Discovery of an Innate Cancer Resistance Gene?
Paul Ehrlich first suggested the idea that the immune system might help to protect humans and other animals against cancer a century ago, but despite the subsequent emergence of Burnet and Thomas’ cancer immune surveillance hypothesis, the field of tumor immunity has enjoyed a checkered history. Several recent reviews have eloquently covered the high and low points that this field has endured (1,2), but the most recent developments in cancer immunology have been remarkable. The identification of highly promising, specific tumor antigens; the confirmation of a measurable—using new methodologies—human response to such antigens; the discovery of pattern recognition receptors (i.e., the Toll family of receptors that recognize motifs on proteins and lipids from bacteria) that link innate and acquired immunity (Box 1), the development of new cancer vaccines; and powerful evidence that validates the theory of tumor immune surveillance, have served to rekindle interest in tumor immunology. Much research and clinical effort has been focused on dendritic cells (DC) as sentinels of the immune response and cytotoxic T cells (CTL) as effectors of adaptive antitumor immunity. However, a new stage of cancer vaccine development is on the horizon with the discovery and characterization of novel, but possibly evolutionarily ancient, tumor rejection mechanisms that involve the innate effector cells of our immune system. Such populations include natural killer (NK) cells and macrophages recognizing newly described stress-induced ligands (3) , NK1.1+ T (NKT) cells restricted by and recognizing glycolipid antigens in the context of the non polymorphic MHC class I-like molecule CD1d, as major regulators of cell-mediated immunity to tumors (4) , eosinophils, and other polymorphonuclear cells. We discuss the concept of immunosurveillance in light of recent observations describing a colony of mice in which innate resistance to implanted tumors is dominantly inherited and age-dependent (5).
Innate vs adaptive immunity.
Innate immunity involves epithelium, phagocytic cells such as neutrophils and macrophages, the complement system, natural killer (NK) cells, and cytokines. These components rapidly respond and protect the host from primary infections, without a requirement for previous priming by specific non-self antigens. Adaptive immunity is mediated by T and B cells that are stimulated by the presentation of processed antigen. In contrast to innate immunity, adaptive immunity is characterized by specificity for distinct macromolecules and “memory.”
Despite the promising responses observed in some patients receiving new therapies that mobilize innate or adaptive or both arms of the immune system, an important question about natural immunity to neoplastic disease still lingers: Is the incidence of cancer controlled solely by the balance between the genetic mutation rate and programmed cell death, or is there a natural contribution from an immune system component? Recent studies have indicated the importance of T cells in the prevention of some hematological and epithelial cancers in mice (6 –8) , and a variety of cytotoxic molecules [i.e., perforin, Fas ligand (FasL), and tumor necrosis factor (TNF)-related apoptosis inducing ligand (TRAIL)] and cytokines (e.g., IFN-γ ) as deemed critical to host prevention of malignancy (7, 9–,12). Rare paraneoplastic neurological degenerations (PNDs) in humans, where tumor suppression is strongly associated with an autoimmune response to neuronal antigens, also compellingly support a natural role for T cells in immunity to cancer (13). Conversely, the contribution of innate cells in tumor surveillance has been comparatively neglected.
Recently, Cui et al. have described an extraordinary strain of mice (spontaneous regression–cancer resistant, SR/CR) that are resistant to many types of transplantable tumors (5). The mouse strain derives from a single male BALB/c mouse initially observed to be completely resistant to the aggressive S180 sarcoma cell line. The resistance trait is inherited in a dominant manner and appears to be controlled by one chromosomal locus. The authors determined that the immune system mediated the antitumor effect because leukocytes adoptively transferred from tumor-primed mice conferred tumor resistance (of the SR/CR strain) to recipient mice (of the original BALB/c strain). In addition, leukocytes from the SR/CR mouse killed various types of tumor cells in vitro. Most interestingly, the ability to resist tumor challenge was not dependent on T cells because SR/CR athymic nude mice, which lack most T cells, also displayed tumor resistance. A massive infiltration of leukocytes consisting predominantly of innate immune-system cells including granulocytes, macrophages, and NK cells was observed in the tumors, and the SR/CR mice also rejected MHC-negative tumor cells, which could not be detected by T cells. Reports of strains of mice that can completely resist tumor growth or metastasis have been extremely rare (14, 15); however, this recent work appears even more unusual because it provides some of the best evidence that the innate immune system can be a powerful means of controlling malignancy.
Several other recent studies have indicated the importance of various lymphocyte subsets of the innate immune system in controlling tumor formation including roles for NK and invariant NKT cells in mediating protection from carcinogen-induced sarcomas (16,17); NK cells in graft-versus-leukemia following mouse or human allogeneic bone marrow transplantation (18); and γδ+ T cells in protection from carcinogen-induced skin papillomas (19). It remains unclear what molecules on the transformed cells are stimulating these innate effector cells to prevent tumor development, but the possibility of stress-induced ligands that react with the activating NKG2D on these effector cells is one possibility that is receiving much attention (3). NKG2D is an activating receptor expressed on human and mouse NK cells, macrophages, γδ+ T cells, and some other T cell subsets that binds several families of receptors only expressed on transformed or stressed cells. The ability of other leukocytes, such as granulocytes, to effect tumor suppression is also an emerging area. Curcio et al. have demonstrated that depletion (i.e., removal) of polymorphonuclear granulocytes completely abolishes protection afforded by a DNA vaccine that prevents mammary tumor growth (20). In concert with a previous study implicating CD4+ T cells (also termed T helper type 2, Th 2) in orchestrating host innate responses to tumors (21), Mattes et al. observed the remarkable antigen-specific Th 2 cell–dependent rejection of established lung metastases by eosinophils (22). It will now be of great interest to see what population of leukocytes (i.e., neutrophil, macrophage, NK cell, or other) is the primary effector that mediates protection in the SR/CR mice. Notably, the SR/CR mouse rejected NK cell-resistant targets that were also FasL- and TRAIL-resistant, thus, it remains to be determined what effector mechanism is operating in the SR/CR mice.
A revised model of cancer immune surveillance. The initial event in priming the immune system to cellular transformation likely consists of a specific danger/alert signature caused by cellular transformation in the context of its surrounding tissue microenvironment. This signature change in the microenvironment may manifest itself as alterations in regulatory cytokines and chemokines produced by tissue cells, the leukocytes with the tissue, and the endothelium supplying nutrients to the tissue. The subsequent immune response to this signature will depend upon the nature of the host cell transformed and transformation process itself. This resulting immune response may be entirely innate (A) or additionally involve components of the adaptive immune response (B). Tissue dendritic cells will likely be central to either type of basic response by virtue of their ability to capture antigen, receive signals through their pattern recognition receptors, and commence crosstalk with other cells in the tissue. Innate responses may include the recruitment of various leukocytes including NK cells, macrophages, and granulocytes (i.e., neutrophils and eosinophils). These events may be sufficiently amplified to control transformation and eliminate the tumor mass by immunological and non-immunological (e.g., vasculature) mechanisms. In other instances, and by as yet undefined molecular cues, dendritic cells within the transformed tissue will be activated and migrate to draining lymph nodes where they will provoke responsiveness of the adaptive immune system including CD4+ and CD8+ T cells. Activated T cells specific for tumor then make their way along a chemokine gradient to the site of tumor growth where they engage in MHC-restricted elimination of tumor cells.
Clearly, the most important outstanding issue is the identity of the single locus dominant gene in the SR/CR mice. There are many possibilities that come to mind, but given that Cui et al. report that immunity can be transferred to other mice, it would seem most likely the gene encodes a ligand uniquely expressed by a subset of leukocytes in the SR/CR mice. Is it a new cell-surface transmembrane ligand expressed by innate cells? If so, is its counter receptor found specifically on tumor cells? Is the expression of a counter receptor induced by stress, or is the receptor a common retroviral or pathogen gene product? Does the ligand, in SR/CR mice, have a role in host-defense against pathogens, and what might be the evolutionary disadvantage of retaining this gene? Do the innate cells of SR/CR mice generate a heightened “danger” response to malignant cells? The next issue is whether the SR/CR mice are also resistant to tumor initiation in situ. A simple set of experiments involving chemical- and oncogene-driven carcinogenesis in these mice will determine just how genetically resistant to cancer these mice might be. Another fascinating observation made by Cui et al. was that the resistance trait deteriorated with age. This suggests that the increased incidence of cancer with age may not be solely due to the accumulation of genetic mutations, but may also involve a contribution from a resistance gene that is switched off with age or is embodied in a leukocyte population that deteriorates with age.
Three other features of tumor rejection by the SR/CR mice were quite remarkable. First, transplanted tumors never escaped in young mice carrying the resistance gene, which suggests that this immune mechanism is not easily evaded by a broad variety of tumors. Second, despite the fact that primary rejection was mediated by innate cells, mice subsequently demonstrated a more effective secondary (memory) response to the same tumor. Exploring whether other mice (including old SR/CR mice) can be protected by priming—and the molecular means by which SR/CR mice become primed—will be of great interest. Finally, SR/CR mice strongly responding to tumor challenge never developed any reported indications of autoimmunity, an indication that innate immunity can operate powerfully without causing self-tissue damage.
It is tempting to speculate that a similar gene exists in humans, and that this may be responsible for protection of rare individuals despite exposure to large quantities of radiation, carcinogens, and other insults. Importantly, the comprehensive nature of the resistance raises the intriguing possibility that this strain of mice may provide insight into the “holy grail” of tumor immunology, that is a panacea for cancer. Discerning the molecular mechanisms responsible for control of tumor growth in the SR/CR mice may lead to the development of new and exciting therapies for cancer.
It is likely that this report will generate much excitement within the research community, and more attention on innate immune mechanisms in tumor immunity. Clearly identifying the relevant leukocytes and the gene responsible for this remarkable effect are high priorities. Scientists from many disciplines all eagerly await further developments in this story.
- © American Society for Pharmacology and Experimental Theraputics 2003
References
Mark Smyth, PhD, (left) is currently Associate Professor and Head of the Cancer Immunology Program at The Peter MacCallum Cancer Centre in Melbourne, Australia. Mark has a keen interest in natural immunity to cancer, and has recently described molecular links between innate and adaptive immunity in tumor rejection. Address correspondence to MS. E-mail m.smyth{at}pcmi.unimelb.edu.au; fax +61 3 9656 1411.
Michael Kershaw, PhD, (right) is currently a Senior Research Fellow at The Peter MacCallum Cancer Centre. His main interests lie in the interaction between tumors and the immune system, and how to utilize knowledge about this field to develop effective immunotherapies for cancer.
