21 was still significantly Subcutaneous administration of IL-21 increased. Now, similar to the results in figure significantly increased the density of tumor- 5a, subcutaneous administration of IL-21 also infiltrating CD8+ T cells (2.7 fold compared showed an increase in the density of CD8+ T to vehicle, p<0.05), but intratumoral cells, although only borderline significant administration of IL-21 showed an even (p=0.06) due to a smaller group size (n=4). stronger increase (10.4 fold compare to The density of CD4+CD25- T cells increased vehicle, p<0.01) (Figure 5a), equivalent to our after both subcutaneous and intratumoral immunohistochemistry results (Figure 4). The administration of IL-21, although not difference between subcutaneous and statistically significant. intratumoral administration in this experiment

We also quantified the density of administration, intratumoral administration of CD4+CD25+FoxP3+ regulatory T cells (Tregs) IL-21 significantly increased the percentage on day 1 and 10 post treatment-start (Figure of tumor-infiltrating CD8+ T cells expressing 5d). On day 1, there was no difference in the CD107a and IFNγ (p<0.05), and their density of Tregs between treatments. expression of granzyme B measured as Interestingly, on day 10 the density of Tregs median fluorescence intensity (MFI) (p<0.05) decreased after intratumoral administration of (Figure 6b). On day 10, the percentage of IL-21 despite the increased CD4+CD25- T cell CD107a expressing cells had generally density, whereas the density of Tregs after decreased, but intratumoral IL-21 still showed subcutaneous administration correlated more a significant increase compared to vehicle with the increased CD4+CD25- T cell density. (p<0.05). The MFI of granzyme B was still Although these differences did not reach slightly increased in response to subcutaneous statistical significance they indicate that IL-21 administration, whereas the expression intratumoral administration of IL-21 can following intratumoral IL-21 treatment now selectively increase the density of certain was similar to vehicle, indicating loss of tumor-infiltrating T cell subsets in RenCa granzyme B expression perhaps due to the tumors, but not Tregs, considerably increasing increased exocytosis activity. However, the the ratio of CD8+ T cells/Tregs. percentage of tumor-infiltrating CD8+ T cells expressing IFNγ following intratumoral IL-21 Intratumoral IL-21 increases IFNγ and was still significantly increased compared to granzyme B expression, as well as both subcutaneous administration and vehicle. degranulation of tumor-infiltrating CD8+ T Taken together, these data indicate that cells intratumoral administration of IL-21 also The significant effects on the densities of increases the activity of RenCa tumor- tumor-infiltrating T cells led us to examine infiltrating CD8+ T cells in comparison to whether these infiltrating T cells also showed subcutaneous administration. increased activation following intratumoral administration of IL-21. In a similar IL-21 enlarges tumor-draining lymph experiment to the one described in figure 1b, nodes by increased naïve lymphocyte we stained TILs from RenCa tumors for IFNγ, numbers and proliferation of activated granzyme B, and the degranulation-associated lymphocytes molecule CD107a on day 1, 5 and 10 post The benefits of intratumoral administration of treatment-start and analyzed the expression of IL-21 on both anti-tumor activity and TILs these molecules by flow cytometry. TILs led us to also investigate the effects on tumor- were gated into CD45+TCRβ+CD8+ T cells draining LNs. The popliteal LNs in mice and in figure 6a representative FACS plots drain the foot pad region and we have from day 5 post treatment-start show the previously shown that IL-21 is found in expression of the three molecules after popliteal LNs after injections in the foot pad vehicle, subcutaneous and intratumoral IL-21 (25). Also, LN-mapping showed that the treatment. On day 1, CD107a expression was popliteal LN was the primary draining LN of similar between treatments and minor tumors in the foot pad (Supplementary figure increases were observed in the expression of 2a). To investigate how intratumoral granzyme B and IFNγ after intratumoral IL- administration of IL-21 modulates tumor- 21, but no significant differences (Figure 6b). draining LNs we treated established RenCa On day 5, subcutaneous administration of IL- tumors in the hind foot pad of mice with 21 showed minor increases in CD107a, intratumoral IL-21 5x/week for 8 days. granzyme B and IFNγ expression, however Macroscopically, the LNs draining IL-21- compared to both vehicle and subcutaneous treated tumors were substantially enlarged

compared to LNs draining vehicle-treated 7f), this could not explain the substantial tumors (Figure 7a). The average number of increase in naïve T cell numbers following cells in the tumor-draining LNs increased IL-21 treatment, suggesting non-proliferative from 2.1x106 in vehicle to 5.7x106 in IL-21 effects of IL-21 on naïve T cells. The increase treated mice (p<0.01) (Figure 7b). in B cells also primarily concerned naïve TCRβ+CD4+CD25-, TCRβ+CD8+ and TCRβ- (CD62L+CD44-) B cells and to a minor CD19+ cells all significantly increased in extend central memory (CD62L+CD44+) B numbers, whereas the number of CD4+CD25+ cells (data not shown). However, the cells were unchanged (Figure 7b). percentage of proliferating B cells decreased Interestingly, T cells gated into naïve and those proliferating were primarily (CD62L+CD44-), central memory activated CD62L+CD44+ and CD62L-CD44+ (CD62L+CD44+) and effector memory B cells (Figure 7e and 7f), indicating that (CD62L-CD44+) subsets showed a particular proliferation could not explain the increase in increase in the number of naïve CD4+ and B cells. Still, IL-21 treatment did increase the CD8+ T cells, and in the central memory number of proliferating CD62L+CD44+ B (CD62L+CD44+) CD8+ T cells (Figure 7c). cells, indicating that IL-21 does stimulate Similar results were found after only 3 days proliferation of activated B cells in tumor- of IL-21 treatment (data not shown). IL-21 draining LNs. injection in the foot pad of naïve, non-tumor- Together, these data indicate that IL-21 bearing mice also increased lymphocyte enlarges RenCa tumor-draining LNs after numbers in popliteal LNs, but less intratumoral injections primarily by pronounced than in the context of tumor increasing naïve lymphocyte numbers and by (Supplementary figure 2b), suggesting that increased proliferation of activated IL-21 activates draining LNs, which is lymphocytes. enhanced by tumor antigen. In contrast, contralateral LNs were largely unaffected by Discussion IL-21 treatment (Supplementary figure 3). To determine the contribution of proliferation IL-21 is currently in development for the to the increase in lymphocyte numbers we treatment of cancer. To this end, it is stained RenCa tumor-draining LN important to understand the anti-tumor lymphocytes with KI67. Representative mechanisms of IL-21, establish the most FACS plots show KI67 expression in T cells + effective routes of administration, and after treatments and subdivision of the KI67 identify potential biological markers of cells based on their CD62L and CD44 efficacy to improve patient benefit. expression (Figure 7d). The percentages of + + + + Historically, IL-21 has shown anti-tumor KI67 CD4 and KI67 CD8 T cells were activity in several different preclinical tumor significantly increased after IL-21 treatment, models, including plasmid gene delivery, indicating that increased proliferation could tumor cell secretion, and in models enhanced have caused the increase in these cells in the by foreign antigens, vaccines and adoptively tumor-draining LN (Figure 7e). However, transferred tumor-specific lymphocytes (21). CD62L and CD44 expression showed that the + Although these results support the use of IL- proliferating KI67 T cells were primarily 21 in cancer immunotherapy, most studies are activated cells with a central memory difficult to translate clinically. More recently, (CD62L+CD44+) and effector memory - + several combination strategies for improving (CD62L CD44 ) phenotype (Figure 7f). IL-21 anti-tumor activity have been Although a minor increase was seen in the investigated (27), but the anti-tumor number of KI67+ naïve (CD62L+CD44-) + mechanism of IL-21 as a single therapy is still CD8 T cells after IL-21 treatment (Figure largely unknown.

Previously, we have used IL-21 protein characteristics with low MHC expression and administered by conventional routes and no transporters associated with antigen shown that subcutaneous administration of processing (TAP)-1 expression (reviewed in IL-21 had anti-tumor activity and was able to 31). We have previously shown that the increase the density of tumor-infiltrating density of tumor-infiltrating CD8+ T cells was CD8+ T cells (25). Here, we have investigated approximately 10-fold higher in RenCa whether local administration of IL-21 further compared to B16 tumors (25), and theses increases the anti-tumor effects of IL-21 findings are supported by the difference in compared to systemic treatment. The clinical MHC class I expression on our B16 and applicability of local therapy is more limited RenCa cells (Supplementary figure 1). Thus, than systemic treatment, but is valid e.g. in adaptive immune responses are likely to have unresectable metastasis in melanoma or as a greater role in immune responses against adjuvant therapy in association with surgery, RenCa tumors compared to B16 tumors. and local therapy lowers the risk of systemic So far, the anti-tumor activity of IL-21 has adverse effects. Results with IL-2, which is mainly been assigned to either NK cells, approved for the treatment of renal cell CD8+ T cells or both depending on the carcinoma and metastatic melanoma, have specific models used (21). Previously we have shown improved responses after intratumoral shown that CD8+ T cells were the main administration in metastatic skin disease of mediators of IL-21 anti-tumor activity using melanoma (28) and less severe side effects B16 melanomas, whereas NK cells were of than systemic treatment (26). This indicates less importance (25). Here, we have extended that local therapy is valid in certain clinical these data to intratumoral injection of IL-21 settings and points out the rationale for where the anti-tumor activity against B16 exploring this route of administration also for melanomas was lost in athymic mice and in IL-21. β2m-deficient mice. The mechanism of the In this study, we demonstrate the first results CD8+ T cell-dependent anti-tumor activity of using local administration of IL-21 protein in IL-21 is not understood in detail; IL-21 two different preclinical solid tumor models – increases cytotoxicity and IFNγ production RenCa renal cell carcinoma and B16 from CD8+ T cells in concert with other melanoma. Our results show that intratumoral stimuli (32-34), and co-stimulates CD8+ T administration of IL-21 potently inhibits B16 cell proliferation and antigen-specific CD8+ T and RenCa tumor growth leading to cell expansion and survival (35-38). However, prolonged long-term survival, particularly in these data have mainly been generated in vitro the RenCa model, compared to a similar dose by stimulation of CD8+ T cells from e.g. by subcutaneous administration. These results spleen or blood, and it remains to be shown indicate that IL-21 has improved anti-tumor whether IL-21 similarly can activate CD8+ T effects when present locally in the tumor cells in vivo and more importantly those environment and that local IL-21 therapy infiltrating tumors. Here, we found that IL-21 could improve patient outcome, where it is increased the surface expression of CD107a, applicable. and intracellular expression of granzyme B Generally, our data showed that RenCa and IFNγ in RenCa tumor-infiltrating CD8+ T carcinomas had increased responses to IL-21 cells analyzed directly ex vivo. Increased therapy compared to B16 melanomas, which expression of granzyme B and IFNγ in CD8+ is consistent with the literature where RenCa T cells have both been found in IL-21 clinical cells have generally been found to be more trials (22;24), suggesting that these molecules immunogenic compared to B16 cells (29;30). are very relevant markers of IL-21 activity. Specifically, B16 melanomas have been CD107a residing in cytotoxic granules is described to have very poor immunogenic exposed on the cell surface upon

degranulation and is associated with tumor vitro (36) and CD8+ T cells in vivo (37) could cytolytic activity (39). Therefore, our findings mediate an increase in the long-term tumor T suggest that IL-21 stimulates cytolytic activity cell densities. However, we observed a directly in tumors. Seeing that IL-21 activates striking increase in T cell densities already 5 tumor-infiltrating CD8+ T cells in vivo is days post treatment-start, making it less likely encouraging, and that it does so even more that increased survival was the main when given locally supports the use of mechanism, and rather points to an active intratumoral administration. response. Third, IL-21 has been shown to Although IL-21 can augment expansion, mediate the release of soluble factors involved activity and survival of CD8+ T cells, these in lymphocyte trafficking such as chemokine cells still need to get into contact with tumor CXC motif ligand (CXCL)9, CXCL10 and cells and be present in proper numbers to CXCL11 (40) and C-C motif ligand (CCL)20 mediate effective killing. Previously, we have (42), and shown up-regulation of relevant shown that subcutaneous administration of chemokine receptors in cancer patients (24), IL-21 increased the density of tumor- indicating that IL-21 could increase T cell infiltrating CD8+ T cells, which also have chemotaxis, but direct evidence of this still been indicated by others (40;41). In this remains. Finally, whether or not IL-21 in it study, we found that intratumoral IL-21 self could act as a T cell attractant is another therapy increased the density of tumor- possibility but this remains speculative. infiltrating CD8+ T cells even further Interestingly, despite increased densities of compared to subcutaneous administration in tumor-infiltrating CD8+ and CD4+CD25- T both B16 and RenCa tumors and increased the cells, the density of tumor-infiltrating Tregs density of CD4+CD25- T cells, suggesting that rather decreased following intratumoral IL-21 part of the increased anti-tumor activity of treatment in RenCa tumors. This indicates intratumoral IL-21 may be ascribed to that IL-21 can selectively increase the density increased TIL numbers. This notion is further of certain T cells in tumors without increasing supported by our finding that the density of the density of Tregs. These findings are tumor-infiltrating CD8+ T cells and to less supported by studies showing that IL-21 has extent CD4+ T cells correlated with tumor no direct effects on Tregs (43;44). Also, inhibition. Furthermore, these data suggest recent data suggest that the percentage of that the density of tumor-infiltrating CD8+ T tumor-infiltrating Tregs is decreased in IL-21 cells might be a relevant biological marker of secreting tumors (45). It is not known what IL-21 anti-tumor activity. the mechanism behind this selectivity is, but The mechanism by which IL-21 increases the IL-21 has been suggested to suppress the density of tumor-infiltrating T cells is still not expression of FoxP3 (46), and the number of clear, and could have several explanations. FoxP3+ cells in culture (38). Although we did First, IL-21 may increase the expansion of observe a lower FoxP3 expression in the antigen-specific CD8+ T cells (33;35;37;38), tumor-infiltrating CD4+CD25+ T cell which could produce a greater number of population following intratumoral IL-21 tumor-reactive cells infiltrating the tumor. treatment, this was not significant (data not Our findings that tumor-draining LNs were shown). enlarged in response to intratumoral IL-21 Nonetheless, our finding that intratumoral treatment with increased proliferation of administration of IL-21 increased the density activated T cells is supportive of this option, of CD8+ T cells in RenCa tumors, while the but the increased LN activation and tumor density of Tregs decreased, suggests a highly infiltration remains to be directly linked. increased tumor-infiltrating CD8+ T cell/Treg Second, the ability of IL-21 to sustain ratio. In the clinic, prognostic evaluations of survival of both CD4+ and CD8+ T cells in TILs have shown that not only the level of

CD8+ T cell infiltration, but especially a high B cells in the anti-tumor activity of IL-21, tumor-infiltrating CD8+ T cell/Treg ratio was which has previously been suggested (54-56), a significant independent prognostic factor of but this remains to be clarified in the B16 and improved overall survival (4;7). RenCa models and was beyond the scope of In general, our data suggest that IL-21 has this paper. improved anti-tumor activity when present in In conclusion, we have shown that compared the local tumor environment. This is to subcutaneous administration, intratumoral supported by studies of IL-21-expressing administration of IL-21 protein has superior tumors where tumors are unable develop anti-tumor activity, which may be ascribed to (40;47-49). Also, we found that in animals increased activation of tumor-infiltrating bearing subcutaneous tumors on both flanks, CD8+ T cells as well as increased density of the tumors injected intratumorally with IL-21 tumor-infiltrating CD8+ T cells, with the latter grew slower than contralateral tumors. Our being strongly associated with tumor growth finding that local administration of IL-21 inhibition. Overall, our data warrant the enlarged tumor-draining LNs, increased the clinical investigation of local IL-21 therapy number of naïve CD4+ and CD8+ T cells and and suggest that the density of tumor- the proliferation of activated T cells, is infiltrating CD8+ T cell is a relevant additional evidence that IL-21 benefits the biological marker of IL-21 anti-tumor effect. local tumor environment. Although IL-21 can co-stimulate proliferation Acknowledgements of naïve T cells (33;50), we mainly found This work was funded by Novo Nordisk A/S increased proliferation of activated T cells, and we would like to thank Birte Jørgensen, pointing to non-proliferative effects of IL-21 Heidi Winther and Tonja Lyngse Jørgensen on LN T cells. IL-21-induced chemokine for technical assistance with the experiments. secretion (40;42) could locally increase the recruitment of T cells to the draining LN. Also, IL-21 has been shown to maintain T cell References expression of CD62L and CCR7 (51;52), thereby keeping T cells in a more precursor- 1. Clemente CG, Mihm MC, Jr., Bufalino R, et al. like state which may favor their retention in Prognostic value of tumor infiltrating secondary lymphoid organs or simply skew lymphocytes in the vertical growth phase of the distribution toward more CD62L+ cells. primary cutaneous melanoma. Cancer. 1996;77:1303-1310. Still, it remains to be fully clarified what the causative mechanism is and whether this 2. Galon J, Costes A, Sanchez-Cabo F, et al. Type, increase in naïve T cells in draining LNs has density, and location of immune cells within any impact on the anti-tumor response. human colorectal tumors predict clinical Similarly, B cell numbers also significantly outcome. Science. 2006;313:1960-1964. increased in tumor-draining LNs following 3. Pages F, Berger A, Camus M, et al. Effector IL-21 treatment, and particularly naïve B cells memory T cells, early metastasis, and survival in (data not shown). But again, proliferation was colorectal cancer. N Engl J Med. mainly seen in activated B cells, showing that 2005;353:2654-2666. IL-21 also has both proliferative and non- 4. Gao Q, Qiu SJ, Fan J, et al. Intratumoral balance proliferative effects on LN B cells. IL-21 co- of regulatory and cytotoxic T cells is associated stimulates B cell proliferation, but induces B with prognosis of hepatocellular carcinoma after cell apoptosis without proper co-stimulation resection. J Clin Oncol. 2007;25:2586-2593. (53), indicating the presence of B cell- activating antigens in the RenCa tumor 5. Naito Y, Saito K, Shiiba K, et al. CD8+ T cells model. This also points to a possible role for infiltrated within cancer cell nests as a

prognostic factor in human colorectal cancer. production in a STAT3-dependent manner. J Cancer Res. 1998;58:3491-3494. Biol Chem. 2007;282:34605-34610.

6. Piersma SJ, Jordanova ES, van Poelgeest MI, et 16. Wurster AL, Rodgers VL, Satoskar AR, et al. al. High number of intraepithelial CD8+ tumor- Interleukin 21 is a T helper (Th) cell 2 cytokine infiltrating lymphocytes is associated with the that specifically inhibits the differentiation of absence of lymph node metastases in patients naive Th cells into interferon gamma-producing with large early-stage cervical cancer. Cancer Th1 cells. J Exp Med. 2002;196:969-977. Res. 2007;67:354-361. 17. Chtanova T, Tangye SG, Newton R, et al. T 7. Sato E, Olson SH, Ahn J, et al. Intraepithelial follicular helper cells express a distinctive CD8+ tumor-infiltrating lymphocytes and a high transcriptional profile, reflecting their role as CD8+/regulatory T cell ratio are associated with non-Th1/Th2 effector cells that provide help for favorable prognosis in ovarian cancer. Proc Natl B cells. J Immunol. 2004;173:68-78. Acad Sci U S A. 2005;102:18538-18543. 18. Coquet JM, Kyparissoudis K, Pellicci DG, et al. 8. Sharma P, Shen Y, Wen S, et al. CD8 tumor- IL-21 Is Produced by NKT Cells and Modulates infiltrating lymphocytes are predictive of NKT Cell Activation and Cytokine Production. J survival in muscle-invasive urothelial Immunol. 2007;178:2827-2834. carcinoma. Proc Natl Acad Sci U S A. 2007;104:3967-3972. 19. Harada M, Magara-Koyanagi K, Watarai H, et al. IL-21-induced Bepsilon cell apoptosis 9. Schumacher K, Haensch W, Roefzaad C, et al. mediated by natural killer T cells suppresses IgE Prognostic significance of activated CD8(+) T responses. J Exp Med. 2006;203:2929-2937. cell infiltrations within esophageal carcinomas. Cancer Res. 2001;61:3932-3936. 20. Spolski R, Leonard WJ. Interleukin-21: basic biology and implications for cancer and 10. Zhang L, Conejo-Garcia JR, Katsaros D, et al. autoimmunity. Annu Rev Immunol. 2008;26:57- Intratumoral T cells, recurrence, and survival in 79. epithelial ovarian cancer. N Engl J Med. 2003;348:203-213. 21. Leonard WJ, Spolski R. Interleukin-21: a modulator of lymphoid proliferation, apoptosis 11. Rosenberg SA, Restifo NP, Yang JC, et al. and differentiation. Nat Rev Immunol. Adoptive cell transfer: a clinical path to 2005;5:688-698. effective cancer immunotherapy. Nat Rev Cancer. 2008;8:299-308. 22. Davis ID, Brady B, Kefford RF, et al. Clinical and biological efficacy of recombinant human 12. Gajewski TF, Meng Y, Blank C, et al. Immune interleukin-21 in patients with stage IV resistance orchestrated by the tumor malignant melanoma without prior treatment: a microenvironment. Immunol Rev. phase IIa trial. Clin Cancer Res. 2009;15:2123- 2006;213:131-145. 2129.

13. Parrish-Novak J, Dillon SR, Nelson A, et al. 23. Davis ID, Skrumsager BK, Cebon J, et al. An Interleukin 21 and its receptor are involved in open-label, two-arm, phase I trial of NK cell expansion and regulation of lymphocyte recombinant human interleukin-21 in patients function. Nature. 2000;408:57-63. with metastatic melanoma. Clin Cancer Res. 2007;13:3630-3636. 14. Bryant VL, Ma CS, Avery DT, et al. Cytokine- mediated regulation of human B cell 24. Frederiksen KS, Lundsgaard D, Freeman JA, et differentiation into Ig-secreting cells: al. IL-21 induces in vivo immune activation of predominant role of IL-21 produced by NK cells and CD8(+) T cells in patients with CXCR5+ T follicular helper cells. J Immunol. metastatic melanoma and renal cell carcinoma. 2007;179:8180-8190. Cancer Immunol Immunother. 2008;57:1439- 1449. 15. Wei L, Laurence A, Elias KM, et al. IL-21 is produced by Th17 cells and drives IL-17 25. Sondergaard H, Frederiksen KS, Thygesen P, et al. Interleukin 21 therapy increases the density

of tumor infiltrating CD8+ T cells and inhibits 37. Moroz A, Eppolito C, Li Q, et al. IL-21 the growth of syngeneic tumors. Cancer enhances and sustains CD8+ T cell responses to Immunol Immunother. 2007;56:1417-1428. achieve durable tumor immunity: comparative evaluation of IL-2, IL-15, and IL-21. J Immunol. 26. Den OW, Jacobs JJ, Battermann JJ, et al. Local 2004;173:900-909. therapy of cancer with free IL-2. Cancer Immunol Immunother. 2008;57:931-950. 38. Li Y, Yee C. IL-21 mediated Foxp3 suppression leads to enhanced generation of antigen-specific 27. Skak K, Kragh M, Hausman D, et al. Interleukin CD8+ cytotoxic T lymphocytes. Blood. 21: combination strategies for cancer therapy. 2008;111:229-235. Nat Rev Drug Discov. 2008;7:231-240. 39. Rubio V, Stuge TB, Singh N, et al. Ex vivo 28. Radny P, Caroli UM, Bauer J, et al. Phase II trial identification, isolation and analysis of tumor- of intralesional therapy with interleukin-2 in cytolytic T cells. Nat Med. 2003;9:1377-1382. soft-tissue melanoma metastases. Br J Cancer. 2003;89:1620-1626. 40. Di Carlo E, Comes A, Orengo AM, et al. IL-21 induces tumor rejection by specific CTL and 29. Scheffer SR, Nave H, Korangy F, et al. IFN-gamma-dependent CXC chemokines in Apoptotic, but not necrotic, tumor cell vaccines syngeneic mice. J Immunol. 2004;172:1540- induce a potent immune response in vivo. Int J 1547. Cancer. 2003;103:205-211. 41. Furukawa J, Hara I, Nagai H, et al. Interleukin- 30. Krup OC, Kroll I, Bose G, et al. Cytokine depot 21 gene transfection into mouse bladder cancer formulations as adjuvants for tumor vaccines. I. cells results in tumor rejection through the Liposome-encapsulated IL-2 as a depot cytotoxic T lymphocyte response. J Urol. formulation. J Immunother. 1999;22:525-538. 2006;176:1198-1203.

31. Overwijk WW. Breaking tolerance in cancer 42. Caruso R, Fina D, Peluso I, et al. A functional immunotherapy: time to ACT. Curr Opin role for interleukin-21 in promoting the Immunol. 2005;17:187-194. synthesis of the T-cell chemoattractant, MIP- 3alpha, by gut epithelial cells. Gastroenterology. 32. Kasaian MT, Whitters MJ, Carter LL, et al. IL- 2007;132:166-175. 21 limits NK cell responses and promotes antigen-specific T cell activation: a mediator of 43. Wuest TY, Willette-Brown J, Durum SK, et al. the transition from innate to adaptive immunity. The influence of IL-2 family cytokines on Immunity. 2002;16:559-569. activation and function of naturally occurring regulatory T cells. J Leukoc Biol. 2008;84:973- 33. Zeng R, Spolski R, Finkelstein SE, et al. 980. Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion and function. J Exp Med. 44. Peluso I, Fantini MC, Fina D, et al. IL-21 2005;201:139-148. counteracts the regulatory T cell-mediated suppression of human CD4+ T lymphocytes. J 34. Liu S, Lizee G, Lou Y, et al. IL-21 synergizes Immunol. 2007;178:732-739. with IL-7 to augment expansion and anti-tumor function of cytotoxic T cells. Int Immunol. 45. Kim-Schulze S, Kim HS, Fan Q, et al. Local IL- 2007;19:1213-1221. 21 Promotes the Therapeutic Activity of Effector T cells by Decreasing Regulatory T 35. Li Y, Bleakley M, Yee C. IL-21 influences the Cells Within the Tumor Microenvironment. Mol frequency, phenotype, and affinity of the Ther. 2008. antigen-specific CD8 T cell response. J Immunol. 2005;175:2261-2269. 46. Nurieva R, Yang XO, Martinez G, et al. Essential autocrine regulation by IL-21 in the 36. Ostiguy V, Allard EL, Marquis M, et al. IL-21 generation of inflammatory T cells. Nature. promotes T lymphocyte survival by activating 2007;448:480-483. the phosphatidylinositol-3 kinase signaling cascade. J Leukoc Biol. 2007. 47. Ma HL, Whitters MJ, Konz RF, et al. IL-21 activates both innate and adaptive immunity to

generate potent antitumor responses that require 52. Ferrari-Lacraz S, Chicheportiche R, Schneiter G, perforin but are independent of IFN-gamma. J et al. IL-21 promotes survival and maintains a Immunol. 2003;171:608-615. naive phenotype in human CD4+ T lymphocytes. Int Immunol. 2008;20:1009-1018. 48. Ugai S, Shimozato O, Kawamura K, et al. Expression of the interleukin-21 gene in murine 53. Jin H, Carrio R, Yu A, et al. Distinct activation colon carcinoma cells generates systemic signals determine whether IL-21 induces B cell immunity in the inoculated hosts. Cancer Gene costimulation, growth arrest, or Bim-dependent Ther. 2003;10:187-192. apoptosis. J Immunol. 2004;173:657-665.

49. Ugai S, Shimozato O, Yu L, et al. Transduction 54. Iuchi T, Teitz-Tennenbaum S, Huang J, et al. of the IL-21 and IL-23 genes in human Interleukin-21 augments the efficacy of T-cell pancreatic carcinoma cells produces natural therapy by eliciting concurrent cellular and killer cell-dependent and -independent antitumor humoral responses. Cancer Res. 2008;68:4431- effects. Cancer Gene Ther. 2003;10:771-778. 4441.

50. Gagnon J, Ramanathan S, Leblanc C, et al. IL-6, 55. Nakano H, Kishida T, Asada H, et al. in synergy with IL-7 or IL-15, stimulates TCR- Interleukin-21 triggers both cellular and humoral independent proliferation and functional immune responses leading to therapeutic differentiation of CD8+ T lymphocytes. J antitumor effects against head and neck Immunol. 2008;180:7958-7968. squamous cell carcinoma. J Gene Med. 2006;8:90-99. 51. Hinrichs CS, Spolski R, Paulos CM, et al. IL-2 and IL-21 confer opposing differentiation 56. Daga A, Orengo AM, Gangemi RM, et al. programs to CD8+ T cells for adoptive Glioma immunotherapy by IL-21 gene-modified immunotherapy. Blood. 2008;111:5326-5333. cells or by recombinant IL-21 involves antibody responses. Int J Cancer. 2007;121:1756-1763.

Figures a 1100 Vehicle IT Figure 1. Intratumoral IL-21 administration ) 3 1000 IL-21 IT increases anti-tumor efficacy in RenCa 900 carcinomas compared to subcutaneous 800 administration. BALB/c mice were injected with 700 2x105 RenCa cells SC in the right (a and b) or in 600 both flanks (c) and randomized prior to treatment- SEM 500 ± start as indicated by arrows. Treatment was started 400 3 3 300 with a mean tumor volume > 40 mm (a) >100 mm 3 200 (b) >50 mm (c). Fifty μg or indicated doses of IL- Mean tumor volume (mm volume tumor Mean 100 *** 21 or vehicle was injected IT or SC 3x/week. Mean 0 * ** ** *** ± SEM, n = 9 (a), n = 8, representative of two 8 10 12 14 16 18 20 22 independent experiments (b), n = 10 (c), # P=0.09, Days post RenCa inoculation *P<0.05, **P<0.01, ***P<0.001 compared to b vehicle control (a and b) or contralateral tumor (c) 900 Vehicle IT by Student’s t-test. ) 3 800 50μg IL-21 SC 700 2μg IL-21 IT μ 600 10 g IL-21 IT 50μg IL-21 IT 500

SEM #

± 400 300 200 * Mean tumor volume (mm volume tumor Mean 100 * * * 0 8 10 12 14 16 18 20 22 24 26 Days post RenCa inoculation c 1200 Contralateral tumor

) 1100 3 IL-21 IT 1000 900 800 700 600 SEM

± 500 400 300 200 Mean tumor volume (mm tumor volume Mean 100 * * 0 8 10 12 14 16 18 20 22 24 26 Days post RenCa inoculation

100 Vehicle IT 3 80 IL-21 IT IL-21 SC 60 40 * 20 Percent < 1000 mm Percent < 1000 Treatment (day 13-29, 3x/w eek) ** 0 0 10 20 30 40 50 Days post RenCa inoculation

Figure 2. Intratumoral IL-21 therapy increases survival compared to subcutaneous administration. BALB/c mice were injected with 2x105 RenCa cells SC in the right flank and randomized prior to treatment-start as indicated. Treatment was started with a mean tumor volume > 40 mm3. Fifty µg IL-21 or vehicle was injected IT or SC 3x/week for the indicated period of time. The chart shows Kaplan-Meyer survival analysis of mice with tumors < 1000 mm3, n = 10, representative of two independent experiments *P<0.05, **P<0.01, compared to IT IL-21 by Mantel Cox Log Rank test.

a b 1300 Vehicle IT 1000 Vehicle IT ) 1200 3 ) 50μg IL-21 SC 3 900 IL-21 IT 1100 2μg IL-21 IT 800 1000 900 10μg IL-21 IT 700 800 50μg IL-21 IT 600 700

500 SEM

SEM 600 ± ± 400 500 300 400 300 200 *** 200 *

*** (mm tumor volume Mean Mean tumor volume (mm tumor volume Mean 100 * ** 100 0 0 2 4 6 8 10 12 14 16 8 10 12 14 16 18 20 22 Days post B16 tumor inoculation Days post B16 tumor inoculation

C57BL/6nu/nu C57BL/6 β2m-/- c d 1100 Vehicle IT 1000 Vehicle IT ) ) 3 3 1000 50μg IL-21 IT 900 50μg IL-21 IT 900 800 800 700 700 600 600 500 SEM SEM

± 500 ± 400 400 300 300 200 200

Mean tumor volume (mm tumor volume Mean 100 (mm tumor volume Mean 100 0 0 6 7 8 9 10 11 12 13 14 15 6 7 8 9 10 11 12 13 14 15 16 Days post B16 tumor inoculation Days post B16 tumor inoculation

Figure 3. Intratumoral IL-21 increases CD8+ T cell-dependent anti-tumor effect in B16 melanomas. C57BL/6 WT (a and b), C57BL/6nu/nu (c) and CD57BL/6 β2m-/- (d) mice were injected with 1x105 B16 melanoma cells SC in the right flank and randomized prior to treatment-start as indicated. Treatment was started with a mean tumor volume > 50 mm3 (a) > 75 mm3 (b), > 60 mm3 (c and d). Fifty μg or indicated doses of IL-21 or vehicle was injected IT or SC 1x/day. Mean ± SEM, n = 15, representative of two independent experiements (a), n = 7 (b), n = 10 (c and d), *P<0.05, **P<0.01, ***P<0.001 compared to vehicle control by Student’s t-test.

a c

Vehicle IT Vehicle IT

b d

IL-21 IT IL-21 IT e f CD8+ T cells CD4+ T cells 15 *** 15 **

AOI 10 AOI 10 2 2

5 5 Cells / mm Cells Cells / mm Cells

0 0 T IT IT IT le I 1 ic -21 icle h IL IL-2 eh Ve V Figure 4. Intratumoral IL-21 increases the density of tumor-infiltrating T cells. Cryo-sections from tumor-biopsies obtained at the end of the experiment shown in figure 3a were stained for CD4+ and CD8+ T cells by immunohistochemistry according to ‘Materials and methods’. Positive cells are visualized with liquid permanent red and indicated by arrows. Nuclei are blue by Mayer’s hematoxylin staining. All positive cells located intratumorally were counted in one section from each biopsy and related to a stereologically measured area of interest (AOI) excl. necrotic areas and non-tumor tissue, such as connective tissue. Representative pictures of cryo-sections at 20× magnification showing tumor-infiltrating CD8+ (a and c) and CD4+ T cells (b and d) in B16 melanomas treated as shown. The bar plots show the density of tumor-infiltrating CD8+ T cells (e) and CD4+ T cells (f) scored from the immunohistochemically stained sections. Bars represent mean±SEM, n=15, **P<0.01, ***P<0.001 by Student’s t-test.

a b

CD8+ T cells

150000 1000000 Vehicle IL-21 SC 100000 IL-21 IT 100000 ** 10000

50000 1000

* tissue tumor g cells/ T + Cells/ g tumor tissue tumor g Cells/ ***Spearman correlation p=0.0005 100 0 CD8 10 100 1000 e T I 1 3 ehicl -2 Tumor volume (mm ) IL V IL-21 SC c d + - CD8+ T cells CD4 CD25 T cells 5000000 800000

4000000 600000

3000000 400000 2000000 Tregs Day 1 1 Day 200000 500000 1000000 Cells / g tissue tumor Cells Cells / tissue g tumor Cells 400000 0 0 le IT IT ic 1 icle 300000 h 21 eh -2 - V IL Ve IL IL-21 SC IL-21 SC 200000 ** 5000000 1000000 * 100000

*** / tissue tumor g Cells 4000000 800000 * 0

3000000 600000

2000000 400000 250000 Day 5 5 Day

1000000 200000 200000 Cells / g tissue tumor Cells tissue / tumor g Cells 0 0 150000 C e le IT l IT ic c SC 1 1 S eh 2 hi 2 100000 V IL-21 L- IL- Ve IL-21 I 50000

800000 tissue /tumor g Cells 5000000 * 0 4000000 600000 21 IT Vehicle IL- IL-21 SC 3000000 400000 2000000 p=0.06 200000 10 Day 1000000 Cells / g tumor tissue Cells Cells / tissue g tumor Cells 0 0 IT IT icle SC 1 icle h 1 -2 h -21 Ve IL Ve IL IL-2 IL-21 SC

Figure 5. The density of tumor-infiltrating CD8+ and CD4+CD25- T cells, but not Tregs increases after intratumoral IL-21 correlating with tumor-growth inhibition. BALB/c mice bearing SC RenCa tumors were treated with vehicle, 50 µg IL-21 SC or IT similar to the experiment in figure 1b. Tumor-biopsies were taken out and weighed at the end of the experiment and the absolute numbers of tumor-infiltrating CD8+ T cells were quantified by flow cytometry and related to biopsy weight (a), and the CD8+ T cell density was related to final tumor volume (b). In a separate but similar experiment tumor-biopsies were taken out on day 1, 5 and 10 post treatment-start and the densities of tumor-infiltrating CD8+ and CD4+CD25- T cells were quantified (c), also on day 1 and day 10 post treatment-start the density of tumor-infiltrating CD4+CD25+FoxP3+ T cells (Tregs) were quantified (d). Bars represent mean±SEM, n=7 (a) and n=4 (c and d), *P<0.05, **P<0.01 by One-way ANOVA with Tukey’s post test compared to vehicle or SC as indicated, and correlation was evaluated by Spearman’s correlation coefficient.

a Day 5 post treatment-start Isotype Vehicle IL-21 SC IL-21 IT

b CD107a Granzyme B IFNγ 8 550 3 500 6 450 2 4 400

cells T +CD8+ 350 γ 1 Day 1 1 Day 2 of CD8+ of T cells 300 %IFN Granzyme B Median FI Granzyme B Median %CD107a+CD8+ T cells 0 250 0

20 2000 20 * * * 15 1500 15 10 1000 10 Day 5 5 Day +CD8+ T T cells +CD8+ γ 5 500 5 of CD8+ of T cells %IFN

FI Granzyme B Median %CD107a+CD8+ T cells 0 0 0 2.5 2000 15 * * 2.0 1500 10 1.5 1000 1.0 Day 10 10 Day 5 500 0.5 CD8+of T cells %IFNy+CD8+ T cells Granzyme B Median FI Granzyme B Median %CD107a+CD8+ T cells 0.0 0 0 e le IT le C IT cl SC c 1 c S 1 1 2 2 Vehi IL-21 IT Vehi IL- Vehi IL- IL-2 IL-21 SC IL-21

Figure 6. Intratumoral IL-21 increases IFNγ and granzyme B expression and degranulation in tumor- infiltrating CD8+ T cells. BALB/c mice bearing SC RenCa tumors were treated with vehicle, 50 µg IL-21 SC or IT similar to the experiment in figure 1b. Tumor-biopsies were taken out on day 1, 5 and 10 post treatment-start and stained for intracellular IFNγ and granzyme B, and for surface expression of CD107a. Representative FACS plots are shown from day 5 post treatment-start (a) and bar plots show the percent of CD107a+, median fluorescence intensity (MFI) of granzyme B expression and percent of IFNγ expressing CD8+ T cells from day 1, 5 and 10 post treatment-start (b). Bars depict mean±SEM, n=4, *p<0.05, by Mann Whitney’s U test compared to all others, except CD107a expression on day 10 where it is only compared to vehicle.

a b c IL-21 7.0 1.5 ** ) ) ** IL-21 IL-21 6 6 6.0 Vehicle 1.2 Vehicle 5.0 4.0 * 0.9

3.0 0.6 ** 2.0 * * * 0.3 Number of cells (10 cells of Number Vehicle (10 cells of Number 1.0 * 0.0 0.0 * +CD4+ +CD8+ -CD19+ β β β Total cells CD4+CD25- CD4+CD25+ TCR TCR TCR CD62L-CD44- CD62L-CD44- Lymphocytes CD62L+CD44- CD62L-CD44+ CD62L+CD44- CD62L-CD44+ CD62L+CD44+ CD62L+CD44+ CD8+ T cells CD4+ T cells d Vehicle IL-21 e 15 IL-21 Vehicle

10 ** ** *

5 * % cells KI67+

0 +CD8+ +CD4+ +CD19+ β β β TCR TCR TCR

f TCRβ+CD8+ T cells TCRβ+CD4+ T cells TCRβ-CD19+ B cells ) ) ) 6 6 0.08 6 0.15 0.20 IL-21 IL-21 IL-21 Vehicle ** Vehicle Vehicle 0.06 0.15 * 0.10 ** * 0.04 * 0.10 0.05 0.02 * 0.05 Number of KI67+ cells (10 cells KI67+ of Number (10 cells KI67+ of Number 0.00 (10 cells KI67+ of Number 0.00 0.00 CD62L-CD44- CD62L-CD44- CD62L-CD44- CD62L+CD44- CD62L-CD44+ CD62L+CD44- CD62L-CD44+ CD62L+CD44- CD62L-CD44+ CD62L+CD44+ CD62L+CD44+ CD62L+CD44+

Figure 7. IL-21 enlarges tumor-draining lymph nodes by increasing naïve lymphocyte numbers and proliferation of activated lymphocytes. BALB/c mice were inoculated with 1x106 RenCa tumor cells SC in the hind foot pad and treated on day 12 post inoculation with vehicle or 50 µg IL-21 IT 5x/week. On day 20 popliteal LNs were taken out and the absolute number of LN lymphocytes was counted and analyzed by flow cytometry. Macroscopic picture of IL-21 and vehicle treated draining LNs (a), the absolute number of the indicated lymphocyte populations in tumor-draining LNs (b), and the absolute number of the indicated T cell subpopulations (c). In an identical experiment LN lymphocytes were stained for intracellular KI67 and analyzed by flow cytometry. Representative FACS plots show KI67 staining in CD8+ and CD4+ T cells from vehicle and IL-21 treated mice with the distribution of CD62L and CD44 expression (d), the percent of KI67+ CD8+ and CD4+ T cells, and CD19+ B cells are shown (e), and the number KI67+ CD8+ and CD4+ T cells and CD19+ B cells are shown based on their CD62L and CD44 expression as indicated (f). Bars represent mean±SEM, n=5, *P<0.05, **P<0.01 compared to vehicle control by Student’s t-test using Bonferroni correction. (b) and (c) are representatives of three independent experiments.

B16 RenCa

Supplementary figure 1. Greater MHC class I expression on RenCa compared to B16 cells. B16 melanoma and RenCa renal cell carcinoma cells were stained with primary unlabelled mouse anti-H-2Kb and H-2Db (B16) and mouse anti-H-2Kd and H-2Dd (RenCa). Primary staining with mouse IgG2a was used as an isotype control. Secondary staining with goat anti-mouse IgG PE was used for detection by flow cytometry. Histogram X-axes depict the level of MHC class I expression as fluorescence intensity of B16 (left) and RenCa (right) after secondary staining. Representative of two independent experiments.

a SC in lower leg IT in foot tumor b

4.0 IL-21 3.5 Vehicle 3.0

cells)

6 2.5 2.0 1.5 1.0

Cell count (10 Cell count 0.5 0.0 CD4+ CD8+

CD19+ Total cells Total CD4+CD25-

CD4+CD25+ Lymphocytes Supplementary figure 2. IL-21 increases lymphocyte numbers in naïve lymph nodes draining the foot pad. BALB/c mice inoculated with 1x106 RenCa tumor cells SC in the foot pad were injected once with 30µL of Pantent Blue V dye 12 days post inoculation either SC in the lower leg or IT and popliteal LNs were taken out 30 min. later. The macroscopic pictures show two representative popliteal LNs after injection with Pantent Blue V dye SC in the lower leg or IT in foot pad tumor as indicated (a). BALB/c mice bearing no tumors were injected in the foot pad with 50 µg IL-21 on days 0, 2 and 4, popliteal LNs were taken out on day 7 and LN lymphocytes were quantified by flow cytometry and the bar plot shows the absolute number of the indicated lymphocyte populations (b). Bars represent mean±SEM, n=5, ÌP<0.05, ÌÌP<0.01 compared to vehicle control by Student’s t-test using Bonferroni correction.

Supplementary figure 3.

0.7 IL-21 Vehicle 0.6

0.5 cells) 6 0.4

0.3

0.2

Cellcount (10 0.1

0.0 Total cells CD4+CD25- TCRb+CD8+ TCRb+CD4+ CD4+CD25+ TCRb-CD19+ Lymphocytes

Supplementary figure 3. Intratumoral IL-21 does not modulate contralateral lymph nodes. BALB/c mice were inoculated with 1x106 RenCa tumor cells SC in the right hind foot pad and treated on day 12 post inoculation with vehicle of 50 µg IL-21 IT 5x/week (from experiment in figure 6). On day 20 the left popliteal LNs (contralateral to the tumor site) were taken out and the absolute number of indicated LN lymphocytes was quantified by flow cytometry.

Paper IV:

Endogenous interleukin 21 restricts CD8+ T cell expansion and is not required for tumor immunity

Søndergaard H., Coquet J.M., Uldrich A.P., McLaughlin N, Godfrey D.I., Sivakumar P.V. Skak

K. and Smyth M.J., J Immunol 2009 Dec 1;183(11):7326-36. Epub 2009 Nov 13.

78 Published November 13, 2009, doi:10.4049/jimmunol.0902697 The Journal of Immunology

Endogenous IL-21 Restricts CD8؉ T Cell Expansion and Is not Required for Tumor Immunity1

Henrik Søndergaard,*‡ Jonathan M. Coquet,† Adam P. Uldrich,* Nicole McLaughlin,* Dale I. Godfrey,† Pallavur V. Sivakumar,§ Kresten Skak,‡ and Mark J. Smyth2*

IL-21 has antitumor activity through actions on NK cells and CD8؉ T cells, and is currently in clinical development for the treatment of cancer. However, no studies have addressed the role of endogenous IL-21 in tumor immunity. In this study, we have studied both primary and secondary immune responses in IL-21؊/؊ and IL-21R؊/؊ mice against several experimental tumors. We found intact immune surveillance toward methylcholanthrene-induced sarcomas in IL-21؊/؊ and IL-21R؊/؊ mice compared with wild-type mice and B16 melanomas showed equal growth kinetics and development of lung metastases. IL-21R؊/؊ mice showed competent NK cell-mediated rejection of NKG2D ligand (Rae1␤) expressing H-2b؊ RMAS lymphomas and sustained transition to CD8؉ T cell-dependent memory against H-2b؉ RMA lymphomas. ␣-Galactosylceramide stimulation showed equal expansion and activation of NKT and NK cells and mounted a powerful antitumor response in the absence of IL-21 signaling, despite reduced expression of granzyme B in NKT, NK, and CD8؉ T cells. Surprisingly, host IL-21 significantly restricted the expansion of Ag-specific CD8؉ T cells and inhibited primary CD8؉ T cell immunity against OVA-expressing EG7 lymphomas, as well as the secondary expansion of memory CD8؉ T cells. However, host IL-21 did not alter the growth of less immunogenic MC38 colon carcinomas with dim OVA expression. Overall, our results show that endogenous IL-21/IL-21R is not required for NK, NKT, and ,CD8؉ T cell-mediated tumor immunity, but restricts Ag-specific CD8؉ T cell expansion and rejection of immunogenic tumors indicating novel immunosuppressive actions of this cytokine. The Journal of Immunology, 2009, 183: 7326–7336.

he novel class I cytokine IL-21 is a member of the com- expressing plasmids and recombinant mouse IL-21. In this study, mon ␥-chain receptor family. IL-21 is primarily produced antitumor effects have been demonstrated both on established s.c. T by activated CD4ϩ T cells and NKT cells (1, 2) and sig- tumors, on lung and liver metastases from i.v. injected tumors, and nals through its unique IL-21R. IL-21R is expressed by most leu- on disseminated tumors (4). In addition to its antitumor effects as kocytes including B, T, NK, NKT, macrophages, and dendritic monotherapy, IL-21 shows additional efficacy when used in com- cells (DCs),3 giving IL-21 substantial effects in both humoral and bination with several other therapies (5). The antitumor activity of cellular immune responses; these effects include costimulation of IL-21 is mainly mediated by NK cells (6–8) and CD8ϩ T cells B cell proliferation, differentiation, and isotype switching, Th17 (9–11) with requirement of IFN-␥, perforin, or both (7, 8, 12–14) ϩ ϩ cell differentiation of CD4 T cells, increased CD8 T cell ex- depending on specific model conditions. Consistently, IL-21 has pansion and effector function, and activation of NK and NKT cells been found to sustain Ag-specific CD8ϩ T cell responses (9), aug- (3). These profound immunomodulatory effects of IL-21 regulates menting both naive and memory CD8ϩ T cell expansion (11, 15) immune responses in a variety of diseases, including infections, and boosting both CD8ϩ T cell and NK cell cytotoxicity in vitro autoimmunity, and cancer (3–5). (11, 16). In a few reports, B cells have also been suggested to be The antitumor effects of IL-21 have been extensively investi- involved in IL-21 antitumor responses with increased production gated in mouse tumor models using a range of different sources of of tumor-specific IgG (17, 18). Moreover, IL-21 has been found to IL-21 delivery such as, IL-21-transfected tumor cell lines, IL-21- modulate the activity and cytokine production of NKT cells and enhance the antitumor effects mediated by NKT cell stimulation *Cancer Immunology Program, Peter MacCallum Cancer Centre, East Melbourne, (2, 8, 19). Based on these data, IL-21 is currently in clinical trials †Department of Microbiology and Immunology, The University of Melbourne, Parkville, Victoria, Australia; ‡Immunopharmacology, Novo Nordisk A/S, Måløv, for the treatment of metastatic melanoma and renal cell carcinoma Denmark; and §Zymogenetics, Seattle, WA 98102 where it has shown an acceptable safety profile with reports of Received for publication August 17, 2009. Accepted for publication October 6, 2009. responding patients along with several indications of in vivo im- The costs of publication of this article were defrayed in part by the payment of page mune activity (20–22). charges. This article must therefore be hereby marked advertisement in accordance The majority of studies on IL-21 in tumor immunology have with 18 U.S.C. Section 1734 solely to indicate this fact. used exogenous sources of IL-21. However, IL-21 also plays an 1 This work was supported by Grant 454569 from the National Health and Medical important role in the pathogenesis of several autoimmune diseases Research Council of Australia Program (to M.J.S., and D.I.G.), by a Cancer Research Institute Postgraduate Scholarship (to J.M.C.), a Doherty Fellowship (to A.P.U.), and and here the role of endogenous IL-21/IL-21R has been widely by National Health and Medical Research Council Research Fellowships (to M.J.S. studied. Particularly, crossing of IL-21R-deficient (IL-21RϪ/Ϫ) and D.I.G.). D.I.G. and M.J.S. have received research support from Novo Nordisk A/S. mice onto diabetes prone NOD mice, spontaneous arthritis suscep- 2 Address correspondence and reprint requests to Dr. Mark J. Smyth, Peter MacCal- tible K/XbN mice, and BXSB-Yaa mice that develop a systemic lum Cancer Centre, St Andrews Place, East Melbourne, 3002, Victoria, Australia. lupus erythematosus-like syndrome all showed significantly re- E-mail address: [email protected] duced disease activity (23–25). Furthermore, collagen-induced ar- 3 Abbreviations used in this paper: DC, dendritic cell; MCA, methylcholanthrene; thritis were reduced in DBA/1 mice treated with IL-21R.Fc (26) ␣GC, alpha-galactosylceramide; MSCV, murine stem cell virus; WT, wild type. and IL-21Ϫ/Ϫ mice showed resistance to DSS- and TNBS-induced Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 colitis (27). To this end, blockade of the host IL-21/IL-21R axis

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902697

The Journal of Immunology 7327 has been suggested as a potential therapeutic opportunity in several MCA-induced fibrosarcoma model of these autoimmune disorders. Male B6 WT, IL-21Ϫ/Ϫ, or IL-21RϪ/Ϫ mice were injected s.c. with 100 ␮l To date, no studies have investigated the role of endogenous of corn oil containing MCA (100 ␮g or 25 ␮g) on the left hind flank and IL-21/IL-21R signaling in the control of tumor initiation, growth were monitored for the onset and progression of tumors on a weekly basis and metastasis. Based on the evident antitumor effect of exogenous until 50 wk of age (30). Survival was plotted using a Kaplan-Meier curve 2 IL-21, we wanted to explore the contribution of host signaling via in which the time when tumors exceeded 150 mm was used as endpoint. the endogenous IL-21/IL-21R axis in both primary and secondary Other experimental tumor models tumor immune responses. In fully backcrossed IL-21 and IL-21R Ϫ/Ϫ Ϫ/Ϫ gene-targeted mice, we investigated the response to several differ- B6 WT, IL-21 , or IL-21R mice were inoculated s.c. in the right flank with 105 B16F10 melanoma, 5 ϫ 106 RMAS (H-2b-negative), 5 ϫ ent experimental tumors controlled by different effector cells or 106 RMA-S MSCV (empty vector transfected), 5 ϫ 106 RMAS-Rae1␤, with responsiveness to IL-21 therapy. We show in this study that 5 ϫ 105 MC38 OVAdim, or 3 ϫ 106 immunogenic EG7 cells. Rechallenges endogenous IL-21 is not required for tumor immunity but reveal a were made more than 40 days post tumor regression with 106 RMA or 3 ϫ 6 novel suppressive effect of IL-21 on CD8ϩ T cell immunity. 10 EG7 cells. In these experiments, tumor size was calculated as a product of two perpendicular diameters measured with a digital caliper approximately three times per week. The termination criterion was a tumor volume Materials and Methods of 150 mm2, which was used as a surrogate survival endpoint in Kaplan- Ϫ/Ϫ Ϫ/Ϫ Mice Meier analysis. In other experiments, B6 WT, IL-21 , or IL-21R mice were injected i.v. in the tail vein on day 0 with 2 ϫ 105 B16 mela- C57BL/6 (B6) wild-type (WT) mice were purchased from the Walter and noma cells either unpulsed or pulsed for 2 days in vitro with 500 ng/ml Eliza Hall Institute of Medical Research (Melbourne, Victoria, Australia), ␣GC. On day 14 postinoculation, mice were sacrificed, lungs were har- or bred in-house at the Peter MacCallum Cancer Centre Animal Facility vested, and the number of surface metastasis per lung was counted using a (Melbourne, Victoria, Australia). B6 IL-21-deficient (IL-21Ϫ/Ϫ) mice, dissecting microscope. originally provided by Dr. P. Sivakumar (Zymogenetics, Seattle, WA), and B6 IL-21RϪ/Ϫ mice, originally provided Dr. W. J. Leonard (National In vivo stimulation of NKT cells

Heart, Blood and Lung Institute, Bethesda, MD) were backcrossed to B6 Ϫ/Ϫ Ϫ/Ϫ Ϫ/Ϫ for 8–12 generations at the Peter MacCallum Cancer Centre Animal Fa- B6 WT, IL-21 , IL-21R , or TCR.J␣18 mice were injected i.p. cility (Melbourne, Victoria, Australia). B6 TCR.J␣18Ϫ/Ϫ mice obtained with 2 ␮g of ␣GC prepared in PBS at 200-␮l doses. After 2, 24, 72, or 144 h postinjection, livers and spleens were harvested to assay for NK, from Dr. M. Taniguchi (Chiba, Japan) were backcrossed for 12 generations ϩ to B6 at the Peter MacCallum Cancer Centre Animal Facility (Melbourne, NKT, and CD8 T cell activation by flow cytometry. NKT cells were identified as TCR␤ϩ␣GC/CD1dtetramerϩ cells, NK cells as Victoria, Australia). Mice age 6–14 wk were used in all experiments in ϩ Ϫ Ϫ ϩ ϩ ϩ accord with animal ethics guidelines of the Peter MacCallum Cancer NK.1.1 TCR␤ ␣GC/CD1dtetramer and CD8 T cells as TCR␤ CD8 Centre. cells. For intracellular staining, cells were cultured in GolgiStop (BD Biosciences) for 2–4 h before being surface stained and subsequently fixed and permeabilized using the BD Cytofix/Cytoperm Plus Fixation/Perme- Cell lines and culture conditions abilization kit (BD Biosciences). In experiments in which ␣GC was used Ϫ/Ϫ Ϫ/Ϫ B16 (F10) melanoma cells (American Type Culture Collection (CRL- in conjunction with chicken OVA, B6 WT, IL-21 , or TCR.J␣18 6322; ATCC), RMA (H-2bϩ), RMAS (H-2bϪ), RMAS murine stem cell were injected with OVA (400 ␮g/mouse) and ␣GC (1 ␮g/mouse) i.v. in virus (MSCV) (empty vector transfected) and stable transfectant RMAS- 200 ␮l of PBS. On days 7 and 35 postinjection, livers and spleens were ϩ Rae1␤, and MC38 OVAdim (28) were all maintained in RPMI 1640 sup- harvested and assayed for CD8 T cell activation by staining with SIIN plemented with 10% heat-inactivated FCS, 100 U/ml penicillin, 100 ␮g/ml FEKL-loaded MHC class I tetramers. For intracellular IFN-␥ staining, cells streptomycin, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 2 were restimulated with SIINFEKL peptide for 12 h in vitro in the presence mM glutamax, and 15 mM HEPES buffer (all from Invitrogen). Chicken of GolgiStop for the last 4 h, and subsequently stained for CD8 and intra- OVA transfected EL-4 T cell lymphoma cells EG7 (CRL-2113; ATCC) cellular IFN-␥ and analyzed by flow cytometry. and a more immunogenic variant of EG7 were grown in similar medium supplemented with 0.4 ␮g/ml geneticin G418 (Invitrogen). Statistics Student’s t test (two-tailed, assuming equal variance), two-tailed Mann- Reagents and Abs Whitney U test or Kruskal-Wallis test with Dunn’s posttest was used for statistical evaluations of differences between WT B6 mice and IL-21Ϫ/Ϫ/ ␣-Galactosylceramide (␣GC), a marine sponge glycolipid that activates Ϫ/Ϫ CD1d-restricted NKT cells (29), was provided by the Pharmaceutical Re- IL-21R mice as indicated in each experiment. Mantel Cox Log Rank search Laboratories (Kirin Brewery), and working concentrations were pre- test was used to evaluate statistical differences in Kaplan-Meier analyses. pared in PBS. Methylcholanthrene (MCA) was purchased from Sigma- Data are generally shown as individual observations or as mean Ϯ SEM Aldrich. For flow cytometric analysis, the following anti-mouse Abs were unless otherwise noted, and p Ͻ 0.05 was considered statistically used: FITC-conjugated, FoxP3 (FJK-16s) and CD44 (IM7) (eBioscience); significant. PE-conjugated CD25 (PC61.5) and TCR-␤ (H57-597) (eBioscience); PE- Cy5.5-conjugated TCR-␤ (H57-597) (eBioscience); PE-Cy7-conjugated, Results CD3 (145-2C11), CD8 (53-6.7), and NK1.1 (PK136) (BD Pharmingen); Endogenous IL-21 does not protect from MCA-induced allophycocyanin-conjugated TCR-␤ (H57-597), CD62L (MEL-14), CD4 carcinogenesis (GK1.5), B220 (RA3-6B2) (eBioscience), allophycocyanin-Cy7-conjugated CD3 (145-2C11), CD8 (53-6.7) and B220 (RA3-6B2) (BD Pharm- To date no studies have evaluated the effect of endogenous IL-21/ ingen); and Pacific blue-conjugated CD4 (RM4-5) (eBioscience). PE-con- IL-21R signaling in tumor immunity or tumor immune surveil- jugated SIINFEKL-loaded MHC class I tetramer from Dr. A. Brooks (University of Melbourne, Parkville, Australia) and ␣GC-loaded CD1d tet- lance. Initially we wanted to study the involvement of the IL-21/ ramer produced in house by K. Kyparissoudis (University of Melbourne, IL-21R axis in tumor immune surveillance and here we used the Parkville, Australia), using a construct originally provided by M. Kronen- chemical carcinogen 3-MCA, which is known for its ability to berg (La Jolla Institute for Allergy and Immunology, La Jolla, CA) were promote fibrosarcoma carcinogenesis (31) and illustrate natural an- ϩ used to detect OVA-specific CD8 T cells and NKT cells, respectively. titumor immunity (32). Groups of B6 WT, IL-21Ϫ/Ϫ, and IL- Furthermore, Fc receptor block (clone 2.4G2; grown in-house) was used in Ϫ/Ϫ all experiments to prevent nonspecific binding of Abs. Cell suspensions 21R mice were inoculated s.c. with 25 or 100 ␮g of MCA and were stained in FACS tubes or 96-well U-bottom plates for 30 min at 4°C observed for fibrosarcoma development over a period of 50 wk in the dark and washed between incubations. Flow cytometric acquisition (Fig. 1). At 100 ␮g of MCA, Ͼ80% of WT mice were sacrificed was performed on a FACSCanto-II or LSR-II (BD Biosciences). Nonviable over the course of the experiment, whereas ϳ50% were sacrificed lymphocytes were excluded on the basis of staining with 7-aminoactino- mycin D (eBioscience) or hydroxystilbamidine methanesulfonate/Fluoro- at the 25-␮g dose. However, at the doses examined in this study, Gold (FG) (Molecular Probes). Analysis was performed using FlowJo soft- neither the onset nor incidence of MCA-induced fibrosarcomas ware (Tree Star) and FACSDiva software (BD Biosciences). were significantly different between the IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ

7328 HOST IL-21 IN TUMOR IMMUNITY

FIGURE 1. Endogenous IL-21 does not protect against MCA-induced carcinogenesis. Cohorts of mice were challenged with 25 or 100 ␮g of the chemical carcinogen 3Ј MCA s.c. and were monitored for sarcoma incidence. Data depict survival of B6 WT, IL-21Ϫ/Ϫ, and IL- 21RϪ/Ϫ mice. Survival was deﬁned as when tumor size Ͼ150 mm2 for n ϭ 16–21 mice per group. Results are representative of two independent experiments.

FIGURE 2. B16 melanoma growth and lung metastases are not con- mice compared with WT, suggesting that natural immune surveil- trolled by IL-21/IL-21R signaling. B6 WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ mice lance against chemically induced fibrosarcomas does not require were injected with B16F10 melanoma cells either s.c. with 1 ϫ 105 cells IL-21 signaling. on the right flank and monitored for tumor growth and survival defined as time when tumor size Ͼ150 mm2 (A) or 2 ϫ 105 cells i.v. and harvesting IL-21/IL-21R signaling does not protect against B16F10 tumor of lungs 14 days later for evaluation of lung metastases (B). Tumor growth growth or lung metastasis curves represent mean Ϯ SEM and Kaplan-Meier survival curves depict 2 In previous studies of IL-21 antitumor activity, we and others have survival defined as time when tumor size Ͼ150 mm for n ϭ 9–12 mice in A. Data show mean Ϯ SEM for n ϭ 6–7 mice in B. found that the B16 melanoma model was sensitive to treatments with IL-21 alone or in different combinations (6, 10, 11, 14, 33). In this study, we used the B16 melanoma model to examine whether low levels of H-2b and a variant transfected to express the NKG2D host-derived IL-21 or IL-21R signaling might play a role in the ligand retinoic acid inducible-1␤ (Rae1␤). RMAS lymphomas are control of B16 tumor growth. WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ mice well known for their NK cell sensitivity both in vitro and in vivo were injected with B16F10 melanoma cells s.c. and monitored for (34) and Rae1␤-transfected variants have been shown to enhance tumor growth and survival (Fig. 2A) or i.v. and evaluated 14 days the NK cell-mediated rejection (35). In this study, we inoculated later for development of lung metastases (Fig. 2B). The results of IL-21RϪ/Ϫ and WT mice s.c. with parental RMAS cells, RMAS the s.c. experiment showed equal growth kinetics of B16F10 in MSCV (RMAS cells transfected with an empty vector), and both WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ with similar survival mea- RMAS-Rae1␤ (Fig. 3A). WT and IL-21RϪ/Ϫ showed equal out- sured as the individual time when tumor size exceeded 150 mm2. growth of the RMAS and RMAS MSCV, whereas RMAS-Rae1␤ Furthermore, after i.v. injection of B16F10, the number of metas- showed equal rejection in both strains (9 of 10 tumors were re- tases observed in the lungs 14 days later was also equivalent be- jected in WT and 9 of 11 tumors were rejected in IL-21RϪ/Ϫ). tween WT and IL-21Ϫ/Ϫ or IL-21RϪ/Ϫ mice. These results suggest These results suggest competent NKG2D-dependent NK cell-me- that endogenous IL-21/IL-21R signaling does not protect against diated tumor rejection despite IL-21R deficiency. B16 melanoma growth or metastasis development. To further explore the functional properties of this NK cell- mediated rejection we next addressed whether or not endogenous Primary tumor rejection by NK cells via NKG2D and transition ϩ IL-21/IL-21R signaling plays a role in the transition from innate to to CD8 T cell immunity does not require IL-21R signaling adaptive immunity, which has previously been suggested (16). In IL-21 has been suggested to mediate tumor rejection through NK this experiment, we rechallenged WT and IL-21RϪ/Ϫ mice that cells and more specifically via the activating receptor NKG2D (6, initially had rejected RMAS-Rae1␤ over 40 days postregression 7). To specifically investigate NK cell-mediated tumor immunity with Ag-processing competent and H-2b-positive cells RMA (Fig. and particularly NKG2D rejection mechanisms, we used the TAP- 3B). As a control, RMA tumors were inoculated in naive WT mice deficient T cell lymphoma cell line RMAS, which expresses very and these tumors all grew out. The RMAS-Rae1␤ immunized

The Journal of Immunology 7329

NKT cell activation primes NK cells and mediates antitumor response in the absence of IL-21/IL-21R function Recently, our lab showed that NKT cells produce IL-21 following ␣GC stimulation and show activation in response to IL-21 stimulation (2). In this experiment, we used ␣GC stimulation as a way to induce IL-21 production and to investigate whether the activation and antitumor response of NKT cells could in part be mediated by IL-21. We injected WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ with 2 ␮g of ␣GC i.p. and quantified the expansion of NKT cells after 2, 24, 72, and 144 h in liver and spleen using flow cytometry where NKT cells were identified as TCR␤ϩ␣GC/CD1dtetramerϩ lymphocytes (Fig. 4A). Two hours after ␣GC stimulation the detection of NKT cells decreased compared with unstimulated mice and almost disappeared after 24 h due to down-regulation of their TCR. However, after 72 h a substantial expansion (4- to 7-fold in both liver and spleen) of the NKT cell population was observed followed by a contraction of the NKT cell population by 144 h (Fig. 4A). NKT cell expansion was found to be similar between WT and IL-21Ϫ/Ϫ/IL-21RϪ/Ϫ mice, suggesting a normal proliferative response of NKT cells without IL-21 signaling. In addition, NKT cells were stained for CD4 and intracellular IFN-␥ expression after ␣GC stimulation (Fig. 4B). These data showed a rapid induction of, and increased proportion of, IFN-␥ expressing NKT cells after 2 and 24 h in both CD4ϩ/CD4Ϫ subsets, which declined again at 72 h and almost returned to baseline by 144 h. FACS plots are only shown for liver NKT cells but similar results were obtained in the spleen (data not shown). No difference was seen between WT and IL-21Ϫ/Ϫ or IL-21RϪ/Ϫ NKT cell IFN-␥ responses after ␣GC stimulation, suggesting that cytokine production from NKT cells in response to ␣GC does not require endogenous IL-21. Because ␣GC also stimulates potent IFN-␥ production by NK cells (36), NK cell activation was also examined by gating on NK cells (␣GC/CD1d tetramerϪNK1.1ϩ␤TCRϪ) and detecting their intracellular IFN-␥ expression (Fig. 4, C and D). Representative FACS plots of gated liver NK cells are shown, but similar results were found in spleen NK cells (data not shown). The results show that NK cell proportions and intracellular IFN-␥ expression was comparable between NK cells from WT, IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice following ␣GC stimulation at all time points (Fig. 4, C and D). Thus, IL-21 did not appear to be important for NK cell IFN-␥ production following ␣GC stimulation. The granularity/cytotoxicity of NK cells (33), CD8ϩ T cells (11) and NKT cells (2) can also be augmented by IL-21. In this exper- FIGURE 3. Primary NKG2D-dependent tumor rejection by NK cells iment, we analyzed the expression of intracellular granzyme B on ϩ Ϫ/Ϫ and transition to CD8 T cell memory is intact in IL-21R mice. A, B6 gated NKT, NK, and CD8ϩ T cells following ␣GC administration WT and IL-21RϪ/Ϫ mice were challenged with 5 ϫ 106 RMAS, RMAS (Fig. 4E). Intracellular granzyme B expression was most strikingly MSCV (empty vector), or RMAS-Rae1␤ s.c. on the right flank and monitored for tumor growth. B, Naive B6 WT mice and WT and IL-21RϪ/Ϫ up-regulated in NK cells within 24 h of stimulation. NKT, NK, and ϩ mice that rejected their primary RMAS-Rae1␤ tumor challenge were re- CD8 T cells from WT mice expressed slightly more intracellular challenged Ͼ40 days postrejection with 106 RMA cells s.c. on the con- granzyme B than IL-21Ϫ/Ϫ mice at the 72 h time point. IL-21 also tralateral flank and monitored for tumor growth. Data in A depict mean Ϯ appeared to be involved in the maintenance of NKT cell granzyme SEM for n ϭ 8–11 mice per group. Data in B show individual tumor B expression at 144 h as well as NK cell granzyme B expression growth curves for n ϭ 6 mice (naive WT), n ϭ 10 mice (immune WT), and at 24 h. These results suggest that endogenous IL-21 is produced n ϭ 9 mice (immune IL-21RϪ/Ϫ). after ␣GC stimulation and may play a role in ␣GC-mediated granzyme B up-regulation and possibly cytotoxicity of NKT, NK, and CD8ϩ T cells. Therefore, to explore the capacity of IL-21Ϫ/Ϫ NKT mice, however, showed potent memory responses regardless of the cell-mediated tumor immune responses, we injected WT and IL- Ϫ/Ϫ lack of IL-21R signaling with initial tumor growth followed by 21 mice i.v. with B16F10 melanoma cells either unpulsed or complete rejection already within 10 days post-rechallenge in both pulsed in vitro for 2 days with 500 ng/ml ␣GC and 14 days later strains. These results suggest that IL-21R signaling is redundant in lungs were harvested to count metastases (Fig. 4F). This experi- the transition from an initial NK cell-mediated immunization to a mental protocol, based on a previously published approach (37), successful CD8ϩ T cell-mediated response. affected neither the in vitro growth nor viability of the tumor cells

7330 HOST IL-21 IN TUMOR IMMUNITY

FIGURE 4. Activated NKT cells prime NK cells and mediate antitumor response in the absence of IL-21 or IL-21R. B6 WT (n ϭ 4 mice), IL-21Ϫ/Ϫ (n ϭ 4 mice), and IL-21RϪ/Ϫ (n ϭ 2–4 mice) were administered 2 ␮g of ␣GC i.p. Liver and spleen specimens were harvested 2, 24, 72, or 144 h later to assay for NKT and NK cell activation by flow cytometry. A, NKT cells were gated as TCR␤ϩ␣GC/CD1dtetramerϩ and the absolute number of NKT cells in the liver (left) and spleen (right) of mice after administration of ␣GC are shown. B, Representative profiles of CD4 vs intracellular IFN-␥ expression of gated NKT cells from the liver following ␣GC. Number represents the percentage of cells in that quadrant. C, Representative density profiles of TCR␤ vs NK1.1 expression on non-NKT cells in the liver following ␣GC administration where gated region depicts NK cells (␣GC/ CD1dtetramerϪNK1.1ϩTCR␤Ϫ). The number represents the percentage of cells in that gate. D, Representative histograms of intracellular IFN-␥ expression by gated NK cells from C. WT and IL-21Ϫ/Ϫ mice were administered 2 ␮g of ␣GC i.p. and liver and spleen specimens were harvested at the specified time points and cells were stained for TCR-␤, ␣GC/CD1dtetramer, NK1.1, CD8, and intracellular granzyme B. E, Representative histograms of n ϭ 4 mice at each time point of granzyme B expression on gated NKT cells (␣GC/CD1d tetramerϩTCR␤ϩ), NK cells (␣GC/CD1d tetramerϪNK1.1ϩ␤TCRϪ) and CD8ϩ T cells (␣GC/CD1d tetramerϪTCR␤ϩCD8ϩ) from WT and IL-21Ϫ/Ϫ mice after ␣GC administration. F, B6 WT and IL-21Ϫ/Ϫ mice were injected i.v. with 2 ϫ 105 B16F10 melanomas either unpulsed or pulsed in vitro for 2 days with 500 ng/ml ␣GC, 14 days later lungs were harvested and lung metastasis were counted. Data depict mean Ϯ SEM for n ϭ 7–8 mice. Representative experiment is of three separate experiments, and Kruskal-Wallis test with Dunn’s .p Ͻ 0.001 ,ءءء ;p Ͻ 0.01 ,ءء .posttest was used to compare B16 Ϯ ␣GC in WT and IL-21Ϫ/Ϫ

(data not shown). The results showed a significant reduction in the or TCR.J␣18Ϫ/Ϫ mice with chicken OVA protein i.p. Ϯ 1 ␮g of number of lung metastases in mice that had received the ␣GC- ␣GC and detected the Ag-specific CD8ϩ T cell response after 7 or pulsed B16F10 cells compared with mice receiving unpulsed cells, 35 days in spleens and livers via SIINFEKL-loaded MHC class I but this reduction was similar in WT and IL-21Ϫ/Ϫ, suggesting that tetramers (Fig. 5A). Immunization with OVA alone did not pro- IL-21 did not contribute to the NKT cell-mediated antitumor reduce a very strong Ag-specific response, whereas addition of ␣GC sponse in this model. Similar results, but with increasing number produced a marked increase in SIINFEKL-specific CD8ϩ T cells of lung metastases, were obtained when cells were pulsed with in both WT and IL-21Ϫ/Ϫ on day 7. Intriguingly, CD8ϩ T cells lower concentrations of ␣GC (data not shown). expanded to a greater proportion in livers of IL-21Ϫ/Ϫ mice ( p Ͻ 0.05), suggesting that IL-21 inhibited the expansion of Ag-specific ϩ CD8 T cell expansion, but not IFN-␥ production is enhanced CD8ϩ T cells. As a control NKT cell-deficient TCR.J␣18Ϫ/Ϫ mice Ϫ/Ϫ in IL-21 mice showed no increased response to OVA plus ␣GC. On day 35, the Stimulation of NKT cells with ␣GC have also been shown to en- level of OVA-specific CD8ϩ T cells were reduced to the levels hance CD8ϩ T cell responses (38–40). To test whether IL-21 pro- seen with OVA immunization alone in both strains. For a more duction from NKT cells might contribute to the generation of functional readout, splenocytes from WT or IL-21Ϫ/Ϫ mice were CD8ϩ T cell-based immune responses, we injected WT, IL-21Ϫ/Ϫ, also re-stimulated on day 7 or 35 with SIINFEKL peptide for 12 h

The Journal of Immunology 7331

FIGURE 6. OVAdim expressing MC38 tumor growth is similar in WT and IL-21Ϫ/Ϫ mice. B6 WT and IL-21Ϫ/Ϫ mice were challenged with 5 ϫ 105 MC38-OVAdim cells s.c. on the right flank and monitored for tumor growth. Survival was defined as time when tumors Ͼ150 mm2. Tumor growth curves represent mean Ϯ SEM and Kaplan-Meier survival curves depict survival defined as time when tumor size Ͼ150 mm2 for n ϭ 12 mice.

growth inhibition (our unpublished observations). Another MC38 variant established with high expression of OVA (MC38 OVAbright) showed complete CD8ϩ T cell-dependent rejection in dim FIGURE 5. CD8ϩ T cell expansion, but not IFN-␥ response is enhanced WT (28), suggesting that MC38 OVA tumors generate an in- in IL-21Ϫ/Ϫ mice. B6 WT, TCR.J␣18Ϫ/Ϫ, or IL-21Ϫ/Ϫ mice were injected sufficient immune response for tumor rejection with the potential i.v. with either 400 ␮g of OVA alone or 400 ␮g of OVA plus 1 ␮g/mouse for modulation in IL-21Ϫ/Ϫ mice. We injected WT and IL-21Ϫ/Ϫ of ␣GC i.p. A, On days 7 or 35, spleens and livers of mice were harvested mice with MC38 OVAdim cells s.c. and monitored tumor growth to detect the Ag-specific CD8ϩ T cell response by SIINFEKL-loaded MHC and survival, defined as time when tumor size Ͼ150 mm2 (Fig. 6). class I tetramer staining. B, On day 7 or day 35 after OVA plus ␣GC The results showed equal tumor growth and survival in WT and Ϫ/Ϫ injection, splenocytes from WT or IL-21 mice were restimulated with IL-21Ϫ/Ϫ mice with a period of tumor growth delay observed in SIINFEKL peptide for 12 h in vitro, stained for CD8 and intracellular both strains. These results suggest that MC38 OVAdim tumor IFN-␥, and the percentages of CD8ϩIFN-␥ϩ were obtained by flow cy- growth is not controlled by IL-21-dependent mechanisms. tometry. Each data point represents individual mice and Mann-Whitney U Ϫ/Ϫ p Ͻ 0.05. IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice show potent CD8ϩ T cell ,ء .test was used to compare WT and IL-21 groups mediated antitumor response against immunogenic EG7 tumors in vitro, stained for CD8 and intracellular IFN-␥ expression and To investigate the CD8ϩ T cell response in another and more analyzed by flow cytometry (Fig. 5B). On day 7 a detectable immunogenic tumor model we used an immunogenic variant of the amount of IFN-␥ was found (0.1%–0.2% of lymphocytes), but OVA-expressing EL-4 lymphoma cell line EG7 grown in our lab- with similar levels in both WT and IL-21Ϫ/Ϫ and on day 35 the oratory. EG7 lymphomas have previously been found to respond to levels of IFN-␥ had dropped to very low levels. Taken together, IL-21 therapy (9). Our EG7 variant gives a minor fraction of spon- these results suggest that an Ag-specific CD8ϩ T cell response can taneous rejections in WT mice (our unpublished observations), so be mounted by NKT cell stimulation without IL-21 production, but this model has an active CD8ϩ T cell driven tumor immune re- that endogenous IL-21 has suppressive effects on Ag-specific sponse that possibly could be modulated by IL-21/IL-21R defi- CD8ϩ T cell expansion at least in liver. ciency. WT and IL-21Ϫ/Ϫ mice (Fig. 7A) or IL-21RϪ/Ϫ mice (Fig. 7B) were inoculated with EG7 cells s.c. and monitored for tumor Primary antitumor response against OVAdim expressing MC38 Ϫ/Ϫ growth and survival defined as the time when individual tumor size colon carcinoma is normal in IL-21 mice exceeded 150 mm2. The results showed, as previously observed, Seeing that both NK cell- and NKT cell-mediated tumor immune spontaneous complete rejection in WT mice of ϳ18% (3/17) and responses were largely unchanged by the lack of endogenous 33% (5/15) in the two separate experiments shown. Interestingly, IL-21 and IL-21R, but that a CD8ϩ T cell response increased in IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice both showed increased rejection IL-21Ϫ/Ϫ mice, we next wanted to explore the functionality of rates of 53% (9/17) and 94% (15/16), respectively. In Kaplan- CD8ϩ T cell-dependent antitumor immune responses without host Meier survival analysis this difference was significant for both IL- IL-21 signaling. Initially, we used a MC38 colon carcinoma cell 21Ϫ/Ϫ and IL-21RϪ/Ϫ when compared with WT ( p Ͻ 0.01). Fur- line transfected to express low levels of OVA (MC38 OVAdim). thermore, we detected the OVA-specific CD8ϩ T cell response in This cell line was recently established in our laboratory and found blood using SIINFEKL-loaded MHC class I tetramers on day 13 to give 100% outgrowth in WT mice, but with a period of tumor postinjection of tumor in the IL-21Ϫ/Ϫ experiment (Fig. 7C) and

7332 HOST IL-21 IN TUMOR IMMUNITY

FIGURE 7. IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice show powerful immune response against immunogenic EG7 tumors. In two separate experiments, B6 WT and IL-21Ϫ/Ϫ (A) or IL-21RϪ/Ϫ (B) mice were challenged with 3 ϫ 106 EG7 cells s.c. on the right flank and monitored for tumor growth and survival, which was defined as time when tumors Ͼ150 mm2. On day 13 postinjection of tumor, blood was drawn from WT and IL-21Ϫ/Ϫ mice and stained with SIINFEKL-loaded MHC class I tetramers (C) and similarly on day 10 postinjection tumor in WT and IL-21RϪ/Ϫ mice (D). Data are individual tumor growth curves showing number of mice in which tumors have grown out or regressed out of total mice in A and B. Kaplan-Meier survival curves depict time when individual tumors exceeded 150 mm2, Mantel Cox Log Rank test was used to evaluate statistical differences in survival between WT and IL-21Ϫ/Ϫ or -p Ͻ 0.01. Dot plots represent the percentage of SIINFEKL-specific cells of total CD8ϩ T cells in blood from individual mice. Mann ,ءء .IL-21RϪ/Ϫ .p Ͻ 0.05 ,ء .Whitney U test was used to compare differences between WT and IL-21Ϫ/Ϫ or IL-21RϪ/Ϫ day 10 postinjection of tumor in the IL-21RϪ/Ϫ experiment (Fig. ory responses. To evaluate the magnitude of the adaptive memory 7D). In general, more than 90% of the detected OVA-specific response in the absence of IL-21, blood samples were collected CD8ϩ T cells had the phenotype CD62LϪCD44ϩ (data not pre-rechallenge and on day 3 post-rechallenge to detect the level of shown), indicating that they were activated effector-memory cells. OVA-specific CD8ϩ T cells using SIINFEKL-loaded MHC class In the IL-21Ϫ/Ϫ experiment the results on day 13 showed a robust I tetramers (Fig. 8B). Before rechallenge all mice displayed a sig- OVA-specific CD8ϩ T cell response with a significant increase in nificant level of circulating SIINFEKL-specific CD8ϩ T cells but the percentage of OVA-specific CD8ϩ T cells in IL-21Ϫ/Ϫ mice with no difference between strains (WT: 1.4%, IL-21Ϫ/Ϫ: 1.5%, compared with WT mice with a mean of 9.6% vs 2.6% ( p Ͻ 0.05), IL-21RϪ/Ϫ: 2.3%, mean % of total CD8ϩ T cells), suggesting that respectively. Similar results were found in IL-21RϪ/Ϫ mice, where IL-21/IL-21R deficiency does not alter the circulating effector- OVA-specific CD8ϩ T cells were increased already on day 10 memory CD8ϩ T cell pool. On day 3 post-rechallenge the level of postinjection of tumor with generally lower percentages as ex- OVA-specific CD8ϩ T cells generally increased in all three strains pected at this earlier time point (mean of 1.1% in IL-21RϪ/Ϫ mice compared with pre-rechallenge (WT: 1.8%, IL-21Ϫ/Ϫ: 2.7%, IL- vs 0.7% in WT mice, ( p Ͻ 0.05). Furthermore, we found that the 21RϪ/Ϫ: 3.7%, mean % of total CD8ϩ T cells). For comparison, level of OVA-specific CD8ϩ T cells showed significant inverse naive WT and IL-21Ϫ/Ϫ or IL-21RϪ/Ϫ mice challenged with EG7 correlation with the tumor size at the time of sampling and it was showed on average Ͻ0.25% OVA-specific CD8ϩ T cells, indicat- also significantly predictive of complete rejection (data not ing that the level of de novo generated OVA-specific CD8ϩ T cells shown). Together, these observations suggest a suppressive role of was very low on day 3 postchallenge of tumor as expected. The endogenous IL-21 and IL-21R signaling in CD8ϩ T cell-depen- average absolute changes of OVA-specific CD8ϩ T cells on day 3 dent immunity toward immunogenic tumors, restricting the expan- relative to pre-rechallenge was 0.4% in WT mice, but was in- sion of Ag-specific CD8ϩ T cells. creased to 1.2% in IL-21Ϫ/Ϫ mice and 1.3% in IL-21RϪ/Ϫ mice. This difference was statistically significant in IL-21Ϫ/Ϫ mice com- ϩ Ϫ/Ϫ CD8 T cell-mediated memory responses are intact in IL-21 pared with WT mice ( p Ͻ 0.05), but not for IL-21RϪ/Ϫ mice. Ϫ/Ϫ and IL-21R mice Overall, these data suggest that CD8ϩ T cell memory is intact in Seeing the suppressive effect of IL-21 on primary tumor immune mice deficient for IL-21 and IL-21R, but that IL-21 also suppresses responses we next explored whether IL-21Ϫ/Ϫ or IL-21RϪ/Ϫ mice the secondary expansion of Ag-specific CD8ϩ T cells. might have altered long-term CD8ϩ T cell-based memory responses. To test this, cohorts of WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ Discussion mice that initially had rejected their primary tumor challenge with In this study, we present the first set of data examining both pri- the immunogenic EG7 tumors were rechallenged with conven- mary and secondary tumor immune responses in mice deficient of tional EG7 tumor cells. In naive WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ IL-21 or IL-21R. Our results show that these mice have intact mice, these tumors all grew out (data not shown). The rechallenged tumor immune surveillance, primary and secondary tumor immu- mice were injected between 40 days and up to 6 mo after their nity, but, in contrast to the literature, reveal a suppressive role for primary tumor rejection and monitored for tumor growth (Fig. 8A). endogenous IL-21 during Ag-specific CD8ϩ T cell expansion and Initial tumor growth was observed in most mice followed by com- reactivity to immunogenic tumors. Exogenous IL-21 has shown plete rejection of all tumors in both WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ significant antitumor activity in numerous experimental mouse tu- mice, within 10–15 days post injection, suggesting that IL-21Ϫ/Ϫ mor studies, either secreted from artificial tumor cells, expressed and IL-21RϪ/Ϫ mice have competent CD8ϩ T cell-mediated mem- by plasmids, or injected as recombinant protein (4). Studies of

The Journal of Immunology 7333

FIGURE 8. CD8ϩ T cell-mediated memory responses are intact in IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice. A, Groups of B6 WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ mice that had initially rejected a primary tumor challenge with 3 ϫ 106 EG7 (immunogenic) cells s.c. on the right flank were rechallenged Ͼ40 days and up to 6 mo postrejection with 3 ϫ 106 EG7 (conventional) s.c. on the flank contralateral to the primary challenge and monitored for tumor growth. B, Pre-rechallenge and on day 3 post-rechallenge blood was drawn and stained with SIINFEKL-loaded MHC-I tetramers. Data in A depict individual tumor growth curves, for n ϭ 8 (WT), n ϭ 13 (IL-21Ϫ/Ϫ), and n ϭ 14 (IL-21RϪ/Ϫ) mice and are representative of three independent experiments. B, Dot plots represent percentage of SIINFEKL-specific cells of total CD8ϩ T cells in blood from individual mice pre-rechallenge and day 3 post-rechallenge and the absolute percentage of change from pre-rechallenge to day 3 post-rechallenge. Kruskal-Wallis test with Dunn’s posttest was used to compare the absolute .p Ͻ 0.05 compared with WT ,ء percentage of change in WT, IL-21Ϫ/Ϫ, and IL-21RϪ/Ϫ mice endogenous IL-21 have established that the cytokine and its responses with different degrees of immune activity. B16 melano- signaling pathway play a significant role in the pathogenesis of mas showed equal s.c. growth kinetics and development of lung several experimental autoimmune diseases, including colitis (27), metastases and RMAS-Rae1␤ tumors were equally rejected in the diabetes (25), arthritis (24), and systemic lupus erythematosus absence of IL-21 signaling. Recombinant IL-21 has previously (23), with experimental autoimmune encephalomyelitis being de- shown antitumor responses in both tumor models; in B16 mela- bated (41, 42). These results highlight that IL-21 is primarily a noma through actions on either NK cells or CD8ϩ T cells (6, 10, proinflammatory cytokine. However, our data now extend these 33), and in RMAS-Rae1␤ tumors through NK cells in an NKG2D- results and propose that endogenous IL-21 might also have poten- dependent manner (7). B16 melanomas are well known for being tially immunosuppressive effects. aggressive and poorly immunogenic tumors, which might not gen- IL-21 therapy has shown antitumor activity through effects on erate a sufficient host immune response for endogenous IL-21 to NK cells and CD8ϩ T cells (4), and augmented antitumor re- play a role and the level of endogenous IL-21 production could in sponses by NKT cells (8). Based on these data we anticipated that this study be insignificant or otherwise redundant. RMAS-Rae1␤ IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice might have reduced immunity to tumors clearly show an active immune response as seen previously tumors particularly mediated by these effector cells. We found in- (35), but in this study it might be that NKG2D stimulation alone is tact tumor immune surveillance toward MCA-induced sarcomas sufficient for tumor rejection and because this response is mainly despite the lack of functional IL-21 or IL-21R. In contrast to trans- NK cell-mediated it could also be that IL-21-producing cells are plantable tumor models, the MCA-induced sarcoma model reflects not properly engaged. However, when we rechallenged mice pos- a more biologically relevant carcinogenesis process, which is sub- trejection of RMAS-Rae1␤ tumor with H-2bϩ RMA cells, which jected to natural tumor immune surveillance mainly conducted by are controlled by T cells but not NK cells, we found powerful but NKT cells and NK cells in a perforin- and IFN-␥-dependent man- equal rejection compared with WT. Although these results suggest ner (43, 44). So, during the course of this model, activation of NK involvement of T cells, under these conditions IL-21 signaling was and NKT cells must be occurring. This seems to be a suitable not essential for the successful transition from innate NK cell- model to test for potential IL-21 involvement as IL-21 is reported mediated immunity to adaptive T cell-mediated immunity, which to increase IFN-␥ production and perforin-dependent antitumor has previously been suggested (16). responses by NK cells as well as stimulate granularity and IFN-␥ A critical factor in these experiments is whether IL-21 is pro- production by NKT cells, which can also produce IL-21 upon ac- duced to any significant amount. To address this proposal, we used tivation (2, 33). the CD1d-restricted ligand ␣GC, which induce IL-21 production We used the B16 melanoma model and H-2bϪ RMAS-Rae1␤ from NKT cells as recently described (2). NKT cells secondarily tumors to detect potential IL-21 involvement in primary tumor enhance IFN-␥ production and antitumor activity of NK cells (45),

7334 HOST IL-21 IN TUMOR IMMUNITY

IL-21 can activate NKT cells (2), and similarly enhance IFN-␥ We mainly found increased CD8ϩ T cell expansion in the liver production and antitumor activity of NK cells (16, 33). So, it could in response to OVA and ␣GC immunization, which is most likely be that IL-21 was a potential link in the direct effects of ␣GC on due to the increased proportion of NKT cells in this organ com- NKT cells and in the sequential effects on NK cells. However, we pared with the spleen. However, we only observed an increase in found that NKT cell expansion and IFN-␥ production in response OVA-specific CD8ϩ T cells on day 7 after ␣GC and OVA immu- to ␣GC stimulation were not influenced by host IL-21 nor was the nization, whereas similar levels were found on day 35 postimmu- concomitant IFN-␥ production from NK cells. These data suggest nization. And, the circulating level of effector/memory CD8ϩ T that IL-21 does not have a major autocrine role during ␣GC acti- cells more than 40 days postrejection of EG7 was similar between vation of NKT cells or in paracrine stimulation of NK cells. Our WT, IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice. These results indicate that finding that IL-21 was required to some extent for increased gran- host IL-21 is not involved in the long-term maintenance of effec- zyme B expression in NKT, NK and CD8ϩ T cells following ␣GC tor/memory CD8ϩ T cells. This suggestion seems to contrast find- stimulation indicates that IL-21 is produced following ␣GC stim- ings with recombinant IL-21 (9), however, in this study tumors ulation and possibly plays a role in the optimal development of were not completely rejected as in our model, but instead showed effector functions in these cells. This finding is supported by sev- chronic growth inhibition maintaining the presence of Ag. Fur- eral studies in both mice and humans showing that IL-21 up-reg- thermore, in studies of viral infections IL-21 was not needed to ϩ ulates granzyme B in NK cells, CD8ϩ T cells and B cells (2, 21, maintain circulating effector/memory CD8 T cells after a re- ϩ 22, 46, 47). However, our finding that ␣GC pulsed B16 melano- solved infection, only for maintaining effector CD8 T cells dur- mas were also strongly rejected in the absence of IL-21 suggests ing chronic infections (53). Taken together, these results indicate ϩ that host IL-21 is not a critical factor in ␣GC-mediated NKT cell- that IL-21 is not required to maintain circulating memory CD8 T ϩ dependent tumor rejection. In contrast, we have previously shown cells in the absence of Ag, but can sustain CD8 T cell responses that the sequential activation of NKT cells and NK cells with ␣GC during Ag persistency. This is also consistent with the findings that followed by recombinant IL-21 stimulation was a very effective TCR stimulation is required for IL-21 production (1, 2). In addition to its effects during primary CD8ϩ T cell-mediated treatment of experimental tumors (8). Interestingly, this study ϩ showed that the additional effects of IL-21 were seen only if the tumor immunity, IL-21 also has a putative role in secondary CD8 treatment was given 3 days after ␣GC administration and not if T cell memory responses (9, 11, 15, 54). However, we found that secondary CD8ϩ T cell memory responses were normal or even IL-21 treatment was started together with or 6 days after ␣GC increased in IL-21Ϫ/Ϫ and IL-21RϪ/Ϫ mice, indicating that endog- administration, showing that the timing or context of IL-21 is es- enous IL-21 might also limit secondary effector CD8ϩ T cell ex- sential for the antitumor response. So, although the IL-21 produc- pansion. Consistently, IL-21 was not required for viral recall re- tion we achieve following ␣GC stimulation might be insignificant sponses (53). However, in this study IL-21 did not limit secondary to add to the antitumor effect, another explanation might be that Ag-specific T cell expansion and although differences might have IL-21 is produced at a suboptimal time point to add to the antitu- been missed due to small group sizes, more work is warranted to mor effect in the models tested. confirm the role of endogenous IL-21 during secondary CD8ϩ T IL-21 mediates antitumor effects through stimulation of CD8ϩ T cell expansion. At present, the mechanism behind the suppressive cells, and is a potent costimulant of Ag-specific CD8ϩ T cell ex- effect of IL-21 is unknown. Recently, it was found that TCR prim- pansion and cytotoxicity (9–11, 48), suggesting that such re- ing in the presence of IL-21 results in IL-10-producing immuno- sponses might be impaired without functional IL-21 signaling. In suppressive T cells (55). So, the production of IL-21 in response to contrast to what is presented in the literature with exogenous IL- Ag could mediate suppression either through inhibition of DCs as 21, we found that endogenous IL-21 has suppressive activity on ϩ ϩ stated above or by induction of IL-10-producing immunosuppres- CD8 T cell responses, restricting Ag-specific CD8 T cell ex- sive T cells. In contrast with this scenario, we did not see any pansion and antitumor activity against immunogenic tumors. suppressive effects of host IL-21 on the growth of less immuno- These results suggest that IL-21 is produced during T cell sensitive genic MC38-OVAdim tumors. This may be due to an insufficient tumor immune responses, but unexpectedly has suppressive rather Ag-induced immune response or other redundant immune suppres- than stimulatory effects. In support of these findings, we have pre- sive mechanisms by this tumor. Ϫ/Ϫ Ϫ/Ϫ viously reported that IL-21 and IL-21R showed exacer- In this study, we anticipated that endogenous IL-21 would to bated responses to experimental autoimmune encephalomyelitis some degree be involved in tumor immunity. However, based on (41). Our results are also supported by studies showing that IL-21 the data presented in this study, we conclude that endogenous can maintain DCs in an immature state inhibiting Ag presentation IL-21 and IL-21R is not required for immune surveillance or tumor and T cell activation (49, 50). Furthermore, they are supported by immunity mediated by NK, NKT, or CD8ϩ T cells, at least in the results showing that the antitumor activity of IL-21 is lost by treat- range of tumor models tested. We have not investigated the role of ing mice around the time of tumor inoculation as compared with endogenous IL-21 in the development of B cell lymphomas or in treatment started just 2 days posttumor inoculation (9). A recent B cell-dependent tumor rejection, in which IL-21 could have a role trio of studies suggests that IL-21 signaling is essential for main- (18, 56). However, our findings suggest that anti-IL-21 treatment, taining CD8ϩ T cell responses during chronic viral infections (51– which has been proposed as a potential treatment strategy for many 53). Interestingly, they also show that host IL-21 limits Ag-specific autoimmune diseases (57), would not compromise tumor immune CD8ϩ T cell expansion during the acute phase of infections, surveillance. Indeed, given the enhanced CD8ϩ T cell response, whereas IL-21 is strictly needed to sustain virus-specific CD8ϩ T short-term blockade of host IL-21 signaling during the early prim- cells during chronic infections (52). Together with our results, ing phase followed by stimulation with recombinant IL-21 therapy these findings support our notion that the timing or context of might even enhance tumor immunity. We have highlighted that the IL-21 is critical for the outcome of an immune response; whereas actions of IL-21 are potentially very context-dependent, but it is endogenous IL-21 produced during the priming phase might limit still unknown which cells produce IL-21 during a tumor immune expansion of Ag-specific CD8ϩ T cells and antitumor activity, response, in which anatomical context, and when does this happen. IL-21 delivered after the initial priming works to sustain or boost Injection of recombinant IL-21 into mice results in high nano- CD8ϩ T cell activity against both viruses and tumors. gram per milliliter serum concentrations that are maintained for

The Journal of Immunology 7335 many hours (10), and generally endogenous serum cytokine con- 13. Di Carlo, E., A. Comes, A. M. Orengo, O. Rosso, R. Meazza, P. Musiani, centrations are only in the picogram range even after stimulation. M. P. Colombo, and S. Ferrini. 2004. IL-21 induces tumor rejection by specific CTL and IFN-␥-dependent CXC chemokines in syngeneic mice. J. Immunol. Thus it is perhaps not surprising that our study highlights the great 172: 1540–1547. difference between studying the function of an administered re- 14. Ma, H. L., M. J. Whitters, R. F. Konz, M. Senices, D. A. Young, M. J. Grusby, M. Collins, and K. Dunussi-Joannopoulos. 2003. IL-21 activates both innate and combinant cytokine and the endogenous cytokine by gene target- adaptive immunity to generate potent antitumor responses that require perforin ϩ ing. CD4 T cells are the most likely producers of IL-21 in our but are independent of IFN-␥. J. Immunol. 171: 608–615. tumor model and the need for TCR stimulation suggests that sec- 15. Allard, E. L., M. P. Hardy, J. Leignadier, M. Marquis, J. Rooney, D. Lehoux, and N. Labrecque. 2007. Overexpression of IL-21 promotes massive CD8ϩ memory ondary lymphoid organs are the primary location for IL-21 pro- T cell accumulation. Eur. J. Immunol. 37: 3069–3077. duction. This suggests that IL-21 could be produced during the 16. Kasaian, M. T., M. J. Whitters, L. L. Carter, L. D. Lowe, J. M. Jussif, early priming phase, which could be the context of its suppressive B. Deng, K. A. Johnson, J. S. Witek, M. Senices, R. F. Konz, et al. 2002. IL-21 limits NK cell responses and promotes antigen-specific T cell activa- actions. However, we believe the generation of reporter mice co- tion: a mediator of the transition from innate to adaptive immunity. Immunity expressing, e.g., GFP along with IL-21 will be an essential tool to 16: 559–569. formally answer these questions. 17. Nakano, H., T. Kishida, H. Asada, M. Shin-Ya, T. Shinomiya, J. Imanishi, T. Shimada, S. Nakai, M. Takeuchi, Y. Hisa, and O. Mazda. 2006. Interleukin-21 In summary, we have demonstrated that endogenous IL-21 sig- triggers both cellular and humoral immune responses leading to therapeutic an- naling is not required for tumor immune surveillance, NK, NKT, titumor effects against head and neck squamous cell carcinoma. J. Gene Med. 8: and CD8ϩ T cell-dependent primary tumor immunity, or for 90–99. ϩ 18. Daga, A., A. M. Orengo, R. M. Gangemi, D. Marubbi, M. Perera, A. Comes, CD8 T cell-dependent secondary memory responses. However, S. Ferrini, and G. Corte. 2007. Glioma immunotherapy by IL-21 gene-modified we found that endogenous IL-21 restricts CD8ϩ T cell expansion cells or by recombinant IL-21 involves antibody responses. Int. J. Cancer 121: and immunity toward immunogenic tumors. These results reveal 1756–1763. 19. Maeda, M., Y. Yanagawa, K. Iwabuchi, K. Minami, Y. Nakamaru, D. Takagi, an unexpected suppressive role for IL-21 in Ag-specific immunity S. Fukuda, and K. Onoe. 2007. IL-21 enhances dendritic cell ability to induce in contrast to its general perception as a proinflammatory cytokine interferon-␥ production by natural killer T cells. Immunobiology 212: 537–547. in both cancer immunotherapy and autoimmunity. 20. Davis, I. D., B. K. Skrumsager, J. Cebon, T. Nicholaou, J. W. Barlow, N. P. Moller, K. Skak, D. Lundsgaard, K. S. Frederiksen, P. Thygesen, and G. A. McArthur. 2007. An open-label, two-arm, phase I trial of recombinant Acknowledgments human interleukin-21 in patients with metastatic melanoma. Clin. Cancer Res. 13: 3630–3636. We thank Michelle Stirling for the breeding and maintenance of the mice 21. Davis, I. D., B. Brady, R. F. Kefford, M. Millward, J. Cebon, B. K. Skrumsager, in these studies and Charlene Guan and Suzanne Medwell for the geno- U. Mouritzen, L. T. Hansen, K. Skak, D. Lundsgaard, et al. 2009. Clinical and typing of mice. We also thank Ralph Rossi for flow cytometry support and biological efficacy of recombinant human interleukin-21 in patients with stage IV Konstantinos Kyparissoudis for providing ␣GC-loaded CD1d tetramer. malignant melanoma without prior treatment: a phase IIa trial. Clin. Cancer Res. 15: 2123–2129. 22. Frederiksen, K. S., D. Lundsgaard, J. A. Freeman, S. D. Hughes, T. L. Holm, Disclosures B. K. Skrumsager, A. Petri, L. T. Hansen, G. A. McArthur, I. D. Davis, and ϩ H.S. and K.S. are employed by Novo Nordisk, and P.V.S. is employed by K. Skak. 2008. IL-21 induces in vivo immune activation of NK cells and CD8 T cells in patients with metastatic melanoma and renal cell carcinoma. Cancer Zymogenetics. D.I.G. and M.J.S. have received research support from Immunol. Immunother. 57: 1439–1449. Novo Nordisk. The remaining authors have no financial conflict of interest. 23. Bubier, J. A., T. J. Sproule, O. Foreman, R. Spolski, D. J. Shaffer, H. C. Morse, III, W. J. Leonard, and D. C. Roopenian. 2009. A critical role for IL-21 receptor signaling in the pathogenesis of systemic lupus erythematosus in BXSB-Yaa References mice. Proc. Natl. Acad. Sci. USA 106: 1518–1523. 1. Parrish-Novak, J., S. R. Dillon, A. Nelson, A. Hammond, C. Sprecher, 24. Jang, E., S. H. Cho, H. Park, D. J. Paik, J. M. Kim, and J. Youn. 2009. A positive J. A. Gross, J. Johnston, K. Madden, W. Xu, J. West, et al. 2000. Interleukin 21 feedback loop of IL-21 signaling provoked by homeostatic CD4ϩ. J. Immunol. and its receptor are involved in NK cell expansion and regulation of lymphocyte 182: 4649–4656. function. Nature 408: 57–63. 25. Spolski, R., M. Kashyap, C. Robinson, Z. Yu, and W. J. Leonard. 2008. IL-21 2. Coquet, J. M., K. Kyparissoudis, D. G. Pellicci, G. Besra, S. P. Berzins, signaling is critical for the development of type I diabetes in the NOD mouse. M. J. Smyth, and D. I. Godfrey. 2007. IL-21 is produced by NKT cells and Proc. Natl. Acad. Sci. USA 105: 14028–14033. modulates NKT cell activation and cytokine production. J. Immunol. 178: 26. Young, D. A., M. Hegen, H. L. Ma, M. J. Whitters, L. M. Albert, L. Lowe, 2827–2834. M. Senices, P. W. Wu, B. Sibley, Y. Leathurby, T. P. Brown, 3. Spolski, R., and W. J. Leonard. 2008. Interleukin-21: basic biology and impli- C. Nickerson-Nutter, J. C. Keith, Jr., and M. Collins. 2007. Blockade of the cations for cancer and autoimmunity. Annu. Rev. Immunol. 26: 57–79. interleukin-21/interleukin-21 receptor pathway ameliorates disease in animal 4. Leonard, W. J., and R. Spolski. 2005. Interleukin-21: a modulator of lymphoid models of rheumatoid arthritis. Arthritis Rheum. 56: 1152–1163. proliferation, apoptosis and differentiation. Nat. Rev. Immunol. 5: 688–698. 27. Fina, D., M. Sarra, M. C. Fantini, A. Rizzo, R. Caruso, F. Caprioli, C. Stolfi, 5. Skak, K., M. Kragh, D. Hausman, M. J. Smyth, and P. V. Sivakumar. 2008. I. Cardolini, M. Dottori, M. Boirivant, F. Pallone, T. T. MacDonald, and Interleukin 21: combination strategies for cancer therapy. Nat. Rev. Drug Discov. G. Monteleone. 2008. Regulation of gut inflammation and th17 cell response by 7: 231–240. interleukin-21. Gastroenterology 134: 1038–1048. 6. Wang, G., M. Tschoi, R. Spolski, Y. Lou, K. Ozaki, C. Feng, G. Kim, 28. Gilfillan, S., C. J. Chan, M. Cella, N. M. Haynes, A. S. Rapaport, K. S. Boles, W. J. Leonard, and P. Hwu. 2003. In vivo antitumor activity of interleukin 21 D. M. Andrews, M. J. Smyth, and M. Colonna. 2008. DNAM-1 promotes acti- mediated by natural killer cells. Cancer Res. 63: 9016–9022. vation of cytotoxic lymphocytes by nonprofessional antigen-presenting cells and 7. Takaki, R., Y. Hayakawa, A. Nelson, P. V. Sivakumar, S. Hughes, M. J. Smyth, tumors. J. Exp. Med. 205: 2965–2973. and L. L. Lanier. 2005. IL-21 enhances tumor rejection through a NKG2D-de- 29. Smyth, M. J., N. Y. Crowe, D. G. Pellicci, K. Kyparissoudis, J. M. Kelly, pendent mechanism. J. Immunol. 175: 2167–2173. K. Takeda, H. Yagita, and D. I. Godfrey. 2002. Sequential production of inter- 8. Smyth, M. J., M. E. Wallace, S. L. Nutt, H. Yagita, D. I. Godfrey, and feron-␥ by NK1.1ϩ T cells and natural killer cells is essential for the antimeta- Y. Hayakawa. 2005. Sequential activation of NKT cells and NK cells provides static effect of ␣-galactosylceramide. Blood 99: 1259–1266. effective innate immunotherapy of cancer. J. Exp. Med. 201: 1973–1985. 9. Moroz, A., C. Eppolito, Q. Li, J. Tao, C. H. Clegg, and P. A. Shrikant. 2004. 30. Swann, J. B., M. D. Vesely, A. Silva, J. Sharkey, S. Akira, R. D. Schreiber, and IL-21 enhances and sustains CD8ϩ T cell responses to achieve durable tumor M. J. Smyth. 2008. Demonstration of inflammation-induced cancer and cancer immunity: comparative evaluation of IL-2, IL-15, and IL-21. J. Immunol. 173: immunoediting during primary tumorigenesis. Proc. Natl. Acad. Sci. USA 105: 900–909. 652–656. 10. Sondergaard, H., K. S. Frederiksen, P. Thygesen, E. D. Galsgaard, K. Skak, 31. Crowe, N. Y., M. J. Smyth, and D. I. Godfrey. 2002. A critical role for natural P. E. Kristjansen, N. Odum, and M. Kragh. 2007. Interleukin 21 therapy increases killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. the density of tumor infiltrating CD8ϩ T cells and inhibits the growth of synge- J. Exp. Med. 196: 119–127. neic tumors. Cancer Immunol. Immunother. 56: 1417–1428. 32. Swann, J. B., and M. J. Smyth. 2007. Immune surveillance of tumors. J. Clin. 11. Zeng, R., R. Spolski, S. E. Finkelstein, S. Oh, P. E. Kovanen, C. S. Hinrichs, Invest. 117: 1137–1146. C. A. Pise-Masison, M. F. Radonovich, J. N. Brady, N. P. Restifo, et al. 2005. 33. Brady, J., Y. Hayakawa, M. J. Smyth, and S. L. Nutt. 2004. IL-21 induces the Synergy of IL-21 and IL-15 in regulating CD8ϩ T cell expansion and function. functional maturation of murine NK cells. J. Immunol. 172: 2048–2058. J. Exp. Med. 201: 139–148. 34. Karre, K., H. G. Ljunggren, G. Piontek, and R. Kiessling. 1986. Selective rejec- 12. Comes, A., O. Rosso, A. M. Orengo, E. Di Carlo, C. Sorrentino, R. Meazza, tion of H-2-deficient lymphoma variants suggests alternative immune defence T. Piazza, B. Valzasina, P. Nanni, M. P. Colombo, and S. Ferrini. 2006. strategy. Nature 319: 675–678. CD25ϩ regulatory T cell depletion augments immunotherapy of micrometas- 35. Hayakawa, Y., J. M. Kelly, J. A. Westwood, P. K. Darcy, A. Diefenbach, tases by an IL-21-secreting cellular vaccine. J. Immunol. 176: 1750–1758. D. Raulet, and M. J. Smyth. 2002. Cutting edge: tumor rejection mediated by

7336 HOST IL-21 IN TUMOR IMMUNITY

NKG2D receptor-ligand interaction is dependent upon perforin. J. Immunol. 169: perforin-mediated cytotoxicity, to anti-metastatic effect of ␣-galactosylceramide. 5377–5381. Eur. J. Immunol. 31: 1720–1727. 36. Carnaud, C., D. Lee, O. Donnars, S. H. Park, A. Beavis, Y. Koezuka, and 46. Casey, K. A., and M. F. Mescher. 2007. IL-21 promotes differentiation of naive A. Bendelac. 1999. Cutting edge: cross-talk between cells of the innate im- CD8 T cells to a unique effector phenotype. J. Immunol. 178: 7640–7648. mune system: NKT cells rapidly activate NK cells. J. Immunol. 163: 47. Skak, K., K. S. Frederiksen, and D. Lundsgaard. 2008. Interleukin-21 activates 4647– 4650. human natural killer cells and modulates their surface receptor expression. Im- 37. Shimizu, K., A. Goto, M. Fukui, M. Taniguchi, and S. Fujii. 2007. Tumor munology 123: 575–583. cells loaded with ␣-galactosylceramide induce innate NKT and NK cell-de- 48. Liu, S., G. Lizee, Y. Lou, C. Liu, W. W. Overwijk, G. Wang, and P. Hwu. 2007. pendent resistance to tumor implantation in mice. J. Immunol. 178: IL-21 synergizes with IL-7 to augment expansion and anti-tumor function of 2853–2861. cytotoxic T cells. Int. Immunol. 19: 1213–1221. 38. Nishimura, T., H. Kitamura, K. Iwakabe, T. Yahata, A. Ohta, M. Sato, K. Takeda, 49. Brandt, K., S. Bulfone-Paus, A. Jenckel, D. C. Foster, R. Paus, and K. Okumura, K. L. Van, T. Kawano, et al. 2000. The interface between innate and R. Ruckert. 2003. Interleukin-21 inhibits dendritic cell-mediated T cell acti- acquired immunity: glycolipid antigen presentation by CD1d-expressing den- vation and induction of contact hypersensitivity in vivo. J. Invest. Dermatol. dritic cells to NKT cells induces the differentiation of antigen-specific cytotoxic 121: 1379–1382. T lymphocytes. Int. Immunol. 12: 987–994. 50. Brandt, K., S. Bulfone-Paus, D. C. Foster, and R. Ruckert. 2003. Interleukin-21 39. Hermans, I. F., J. D. Silk, U. Gileadi, M. Salio, B. Mathew, G. Ritter, R. Schmidt, inhibits dendritic cell activation and maturation. Blood 102: 4090–4098. A. L. Harris, L. Old, and V. Cerundolo. 2003. NKT cells enhance CD4ϩ and 51. Yi, J. S., M. Du, and A. J. Zajac. 2009. A vital role for interleukin-21 in the CD8ϩ T cell responses to soluble antigen in vivo through direct interaction with control of a chronic viral infection. Science 324: 1572–1576. dendritic cells. J. Immunol. 171: 5140–5147. 40. Stober, D., I. Jomantaite, R. Schirmbeck, and J. Reimann. 2003. NKT cells pro- 52. Elsaesser, H., K. Sauer, and D. G. Brooks. 2009. IL-21 is required to control vide help for dendritic cell-dependent priming of MHC class I-restricted CD8ϩ chronic viral infection. Science 324: 1569–1572. T cells in vivo. J. Immunol. 170: 2540–2548. 53. Frohlich, A., J. Kisielow, I. Schmitz, S. Freigang, A. T. Shamshiev, J. Weber, 41. Coquet, J. M., S. Chakravarti, M. J. Smyth, and D. I. Godfrey. 2008. Cutting B. J. Marsland, A. Oxenius, and M. Kopf. 2009. IL-21R on T cells is critical for edge: IL-21 is not essential for Th17 differentiation or experimental autoimmune sustained functionality and control of chronic viral infection. Science 324: encephalomyelitis. J. Immunol. 180: 7097–7101. 1576–1580. 42. Nurieva, R., X. O. Yang, G. Martinez, Y. Zhang, A. D. Panopoulos, L. Ma, 54. Rodrigues, L., S. Nandakumar, C. Bonorino, B. T. Rouse, and U. Kumaraguru. K. Schluns, Q. Tian, S. S. Watowich, A. M. Jetten, and C. Dong. 2007. Essential 2009. IL-21 and IL-15 cytokine DNA augments HSV specific effector and mem- ϩ autocrine regulation by IL-21 in the generation of inflammatory T cells. Nature ory CD8 T cell response. Mol. Immunol. 46: 1494–1504. 448: 480–483. 55. Spolski, R., H. P. Kim, W. Zhu, D. E. Levy, and W. J. Leonard. 2009. IL-21 43. Street, S. E., E. Cretney, and M. J. Smyth. 2001. Perforin and interferon-␥ ac- mediates suppressive effects via its induction of IL-10. J. Immunol. 182: tivities independently control tumor initiation, growth, and metastasis. Blood 97: 2859–2867. 192–197. 56. Gelebart, P., Z. Zak, M. Anand, J. en-Bard, H. M. Amin, and R. Lai. 2009. 44. Smyth, M. J., N. Y. Crowe, and D. I. Godfrey. 2001. NK cells and NKT cells Interleukin-21 effectively induces apoptosis in mantle cell lymphoma through a collaborate in host protection from methylcholanthrene-induced fibrosarcoma. STAT1-dependent mechanism. Leukemia 23: 1836–1846. Int. Immunol. 13: 459–463. 57. Ettinger, R., S. Kuchen, and P. E. Lipsky. 2008. Interleukin 21 as a target of 45. Hayakawa, Y., K. Takeda, H. Yagita, S. Kakuta, Y. Iwakura, K. L. Van, I. Saiki, intervention in autoimmune disease. Ann. Rheum. Dis. 67 (Suppl. 3): and K. Okumura. 2001. Critical contribution of IFN-␥ and NK cells, but not iii83–iii86.

Chapter 4 – Discussion

The primary objective of this thesis was to evaluate the anti-tumor effects of IL-21 protein therapy in relevant preclinical cancer models and secondarily investigate the role of endogenous IL-21 in tumor immunity.

In Paper II and III, we have demonstrated that IL-21 significantly inhibited growth of B16 melanoma and RenCa renal cell carcinoma. These models were chosen based on their wide usage in cancer therapy studies and responsiveness to immunotherapy (Sayers et al., 1998; Wigginton et al., 1996; DeMatos et al., 1998; van et al., 1999). In order to mimic conditions in humans, where the immune system poorly recognizes tumors, B16 and RenCa cells were inoculated in their respective syngeneic hosts, C57BL/6 and BALB/c mice, providing low intrinsic immune reactivity. Furthermore, these mouse strains are fully immunocompetent in contrast to e.g. xenograft models allowing the entire immune repertoire to come in to play as it would in humans. However, the syngeneic interaction between the tumor and host also makes it challenging to mount an immune response and these models therefore represent reasonable models for the testing of tumor immunotherapy. In Paper II and III RenCa tumors were generally found to be more responsive to IL-21 therapy compared to B16 tumors and showed higher densities of tumor infiltrating T cells. These observations were expected based on the literature showing increased immunogenicity of RenCa compared to B16 (Krup et al., 1999; Scheffer et al., 2003), and in Paper III these observations were extended to our models, where increased expression of MHC class I was found on RenCa compare to B16 cells. Nevertheless, B16 and RenCa are both aggressive tumors; in Paper II and III vehicle treated tumors reached a tumor size of ~1000 mm3 within 16-22 days and both tumors have been found to kill their host within approximately 25 days post intraperitoneal injection (Krup et al., 1999). Thus, B16 and RenCa are aggressive preclinical tumor models with different inherent immunogenicity and it was encouraging that significant anti-tumor effect was obtained in both models despite their limited therapeutic window. The use of therapeutic administration of IL-21 in both Paper II and III, where treatment of even larger established tumors results in significant tumor-growth inhibition, further strengthens the results. Based on these model characteristics, we consider the results obtained in Paper II and III clinically relevant.

However, several limitations to these models should be acknowledged. First, both models rely on transplantation of 105 or more tumor cells, which poorly reflect the slow development of human cancer, where the immune system gradually tolerates a tumor. Instead, the injection of tumor cells could generate considerable cell death, which abruptly exposes the immune system

to potentially immunogenic antigens. Second, tumor cells were injected subcutaneously, which may represent a site with enhanced immunosurveillance because of its critical junction between the host and the environment (Girardi, 2007), and although this site has some anatomical relevance for melanomas, it has much less so for renal cancer. Third, the transplantable tumors used here have a low frequency of spontaneous metastasis, and the effect of IL-21 on spontaneous metastasis was therefore not evaluated, although this is the predominant cause of death in humans. An alternative to transplantable models could have been transgenic cancer models, which more closely mimic the slow process of human cancer development and now include many organ-specific models (Ostrand-Rosenberg, 2004). However, cancers in these models often take months to develop with differential onset and disease penetrance, making controlled pharmacological testing of drugs very difficult. Since there are pros and cons for all animal models results should preferably be reproduced in more than one model. Thus, it was encouraging to show qualitatively similar anti-tumor effect and similar effect on the density of tumor infiltrating T cells in both B16 and RenCa tumors, although of different magnitudes.

In Paper II the anti-tumor efficacy of subcutaneous administration appeared greater than intraperitoneal administration particularly in the RenCa model and in Paper III intratumoral administration showed superior long-term tumor-growth inhibition compared to subcutaneous administration in RenCa. To compare the anti-tumor efficacies in Paper II and III across the different administration routes and various starting tumor sizes the classical T/C% value has been used (see Table 1 and 2). The T/C% value describes the ratio of the mean endpoint tumor sizes of treated over controls, hence a lower T/C% value equals better efficacy. It has traditionally been used to compare the efficacy of chemotherapies in preclinical studies and shown to be predictive of their response rates in clinical trials against certain cancers (Voskoglou-Nomikos et al., 2003). T/C% values support the notion that intraperitoneal administration is less efficacious compared to subcutaneous administration and that intratumoral administration has the greatest efficacy of the evaluated administration routes. These data suggest that intratumoral administration is a more effective administration route for IL-21. Unfortunately, intratumoral administration is not very practical, since metastatic cancer rarely is available for intratumoral injections. However, intratumoral injections could help increase immune responses toward primary tumors, if such are accessible, and results from IL-2 clinical trials have shown that intratumoral administration could increase the response rates and lower side effects in soft tissue skin metastasis of melanoma (Radny et al., 2003). The T/C% values also show that subcutaneous administration is a feasible administration route for IL-21. Thus, subcutaneous administration of IL-21 deserves clinical investigation since it

could potentially lower adverse events while sustaining efficacy as previously shown with IL-2 (Geertsen et al., 2004; Negrier et al., 2008). Also, subcutaneous administration might allow outpatient treatment given the moderate toxicity of IL-21 (Davis et al., 2007; Davis et al., 2009).

Table 1. Summary of efficacy in B16 melanomas

Treatment start (~mean tumor size) Paper/Figure Dose & schedule T/C%* P-value$ & route of administration Early treatment (~5mm3) I.p. II/2A 50 μg daily 30 <0.001 S.c. II/2A 50 μg daily 33 <0.001 Late treatment (~50mm3) I.p II/2B 50 μg daily 73 0.19 S.c. II/2B 50 μg daily 63 <0.05 I.t. III/3A 50 μg daily 42 <0.001 Late treatment (~70mm3) S.c. III/3B 50 μg daily 60 0.24 I.t. III/3B 50 μg daily 42 <0.05 I.t. III/3B 10 μg daily 56 0.14

I.t. III/3B 2 μg daily 83 0.59

* T/C% = (Treated mean endpoint tumor size)/(Control mean endpoint tumor size)x100%. $ Student’s t-test.

Table 2. Summary of efficacy in RenCa renal cell carcinomas

Treatment start (~mean tumor size) Paper/Figure Dose & schedule T/C%* P-value$ & route of administration Early treatment (~5mm3) I.p. II/2C 50 μg 3x/week 70 0.06 S.c. II/2C 50 μg 3x/week 43 <0.001 Late treatment (~50mm3) I.p II/2D 50 μg 3x/week 76 0.26

S.c. II/2D 50 μg 3x/week 50 <0.05

I.t. III/1A 50 µg 3x/week 23 <0.001 Late treatment (~100mm3) S.c. III/1B 50 µg 3x/week 61 0.09 I.t. III/1B 50 µg 3x/week 44 <0.05 I.t. III/1B 10 µg 3x/week 66 0.21

I.t. III/1B 2 µg 3x/week 96 0.88

*T/C% = (Treated mean endpoint tumor size)/(Control mean endpoint tumor size)x100%. $Student’s t-test.

From table 1 and 2 it can be appreciated that earlier treatment start of IL-21 generally improves the efficacy, particularly evident in B16 melanomas. Also, in the long-term tumor- growth inhibition study of RenCa tumors in Paper III (Figure 2), it was seen that discontinuation of IL-21 abruptly increased tumor-growth (data not shown). These findings

suggest that earlier treatment start with smaller tumor burdens and continued presence of IL- 21 benefit the anti-tumor effect. Unfortunately, this notion contrasts the general practice in modern cancer therapy, where new experimental treatments must initially show efficacy in pre-treated patients with end-stage disease in order to be continued, and this has also been the case for IL-21 (Davis et al., 2009). Undoubtedly, many experimental treatments would benefit from trials in patients with less advanced disease. Cytokines are no exception, since they function to boost weak inherent immune responses, which might become progressively more exhausted as disease progresses caused by the tumor-intrinsic immunosuppression as presented in figure 1 of this thesis.

In Paper II and III, subcutaneous and intratumoral administration of IL-21 significantly increased the density of tumor infiltrating CD8+ T cells, which were required for and correlated with the anti-tumor effect. Furthermore, the increased effect of intratumoral administration of IL-21 was associated with further increased density and activity of tumor infiltrating CD8+ T cells. These results suggest that the density of tumor infiltrating CD8+ T cells might represent a relevant biological marker of IL-21 efficacy that could prove clinically useful. Although the mechanism behind this increase in CD8+ T cell infiltration still needs to be clarified, these results also suggest that IL-21 is able to overcome a critical barrier for successful tumor immunotherapy, namely low immune cell infiltration and function as outlined in figure 1 of this thesis. Furthermore, in Paper III intratumoral IL-21 adminitration showed an increase in the

+ density of tumor infiltrating CD4 T cells but not Tregs, suggesting that IL-21 also is able to selectively increase certain T cells subsets in the tumor environment, without increasing suppressive T cells. Results from cancer patients have clearly shown what benefit increased CD8+ T cell infiltration

+ in tumors and particularly an increased CD8 T cell/Treg ratio could have, if these findings are translatable to humans (Clemente et al., 1996; Galon et al., 2006; Pages et al., 2005; Gao et al., 2007; Naito et al., 1998; Piersma et al., 2007; Sato et al., 2005; Sharma et al., 2007; Schumacher et al., 2001; Zhang et al., 2003). However, this remains an open question since the evaluation of tumor infiltrating lymphocytes in tumor biopsies remains to be included in IL- 21 clinical trials. But, other signs of relevant immune activation from Paper II and III seem to be translatable. In Paper III we show that the expression of granzyme B and IFNγ was increased in tumor infiltrating CD8+ T cells. Consistently, IL-21 administration in clinical trials increased granzyme B and IFNγ expression in CD8+ T cells from PBMCs (Davis et al., 2009; Frederiksen et al., 2008). In addition, IL-21 administration in clinical trials increased the fractions of CD62L+CD4+ and CD8+ T cells in PBMCs (Davis et al., 2009). This was thought to be associated with the frequent finding of lymphopenia in treated patients caused by increased

secondary lymphoid redistribution. In Paper III, intratumoral administration of IL-21 resulted in increased numbers of CD62L+CD4+ and CD8+ T cells in tumor-draining lymph nodes caused primarily by non-proliferative effects, indicating analogous activities in our model. Still, none of these clinical biomarkers were associated with anti-tumor responses showing the need for new predictive markers of IL-21 anti-tumor efficacy.

In Paper IV, the quest to unveil the role of endogenous IL-21 in cancer control surprisingly led to the conclusion that endogenous IL-21 is not required for tumor immunity, but rather restricted CD8+ T cell expansion and the control of immunogenic tumors. Considering the literature, these results were counterintuitive, since endogenous IL-21 mainly has been shown to promote proinflammatory conditions and autoimmunity as described in Paper I. However, as it was highlighted in Paper I, the actions of IL-21 are very context dependent and both the timing of IL-21 administration and the level of co-stimulation can greatly influence the outcome of IL-21 stimulated immune responses. This notion is exemplified by the effect IL-21 has on B cells; IL-21 stimulation of unstimulated B cells induces apoptosis, whereas the addition of B cell receptor and CD40 co-stimulation enables IL-21 to drive significant B cell differentiation (Ettinger et al., 2005). Very recently, it was found that endogenous IL-21 has a critical role in CD8+ T cell control of chronic viral infections (Elsaesser et al., 2009; Frohlich et al., 2009; Yi et al., 2009), but in one of these studies it was also shown that endogenous IL-21 restricted antigen-specific CD8+ T cell expansion during the acute phase of infections (Elsaesser et al., 2009). These results are further evidence of the complex role IL-21 has on immune responses and support a dichotomous role of IL-21 in CD8+ T cell immunity as suggested here in Paper IV. However, they also suggest that endogenous IL-21 could play a role in a more chronic CD8+ T cell-controlled tumor model, perhaps not thoroughly addressed in Paper IV. Still, the MCA sarcoma model used in Paper IV is a more chronic tumor model that apart from the generation of NK and NKT cell-controlled tumors (Crowe et al., 2002; Smyth et al., 2001) also has been shown to give rise to occult cancers contained by adaptive immunity (Koebel et al., 2007). Despite the unchanged incidence and tumor-growth of MCA sarcomas in IL-21-deficient mice, focus on more chronic tumor models controlled by CD8+ T cells in future studies could contain a role for IL-21.

The results in Paper IV suggest that neutralization of IL-21, which has been suggested as a potential treatment for several autoimmune diseases as reviewed in Paper I, does not overtly risk more frequent cancer development or compromised immunity toward cancers. Rather, based on the findings in this thesis, short-term and well-timed neutralization of host IL-21 followed by administration of IL-21 protein might even enhance tumor immunity.

Dichotomous features are not unique to IL-21. IL-2, which is approved as therapy for metastatic melanoma and advanced renal cell carcinoma, shows a very overt paradox in its actions. Originally, IL-2 was identified as a critical T cell growth factor in vitro and therefore assumed to be an important driver of T cell expansion in vivo, able to kick start tumor immune responses. However, this theory was questioned when the main phenotype found in IL-2- deficent mice was lymphoproliferation with lethal multi-organ autoimmunity. The discovery of

Treg cell biology has now clarified that IL-2 is indispensable for Treg cell development, but dispensable for immunity, and that Treg deficiency is responsible for the phenotype in IL-2- deficient mice (reviewed by Malek and Bayer, 2004). Although the in vivo biology of IL-2 is still incompletely understood, IL-2 has also shown important functions for CD8+ T cell immunity in vivo (Antony et al., 2006), which could explain the anti-tumor effects seen in patients. In a few selected patients, IL-2 shows remarkable clinical responses, while most do not seem to respond at all (Atkins et al., 1999; McDermott and Atkins, 2006; Klapper et al., 2008). The opposing actions of IL-2 combined with the incompletely understood immune status of patients prior to treatment are perhaps the reason for these inconsistencies. Arguably, IL-2 is beneficial for certain cancer patients, but as a cytokine for cancer immunotherapy it may be far from optimal and the exploration of alternatives is clearly warranted.

IL-2 has been known for more than 25 years, and despite extensive studies, its critical activity on Tregs was just realized in the last decade. IL-21 is only 9 years old and has so far shown pleiotropic actions mainly of proinflammatory nature. But, beside the findings in this thesis immunosuppressive activities of IL-21 are now beginning to emerge (Spolski et al., 2009), and it is likely that more are awaiting discovery. As presented in this thesis, endogenous IL-21- signaling limits CD8+ T cell expansion and tumor immunity whereas the administration of recombinant IL-21 protein enhances CD8+ T cell-mediated anti-tumor immune responses. Obviously, these paradoxical functions of IL-21 should be the focus of future studies, but acknowledging that IL-21 has opposing activities will also increase the need to carefully determine when and who to treat with IL-21. Clarification of the biology behind the immunosuppressive activities of IL-21 will be detrimental to its use with consideration for both the timing and duration of IL-21 administration. Because immunotherapy is relatively new in cancer therapy, traditional evaluations of predictive factors for treatment efficacy have mainly focused on non-immunological parameters (Motzer et al., 2004). However, greater focus on clarifying patient immune status prior to treatment with identification of immunological factors predictive of treatment efficacy might help to better select patients that would benefit from immunotherapy including IL-21. Recently, such efforts have been employed and shown to be relevant for the outcome of IL-2 therapy (Donskov and von der, 2006; Jensen et al., 2009).

Based on the findings in this thesis, characterizing the level and proportions of different tumor infiltrating lymphocytes might be interesting to explore for this purpose.

Collectively, the findings in this thesis outline the main challenge for the use of cytokines as therapies or as therapeutic targets – their pleiotropic and dichotomous actions. A more thorough understanding of the biology of cytokines is essential to yield therapeutic benefit of their administration or neutralization in human disease. This thesis has contributed with new insights into IL-21 anti-tumor biology and shown the role of IL-21 as a cytokine in host tumor immunity. Specifically, IL-21 showed significant anti-tumor effects in preclinical models with clinically relevant immunomodulatory effects, whereas host IL-21 was not required for tumor immunity but showed a suppressive role during CD8+ T cell-dependent tumor immunity. These results support the future development of IL-21 in oncology, but highlight the need for continued studies of IL-21 anti-tumor biology to fully understand its pleiotropic actions.

Chapter 5 – Conclusion

From part 1 in this thesis it can be concluded that: 1. IL-21 protein monotherapy therapy can inhibit established syngeneic tumor growth in two aggressive subcutaneous tumor models with different immunogenicity: B16 melanoma and RenCa renal cell carcinoma. 2. Subcutaneous administration of IL-21 has improved pharmacokinetics and at least as good or better efficacy compared to intraperitoneal administration, suggesting that subcutaneous administration of IL-21 could be applicable in future clinical trials. 3. CD8+ T cells are essential for the anti-tumor effect of IL-21 in the models used and IL-21 increases the density of tumor infiltrating CD8+ T cells, which correlates with tumor- growth inhibition, suggesting that the density of tumor infiltrating CD8+ T cells could be a relevant biological marker of IL-21 anti-tumor activity. 4. Compared to subcutaneous administration, intratumoral administration of IL-21 increases long-term tumor-growth inhibition, and increases the density, exocytotic activity, and granzyme B and IFNγ expression of tumor infiltrating CD8+ T cells. Intratumoral administration of IL-21 selectively increases the density of tumor infiltrating CD4+ T cells

and not Tregs, and increases the number of naïve T cells and proliferation of activated T cells in tumor draining lymph nodes. Together, these data suggest that local IL-21 administration benefits the tumor environment, can overcome tumor infiltrating CD8+ T cell disability and deserves to be clinically investigated when this administration route is accessible.

From part 2 in this thesis it can be concluded that: 1. Endogenous IL-21 is not required for tumor immunosurveillance, NK, NKT and CD8+ T cell-dependent primary tumor immunity, or for secondary CD8+ T cell memory responses. 2. Rather, endogenous IL-21 restricts CD8+ T cell expansion and immunity toward immunogenic tumors, indicating an unexpected immunosuppressive role of IL-21 in CD8+ T cell immunity.

Taken together, the results presented in this thesis support the clinical investigation of IL-21 protein therapy for the treatment of cancer, but reveal a dichotomous role for IL-21 in tumor immunity.

100

Chapter 6 – Future perspectives

The ultimate goal of any cancer therapy is to cure the disease. While the finding in this thesis that IL-21 significantly inhibited tumor growth and increased tumor infiltrating CD8+ T cells is encouraging, IL-21 was not curative in our models. In early clinical trials, IL-21 has shown moderate response rates, and although a couple of complete responses were achieved, IL-21 alone was not curative (Davis et al., 2007; Davis et al., 2009). A likely way forward for IL-21 is exploration of combination strategies. For this purpose, a better understanding of IL-21 biology and its dichotomous actions as highlighted in this thesis will be essential to ensure the basis for a rational selection of combination partners. Specific questions generated by this thesis include; what is the causative mechanism behind the IL-21-induced increase in tumor infiltrating CD8+ T cells? What circumstances determine whether IL-21 suppresses or promotes tumor immunity? Where and when is IL-21 produced and by what cells during tumor immune responses? Answers to these questions may help to clarify the best way to apply IL-21, expand the knowledge about what triggers IL-21 production physiologically and the context in which it is produced, and potentially identify new targets or combination options to pursue.

Results from the treatment of human immunodeficiency virus (HIV), another highly mutating human threat, clearly show the shortage of single targeted therapies and the success obtained by multi-combinations (Palella, Jr. et al., 1998). These findings indicate that to restrain a mutating enemy it is best to attack from several fronts. For this reason, it is very likely that the ultimate goal in cancer therapy will be achieved through multi-combination therapy.

An important feature of a drug to be used for combination therapy is that the drug is well tolerated and so far this has been the case in the early clinical trials of IL-21 (Davis et al., 2007; Davis et al., 2009). In cancer, most novel treatments are initially tested in end-stage and pre-treated patients, so it is likewise important that a new drug can potentially work in concert with conventional drugs. Here, we have recently shown in experimental models that IL-21 has additive effects in combination with IFNα and is feasible in combination with several chemotherapies (Eriksen et al., 2009; Skak et al., 2009 (Cytokine) in press doi:10.1016/j.cyto.2009.07.039). Obviously, the ability to provide additive or synergistic effects to other therapies is vital for a combination drug, and here IL-21 has shown additional anti-tumor effects together with several interesting combination partners as recently reviewed (Skak et al., 2008). Since IL-21 does not rely on a specific target for its actions, it represents an attractive partner to many targeted therapies. Experimental evidence of this has been shown using a triple immune-stimulating antibody cocktail together with IL-21, which resulted in cures of large

101

established tumors (Smyth et al., 2008). Such results show that clinical investigation of multi- combination therapy should be prioritized and that IL-21 is an attractive partner for this purpose.

In humans, the currently most successful immunotherapeutic protocol relies on a combination of non-myeloablative radio- or chemotherapy followed by adoptive transfer of ex vivo expanded antigen-specific tumor infiltrating CD8+ T cells, which has shown remarkable response rates and cures in selected patients with metastatic melanoma (Rosenberg et al., 2008). Traditionally, IL-2 has been used for the ex vivo expansion of tumor-reactive T cells, but interestingly, culture experiments with IL-21 resulted in less differentiated CD8+ T cells with a more robust in vivo expansion capacity and anti-tumor effect (Hinrichs et al., 2008). These results highlight another interesting aspect of IL-21 biology that deserves future attention.

The implications of IL-21 in autoimmune diseases as outlined in Paper I and its emerging immunosuppressive activities, further emphasize the multifaceted research needed to comprehend IL-21. Only the future will tell whether IL-21 ends up as a cancer therapy, will be part of a multi-combination strategy or perhaps be targeted for neutralization in autoimmune diseases. Continued research to fully understand the biology of this fascinating cytokine will be essential to make the most of it.

102

References

1. Antony,P.A., Paulos,C.M., Ahmadzadeh,M., Akpinarli,A., Palmer,D.C., Sato,N., Kaiser,A., Hinrichs,C.S., Klebanoff,C.A., Tagaya,Y., and Restifo,N.P. (2006). Interleukin-2-dependent mechanisms of tolerance and immunity in vivo. J. Immunol. 176, 5255-5266.

2. Atkins,M.B., Lotze,M.T., Dutcher,J.P., Fisher,R.I., Weiss,G., Margolin,K., Abrams,J., Sznol,M., Parkinson,D., Hawkins,M., Paradise,C., Kunkel,L., and Rosenberg,S.A. (1999). High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J. Clin. Oncol. 17, 2105-2116.

3. Blattman,J.N. and Greenberg,P.D. (2004). Cancer immunotherapy: a treatment for the masses. Science 305, 200-205.

4. Bubier,J.A., Sproule,T.J., Foreman,O., Spolski,R., Shaffer,D.J., Morse,H.C., III, Leonard,W.J., and Roopenian,D.C. (2009). A critical role for IL-21 receptor signaling in the pathogenesis of systemic lupus erythematosus in BXSB-Yaa mice. Proc. Natl. Acad. Sci. U. S. A 106, 1518-1523.

5. Burnet,F.M. (1970). The concept of immunological surveillance. Prog. Exp. Tumor Res. 13, 1-27.

6. Cabrera,T., Angustias,F.M., Sierra,A., Garrido,A., Herruzo,A., Escobedo,A., Fabra,A., and Garrido,F. (1996). High frequency of altered HLA class I phenotypes in invasive breast carcinomas. Hum. Immunol. 50, 127-134.

7. Clemente,C.G., Mihm,M.C., Jr., Bufalino,R., Zurrida,S., Collini,P., and Cascinelli,N. (1996). Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer 77, 1303-1310.

8. Coquet,J.M., Kyparissoudis,K., Pellicci,D.G., Besra,G., Berzins,S.P., Smyth,M.J., and Godfrey,D.I. (2007). IL- 21 Is Produced by NKT Cells and Modulates NKT Cell Activation and Cytokine Production. J. Immunol. 178, 2827-2834.

9. Crowe,N.Y., Smyth,M.J., and Godfrey,D.I. (2002). A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. J. Exp. Med. 196, 119-127.

10. Davis,I.D., Brady,B., Kefford,R.F., Millward,M., Cebon,J., Skrumsager,B.K., Mouritzen,U., Hansen,L.T., Skak,K., Lundsgaard,D., Frederiksen,K.S., Kristjansen,P.E., and McArthur,G. (2009). Clinical and biological efficacy of recombinant human interleukin-21 in patients with stage IV malignant melanoma without prior treatment: a phase IIa trial. Clin. Cancer Res. 15, 2123-2129.

11. Davis,I.D., Skrumsager,B.K., Cebon,J., Nicholaou,T., Barlow,J.W., Moller,N.P., Skak,K., Lundsgaard,D., Frederiksen,K.S., Thygesen,P., and McArthur,G.A. (2007). An open-label, two-arm, phase I trial of recombinant human interleukin-21 in patients with metastatic melanoma. Clin. Cancer Res. 13, 3630-3636.

12. DeMatos,P., Abdel-Wahab,Z., Vervaert,C., Hester,D., and Seigler,H. (1998). Pulsing of dendritic cells with cell lysates from either B16 melanoma or MCA-106 fibrosarcoma yields equally effective vaccines against B16 tumors in mice. J. Surg. Oncol. 68, 79-91.

13. Donskov,F. and von der,M.H. (2006). Impact of immune parameters on long-term survival in metastatic renal cell carcinoma. J. Clin. Oncol. 24, 1997-2005.

14. Dougan,M. and Dranoff,G. (2009). Immune therapy for cancer. Annu. Rev. Immunol. 27, 83-117.

15. Dunn,G.P., Bruce,A.T., Ikeda,H., Old,L.J., and Schreiber,R.D. (2002). Cancer immunoediting: from immunosurveillance to tumor escape. Nat. Immunol. 3, 991-998.

16. Dunn,G.P., Old,L.J., and Schreiber,R.D. (2004). The immunobiology of cancer immunosurveillance and immunoediting. Immunity. 21, 137-148.

17. Elsaesser,H., Sauer,K., and Brooks,D.G. (2009). IL-21 is required to control chronic viral infection. Science 324, 1569-1572.

103

18. Eriksen,K.W., Sondergaard,H., Woetmann,A., Krejsgaard,T., Skak,K., Geisler,C., Wasik,M.A., and Odum,N. (2009). The combination of IL-21 and IFN-alpha boosts STAT3 activation, cytotoxicity and experimental tumor therapy. Mol. Immunol. 46, 812-820.

19. Ettinger,R., Kuchen,S., and Lipsky,P.E. (2008). Interleukin 21 as a target of intervention in autoimmune disease. Ann. Rheum. Dis. 67 Suppl 3, iii83-iii86.

20. Ettinger,R., Sims,G.P., Fairhurst,A.M., Robbins,R., da Silva,Y.S., Spolski,R., Leonard,W.J., and Lipsky,P.E. (2005). IL-21 induces differentiation of human naive and memory B cells into antibody-secreting plasma cells. J. Immunol. 175, 7867-7879.

21. Fina,D., Sarra,M., Fantini,M.C., Rizzo,A., Caruso,R., Caprioli,F., Stolfi,C., Cardolini,I., Dottori,M., Boirivant,M., Pallone,F., MacDonald,T.T., and Monteleone,G. (2008). Regulation of gut inflammation and th17 cell response by interleukin-21. Gastroenterology 134, 1038-1048.

22. Frederiksen,K.S., Lundsgaard,D., Freeman,J.A., Hughes,S.D., Holm,T.L., Skrumsager,B.K., Petri,A., Hansen,L.T., McArthur,G.A., Davis,I.D., and Skak,K. (2008). IL-21 induces in vivo immune activation of NK cells and CD8(+) T cells in patients with metastatic melanoma and renal cell carcinoma. Cancer Immunol. Immunother. 57, 1439-1449.

23. Frese,K.K. and Tuveson,D.A. (2007). Maximizing mouse cancer models. Nat. Rev. Cancer 7, 645-658.

24. Frohlich,A., Kisielow,J., Schmitz,I., Freigang,S., Shamshiev,A.T., Weber,J., Marsland,B.J., Oxenius,A., and Kopf,M. (2009). IL-21R on T cells is critical for sustained functionality and control of chronic viral infection. Science 324, 1576-1580.

25. Gajewski,T.F., Meng,Y., Blank,C., Brown,I., Kacha,A., Kline,J., and Harlin,H. (2006). Immune resistance orchestrated by the tumor microenvironment. Immunol. Rev. 213, 131-145.

26. Galon,J., Costes,A., Sanchez-Cabo,F., Kirilovsky,A., Mlecnik,B., Lagorce-Pages,C., Tosolini,M., Camus,M., Berger,A., Wind,P., Zinzindohoue,F., Bruneval,P., Cugnenc,P.H., Trajanoski,Z., Fridman,W.H., and Pages,F. (2006). Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313, 1960-1964.

27. Gao,Q., Qiu,S.J., Fan,J., Zhou,J., Wang,X.Y., Xiao,Y.S., Xu,Y., Li,Y.W., and Tang,Z.Y. (2007). Intratumoral balance of regulatory and cytotoxic T cells is associated with prognosis of hepatocellular carcinoma after resection. J. Clin. Oncol. 25, 2586-2593.

28. Garcia-Lora,A., Algarra,I., and Garrido,F. (2003). MHC class I antigens, immune surveillance, and tumor immune escape. J. Cell Physiol 195, 346-355.

29. Geertsen,P.F., Gore,M.E., Negrier,S., Tourani,J.M., and von der,M.H. (2004). Safety and efficacy of subcutaneous and continuous intravenous infusion rIL-2 in patients with metastatic renal cell carcinoma. Br. J. Cancer 90, 1156-1162.

30. Girardi,M. (2007). Cutaneous perspectives on adaptive immunity. Clin. Rev. Allergy Immunol. 33, 4-14.

31. Hanahan,D. and Weinberg,R.A. (2000). The hallmarks of cancer. Cell 100, 57-70.

32. Hicklin,D.J., Wang,Z., Arienti,F., Rivoltini,L., Parmiani,G., and Ferrone,S. (1998). beta2-Microglobulin mutations, HLA class I antigen loss, and tumor progression in melanoma. J. Clin. Invest 101, 2720-2729.

33. Hinrichs,C.S., Spolski,R., Paulos,C.M., Gattinoni,L., Kerstann,K.W., Palmer,D.C., Klebanoff,C.A., Rosenberg,S.A., Leonard,W.J., and Restifo,N.P. (2008). IL-2 and IL-21 confer opposing differentiation programs to CD8+ T cells for adoptive immunotherapy. Blood 111, 5326-5333.

34. Jensen,H.K., Donskov,F., Nordsmark,M., Marcussen,N., and von der,M.H. (2009). Increased intratumoral FOXP3-positive regulatory immune cells during interleukin-2 treatment in metastatic renal cell carcinoma. Clin. Cancer Res. 15, 1052-1058.

35. Khong,H.T. and Restifo,N.P. (2002). Natural selection of tumor variants in the generation of "tumor escape" phenotypes. Nat. Immunol. 3, 999-1005.

104

36. Klapper,J.A., Downey,S.G., Smith,F.O., Yang,J.C., Hughes,M.S., Kammula,U.S., Sherry,R.M., Royal,R.E., Steinberg,S.M., and Rosenberg,S. (2008). High-dose interleukin-2 for the treatment of metastatic renal cell carcinoma : a retrospective analysis of response and survival in patients treated in the surgery branch at the National Cancer Institute between 1986 and 2006. Cancer 113, 293-301.

37. Koebel,C.M., Vermi,W., Swann,J.B., Zerafa,N., Rodig,S.J., Old,L.J., Smyth,M.J., and Schreiber,R.D. (2007). Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450, 903-907.

38. Krup,O.C., Kroll,I., Bose,G., and Falkenberg,F.W. (1999). Cytokine depot formulations as adjuvants for tumor vaccines. I. Liposome-encapsulated IL-2 as a depot formulation. J. Immunother. 22, 525-538.

39. Leonard,W.J. and Spolski,R. (2005). Interleukin-21: a modulator of lymphoid proliferation, apoptosis and differentiation. Nat. Rev. Immunol. 5, 688-698.

40. Malek,T.R. and Bayer,A.L. (2004). Tolerance, not immunity, crucially depends on IL-2. Nat. Rev. Immunol. 4, 665-674.

41. McDermott,D.F. and Atkins,M.B. (2006). Interleukin-2 therapy of metastatic renal cell carcinoma--predictors of response. Semin. Oncol. 33, 583-587.

42. Mellor,H.R. and Callaghan,R. (2008). Resistance to chemotherapy in cancer: a complex and integrated cellular response. Pharmacology 81, 275-300.

43. Mestas,J. and Hughes,C.C. (2004). Of mice and not men: differences between mouse and human immunology. J. Immunol. 172, 2731-2738.

44. Motzer,R.J., Bacik,J., Schwartz,L.H., Reuter,V., Russo,P., Marion,S., and Mazumdar,M. (2004). Prognostic factors for survival in previously treated patients with metastatic renal cell carcinoma. J. Clin. Oncol. 22, 454- 463.

45. Naito,Y., Saito,K., Shiiba,K., Ohuchi,A., Saigenji,K., Nagura,H., and Ohtani,H. (1998). CD8+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res. 58, 3491-3494.

46. Negrier,S., Perol,D., Ravaud,A., Bay,J.O., Oudard,S., Chabaud,S., Fargeot,P., Delva,R., Deplanque,G., Gravis,G., and Escudier,B. (2008). Randomized study of intravenous versus subcutaneous interleukin-2, and IFNalpha in patients with good prognosis metastatic renal cancer. Clin. Cancer Res. 14, 5907-5912.

47. Nurieva,R., Yang,X.O., Martinez,G., Zhang,Y., Panopoulos,A.D., Ma,L., Schluns,K., Tian,Q., Watowich,S.S., Jetten,A.M., and Dong,C. (2007). Essential autocrine regulation by IL-21 in the generation of inflammatory T cells. Nature 448, 480-483.

48. Ostrand-Rosenberg,S. (2004). Animal models of tumor immunity, immunotherapy and cancer vaccines. Curr. Opin. Immunol. 16, 143-150.

49. Pages,F., Berger,A., Camus,M., Sanchez-Cabo,F., Costes,A., Molidor,R., Mlecnik,B., Kirilovsky,A., Nilsson,M., Damotte,D., Meatchi,T., Bruneval,P., Cugnenc,P.H., Trajanoski,Z., Fridman,W.H., and Galon,J. (2005). Effector memory T cells, early metastasis, and survival in colorectal cancer. N. Engl. J. Med. 353, 2654-2666.

50. Palella,F.J., Jr., Delaney,K.M., Moorman,A.C., Loveless,M.O., Fuhrer,J., Satten,G.A., Aschman,D.J., and Holmberg,S.D. (1998). Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N. Engl. J. Med. 338, 853-860.

51. Parrish-Novak,J., Dillon,S.R., Nelson,A., Hammond,A., Sprecher,C., Gross,J.A., Johnston,J., Madden,K., Xu,W., West,J., Schrader,S., Burkhead,S., Heipel,M., Brandt,C., Kuijper,J.L., Kramer,J., Conklin,D., Presnell,S.R., Berry,J., Shiota,F., Bort,S., Hambly,K., Mudri,S., Clegg,C., Moore,M., Grant,F.J., Lofton-Day,C., Gilbert,T., Rayond,F., Ching,A., Yao,L., Smith,D., Webster,P., Whitmore,T., Maurer,M., Kaushansky,K., Holly,R.D., and Foster,D. (2000). Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature 408, 57-63.

52. Piersma,S.J., Jordanova,E.S., van Poelgeest,M.I., Kwappenberg,K.M., van der Hulst,J.M., Drijfhout,J.W., Melief,C.J., Kenter,G.G., Fleuren,G.J., Offringa,R., and van der Burg,S.H. (2007). High number of intraepithelial CD8+ tumor-infiltrating lymphocytes is associated with the absence of lymph node metastases in patients with large early-stage cervical cancer. Cancer Res. 67, 354-361.

105

53. Rabinovich,G.A., Gabrilovich,D., and Sotomayor,E.M. (2007). Immunosuppressive strategies that are mediated by tumor cells. Annu. Rev. Immunol. 25, 267-296.

54. Radny,P., Caroli,U.M., Bauer,J., Paul,T., Schlegel,C., Eigentler,T.K., Weide,B., Schwarz,M., and Garbe,C. (2003). Phase II trial of intralesional therapy with interleukin-2 in soft-tissue melanoma metastases. Br. J. Cancer 89, 1620-1626.

55. Rosenberg,S.A. (2001). Progress in human tumour immunology and immunotherapy. Nature 411, 380-384.

56. Rosenberg,S.A., Restifo,N.P., Yang,J.C., Morgan,R.A., and Dudley,M.E. (2008). Adoptive cell transfer: a clinical path to effective cancer immunotherapy. Nat. Rev. Cancer 8, 299-308.

57. Sato,E., Olson,S.H., Ahn,J., Bundy,B., Nishikawa,H., Qian,F., Jungbluth,A.A., Frosina,D., Gnjatic,S., Ambrosone,C., Kepner,J., Odunsi,T., Ritter,G., Lele,S., Chen,Y.T., Ohtani,H., Old,L.J., and Odunsi,K. (2005). Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc. Natl. Acad. Sci. U. S. A 102, 18538-18543.

58. Savage,P., Stebbing,J., Bower,M., and Crook,T. (2009). Why does cytotoxic chemotherapy cure only some cancers? Nat. Clin. Pract. Oncol. 6, 43-52.

59. Sayers,T.J., Brooks,A.D., Lee,J.K., Fenton,R.G., Komschlies,K.L., Wigginton,J.M., Winkler-Pickett,R., and Wiltrout,R.H. (1998). Molecular mechanisms of immune-mediated lysis of murine renal cancer: differential contributions of perforin-dependent versus Fas-mediated pathways in lysis by NK and T cells. J. Immunol. 161, 3957-3965.

60. Scheffer,S.R., Nave,H., Korangy,F., Schlote,K., Pabst,R., Jaffee,E.M., Manns,M.P., and Greten,T.F. (2003). Apoptotic, but not necrotic, tumor cell vaccines induce a potent immune response in vivo. Int. J. Cancer 103, 205-211.

61. Schumacher,K., Haensch,W., Roefzaad,C., and Schlag,P.M. (2001). Prognostic significance of activated CD8(+) T cell infiltrations within esophageal carcinomas. Cancer Res. 61, 3932-3936.

62. Shankaran,V., Ikeda,H., Bruce,A.T., White,J.M., Swanson,P.E., Old,L.J., and Schreiber,R.D. (2001). IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature 410, 1107-1111.

63. Sharma,P., Shen,Y., Wen,S., Yamada,S., Jungbluth,A.A., Gnjatic,S., Bajorin,D.F., Reuter,V.E., Herr,H., Old,L.J., and Sato,E. (2007). CD8 tumor-infiltrating lymphocytes are predictive of survival in muscle-invasive urothelial carcinoma. Proc. Natl. Acad. Sci. U. S. A 104, 3967-3972.

64. Skak,K., Kragh,M., Hausman,D., Smyth,M.J., and Sivakumar,P.V. (2008). Interleukin 21: combination strategies for cancer therapy. Nat. Rev. Drug Discov. 7, 231-240.

65. Smyth,M.J., Cretney,E., Kershaw,M.H., and Hayakawa,Y. (2004). Cytokines in cancer immunity and immunotherapy. Immunol. Rev. 202, 275-293.

66. Smyth,M.J., Crowe,N.Y., and Godfrey,D.I. (2001). NK cells and NKT cells collaborate in host protection from methylcholanthrene-induced fibrosarcoma. Int. Immunol. 13, 459-463.

67. Smyth,M.J., Teng,M.W., Sharkey,J., Westwood,J.A., Haynes,N.M., Yagita,H., Takeda,K., Sivakumar,P.V., and Kershaw,M.H. (2008). Interleukin 21 enhances antibody-mediated tumor rejection. Cancer Res. 68, 3019- 3025.

68. Spolski,R., Kashyap,M., Robinson,C., Yu,Z., and Leonard,W.J. (2008). IL-21 signaling is critical for the development of type I diabetes in the NOD mouse. Proc. Natl. Acad. Sci. U. S. A 105, 14028-14033.

69. Spolski,R., Kim,H.P., Zhu,W., Levy,D.E., and Leonard,W.J. (2009). IL-21 mediates suppressive effects via its induction of IL-10. J. Immunol. 182, 2859-2867.

70. Spolski,R. and Leonard,W.J. (2008). Interleukin-21: basic biology and implications for cancer and autoimmunity. Annu. Rev. Immunol. 26, 57-79.

71. Sutherland,A.P., Van,B.T., Wurster,A.L., Suto,A., Michaud,M., Zhang,D., Grusby,M.J., and von,H.M. (2009). Interleukin-21 is required for the development of type 1 diabetes in NOD mice. Diabetes 58, 1144-1155.

106

72. van der,B.P., Traversari,C., Chomez,P., Lurquin,C., De Plaen,E., Van den,E.B., Knuth,A., and Boon,T. (1991). A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 254, 1643-1647.

73. van,E.A., Hurwitz,A.A., and Allison,J.P. (1999). Combination immunotherapy of B16 melanoma using anti- cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and granulocyte/macrophage colony-stimulating factor (GM-CSF)-producing vaccines induces rejection of subcutaneous and metastatic tumors accompanied by autoimmune depigmentation. J. Exp. Med. 190, 355-366.

74. Voskoglou-Nomikos,T., Pater,J.L., and Seymour,L. (2003). Clinical predictive value of the in vitro cell line, human xenograft, and mouse allograft preclinical cancer models. Clin. Cancer Res. 9, 4227-4239.

75. Wigginton,J.M., Komschlies,K.L., Back,T.C., Franco,J.L., Brunda,M.J., and Wiltrout,R.H. (1996). Administration of interleukin 12 with pulse interleukin 2 and the rapid and complete eradication of murine renal carcinoma. J. Natl. Cancer Inst. 88, 38-43.

76. Yee,C., Thompson,J.A., Byrd,D., Riddell,S.R., Roche,P., Celis,E., and Greenberg,P.D. (2002). Adoptive T cell therapy using antigen-specific CD8+ T cell clones for the treatment of patients with metastatic melanoma: in vivo persistence, migration, and antitumor effect of transferred T cells. Proc. Natl. Acad. Sci. U. S. A 99, 16168-16173.

77. Yi,J.S., Du,M., and Zajac,A.J. (2009). A vital role for interleukin-21 in the control of a chronic viral infection. Science 324, 1572-1576.

78. Young,D.A., Hegen,M., Ma,H.L., Whitters,M.J., Albert,L.M., Lowe,L., Senices,M., Wu,P.W., Sibley,B., Leathurby,Y., Brown,T.P., Nickerson-Nutter,C., Keith,J.C., Jr., and Collins,M. (2007). Blockade of the interleukin-21/interleukin-21 receptor pathway ameliorates disease in animal models of rheumatoid arthritis. Arthritis Rheum. 56, 1152-1163.

79. Zhang,L., Conejo-Garcia,J.R., Katsaros,D., Gimotty,P.A., Massobrio,M., Regnani,G., Makrigiannakis,A., Gray,H., Schlienger,K., Liebman,M.N., Rubin,S.C., and Coukos,G. (2003). Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N. Engl. J. Med. 348, 203-213.

107