Improving cytokine immunotherapy
Our research focuses on the study of cytokine’s function within the immune system in order to better understand their biology and improve cytokine-mediated immunotherapy. A well-studied example is interleukin-2 (IL-2). Due to its ability to stimulate anti-tumor immune cells, high-dose IL-2 treatment was the first approved immunotherapy used against metastatic cancer in humans. However, the high doses of IL-2 necessary to achieve clinical response lead to severe adverse events and the stimulation of immunosuppressive cells. We were able to address these shortcomings associated with IL-2 immunotherapy with the generation and study of anti-human IL-2 monoclonal antibodies (mAbs). One of these mAbs termed NARA1 directs IL-2 specifically to anti-tumor immune cells; improving treatment efficacy and reducing unwanted adverse effects (see also “Cancer immunotherapy”). Another type of mAb called UFKA1 directs IL-2 to so-called regulatory T cells. Selective stimulation of these regulatory T cells reduces disease activity in different models of autoimmune diseases (see also “Autoimmune disease and graft rejection”).
The explanation for the above-mentioned effects and the, somewhat, contradictory role of IL-2 relies in the composition and biology of the IL-2 receptor (IL-2R). An intermediate-affinity dimeric IL-2R, consisting of IL-2Rβ (CD122) and the common gamma chain (γc) is mainly expressed on antigen-experienced (memory) CD8+ T cells, carrying background CD25 but high CD122 levels, and on NK cells: cell types known by their capacity to mediate anti-tumor responses. On the other hand, the preferential expression of high-affinity trimeric IL-2R, consisting of IL-2Rβ, γc and additionally IL-2Rα (CD25) on regulatory T cells, can potently suppress overt immune stimulation.
The crystal structure of the human IL-2/NARA1 (Fab) complex (Figure 1A) revealed that NARA1 covers a large and essential domain of the IL-2 interaction site with the IL-2Rα. It efficiently blocks the association of IL-2 with the α subunit while still allowing contact with the β and γ subunits. This leads to the selective stimulation of cells expressing high levels of CD122 as shown in figure 1B.
Figure 1. IL-2/NARA1 complex shows strong analogy of NARA1 to CD25 leading to efficient expansion of CD122high cells (A). Overlay of NARA1 on hIL-2/hIL-2R quaternary complex (IL-2, grey; CD25, red; CD122, purple; yc, gold) with NARA1 heavy and light chains in blue and cyan, respectively. (B) Mice received one, two, three or four injections of either PBS, hIL-2 (1.5 μg) or hIL-2/NARA1 complex (1.5 μg/15 μg). On the first day of injection, animals also received bromodeoxyuridine (BrdU; 0.8 mg/ml) in their drinking water. Mice were euthanized and analyzed for proliferating (BrdU+) T cell subsets, including CD4+ CD25high and CD8+ CD122high CD44high (memory) T cells. Percentages of proliferative cells (% BrdU+) are shown.
The term ‘cancer immunotherapy’ appeared for the first time in 1891, when William Coley began to study intratumoral injections of live inactivated Streptococcus pyogenes and Serratia marcescens, later called Coley’s toxins, and observed complete tumor remission of sarcomas in patients. Since then, a large number of research studies contributed to the understanding of the necessary key steps for inducing an effective anti-tumor immune response. This includes the stimulation of innate immune recognition in order to promote antigen-presentation and thus generating protective T cell responses, which have the capacity to overcome immunosuppression at the tumor site. Additionally, antibody-, natural killer cell- (NK) and natural killer T (NKT) cell-dependent immune responses can also be triggered. Cancer immunosurveillance occurs during the elimination phase of cancer cells where innate and adaptive immunity coordinate to scan for the presence of abnormal cells and destroy early tumors, before they become clinically apparent. In some cases, cancer cells escape immune control and survive. This can lead to the equilibrium phase where the surviving transformed cells are maintained in a state of immune mediated dormancy. Unfortunately, cancer cells can evade immune recognition by creating an immunosuppressive environment.
Understanding the mechanisms of tumor immune evasion and immune suppression have suggested various pathways for therapeutic intervention, in order to, either stimulate immune effector mechanisms or to counteract immunosuppressive mechanisms. The strategies include the administration of recombinant cytokines, the development of cancer vaccines and antibodies, and adoptive T cell therapy. We recently generated a specific monoclonal antibody (mAb) against interleukin-2 (IL-2) termed NARA1, increasing the activity and specificity of the cytokine. By binding IL-2 to NARA1 we managed to address the shortcomings associated with IL-2 immunotherapy such as the development of adverse effects and the stimulation of immunosuppressive cells. This led to potent tumor control in various melanoma models correlating with a favorable ratio of CD8+ T cells over CD4+ T regulatory cells (Figure 1). We also focus on the study of strategical combinations in order to prolong the immune mediated control. One example is the combination of IL-2/NARA1 complexes and Ezh2 inhibition, which can reverse the mechanism of adaptive resistance observed in melanoma (Figure 2).
Figure 1. Use of the IL-2/NARA1 complex mediates a efficient anti-tumor immune response in different melanoma models. (A) Animals were injected intradermally with B16-F10 melanoma cells, followed by 4 injections of PBS , hIL-2, or hIL-2/NARA1 complex. Shown is tumor volume, with arrow indicating start of treatment. (B) Ratio of CD8+ T cells to CD4+ CD25high FoxP3+ T regulatory cells in tumor-infiltrating lymphocytes (TILs) from mice treated as in A. (C) Skin melanoma-free survival and (D) lung metastasis counts of Tyr::N-RasQ61K Ink4a–/– mice, developing spontaneous melanoma, treated with 10 weekly courses of 4 injections of PBS, hIL-2 or hIL-2/NARA1 complex. Shown are representative pictures of lung metastases.
Figure 2. Ezh2 inactivation synergizes with IL-2/NARA1 (IL-2cx) immunotherapy. Tumor growth kinetics in mice harboring B16-F10 melanoma cells receiving 4 injections of PBS, IL-2cx or daily injections of GSK503 with or without IL-2cx. Black arrows mark time-point of treatment start.
Autoimmune disease and graft rejection.
Regulatory T cells (Tregs) have unique capability to suppressive immune responses. They are characterized by the expression of the transcription factor Forkhead-Box-Protein P3 (Foxp3) and the high expression of IL-2 receptor alpha chain (IL2RA, CD25). The importance of Tregs in keeping immune response in balance is demonstrated in patients having mutations in the foxp3 gene. By combining IL-2 with JES6-1 antibody forming IL-2/JES6-1 complexes, one can preferentially stimulate Tregs. Indeed, by injecting IL-2/JES6-1 complexes on three consecutive days, an over 10-foldexpansion of Tregs can be achieved. In our laboratory, we are investigating how the in vivo induction of Tregs can be applied in different clinical situations. With the first step, we were able to demonstrate the positive effect of this approach in a murine model of Multiple Sclerosis (Experimental Autoimmune Encephalomyelitis, EAE). The same treatment strategy was also successfully used to achieve long-term graft acceptance in the model of pancreatic islets transplantation in mice. Our current works focus on improving this treatment strategy and investigation the mechanism of graft protection provided by regulatory T cells.
Figure 1. (A) In vivo expansion of T reg cells: in the spleen, mesenteric LN (MLN), liver and bone marrow after injection of IL-2/JES6-1. (B) Impaired induction of EAE in mice treated with IL-2/JES6-1. EAE was induced on day 0 by immunization of MOG35-55 in CFA. Mice were treated with combinations of 1 μg IL-2, 5 μg JES6-1. Disease score of mice treated on days 3, 2, and 1 with PBS, IL-2, or IL-2/JES6-1. (C) Long-term graft survival in pancreatic islets transplantation in mice: Streptozotocin-induced diabetic C57BL/6 (H2b) mice were treated with PBS, 1 μg IL-2, or 1 μg/5 μg IL-2/JES6-1 on three consecutive days (days 3, 2, and 1). On day 0, mice were transplanted with BALB/c (H2d) islets, and BGL s blood glucose were monitored as a measure of graft function and survival. Grafts were considered rejected after two consecutive BGLs >16 mmol/liter after a period of normoglycemia.
Mechanism of action
As for the mechanism of action, we have recently shown that IL-2/mAbCD25 complexes function primarily by extending the half-life of IL-2. Conversely, IL-2/mAbCD122 complexes exert their in vivo activity by preventing the interaction of IL-2 with CD25, the non-signaling a subunit of the IL-2 receptor, in addition to a slight increase in IL-2 half-life (Figures 4 and 5).
Figure 1. Prolonging IL-2 half-life is not sufficient to mimic IL-2/mAbCD122 complexes. IL-2/mAbCD122 complexes induce increased proliferation of CD8+ T cells in comparison to IL-2-fusion protein (IL-2-FP) in vivo (A), despite similar half-lives (B). Shown is in (A) proliferation of CFSE-labeled CD8+ T cells in mice treated with PBS, IL-2, IL-2/mAbCD122, or IL-2-FP. In (B), serum samples of mice injected with IL-2 vs. IL-2/mAbCD122 (upper) or IL-2-FP (lower) were collected at the indicated time-points after injection and assayed on IL-2-sensitive CTLL cells.
Figure 2. PIL-2 requires prolonged half-life and CD25 blockade to mimic the activity of IL-2/mAbCD122 complexes. Prolonged IL-2 half-life of IL-2-FP in addition to blocking of IL-2 receptor α (CD25) molecules by an anti-CD25 antibody (αCD25) exerts a comparable effect on the proliferation of CD8+ T cells as IL-2/mAbCD122 complexes in vivo.
Psoriasis, a T cell-mediated disease
Psoriasis is a common, chronic, immune-mediated disease characterized by scaling erythematous skin plaques and is frequently associated with arthritis. Pathogenesis involves both the innate and adaptive immune system, involving aberrant activation of cytotoxic CD8+ T cells, CD4+ and gd T cells and induction of proinflammatory cytokines, such as interferon-gamma (IFN-γ), IL-17 IL-22, IL-23, and tumor necrosis factor-alpha(TNF-α). Exaggerated responses of effector T cells are usually held in check by the action of regulatory T cells (Treg), which are less efficient in psoriasis. In a recent publication, our group revealed a novel mechanism of immune modulation in psoriasis, through IL-15 receptor alpha (Rα) release from epithelial stroma. Soluble IL-15Rα reduces the availability of IL-15, dampening the inflammatory response. In line with this observation, we have shown that IL-15 levels in serum of patients correlate with disease severity, whereas the soluble receptor (sIL-15Rα) has an opposite trend. Such correlations can potentially serve as markers for disease severity and treatment response.
In order to better understand the complexity of this disorder and identify the mechanisms involved in maintenance and development of the disease we established a human-skin-graft-onto-AGR-mouse psoriasis model. This allows us to identify the kinetics of immune system infiltration into the skin and the mechanisms involved in lesion formation with particular interest for the tissue-specific mechanisms in regulating the cytokine production.
Macroscopic and microscopic aspect of symptomless uninvolved psoriasis skin before and 8 weeks after grafting onto AGR mice. Based on macroscopic and microscopic assessment before or on the day of grafting, symptomless uninvolved skin from a psoriasis patient (termed PN skin) was comparable to normal human skin and contained only a few CD3+ T cells (both CD4+ and CD8+ cells) and CD1d+ antigen-presenting cells (APC), all confined to the dermis, while pro-inflammatory cytokines, including tumor necrosis factor- (TNF) a, interleukin- (IL) 12, and interferon- (IFN) g, were hardly present (upper rows in both panels). Conversely, at 8 weeks after transplantation onto AGR mice, the PN skin graft became reddish, thickened and scaly, and showed the presence of numerous CD4+