Immune escape as a fundamental trait of cancer: focus on IDO

Abstract

Immune escape is a critical gateway to malignancy. The emergence of this fundamental trait of cancer represents the defeat of immune surveillance, a potent, multi-armed and essential mode of cancer suppression that may influence the ultimate clinical impact of an early stage tumor. Indeed, immune escape may be a central modifier of clinical outcomes, by affecting tumor dormancy versus progression, licensing invasion and metastasis and impacting therapeutic response. Although relatively little studied until recently, immune suppression and escape in tumors are now hot areas with clinical translation of several new therapeutic agents already under way. The interconnections between signaling pathways that control immune escape and those that control proliferation, senescence, apoptosis, metabolic alterations, angiogenesis, invasion and metastasis remain virtually unexplored, offering rich new areas for investigation. Here, an overview of this area is provided with a focus on the tryptophan catabolic enzyme indoleamine 2,3-dioxygenase (IDO) and its recently discovered relative IDO2 that are implicated in suppressing T-cell immunity in normal and pathological settings including cancer. Emerging evidence suggests that during cancer progression activation of the IDO pathway might act as a preferred nodal modifier pathway for immune escape, for example analogous to the PI3K pathway for survival or the VEGF pathway for angiogenesis. Small molecule inhibitors of IDO and IDO2 heighten chemotherapeutic efficacy in mouse models of cancer in a nontoxic fashion and an initial lead compound entered phase I clinical trials in late 2007. New modalities in this area offer promising ways to broaden the combinatorial attack on advanced cancers, where immune escape mechanisms likely provide pivotal support.

Immune escape is the terminal stage of immunoediting, a process that coevolves with oncogenesis

The tumor microenvironment has a vital role in driving malignant development. Immune cells in the microenvironment are particularly important, resulting in both positive and negative consequences for tumor growth. Tumor cells themselves express many aberrant antigens that can promote eradication through immune surveillance. However, this mode of tumor suppression has negative repercussions in the long term, as it applies a selection for cells that evolve mechanisms of immune evasion. This process, termed immunoediting (Dunn et al., 2004), is one of the leading concepts to emerge in the field of cancer research this century, warranting the attention of all investigators whether or not they are immunologists. In essence, immunoediting is a conceptual veneer atop oncogenesis. Neoantigens presented by transformed cells that have accumulated mutations and epigenetic alterations in oncogenes, tumor suppressor genes and DNA repair genes attract the attention of the adaptive immune system, but by doing so they also generate a powerful selection for cells that can evade immune destruction, which emerge as a result of the genomic instability engendered by these mutations (Figure 1). As a consequence, tumor cells that gain a growth or survival edge in a hostile immune microenvironment also gain the opportunity to evolve ways to ‘sculpt’ the immune system to retain supportive elements and escape hostile elements. As suggested elsewhere (Prendergast and Jaffee, 2007), genes driving immune escape may be synonymous with genes defining the peculiar nature of the ‘smouldering’ inflammatory state that supports cancer progression (Balkwill et al., 2005). In any case, the immune cells in a tumor microenvironment are important, given striking evidence that the status of these cells may provide better prognostic information than traditional histological cues (Galon et al., 2006).

Figure 1
figure1

Integration of immunoediting and oncogenesis during cancer progression. Oncogenesis initiates the formation of transformed cells that are attacked by immune cells as the result of neoantigen presentation. This process of immune surveillance imposes a selection for transformed cells that acquire tactics to escape control. Genetic instability driven by oncogenesis facilitates evolution of strategies for immune evasion or suppression, the latter of which may help tilt the tumor microenvironment from hostile to supportive for tumor cells. As the battle evolves between tumor cells and the immune system, representing the process of immunoediting, a state of immune equilibrium may be achieved that corresponds to a dynamic but dormant disease that is clinically occult. Further iteration of evasion mechanisms to defeat different elements of the immune system may ultimately drive immune suppression beyond the local microenvironment, achieving immune escape and thereby licensing invasive and metastatic behaviors that define progressive disease. Invasion and metastasis may be derivative of immune escape since the former are presumably impotent while immune surveillance prevails beyond the local microenviroment. While mutations may initiate cancer, modifiers and microenvironmental factors including those sculpted by the process of immunoediting may dictate the course of disease.

An integrated model of oncogenesis and immunoediting offers other potential explanative powers as well. First, it may provide a deeper understanding of why acquiring the cell-intrinsic traits of cancer—immortalization, tumor suppressor inactivation, growth deregulation, metabolic alterations and resistance to cell death—may yield only unstable, dormant or benign tumors if the immune-shaped cell-extrinsic traits—angiogenesis, invasion, metastasis and immune escape—do not fully evolve (Figure 2). The generation of such occult or benign tumors represent the achievement of a state of immune equilibrium by the tumor in which it can stave off but not defeat immune surveillance (Koebel et al., 2007). This ‘fight to a draw’ between the tumor and the immune system offers an explanation for the clinical phenomenon of tumor dormancy in which a tumor can persist in a dynamic and viable yet occult pattern for many years (Curtis et al., 1997). For example, most smokers by the age of 50 may have small dormant lung tumors, only a minority of which will ever develop into clinically overt disease (Swensen et al., 2005). In such a state, further iteration of mechanisms to strengthen localized and peripheral immunosuppression could be crucial, since the achievement of immune escape may license invasion and metastasis. Put another way, invasion and metastasis may be impossible while immune surveillance prevails outside the immediate microenvironment of the localized primary tumor. Given the documentation of germ line modifier effects for metastasis (Hunter and Crawford, 2006), it is conceivable that the genetic background of the host in terms of its immune responsiveness may have a dominant impact on the clinical course of disease (for example, Engels et al., 2007). In summary, one of the general implications of an integrated oncogenesis and immunoediting model of cancer is that cancer as a clinical disease (as opposed to an occult pathology) might be better understood as a progressive disruption of the immune system rather than a disorder of cell growth, survival and movement. In this sense, immune escape may be a pivotal trait of cancer if it acts as an essential gateway to the manifestation of clinical disease.

Figure 2
figure2

Immune escape is a fundamental trait of cancer. Cancer is initiated by genetic and epigenetic changes to a normal cell that permit it to acquire a set of cell-intrinsic traits, including immortalization, sustained growth, loss of suppressor genes, resistance to programmed cell death and metabolic alterations (right side cluster). For an initiated cancer to progress from benign to malignant status, the cancer cell must master restrictions in its microenvironment. To do so it must acquire a set of so-called cell-extrinsic traits, including the capacity for invasion, angiogenesis, metastasis and immune escape (left side cluster). Selective pressures against an initiated cancer applied as a result of immune surveillance leads initially to the evolution of immune equilibrium, representing an occult state of tumor dormancy (Koebel et al., 2007), and subsequently to immune escape, which represents the culmination of immunoediting (Dunn et al., 2004). While immune escape mechanisms would be expected to interface or overlap with mechanisms driving the other traits of cancer, the weighting between traits is unlikely to be similar. Figure was modified from its original presentation in the seminal review by Hanahan and Weinberg, 2000.

Immune escape is a modifier of cancer progression

Immune escape was not widely recognized among cancer geneticists or molecular cell biologists as a fundamental trait of cancer until recently (Prendergast and Jaffee, 2007). In the late 1960s, studies of immune deficient nude mice, newly developed at the time, argued that they had no increased susceptibility to spontaneous cancers. An influential interpretation of these findings was that immunity was not a critical restraint to tumorigenesis in mammals. However, this interpretation was flawed by the lack of knowledge that nude mice retain natural killer cells (NK cells), which have potent antitumor activity. Studies of mutated oncogenes discovered in the late 1970s and 1980s tended to reinforce the notion that immunity was not critical for tumorigenesis, based on demonstrations that malignant cells could be created from normal cells in vitro. Through the 1990s, the perspective of many cancer geneticists and cancer cell biologists was that unmutated genes had relatively limited roles in cancer pathophysiology, and it was apparent that overt immune regulatory genes were not mutated in cancer. Later, studies in transgenic mouse models and sound clinical documentation of the reality of dormant cancers forced a greater appreciation of how inflammation and immunity contributed to tumorigenesis. For example, tumors arising in patients who had received a transplant from a donor who years earlier had been cured of cancer were found to be derived from the donor (Curtis et al., 1997), arguing that occult tumor cells from the transplant could be immunologically managed for long periods as dormant disease until they were moved to an immunosuppressed organ recipient (here, it should be emphasized that the clinical ‘cure’ achieved in the donor was simply a reduction of the disease to a dormant occult state). In mice, tactics to genetically ablate T-cell function dramatically increased the incidence of spontaneous solid tumors (Shankaran et al., 2001). These findings demonstrated that adaptive immunity performs an essential tumor suppressor function in mammals. Furthermore, they implied that immune escape is essential for the formation of a tumor. Recent experimental findings directly corroborate the notion that tumor cells can exist in an occult state of immune equilibrium for long periods. In a classical model of chemical carcinogenesis, Schreiber and Smyth and colleagues showed that immune depletion will reveal tumors in mice that will remain tumor free after low doses of carcinogen that are insufficient to trigger tumor formation during the host's lifespan (Koebel et al., 2007). Evidence of transformed but dormant tumor cells was obtained in animals along with a demonstration that such cells are more immunogenic than those present in frank tumors. Thus, along with other cell-extrinsic traits of cancer, immune escape is an essential trait for the development of progressive disease, acting as a biological modifier to dictate the outcome of an oncogenically initiated lesion that may otherwise be eliminated or be present in an extended occult state of immune equilibrium corresponding to dormancy. Here, the definition of a cancer modifier is broadly defined in pathological terms as a gene that is phenotypically silent unless evaluated in the context of cancer. By this definition, many genes influencing cell-extrinsic traits of cancer—angiogenesis, invasion, metastasis and immune escape—are understood as modifiers that dominate tumor outcomes. In short, while mutations in oncogenes and tumor suppressors start cancer, modifiers and the microenvironment which act later may dictate their clinical relevance (Figure 1). Learning how immune escape evolves during the integrated processes of oncogenesis and immunoediting may therefore yield more powerful insights into cancer pathophysiology and therapy than achieved to date.

Emerging mechanisms of immune escape

The field of tumor immunology, which for historical reasons has been mostly separated from mainstream cancer genetics and cell biology for years (Prendergast and Jaffee, 2007), is presently undergoing a revolution in perspective not unlike that which occurred in cancer genetics during the early 1990s. At that time, it was learned that tumor suppressors were dominant to oncogenes and that oncogene-driven proliferation was contingent upon ablating suppressor functions (indeed, oncogenes were found to drive senescence or apoptosis if suppressor functions like p53 were intact). In an analogous fashion, it has been known that tumor-mediated immune suppression overrides tactics to stimulate antitumor immunity: triggering an effective response to a tumor neoantigen relies upon ablating an immune suppression function(s). In short, ‘getting off the brakes’ of a suppressor is essential to ‘get on the gas’ of an activator, whether one is concerned with cell proliferation or effective immune responses in cancer.

A variety of active mechanisms of immune suppression that are elaborated by tumors to drive immune escape have emerged (Zou, 2005; Gabrilovich, 2007). While tumor cells employ direct strategies for immune escape, such as the IDO (indoleamine 2,3-dioxygenase) mechanism discussed below, many strategies are based upon the subversion of immune cells recruited by the tumor cell to its immediate microenvironment and to the tumor-draining lymph nodes, where tumor neoantigens scavenged by antigen-presenting cells are cross-presented to the adaptive immune system. Cells that may contribute to crucial escape mechanisms include regulatory dendritic cells (DCs), which create tolerance to the antigen the DC is presenting; T regulatory cells (Tregs), which propagate antigenic tolerance (these cells are conceptually similar to what was formerly termed suppressor T cells) and myeloid-derived suppressor cells (MDSCs), which likely comprise a variety of subtypes including perhaps the so-called M2 macrophages, which secrete pro-inflammatory factors supporting angiogenesis, invasion and metastasis (Lewis and Pollard, 2006). Another cell type that might contribute to the evolution of tumoral immune suppression are mast cells, which through their ability to instruct Treg formation (Lu et al., 2006) may serve an important role in epithelial tumorigenesis (Coussens et al., 1999; Muto et al., 2007; Soucek et al., 2007). As a group, these immune cells are quite abundant in tumors, tumor-draining lymph nodes and/or peripheral tissues, such as the blood and spleen in tumor-bearing animals, especially in progressive disease.

In regulatory DCs that instruct CD4+ T cells and prompt tolerance to antigens, B7 receptor signaling pathways are vitally important to immune suppression and tolerance. T-cell contacts made through this coregulatory system operate in parallel to those made by the T-cell receptor. The DC presents two ligands to the T cell, a processed antigen bound to MHC proteins and a B7 molecule that binds CD28 on the T cell. Coregulation by the B7 signal tells the T cell whether to respond to the antigen being presented or not: while stimulatory B7 signals activate a T cell to the antigen, inhibitory B7 signals instruct the T cell to become tolerant to the antigen. In essence, inhibitory B7 ligands define a regulatory DC that will tolerize a T cell to the antigen that is being presented. B7 signaling is complex because it operates in both directions, with the ligand on the DC surface and the receptor on the T cell each sending signals into each cell. This system is vital to limit reactions to self antigens or nonself antigens where tolerance is important (for example, paternal antigens in pregnancy and environmental allergens). In cancer, regulatory DCs are important because they may instruct and recruit Tregs that propagate immune suppression to tumor antigens. Precisely how tumors recruit and subvert DCs and Tregs remains obscure. Nevertheless, several important inhibitory B7 receptor signaling pathways have been implicated in tumoral immune escape, including, for example, those involving the T-cell coreceptors CTLA-4, the prototypical family member, PD-1 and OX-40 (Iwai et al., 2002; Blank et al., 2006; Colombo and Piconese, 2007; Gabrilovich, 2007). These cell surface molecules present potential therapeutic targets, which could be antagonized by monoclonal antibodies, and in fact clinical development of receptor antagonist antibodies is already underway. Early work on CTLA-4 antagonists in the clinic suggests that augmentation of tumor immunity may occur along with a unique profile of immune-related adverse events, including colitis, hepatitis and endocrinopathies (J Wolchok (MSKCC), personal communication). Such observations touch an old discussion in tumor immunology, which is whether it will be possible (or desirable) to separate antitumor and antiself events, and if not, how one will therapeutically manage the inescapable balancing act entailed. It is with regard to such questions that IDO poses such an interesting subject, because unlike CTLA-4, its disruption by pharmacological or genetic means in the mouse does not cause fulminant autoimmunity.

IDO mediates an important mechanism of immune escape

An interesting development in the mechanistic studies of immune escape has been the discovery of an interface with metabolic alterations, another trait of cancer (Figure 2). Specifically, important roles in mediating immune tolerance to antigens have been revealed which involve microenvironmental catabolism of the essential amino acids tryptophan and arginine, carried out by the enzymes IDO and arginase I, respectively (Bronte and Zanovello, 2005; Muller et al., 2005b; Muller and Prendergast, 2007; Munn and Mellor, 2007; Popovic et al., 2007). At present, the role of IDO and tryptophan catabolism has been more deeply characterized. IDO is a single-chain oxidoreductase that catalyzes the degradation of tryptophan to kynurenine, the first step in biosynthesis of the central metabolic regulator nicotinamide adenine dinucleotide. In mammals, IDO does not catabolize excess dietary tryptophan, which is performed by the liver enzyme tryptophan dioxygenase, and nicotinamide adenine dinucleotide levels are not maintained by synthesis but by salvage. The question why IDO was conserved in mammals was addressed by a set of seminal studies from Munn and Mellor and their colleagues, which revealed that IDO modulates immunity by suppressing T-cell activation (Mellor and Munn, 2004), initially demonstrated in the setting of allogeneic pregnancy (Munn et al., 1998). T cells appear to be preferentially sensitive to IDO activation, such that when starved for tryptophan they cannot divide and as a result cannot become activated by an antigen presented to them. T cells are also preferentially sensitive to kynurenines and other downstream catabolites generated by the IDO pathway (Fallarino et al., 2002), which along with tryptophan restriction appears to be important for induction of Tregs and immune suppression (Fallarino et al., 2006; Munn and Mellor, 2007).

In T cells, tryptophan starvation triggers a Gcn2-dependent stress signaling pathway that alters eIF2α phosphorylation and translational initiation at the ribosome leading to a cell growth arrest (Munn et al., 2005). Additional effects of this tryptophan catabolism signaling pathway may be mediated by activation of LIP (Metz et al., 2007), an alternately initiated isoform of the b/ZIP transcription factor NF-IL6/CEBP-β that participates in stress responses (Figure 3). LIP encodes the bZIP region that mediates dimerization and DNA binding, but because it does not include a transactivation domain, it acts as a natural dominant inhibitor of NF-IL6/CEBP-β. By interfering with gene regulation, induction of LIP as a result of activating the Gcn2-eIF2α pathway may promote Treg function and immunosuppression, for example, by altering expression of key cytokines such as IL-10 or TGF-β (Huber and Schramm, 2006). The Gcn2-LIP pathway is also induced in tumor cells but its pathophysiological relevance is yet to be determined (R Metz, unpublished observations).

Figure 3
figure3

IDO (indoleamine 2,3-dioxygenase) and IDO2 control a tryptophan catabolism signaling pathway. (a) From tryptophan starvation to LIP activation. By catabolizing the essential amino acid tryptophan, IDO and IDO2 generate kynurenines and other reaction products that can modulate T-cell immunity as well as a local microenvironment that is starved for tryptophan. Little is known as yet of the precise mechanistic effects of the tryptophan catabolites generated. Elaboration of the starvation condition triggers a stress response in local T cells through Gcn2, which responds to amino-acid deprivation by phosphorylating the translation initiation factor eIF2α, leading to a blockade of most translation initiation with the exception of certain factors such as LIP involved in mediating responses to the stress. (b) LIP is a dominant negative isoform of the immune regulatory b/ZIP transcription factor NF-IL6, also known as CEBPβ. LIP is an alternately translated isoform of the transcription factor NF-IL6/CEBPβ implicated in regulating proliferation and immune response. Starvation responses switch NF-IL6/CEBPβ expression from LAP isoforms to the LIP isoform through the use of a downstream translation start site in the mRNA. Encoding only a b/ZIP dimerization domain, LIP functions as a ‘natural’ dominant negative molecule that disrupts NF-IL6/CEBPβ function by competing with LAP isoforms for binding to target gene promoters. Both IDO and IDO2 can switch on LIP, but subsequent restoration of tryptophan levels will only switch it off in the case of IDO, offering a possible mechanism for distal propagation of immune suppression away from the local tumor microenvironment (Figure 5). NF-IL6/CEBPβ target genes with relevance to the function of IDO include the immune suppressive cytokines IL-6, IL-10 and TGF-β, which may be upregulated as a result of LIP induction.

The physiological role of IDO in immunosuppression is dramatically illustrated by the ability of the specific, bioactive IDO inhibitor 1-methyl-tryptophan (1MT) to elicit MHC-restricted T cell-mediated rejection of allogeneic mouse concepti (Munn et al., 1998; Mellor et al., 2001). Strikingly, this tolerance mechanism is exploited in a variety of pathological settings, including cancer (Muller and Prendergast, 2007; Munn and Mellor, 2007). Most human tumors overexpress IDO (Uyttenhove et al., 2003), a finding that probably explains early clinical observations of grossly elevated tryptophan catabolism in cancer patients (Rose and Sheff, 1967). IDO elevation also occurs in a subset of plasmacytoid DCs in tumor-draining lymph nodes (Munn et al., 2004). Recent studies in IDO knockout mice suggest that both sites may be relevant: activation of IDO in either tumor cells or nodal regulatory DCs each appears to be sufficient to facilitate tumoral immune escape (Hou et al., 2007; GC Prendergast, JB DuHadaway, DH Munn and AJ Muller, unpublished observations). In regulatory DCs, evidence is mounting rapidly in favor of the idea that IDO induction acts in a forward feedback loop to instruct naive CD4+ T cells to become Tregs, but also to mediate the ability of Tregs to recruit naive DCs to regulatory status (Fallarino et al., 2003, 2006; Mellor et al., 2004; Curti et al., 2007; Sharma et al., 2007) (Figure 4). Thus, by reinforcing a cycle of antigen tolerance IDO may help tilt the tumor microenvironment from hostile to supportive for tumor cells, and also elaborate a peripheral mechanism of immune escape that could facilitate progression to invasive status.

Figure 4
figure4

IDO (indoleamine 2,3-dioxygenase) activation in tumor cells and dendritic cells suppresses T-cell immunity in cancer. IDO is upregulated in both tumor cells and in a class of regulatory dendritic cells (reg DCs) found within tumor-draining lymph nodes. (Left side of figure) In the reg DC, expression of IDO is prompted by an autocrine interferon loop under the control of the CTLA-4 pathway, a canonical B7 inhibitory receptor signaling pathway. When a T regulatory cell (Treg) induces IDO in a DC it is converted to a regulatory (inhibitory) state for antigen presentation to T cells. However, a forward feedback loop may also operate in which an IDO+ DC can recruit a CD4+ T cell to become a Treg. Thus, in tumor-draining lymph node, where tumor neoantigens scavenged by DCs are cross-presented to the immune system, IDO drives a positive loop for production of Tregs and reg DCs that suppress tumor immunity. (Right side of figure) In the tumor cell, IDO overexpression may limit the proliferation and cytotoxic activity of cytotoxic effector T cells, providing a direct mechanism of immune escape. Here, IDO dysregulation is facilitated by attenuation of the cancer suppression gene Bin1, which supports immune surveillance against tumor cells by restricting IDO expression there. Unlike IDO2, which is expressed predominantly in DCs and other antigen presenting cells, IDO is widely expressed and often highly overexpressed in many types of tumor cells.

Furthering the notion that IDO has a central role in immune control, recent evidence suggests that the action of widely used immunosuppressant, dexamethasone, relies upon induction of IDO in DCs by reverse signaling through the glucocorticoid-induced B7 inhibitory T-cell coreceptor (Grohmann et al., 2007). As noted above, an interesting aspect of IDO is its systemic inactivation at the organism level, either pharmacologically or genetically, does not appear to cause autoimmunity. These findings imply that IDO is not involved in tolerance to self but in tolerance to nonself antigens where immune non-responsiveness may be important (for example, such as fetal antigens or environmental antigens in the skin or mucosa), an idea that is developed further elsewhere (Mellor and Munn, 2008). In this case, there are major implications for cancer, as triggering immunity to tumor neoantigens but not normal self antigens may be quite useful.

IDO is a key regulatory target for cancer suppression gene Bin1

Recent investigations into how Bin1 attenuation promotes cancer yielded an initial clue into how IDO may become upregulated during tumor formation (Muller et al., 2005a). Bin1 encodes the prototypical member of the BAR family of adapter proteins originally discovered and named by our laboratory about a decade ago (Sakamuro et al., 1996). The canonical function of BAR adapters is to coordinate membrane dynamics and signaling processes (Ren et al., 2006). However, family members including Bin1 and APPL proteins also localize to the nucleus where they have been implicated in transcriptional repression (Sakamuro et al., 1996; Elliott et al., 1999; Miaczynska et al., 2004; Ramalingam and Prendergast, 2007). In support of a complex integrative function, >10 alternately spliced isoforms of Bin1 have been characterized (variously termed amphiphysin II or 2 (amph II or 2), ALP or SH3P9). Nuclear localizing Bin1 proteins that are ubiquitous outside the central nervous system interact with c-Myc and suppress its oncogenic activity, both by limiting transformation and facilitating programmed cell death (Sakamuro et al., 1996; Elliott et al., 1999, 2000; Ge et al., 1999, 2000a; DuHadaway et al., 2001, 2003). Genetic studies in fission yeast point to a conserved stress response function that is linked to transcriptional repression through a Rad6-Set1 chromatin modification pathway (Routhier et al., 2003; Ramalingam et al., 2007; Ramalingam and Prendergast, 2007). Attenuated expression of Bin1 occurs in many human cancers including in breast, prostate, lung and colon cancers, neuroblastoma and melanoma (Ge et al., 1999; Ge et al., 2000a, 2000b; Tajiri et al., 2003; Chang et al., 2007a). Gene knockout experiments in the mouse indicate that Bin1 has an essential role in anti-inflammation and cancer suppression. In particular, genetic ablation of Bin1 strongly increases the incidence of lung and liver carcinomas during aging (Chang et al., 2007a). Loss of Bin1 will also cooperate with activated ras to drive the progression of breast and colon carcinomas, supporting a model where Bin1 acts to limit the oncogenicity of c-Myc (Chang et al., 2007b).

While using genetically deficient cells to investigate how Bin1 mediates cancer suppression, we found unexpectedly that its presence strongly supports T-cell-mediated immune surveillance (Muller et al., 2005a). In syngeneic animals, engraftment of c-myc+ras-transformed skin epithelial cells produced small slow-growing nodules if Bin1 was maintained, but large aggressive tumors if Bin1 was deleted. Strikingly, in immunodeficient or T-cell-depleted syngeneic hosts, this benefit was abolished, meaning that much of the in vivo benefit to Bin1 attenuation was mediated by escape from T-cell immunity. This finding suggested a basis for understanding why Bin1 attenuation occurs so commonly in cancer cells, namely, that a benefit to immune escape could be gained during immunoediting by selecting for cells with reduced Bin1 function. By inference, loss of a nuclear-based function is critical, because inactivation by a specific RNA missplicing event in cancer cells (exon 12A missplicing) generates inactive suppressor proteins that are excluded from the nucleus (Ge et al., 1999). The significance of this mechanism of Bin1 attenuation is supported by recent findings that it is mediated by the oncogenic factor ASF (Karni et al., 2007) and that it represents one of the more common missplicing events in human cancer (Xu and Lee, 2003; Roy et al., 2005).

In exploring how Bin1 loss affects immune escape, we found that Bin1–/– cells displayed elevated IDO and this elevation was essential to gain the benefits of Bin1 loss to tumor growth (Muller et al., 2005a). The requirement for IDO activity was demonstrated by using 1MT to inhibit the enzyme: 1MT treatment phenocopied Bin1, restricting tumor outgrowth in immunocompetent syngeneic animals. In contrast 1MT had no effect in either nude mice or syngeneic animals depleted of CD4+ or CD8+ T cells. In demonstrating the ability of 1MT to inhibit tumor growth in a T-cell-dependent manner (that is, by de-repressing IDO-mediated immunosuppression), these findings extended earlier evidence that IDO overexpression could facilitate tumor growth (Friberg et al., 2002; Uyttenhove et al., 2003). Thus, Bin1 appears to support immune surveillance by inhibiting an inhibitor of T cell immunity in IDO. Using mice that are genetically deficient in IDO, evidence has been obtained that its dysregulation in Bin1–/– tumor cells is critical: Bin1 null tumors grow equally aggressively in hosts that are either wild type or IDO nullizygous, yet inhibition of the IDO expressed by Bin1–/– tumor cells in the nullizygous host is still sufficient to prevent tumor outgrowth (GC Prendergast, JB DuHadaway, AL Mellor and AJ Muller; unpublished observations). Thus, IDO restriction by Bin1 in the tumor cell contributes to immune surveillance (Figure 3).

The restrictive effect of Bin1 on IDO occurred chiefly at the level of IDO transcription. IDO elevation in Bin1–/– cells depends on increased NF-κB and STAT activity, based on the ability of inhibitors of these transcription factors to phenocopy Bin1 action. These observations imply that Bin1 may limit IDO transcription by restricting the nuclear trafficking or activity of NF-κB and STAT, and some experimental support for this model exists (Bild et al., 2002; Muller et al., 2004). Considering potential intersections with c-Myc, while IDO is not a canonical Myc target gene, it is interesting that transcriptional activation by NF-κB and STAT is antagonized by c-Myc. Furthermore, two recent findings prompt an evaluation of possible effects of c-Myc on IDO transcription by Bin1. First, fission yeast studies suggest that Bin1 supports Rad6/Set1-dependent histone modifications supporting formation of silent heterochromatin that are antagonized by c-Myc (Ramalingam et al., 2007). Second, c-Myc is now recognized to control up to 10–20% of all genes in the genome and its activity has been ascribed recently to more global than local effects on chromatin (Knoepfler et al., 2006). Lastly, among its other oncogenic properties c-Myc has been found to drive immune escape by an undefined mechanism that alters the interferon response (Schlee et al., 2007). This latter feature is relevant because IDO is a major transcriptional target of interferon signaling and this pathway of IDO activation is known to be under Bin1 control (Muller et al., 2005a). While further work is needed to understand the regulatory mechanisms linking Bin1 to the IDO promoter, testable models integrating the intersection between trafficking/signaling and nuclear roles of Bin1 now exist.

Anticancer properties of small molecule inhibitors of IDO

The remaining part of the review provides an overview of observations and lessons emerging from translational research on IDO inhibitors; this area has been reviewed recently in more depth elsewhere (Muller and Scherle, 2006; Muller and Prendergast, 2007). The chemically simple small molecule inhibitor 1MT has been used widely to probe the role of IDO in cancer and other settings. Two initial studies demonstrated that 1MT can limit the growth of tumors where IDO is overexpressed (Friberg et al., 2002; Uyttenhove et al., 2003). However, the lack of pharmacokinetic analysis in these studies made it unclear whether dosing might have been insufficient to trigger regression of established tumors, as compared to merely limiting tumor growth. Investigations in our laboratory determined that 1MT was stable in serum and that it accumulated to levels consistent with in vivo inhibition of IDO when dosed under conditions that could trigger rejection of an allogeneic fetus (Muller et al., 2005a). Under the same conditions, we confirmed that 1MT only retarded tumor outgrowth in an established autochthonous mouse model of breast cancer, the MMTV-neu transgenic mouse, where spontaneous and individually immunoedited mammary adenocarcinomas that arise closely resemble human HER2+ disease (Muller et al., 2005a). 1MT by itself was unable to trigger tumor regression, suggesting limited efficacy as a monotherapy. In contrast, combining 1MT with paclitaxel or several other cytotoxic chemotherapeutic agents used for breast cancer treatment in the clinic caused rapid regression of established tumors that in the MMTV-neu model responded poorly to single-agent therapy (Muller et al., 2005a). These effects could not be explained by drug–drug interaction, that is, by 1MT increasing the effective dose of the cytotoxic agent, because therapeutic efficacy was increased in the absence of increased side effects (for example, in the case of paclitaxel, it would be reflected by neuropathy as indicated by hind leg dragging in affected mice). In subsequent dose, schedule and route of administration experiments, it was found that a twice daily oral dose of 1MT for as little as 4–5 days was sufficient to produce tumor regressions determined at 2 weeks after enrolling tumor-bearing mice on trial (Hou et al., 2007). Important evidence that IDO targeting is essential to 1MT action was provided by the demonstration that 1MT lacks antitumor activity in mice that are genetically deficient for IDO (Hou et al., 2007). This key control experiment provides a genetic validation of IDO as an essential target for inhibition by 1MT, with the caveat that other targets acting upstream of IDO might also score similarly.

Why would IDO inhibition cooperate with chemotherapy? It is notable that agents that cooperate with 1MT include ‘immunogenic’ chemotherapeutics that trigger cell surface trafficking of calreticulin and immune responses to a dying tumor cell (Machiels et al., 2001; Obeid et al., 2007). Indeed, challenging old assumptions that the therapeutic efficacy of chemotherapeutics is based on the efficiency of cancer cell killing, striking evidence was provided recently that efficacy depends on activation of Toll-like receptor TLR4 by ‘alarmin’ proteins that are released by cells killed by chemotherapy (Apetoh et al., 2007). Since IDO can be activated by certain TLR ligands (Wingender et al., 2006), it is conceivable that 1MT may cooperate with immunogenic chemotherapy by relieving an inhibitory effect on TLR signaling in this pathway or by facilitating the movement of proapoptotic/proimmunogenic calreticulin from the endoplasmic reticulum to the cell surface (Obeid et al., 2007). Irrespective of the mechanism, experiments in the MMTV-neu model confirmed the expectation that immunodepletion of CD4+ or CD8+ T cells before treatment abolished 1MT efficacy, establishing that T cells are needed to mediate the effects of IDO inhibition.

Further support of the concept that IDO inhibition is an effective adjuvant to chemotherapy is provided by the similar patterns of efficacy seen with novel small molecule inhibitors of IDO that have been identified recently. When delivered continuously to counter its rapid clearance, the nontoxic inhibitor methyl-thiohydantoin tryptophan displays greater potency but the same pattern of efficacy as 1MT (Muller et al., 2005a). Similarly, inhibitors of IDO based on the natural product brassinin, a plant phytoalexin with known cancer prevention activity in rodents (Mehta et al., 1995), display antitumor efficacy in the same patterns as 1MT (Gaspari et al., 2006; Banerjee et al., 2007). As before, these effects are abolished against IDO non-expressing tumor cells that are engrafted into either nude mice or IDO knockout mice. In summary, a firm preclinical foundation has been established for the concept that IDO inhibitors will offer utility to enhance cancer treatment.

IDO2 is a biochemically preferred target for inhibition by clinical lead compound D-1MT

The compound 1MT has a single chiral center and it has been used by most investigators in the field as a racemic mixture of the component stereoisomers D-1MT and L-1MT. A single chemical species is typically desirable for clinical translation. Based on its more robust preclinical efficacy, the D stereoisomer of 1MT was chosen recently as a lead compound for clinical development. D-1MT has superior antitumor activity relative to the L stereoisomer in most preclinical models and IDO is genetically required for the antitumor activity of D-1MT (Hou et al., 2007). However, at the level of biochemical specificity the distinction between the two isomers is more complicated, with the L isomer but not the D isomer exhibiting activity as a catalytic inhibitor of the purified enzyme (Hou et al., 2007). In DCs it is nonetheless clear that both isomers can block tryptophan catabolism, with the D isomer being relatively more active than the L isomer (Hou et al., 2007). Nevertheless, in cells other than DCs, some uncertainty exists in how D-1MT may be working.

Three solutions that rectify these observations have been envisaged. First, it is possible that D-1MT is racemized in vivo and that its superior antitumor properties relative to L-1MT reflect better pharmacokinetics. However, no evidence of racemization is known in dosed animals to date and L-1MT is actually the more stable of the two racemers in serum. Second, D-1MT might inhibit a variant isoform of IDO that is expressed selectively by DCs, for example, an alternate spliced or modified form of IDO. By western analysis, one antibody raised to an N-terminal IDO peptide recognizes a slightly larger species in DCs that is constitutive and not interferon-inducible like the smaller species visualized that is more consistent with the known IDO isoform (Hou et al., 2007). This observation suggests that a second IDO-related protein may exist in DCs that may not be expressed as widely as IDO in other cell types. Lastly, one could envisage that D-1MT acts through a different target to mediate its antitumor properties.

This last explantion has been corroborated recently by the discovery of a novel IDO-related enzyme, termed IDO2. Significantly, IDO2 is a preferential target for biochemical inhibition by D-1MT compared to the known IDO enzyme (Metz et al., 2007). IDO2 is the product of a distinct gene, also described as INDOL1 (Ball et al., 2007), that lies immediately downstream of the IDO gene itself in mammalian genomes. The region of the human genome where the IDO2 gene is located is misannotated and as a result the IDO2 nomenclature is preferred to distinguish fully annotated cDNA and genomic sequences in public databases, including several alternate splice isoforms that exist, as compared to misannotated and partial sequences based on the INDOL1 nomenclature that are also found in the databases. Human IDO2 appears to include a unique 5′ exon of uncertain significance. Notably, full-length IDO2 is slightly larger than IDO, in both mouse and human, and IDO2 is expressed by DCs in an otherwise much narrower pattern of expression compared to IDO itself (Metz et al., 2007). These findings are consistent with other evidence mentioned above that DCs have a unique IDO-related species that is inhibitable by D-1MT, consistent with IDO2.

As noted above, the catalytic activity of the IDO2 enzyme is inhibited by D-1MT but not by L-1MT, which is precisely opposite the pattern displayed by IDO (Metz et al., 2007). One implication is that the results of the majority of preclinical studies that have used the racemic D, L-1MT mixture should be re-interpreted in light of the possibility that both IDO and IDO2 were inhibited in these studies. A second implication is that inhibiting both enzymes may be beneficial for antitumor efficacy. For example, methyl-thiohydantoin tryptophan inhibits both IDO and IDO2 (Metz et al., 2007) and displays as much or more antitumor efficacy compared to D and L-1MT, when dosed appropriately (Muller et al., 2005a). A third implication is that the action of IDO2 and IDO inhibition may be linked in a common pathway. As noted above, D-1MT lacks antitumor activity in mice that are genetically deficient for IDO, implying that IDO is an essential mediator of D-1MT action. However, D-1MT is not a biochemical inhibitor of IDO catalytic activity, suggesting that IDO inhibition may be indirectly rather than directly required for D-1MT action. Recognizing the ability of the D racemer to inhibit IDO2, it is possible to rectify these observations by proposing that IDO2 acts upstream of IDO in a common pathway of immune suppression. In this model, IDO mediates IDO2 function such that inhibition of IDO2 by D-1MT will be ineffectual if IDO is absent. The choice of D-1MT as a clinical lead makes careful consideration of the above implications important for future work on IDO2.

In catabolic signaling, IDO and IDO2 exhibit a difference that may have key functional consequences. In cells where LIP is activated due to IDO, restoring tryptophan can relieve this signal and switch off LIP. In contrast, LIP activation by IDO2 is stable and cannot be relieved by restoring tryptophan. As proposed elsewhere (Metz et al., 2007), this signaling difference may have implications for immune control, namely, that stable activation of LIP in DCs by IDO2 may provide a mechanism to propagate tolerance from a local to a peripheral immune environment, away from an initial site of tryptophan catabolism, for example, to support distal metastasis (Figure 5). In any case, differences in LIP response after its induction by IDO versus IDO2 suggest that the functions of these enzymes may be distinct, even if the outcomes for eliciting local tolerance are similar.

Figure 5
figure5

Model for IDO2 in propagating antigenic tolerance from local to distal sites in cancer. In contrast to IDO (indoleamine 2,3-dioxygenase), which is highly expressed in both tumor cells and antigen presenting cells, IDO2 is expressed only in antigen-presenting DCs. IDO expression in cancer cells may be responsible for generation of a tryptophan-starved microenvironment that mediates immunosuppression in the tumor. In the model, IDO2 is essential for conversion of a naive DC to a regulatory DC (reg DC) that can tolerize T cells to a neoantigen. Upregulation of IDO2 in tumor antigen-scavenging DCs by some mechanism would trigger stable induction of LIP. Upon movement away from the local tumor microenviroment to the tumor-draining lymph node (TDLN), these tolerizing reg DCs would recruit naive T cells to become T regulatory cells (Tregs) specific for the tumor antigen being presented on the reg DC. Iteration of a positive feedback would ensue with the instructive action of these Tregs on naive DCs to convert them to reg DCs in an IDO-dependent manner. Two central predictions of this model are that IDO2 is sufficient and perhaps essential along with IDO for production of DCs with regulatory function, and that IDO2 is crucial for cancer progression (that is invasion and metastasis) in an immune competent animal.

Inactivating polymorphisms in the IDO2 gene are common in human populations

The coding region of the human IDO2 gene includes two nonsynonymous SNPs (single-nucleotide polymorphisms) that strongly ablate enzymatic activity. One SNP affects a critical contact between IDO2 and the indole ring of tryptophan, reducing catalytic activity to approximately 90%, and the second SNP generates a premature stop codon that completely abolishes activity (Metz et al., 2007). Notably, both SNPs are commonly represented in human populations, such that as many as 50% of individuals of the European or Asian descent and 25% of individuals of the African descent may lack functional IDO2 alleles (Metz et al., 2007). This distribution may have a significant bearing on the interpretation of clinical responses to D-1MT (or other drug-like inhibitors of IDO2), as well as on the possible role of IDO2 as a genetic modifier of immune response in cancer and other diseases marked by immune suppression. Thus, knowing the genetic status of IDO2 in individuals enrolled in D-1MT trials may be important to interpret clinical responses. Assuming that IDO2 contributes to immune control like IDO, individuals harboring catalytically inactive alleles of IDO2 may display a reduced capacity for immune suppression and therefore a reduced susceptibility to cancer or cancer progression. In the latter scenario if deficiencies in IDO2 activity reinforce dormancy and immune equilibrium, they would be disfavoring the chance of immune escape and malignant progression. In future efforts it will be interesting to determine the relevance of IDO2 to tolerance and immune suppression in cancer as well as other diseases.

References

  1. Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A et al. (2007). Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 13: 1050–1059.

    CAS  Article  Google Scholar 

  2. Balkwill F, Charles KA, Mantovani A . (2005). Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer Cell 7: 211–217.

    CAS  Article  Google Scholar 

  3. Ball HJ, Sanchez-Perez A, Weiser S, Austin CJ, Astelbauer F, Miu J et al. (2007). Characterization of an indoleamine 2,3-dioxygenase-like protein found in humans and mice. Gene 396: 203–213.

    CAS  Article  Google Scholar 

  4. Banerjee T, DuHadaway JB, Gaspari P, Sutanto-Ward E, Munn DH, Mellor AL et al. (2008). Antitumor properties of chemopreventive natural product brassinin are based upon inhibition of indoleamine 2,3-dioxygenase (IDO). Oncogene (in press).

  5. Bild AH, Turkson J, Jove R . (2002). Cytoplasmic transport of Stat3 by receptor-mediated endocytosis. EMBO J 21: 3255–3263.

    CAS  Article  Google Scholar 

  6. Blank C, Kuball J, Voelkl S, Wiendl H, Becker B, Walter B et al. (2006). Blockade of PD-L1 (B7-H1) augments human tumor-specific T cell responses in vitro. Int J Cancer 119: 317–327.

    CAS  Article  Google Scholar 

  7. Bronte V, Zanovello P . (2005). Regulation of immune responses by L-arginine metabolism. Nat Rev Immunol 5: 641–654.

    CAS  Article  Google Scholar 

  8. Chang M, Boulden J, Katz JB, Wang L, Meyer TJ, Soler AP et al. (2007a). Bin1 ablation increases susceptibility to cancer during aging, particularly lung cancer. Cancer Res 67: 7605–7612.

    CAS  Article  Google Scholar 

  9. Chang M, Boulden J, Sutanto-Ward E, Duhadaway JB, Soler AP, Muller AJ et al. (2007b). Bin1 ablation in mammary gland delays tissue remodeling and drives cancer progression. Cancer Res 67: 100–107.

    CAS  Article  Google Scholar 

  10. Colombo MP, Piconese S . (2007). Regulatory-T-cell inhibition versus depletion: the right choice in cancer immunotherapy. Nat Rev Cancer 7: 880–887.

    CAS  Article  Google Scholar 

  11. Coussens LM, Raymond WW, Bergers G, Laig-Webster M, Behrendtsen O, Werb Z et al. (1999). Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev 13: 1382–1397.

    CAS  Article  Google Scholar 

  12. Curti A, Pandolfi S, Valzasina B, Aluigi M, Isidori A, Ferri E et al. (2007). Modulation of tryptophan catabolism by human leukemic cells results in the conversion of CD25- into CD25+ T regulatory cells. Blood 109: 2871–2877.

    CAS  PubMed  PubMed Central  Google Scholar 

  13. Curtis RE, Rowlings PA, Deeg HJ, Shriner DA, Socie G, Travis LB et al. (1997). Solid cancers after bone marrow transplantation. N Engl J Med 336: 897–904.

    CAS  Article  Google Scholar 

  14. DuHadaway JB, Du W, Liu A-X, Baker J, Donover PS, Sharp DM et al. (2003). Transformation selective apoptosis by farnesyltransferase inhibitors requires Bin1. Oncogene 22: 3578–3588.

    CAS  Article  Google Scholar 

  15. DuHadaway JB, Sakamuro D, Ewert DL, Prendergast GC . (2001). Bin1 mediates apoptosis by c-Myc in transformed primary cells. Cancer Res 16: 3151–3156.

    Google Scholar 

  16. Dunn GP, Old LJ, Schreiber RD . (2004). The immunobiology of cancer immunosurveillance and immunoediting. Immunity 21: 137–148.

    CAS  Article  Google Scholar 

  17. Elliott K, Ge K, Du W, Prendergast GC . (2000). The c-Myc-interacting adapter protein Bin1 activates a caspase-independent cell death program. Oncogene 19: 4669–4684.

    CAS  Article  Google Scholar 

  18. Elliott K, Sakamuro D, Basu A, Du W, Wunner W, Staller P et al. (1999). Bin1 functionally interacts with Myc in cells and inhibits cell proliferation by multiple mechanisms. Oncogene 18: 3564–3573.

    CAS  Article  Google Scholar 

  19. Engels EA, Wu X, Gu J, Dong Q, Liu J, Spitz MR . (2007). Systematic evaluation of genetic variants in the inflammation pathway and risk of lung cancer. Cancer Res 67: 6520–6527.

    CAS  Article  Google Scholar 

  20. Fallarino F, Grohmann U, Hwang KW, Orabona C, Vacca C, Bianchi R et al. (2003). Modulation of tryptophan catabolism by regulatory T cells. Nat Immunol 4: 1206–1212.

    CAS  Article  Google Scholar 

  21. Fallarino F, Grohmann U, Vacca C, Bianchi R, Orabona C, Spreca A et al. (2002). T cell apoptosis by tryptophan catabolism. Cell Death Diff 9: 1069–1077.

    CAS  Article  Google Scholar 

  22. Fallarino F, Grohmann U, You S, McGrath BC, Cavener DR, Vacca C et al. (2006). The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor zeta-chain and induce a regulatory phenotype in naive T cells. J Immunol 176: 6752–6761.

    CAS  Article  Google Scholar 

  23. Friberg M, Jennings R, Alsarraj M, Dessureault S, Cantor A, Extermann M et al. (2002). Indoleamine 2,3-dioxygenase contributes to tumor cell evasion of T cell-mediated rejection. Int. J Cancer 101: 151–155.

    CAS  Article  Google Scholar 

  24. Gabrilovich DI . (2007). Molecular mechanisms and therapeutic reversal of immune suppression in cancer. Curr Cancer Drug Targets 7: 1.

    CAS  PubMed  Google Scholar 

  25. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C et al. (2006). Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313: 1960–1964.

    CAS  Article  Google Scholar 

  26. Gaspari P, Banerjee T, Malachowski WP, Muller AJ, Prendergast GC, Duhadaway J et al. (2006). Structure-activity study of brassinin derivatives as indoleamine 2,3-dioxygenase inhibitors. J Med Chem 49: 684–692.

    CAS  Article  Google Scholar 

  27. Ge K, DuHadaway J, Du W, Herlyn M, Rodeck U, Prendergast GC . (1999). Mechanism for elimination of a tumor suppressor: aberrant splicing of a brain-specific exon causes loss of function of Bin1 in melanoma. Proc Natl Acad Sci USA 96: 9689–9694.

    CAS  Article  Google Scholar 

  28. Ge K, DuHadaway J, Sakamuro D, Wechsler-Reya R, Reynolds C, Prendergast GC . (2000a). Losses of the tumor suppressor Bin1 in breast carcinoma are frequent and reflect deficits in a programmed cell death capacity. Int. J Cancer 85: 376–383.

    CAS  Article  Google Scholar 

  29. Ge K, Minhas F, DuHadaway J, Mao N-C, Wilson D, Sakamuro D et al. (2000b). Loss of heterozygosity and tumor suppressor activity of Bin1 in prostate carcinoma. Int J Cancer 86: 155–161.

    CAS  Article  Google Scholar 

  30. Grohmann U, Volpi C, Fallarino F, Bozza S, Bianchi R, Vacca C et al. (2007). Reverse signaling through GITR ligands enables dexamethasone to activate IDO allergy. Nature Med 13: 579–586.

  31. Hanahan D, Weinberg RA . (2000). The hallmarks of cancer. Cell 100: 57–70.

    CAS  Article  Google Scholar 

  32. Hou DY, Muller AJ, Sharma MD, DuHadaway J, Banerjee T, Johnson M et al. (2007). Inhibition of indoleamine 2,3-dioxygenase in dendritic cells by stereoisomers of 1-methyl-tryptophan correlates with antitumor responses. Cancer Res 67: 792–801.

    CAS  Article  Google Scholar 

  33. Huber S, Schramm C . (2006). TGF-beta and CD4+CD25+ regulatory T cells. Front Biosci 11: 1014–1023.

    CAS  Article  Google Scholar 

  34. Hunter KW, Crawford NP . (2006). Germ line polymorphism in metastatic progression. Cancer Res 66: 1251–1254.

    CAS  Article  Google Scholar 

  35. Iwai Y, Ishida M, Tanaka Y, Okazaki T, Honjo T, Minato N . (2002). Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. Proc Natl Acad Sci USA 99: 12293–12297.

    CAS  Article  Google Scholar 

  36. Karni R, de Stanchina E, Lowe SW, Sinha R, Mu D, Krainer AR . (2007). The gene encoding the splicing factor SF2/ASF is a proto-oncogene. Nat Struct Mol Biol 14: 185–193.

    CAS  Article  Google Scholar 

  37. Knoepfler PS, Zhang XY, Cheng PF, Gafken PR, McMahon SB, Eisenman RN . (2006). Myc influences global chromatin structure. EMBO J 25: 2723–2734.

    CAS  Article  Google Scholar 

  38. Koebel CM, Vermi W, Swann JB, Zerafa N, Rodig SJ, Old LJ et al. (2007). Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450: 903–907.

    CAS  Article  Google Scholar 

  39. Lewis CE, Pollard JW . (2006). Distinct role of macrophages in different tumor microenvironments. Cancer Res 66: 605–612.

    CAS  Article  Google Scholar 

  40. Lu LF, Lind EF, Gondek DC, Bennett KA, Gleeson MW, Pino-Lagos K et al. (2006). Mast cells are essential intermediaries in regulatory T-cell tolerance. Nature 442: 997–1002.

    CAS  Article  Google Scholar 

  41. Machiels JP, Reilly RT, Emens LA, Ercolini AM, Lei RY, Weintraub D et al. (2001). Cyclophosphamide, doxorubicin, and paclitaxel enhance the antitumor immune response of granulocyte/macrophage-colony stimulating factor-secreting whole-cell vaccines in HER-2/neu tolerized mice. Cancer Res 61: 3689–3697.

    CAS  PubMed  Google Scholar 

  42. Mehta RG, Liu J, Constantinou A, Thomas CF, Hawthorne M, You M et al. (1995). Cancer chemopreventive activity of brassinin, a phytoalexin from cabbage. Carcinogenesis 16: 399–404.

    CAS  Article  Google Scholar 

  43. Mellor AL, Chandler P, Baban B, Hansen AM, Marshall B, Pihkala J et al. (2004). Specific subsets of murine dendritic cells acquire potent T cell regulatory functions following CTLA4-mediated induction of indoleamine 2,3 dioxygenase. Int Immunol 16: 1391–1401.

    CAS  Article  Google Scholar 

  44. Mellor AL, Munn DH . (2004). IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat Rev Immunol 4: 762–774.

    CAS  Article  Google Scholar 

  45. Mellor AL, Munn DH . (2008). Creating immune privilege: active local suppression that benefits friends, but protects foes. Nat Rev Immunol 8: 74–80.

    CAS  Article  Google Scholar 

  46. Mellor AL, Sivakumar J, Chandler P, Smith K, Molina H, Mao D et al. (2001). Prevention of T cell-driven complement activation and inflammation by tryptophan catabolism during pregnancy. Nat. Immunol 2: 64–68.

    CAS  Article  Google Scholar 

  47. Metz R, DuHadaway JB, Kamasani U, Laury-Kleintop L, Muller AJ, Prendergast GC . (2007). Novel tryptophan catabolic enzyme IDO2 is the preferred biochemical target of the antitumor IDO inhibitory compound D-1MT. Cancer Res 67: 7082–7087.

    CAS  Article  Google Scholar 

  48. Miaczynska M, Christoforidis S, Giner A, Shevchenko A, Uttenweiler-Joseph S, Habermann B et al. (2004). APPL proteins link Rab5 to nuclear signal transduction via an endosomal compartment. Cell 116: 445–456.

    CAS  Article  Google Scholar 

  49. Muller AJ, DuHadaway JB, Donover PS, Sutanto-Ward E, Prendergast GC . (2004). Targeted deletion of the suppressor gene Bin1/Amphiphysin2 enhances the malignant character of transformed cells. Cancer Biol Ther 3: 1236–1242.

    CAS  Article  Google Scholar 

  50. Muller AJ, DuHadaway JB, Sutanto-Ward E, Donover PS, Prendergast GC . (2005a). Inhibition of indoleamine 2,3-dioxygenase, an immunomodulatory target of the tumor suppressor gene Bin1, potentiates cancer chemotherapy. Nature Med 11: 312–319.

    CAS  Article  Google Scholar 

  51. Muller AJ, Malachowski WP, Prendergast GC . (2005b). Indoleamine 2,3-dioxygenase in cancer: targeting pathological immune tolerance with small-molecule inhibitors. Expert Opin Ther Targets 9: 831–849.

    CAS  Article  Google Scholar 

  52. Muller AJ, Prendergast GC . (2007). Indoleamine 2,3-dioxygenase in immune suppression and cancer. Curr Cancer Drug Targets 7: 31–40.

    CAS  Article  Google Scholar 

  53. Muller AJ, Scherle PA . (2006). Targeting the mechanisms of tumoral immune tolerance with small-molecule inhibitors. Nat Rev Cancer 6: 613–625.

    CAS  Article  Google Scholar 

  54. Munn DH, Mellor AL . (2007). Indoleamine 2,3-dioxygenase and tumor-induced tolerance. J Clin Invest 117: 1147–1154.

    CAS  Article  Google Scholar 

  55. Munn DH, Sharma MD, Baban B, Harding HP, Zhang Y, Ron D et al. (2005). GCN2 kinase in T cells mediates proliferative arrest and anergy induction in response to indoleamine 2,3-dioxygenase. Immunity 22: 633–642.

    CAS  Article  Google Scholar 

  56. Munn DH, Sharma MD, Hou D, Baban B, Lee JR, Antonia SJ et al. (2004). Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes. J Clin Invest 114: 280–290.

    CAS  Article  Google Scholar 

  57. Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B et al. (1998). Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 281: 1191–1193.

    CAS  Article  Google Scholar 

  58. Muto S, Katsuki M, Horie S . (2007). Decreased c-kit function inhibits enhanced skin carcinogenesis in c-Ha-ras protooncogene transgenic mice. Cancer Sci 98: 1549–1556.

    CAS  Article  Google Scholar 

  59. Obeid M, Tesniere A, Ghiringhelli F, Fimia GM, Apetoh L, Perfettini JL et al. (2007). Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat Med 13: 54–61.

    CAS  Article  Google Scholar 

  60. Popovic PJ, Zeh III HJ, Ochoa JB . (2007). Arginine and immunity. J Nutr 137: 1681S–1686 S.

    CAS  Article  Google Scholar 

  61. Prendergast GC, Jaffee EM . (2007). Cancer immunologists and cancer biologists: why we didn′t talk then but need to now. Cancer Res 67: 3500–3504.

    CAS  Article  Google Scholar 

  62. Ramalingam A, Farmer GE, Stamato TD, Prendergast GC . (2007). Bin1 interacts with and restrains the DNA end-binding protein complex Ku. Cell Cycle 6: 1914–1918.

    CAS  Article  Google Scholar 

  63. Ramalingam A, Prendergast GC . (2007). Bin1 homolog hob1 supports a rad6-set1 pathway of transcriptional repression in fission yeast. Cell Cycle 6: 1655–1662.

    CAS  Article  Google Scholar 

  64. Ren G, Vajjhala P, Lee JS, Winsor B, Munn AL . (2006). The BAR domain proteins: molding membranes in fission, fusion, and phagy. Microbiol Mol Biol Rev 70: 37–120.

    CAS  Article  Google Scholar 

  65. Rose D, Sheff MD . (1967). Tryptophan metabolism in carcinoma of the breast. Lancet 1: 239–241.

    CAS  Article  Google Scholar 

  66. Routhier EL, Donover PS, Prendergast GC . (2003). hob1+, the homolog of Bin1 in fission yeast, is dispensable for endocytosis but required for the response to starvation or genotoxic stress. Oncogene 22: 637–648.

    CAS  Article  Google Scholar 

  67. Roy M, Xu Q, Lee C . (2005). Evidence that public database records for many cancer-associated genes reflect a splice form found in tumors and lack normal splice forms. Nucleic Acids Res 33: 5026–5033.

    CAS  Article  Google Scholar 

  68. Sakamuro D, Elliott K, Wechsler-Reya R, Prendergast GC . (1996). BIN1 is a novel MYC-interacting protein with features of a tumor suppressor. Nature Genet 14: 69–77.

    CAS  Article  Google Scholar 

  69. Schlee M, Schuhmacher M, Holzel M, Laux G, Bornkamm GW . (2007). c-MYC impairs immunogenicity of human B cells. Adv Cancer Res 97: 167–188.

    CAS  Article  Google Scholar 

  70. Shankaran V, Ikeda H, Bruce AT, White JM, Swanson PE, Old LJ et al. (2001). IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature 410: 1107–1111.

    CAS  Article  Google Scholar 

  71. Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H et al. (2007). Plasmacytoid dendritic cells from mouse tumor-draining lymph nodes directly activate mature Tregs via indoleamine 2,3-dioxygenase. J Clin Invest 117: 2570–2582.

    CAS  Article  Google Scholar 

  72. Soucek L, Lawlor ER, Soto D, Shchors K, Swigart LB, Evan GI . (2007). Mast cells are required for angiogenesis and macroscopic expansion of Myc-induced pancreatic islet tumors. Nat Med 13: 1211–1218.

    CAS  Article  Google Scholar 

  73. Swensen SJ, Jett JR, Hartman TE, Midthun DE, Mandrekar SJ, Hillman SL et al. (2005). CT screening for lung cancer: five-year prospective experience. Radiology 235: 259–265.

    Article  Google Scholar 

  74. Tajiri T, Liu X, Thompson PM, Tanaka S, Suita S, Zhao H et al. (2003). Expression of a MYCN-interacting isoform of the tumor suppressor BIN1 is reduced in neuroblastomas with unfavorable biological features. Clin Cancer Res 9: 3345–3355.

    CAS  PubMed  Google Scholar 

  75. Uyttenhove C, Pilotte L, Theate I, Stroobant V, Colau D, Parmentier N et al. (2003). Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 9: 1269–1274.

    CAS  Article  Google Scholar 

  76. Wingender G, Garbi N, Schumak B, Jungerkes F, Endl E, von Bubnoff D et al. (2006). Systemic application of CpG-rich DNA suppresses adaptive T cell immunity via induction of IDO. Eur J Immunol 36: 12–20.

    CAS  Article  Google Scholar 

  77. Xu Q, Lee C . (2003). Discovery of novel splice forms and functional analysis of cancer-specific alternative splicing in human expressed sequences. Nucleic Acids Res 31: 5635–5643.

    CAS  Article  Google Scholar 

  78. Zou W . (2005). Immunosuppressive networks in the tumour environment and their therapeutic relevance. Nat Rev Cancer 5: 263–274.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

I thank my colleagues and collaborators Alexander Muller, Lisa Laury-Kleintop and Laura Mandik-Nayak (Lankenau Institute for Medical Research (LIMR)); Richard Metz (LIMR Development Inc. (LDI)); David Munn and Andrew Mellor (Medical College of Georgia); and Charles Link, Mario Mautino and Nick Vahanian (New Link Genetics Corporation) for productive ongoing discussions of IDO and immune pathobiology in cancer and other settings. I acknowledge Richard Metz for contributing founding ideas about the function of IDO2 as discussed in our earlier publication and elaborated further in the model in Figure 5. My laboratory is supported by the National Cancer Institute R01 Grants CA82222, CA100123, CA109542 and by funds provided by the Lankenau Hospital Foundation. I declare competing interests as a consultant and major shareholder for New Link Genetics Corporation, which supports IDO and IDO2 research in my laboratory and is developing IDO and IDO2 related technology for the therapy of cancer and other chronic diseases associated with pathological immune suppression.

Author information

Affiliations

Authors

Corresponding author

Correspondence to G C Prendergast.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Prendergast, G. Immune escape as a fundamental trait of cancer: focus on IDO. Oncogene 27, 3889–3900 (2008). https://doi.org/10.1038/onc.2008.35

Download citation

Keywords

  • immunosuppression
  • immunotherapy
  • indoleamine 2,3-dioxygenase
  • Bin1
  • IDO2

Further reading