The concept of DC-based tumour vaccine has been tested both clinically and experimentally for the past two decades. Even though only limited success has been achieved to date, DC vaccination remains a promising immunological approach against tumours and deserves further exploration. It aims to elicit and establish specific immunity to destroy tumours. By such an approach, DC are used not only as a vector to deliver tumour antigens, but also as a “natural adjuvant” to boost vaccine efficacy. Tumours are however of mutated “self”, to which the host immune system is essentially tolerated in the absence of external perturbation otherwise. Such a live cell-based approach is unfortunately extremely sensitive to, hence its efficacy inevitably limited by, the tumour microenvironment. Certain immunosuppressive mechanisms triggered by the tumour cells are therefore major obstacles against successful DC vaccination. Attempts have since been made in order to overcome these hurdles. This brief review summarises some of the earlier and current findings, and compares the effectiveness of various approaches used in these studies. It focuses particularly on strategies aimed at enhancing DC immunogenicity, through molecular modifications and functional conditioning of the cell vectors, targeting both the positive and negative regulators of DC functions. By dissecting the roles of DC in immunity versus tolerance induction, and the very mechanisms underlying autoimmunity, we examine further and try to explain how the suppressed or “misguided” immunity may be alternatively switched-on and more effectively redirected for cancer therapy.
The immune system, in particular the adaptive arm, plays evidently important roles in restricting tumour growth and development 1. T lymphocytes are known to be essential in mediating the anti-tumour immune responses 2–4. Tumours are, however, clones of mutated cells that have arisen from the body's own tissues. To prevent autoimmunity, it is believed that the immune system needs to be “educated” early in life (thymic selection) 5, 6, and continuously through adulthood (peripheral tolerance mechanisms) 7, during which T cells with potential self-reactivities are largely removed or immunologically “silenced”. Although the mutations may give rise to the so-called tumour-specific antigens (TSA) or tumour-associated antigens (TAA), most of these newly derived or altered “self” neoantigens, and those aberrantly expressed non-mutated self antigens, remain poorly immunogenic when presented to the host immune system 8. Many TAA-specific T and B lymphocytes have been identified in cancer patients 4, 9, but these TAA-specific cells are often found in an unresponsive or anergised state. Moreover, tumours may also evade the immune system by interacting actively with host immune cells to block their functions 1, 8, 10. It has become a central question in tumour immunology as to how these TAA-specific clones are tolerated or suppressed, and whether they can be re-activated to induce effective anti-tumour immunity 11, 12.
The initial idea of DC-based tumour immunotherapy was prompted by the understanding that DC could be a potent antigen presenting cell (APC) for T-cell activation 11. Owing to their unique immunobiological properties, DC serve as a crucial link between the innate and adaptive immune systems. DC are widely distributed in various tissues and organs throughout the body, and are very efficient in antigen uptake, processing and presentation 13. DC also constitutively express MHC class I and class II molecules, which can be highly up-regulated on mature or activated DC, and are able to present antigens effectively to both CD4+ (helper) and CD8+ (cytotoxic) T cells. Importantly, unlike tissue macrophages, DC naturally exhibit migratory properties. Upon uptake of antigens in the peripheral non-lymphoid tissues, DC migrate to the T-cell areas of secondary lymphoid organs, where naïve T cells preferentially home to. In other words, DC are in the position, and in theory the only cell type, capable of activating naïve T cells in vivo, and are thus crucial in the initiation of adaptive immune responses 14. These, together with the fact that DC or DC-like cells could be generated in vitro in large numbers 15–17, and readily loaded with either defined or even un-defined tumour antigens 18, led to the attractive concept of using DC therapeutically as an immunogenic cell vector for vaccine delivery 11, 19–23.
Over the past two decades, the DC therapy has attracted intense interest in cancer research. Despite some favourable findings from studies in various experimental models, clinical application has thus far been limited by a lack of achievable general efficacy and consistency, and the outcomes from many clinical trials had not been met with initial expectations 24, 25. Indeed, since the early proof-of-concept studies in animal models reported nearly two decades ago 11, 19, 20, the promise remains to be delivered clinically. In a recent update by Gerold Schuler, current progress and several important issues regarding clinical applications of DC in cancer therapy have been discussed 26. The present review will however focus on the roles of DC in immunity versus tolerance induction, and in particular how recent advances in our understanding of the basic immunological mechanisms underlying self-reactivity may offer valuable new insights in the ways forward.
DC heterogeneity and roles in immunity versus tolerance induction
Our understanding of the basic immunobiological properties of DC has been significantly advanced over the years. This has not only provided good explanations for the problems encountered, but also stimulated many new ideas regarding the potential ways forward aimed to improve DC therapy in a more fundamental way. The important issues lie within DC heterogeneity and functional plasticity, and hence their immunogenic versus tolerogenic properties or potentials.
It has gradually become clear that DC are not a homogeneous population, and questions have also been raised about the origin and nature of the monocyte-derived, DC-like cells generated in vitro 27. The ability of these cells to provide activation signals, of both antigen-specific and non-specific triggers, can vary vastly among DC subsets or lineages, and depends on their functional status 28–31. Among them, a unique human DC subset (CD11c+CD141+), with superior antigen cross-presentation capacity and expressing the XC chemokine receptor 1 (XCR1+), has recently been identified by several groups as the homologue of mouse CD8α+ DC 32–35. As with their murine counterparts, this type of DC was found to be effective activators of CD8+ cytotoxic T cells, which may have important implications in the design of new human DC vaccines. Moreover, in addition to subset-dependence, the functional properties of DC are also associated with the maturation status of the cell. Immature DC are in a so-called “antigen-uptake mode”, with low cell surface expression of MHC class I and class II molecules, which can be rapidly enhanced upon exposure to maturation or activation signals, acquiring subsequently the “antigen-presenting mode”. The low MHC expression may therefore affect the ability of immature DC to present antigen to T cells. Under certain conditions, DC can even exert tolerogenic effects by producing immunosuppressive molecules, or by inducing regulatory T cells, to inhibit the immune system 1, 8, 24, 36. The concept of tolerogenic DC has become far more appreciated. It is now recognised that while immunogenic DC play an important role in host defence, their tolerogenic counterparts are crucial for the maintenance of self-tolerance, being part of a built-in mechanism to avoid autoimmunity 37. It has been demonstrated that, under the tumourigenic microenvironment, the host DC possessed a typical tolerogenic, or regulatory, phenotype 38. DC, as a double-edged sword, can therefore induce either active immunity or tolerance depending on their functional conditions. The types and functional status of DC, hence the immunogenic “quality” or nature of the cell vectors employed for tumour vaccine delivery, are therefore of critical importance.
Molecular and functional conditioning of DC for tumour vaccine development
Various attempts have subsequently been made in order to generate DC with a highly immunogenic phenotype. These include strategies to enhance and stabilise the abilities of DC to provide the antigen-specific (“Signal 1”) and antigen-non-specific but essential co-stimulatory (“Signals 2, 3 …”) signals for T-cell activation 39.
DC functionally conditioned for efficient presentation of tumour antigens
Antigen specificity and memory are two essential features of adaptive immunity. A lack of presentation of tumour antigens by DC in vivo in patients with cancer has long been suggested based on findings from early studies in animal models 11, 40, 41. In support of this, abnormalities in DC functional phenotype, with a downregulated expression of MHC class I and class II molecules, have been further demonstrated in cancer-bearing individuals 42. These findings could thus explain at least in part the insufficient induction of T-cell-mediated anti-tumour immunity observed in patients with cancer 40, 43. Indeed, the very objective initially proposed for DC-based tumour therapy was to improve the in vivo presentation of tumour antigens, in an attempt to expand those rare tumour-specific T cells in these patients 11.
To maximise the efficiency and stability of antigen presentation by DC, several strategies have been developed. These include the use of various forms of tumour antigens for DC loading, means by which DC were loaded with tumour antigens, and ways through which the antigen-loaded DC were delivered into the patients 11, 44. Moreover, DC transduced with tumour-derived RNA 45, DNA 46 or fused directly with tumour cells 47 have also been tested and shown to be more effective in delivering the tumour-specific signals, and for the induction of anti-tumour responses in vitro and in vivo. One important issue which was not sufficiently addressed in these early studies, however, was about the abilities of DC to deliver the essential co-stimulatory signals, i.e. in addition to the antigen-specific triggers, for T-cell activation.
DC functionally conditioned to enhance essential co-stimulatory signals
Although the main function of DC is to present antigens to T cells, what make DC special are their potent immunological adjuvanticity and diversified regulatory capacities 7, 14. Importantly, DC can provide both activating and inactivating co-stimulatory signals to the T cells they interact with. These include both the cell surface membrane-bound (e.g. B7) and soluble (e.g. cytokines) molecules. Antigen recognition by T cells in the absence of certain essential co-stimulatory signals may result in T-cell deletion or anergy, and the induction of regulatory T cells 48. The expression or level of expression of these co-stimulatory molecules on DC is again found to be directly associated with the maturation or activation status of the cells. Immature DC are characterised by low surface expression of not only MHC (class I, class II) but also B7 (CD80, CD86) and CD40 molecules 48. The functional status determines the ability of DC to produce cytokines too, including not only those known to be important in anti-tumour responses (Th1 type), but also potent immunosuppressive cytokines that may instead down-regulate host immunity.
To optimise DC immunogenicity, subsequent attentions have therefore been shifted to focus on the enhancement and stabilisation of these immunogenic costimulatory molecules associated with DC functions. One of the initial strategies was to enhance their expression immunologically by factors that induce DC maturation (e.g. inflammatory stimuli or cytokines) 49, 50. However, there is also evidence that even fully mature DC by this approach may promote regulatory T-cell expansion 51. Another strategy is through molecular modification of the cells, e.g. by selective over-expression (transfection) of genes encoding the Th1 cytokines (e.g. IL-12) 52, CD40 or CD40 ligands 53, 54 and the B7 (CD80, CD86) molecules essential for activating T as well as B cells. DC over-expressing, or even tumour cells transfected to express, some of these molecules either individually or in combination, have been shown to possess increased abilities to stimulate allogeneic T responses in vitro, and to induce tumour-specific immunity in vivo 52, 53, 55 (To et al., unpublished observations from our laboratory). These findings indicate that DC can indeed be genetically modified and functionally conditioned to acquire an enhanced immunogenic phenotype. However, the relatively increased immunogenic properties of DC are often limited, and could be rapidly down-regulated again upon their interactions with certain tumour cells or by tumour-derived factors. The key limiting factor is thus again about the immunosuppressive tumour microenvironment such a live cell approach is directly exposed and sensitive to.
Guiding the “misguided” – Targeting the negative regulators of DC functions
Insights from studies of the underlying autoimmune mechanisms
Recent advance in our understanding of autoimmune mechanisms has offered valuable new insights as to how the “misguided” immunity could be more effectively redirected for cancer treatment. This relates particularly to findings about the roles of DC in the induction and regulation of autoimmune responses. DC, and their complex interactions with dying cells, are evidently involved in triggering systemic autoimmunity in mouse models 56, 57. However, susceptibility to the development of a lupus-like clinical disease appeared to depend strictly on the genetic background of the mice, which was associated with the induction of certain pathogenic Th1-mediated auto-antibodies. The disease induction was found to be tightly controlled by certain immune regulatory mechanisms. Among them, an essential protective role of interleukin 10 (IL-10) was demonstrated in the resistant mouse strain 56, and this has also been further confirmed using IL-10-deficient mice (Ling et al., unpublished observations from our laboratory). IL-10 is a potent immunosuppressive cytokine secreted by a variety of immune cell types including DC 58, 59, which can effectively inhibit T-cell activation, while DC differentiation and functional activities are in return tightly regulated by this very cytokine 59–61. IL-10 is also known to be very important in the protection against inflammatory bowel disease (IBD), another Th1/Th17-mediated autoimmune disease in the gut 62.
During autoimmune or overtly persistent immunological responses, many regulatory mechanisms are triggered (many of which involve the induction of IL-10), in an attempt to limit the ongoing harmful inflammatory reactions 59. Such a negative feedback regulatory mechanism is known to be crucial in protecting normal individuals from immune-mediated diseases, which is also a good example of the “Yin-Yang” balance within the context of immunology. Chronic or persistent inflammation has been associated with tumour development too, although the causal relationship remains to be fully understood. Triggering of neoplastic transformation or production of inflammatory mediators that may promote cancer cell survival, proliferation and invasion are among the possible mechanisms proposed 63. The ongoing chronic inflammatory conditions may also reflect a desperate attempt of the host immune system to mount anti-tumour responses, which could be a consequence of the continuous, yet largely futile triggering by those poorly immunogenic TAA. As a result of the negative feedback loop, an excessive production of anti-inflammatory or immunosuppressive molecules followed by the exhaustion of the immune effector cells may instead lower the ability of the host immune system to mount specific anti-tumour responses. The brief but vivid description of tumours being “wounds that do not heal” by Dworak many years ago is indeed a plausible immunological definition of cancers 64.
Moreover, tumours may also produce various immunosuppressive factors, including IL-10, to suppress host immunity directly 65–67. Under the influence of the tumourigenic microenvironment, as mentioned above, the host DC may acquire a tolerogenic phenotype. These tumour-conditioned DC could, in return, produce a variety of immunosuppressive molecules too, thus further promoting tumour immune escape 38.
A crucial role of DC-derived IL-10 in inhibiting successful DC-based tumour immunotherapy
A crucial role of IL-10, particularly DC-derived IL-10 (DC-IL-10), in inhibiting successful DC-based tumour immunotherapy has recently been demonstrated in mouse and rat models of hepatoma and melanoma 68, 69. In these studies, we showed that DC generated from IL-10 knock-out mice (IL-10−/− DC), or knocked down of the endogenous IL-10 by siRNA, were superior over conventional DC as the vectors for vaccine delivery. In the absence of IL-10, DC were found to be highly immunogenic expressing enhanced levels of surface MHC class II molecules and secreting increased amounts of the Th1 type of cytokines (IL-12, IFN-γ) 68. The IL-10-deficient DC also migrated much more rapidly to the T-cell areas of draining lymph nodes (unpublished observations from our laboratory). By inducing tumour-specific killing and through the establishment of immunological memory, the vaccines delivered by IL-10−/− DC could evoke strong therapeutic and protective immunity against the tumours. In particular, the effects on liver cancers are most encouraging 68. This is because liver is an organ best known for its tolerogenic microenvironment, which also explains why DC in the liver are generally tolerogenic in nature.
The superiority of IL-10−/− DC for vaccine delivery is thus well explained immunologically by their improved abilities to provide both the antigen-specific and essential co-stimulatory signals 68, and to reach rapidly the secondary lymphoid organs where adaptive immune responses are initiated. The findings are also in agreement with several previous studies on the role of suppressor of cytokine signalling (SOCS) molecules in regulating DC immunogenicity. SOCS are a group of intracellular negative regulators of JAK/STAT signalling, and the expression of some of its members (SOCS1 and SOCS3) is also associated with IL-10 receptor triggering 70. The SOCS1 molecule, for example, is a potent suppressor of DC and macrophage activation 71–73. DC from SOCS1-deficient mice are hyper-responsive in vitro, and spontaneously activated in vivo. Interestingly, SOCS1−/− mice also develop a spontaneous lupus-like disease indicating a crucial role of this molecule in regulating self-reactivity 71. Most importantly, it has been demonstrated that the inhibition of SOCS1 could enhance significantly the abilities of DC to present tumour antigens, to produce IL-12 and to induce effectively anti-tumour responses 73–75.
The lack of IL-10 could therefore potentially render DC resistant to the tolerogenic tumour microenvironment, hence to the conversion of “regulatory” or “tolerogenic” DC 38. This may have further impact on DC functions by alleviating certain inhibitory signals through other negative receptors expressed on DC. DC-derived Ig receptor 2 (DIgR2) is, for example, an inhibitory receptor associated with immunoreceptor tyrosine-based inhibitory motifs (ITIM), which could be up-regulated on DC in response to IL-10. It has recently been demonstrated that selective blocking DIgR2 on DC could enhance their immunogenicity in vitro, and tumour vaccines delivered by the DIgR2-silenced DC elicited potent anti-tumour immune responses in vivo in mouse models 76.
In conclusion, emerging evidence indicates that one of the most effective ways to enhance the efficacy of DC-based tumour immunotherapy is by targeting the negative arm of immune regulation. The removal of DC-IL-10, in particular, breaks directly and effectively the negative feedback loop thus alleviating the immunosuppressive impacts of tumours on the host immune system. It allows the generation of immunologically optimised DC vectors, which can provide potentially both strong antigen-specific triggers and essential co-stimulatory signals, for inducing tumour-specific immunity even under the highly immunosuppressive tumourigenic microenvironment (Fig. 1).
For future clinical applications, this may be achieved by the use of small interfering RNA (siRNA) for knocking down DC IL-10 expression 77, subjected to highly optimised conditions including kinetics of DC differentiation and maturation, and dosage and type of the transfection reagents employed 69 (http://www.wipo.int/pctdb/en/wo.jsp?WO=2008071093). The idea of generating human embryonic stem-cell derived DC (esDC) cell lines 78 devoid of the IL-10 gene 69 can be tested too. Future studies should also be designed to remove other immunosuppressive molecules associated with DC functions, such as indoleamine-2,3-dioxygenase (IDO) 79, transforming growth factor-β (TGF-β) 80, arginase I and prostaglandin E2 (PGE2) 38, galectin and IL-27 81 and IL-35 82, 83. The risk of using these artificially modified highly immunogenic cells is of course not without concern; however, this may be largely avoided by identification and combination of highly selective immunogenic TAA epitopes for DC antigen presentation and, potentially, by co-introduction of a drug-sensitive “suicide” gene 84, e.g. into the proposed IL-10-deficient esDC 69, as a method of therapeutic end point control. The novel DC vaccines should potentially elicit tumour-specific immunity more effectively, while minimising the impacts of negative feedback loops due to overall host responses to a generalised self-reactivity.
FPH is currently supported by Higher Education Funding Council UK, and has received research funding support from Arthritis Research UK and Hong Kong Research Grant Committee (PIs), the MacFeat Bequest Fund and the Li Ka Sheng Academic Foundation (Fellowship). YXC is currently affiliated to the Xiang Ya School of Medicine, Central South University, China, and has received funding support from the Cheng Yu Tong Academic Foundation (Visiting Scholarship).
Conflict of interest: The authors declare no financial or commercial conflict of interest.