Dendritic cells are a special subset of leukocytes that form a complex network of antigen-presenting cells (APC) throughout the body. They play a principal role in the initiation of immune responses to invading microorganisms (bacteria, fungi, and viruses), malignant cells, and allografts by activating naЁive lymphocytes, by interaction with innate cells, and by the secretion of cytokines. At certain developmental stages they grow branched projections, the dendrites, hence the cell’s name.
Origin and function
Dendritic cells (DC) were characterized for the first time by Steinman in 1973 based on their distinct morphology with different cytoplasmic extensions, such as dendrites, pseudopodia, and lamellipodia, which give the cell its star-shaped feature. Due to their pronounced morphology, DC have a large surface, ensuring close contact with neighboring cells.
Variations among the tissue distribution of DC and differences in their phenotype and function indicate the existence of heterogenous populations of DC. DC originate from different hematopoietic lineages in the bone marrow (Table 1). A myeloid progenitor cell can differentiate in vivo to different DC populations: Langerhans cells that migrate to the skin epidermis and interstitial DC that migrate to the skin dermis and various other tissues (airways, liver, and intestine). Circulating, or migrating, DC are found in the blood and in the afferent lymphatics, respectively (the latter called veiled cells). Interdigitating DC are found in the paracortex of lymph nodes in close proximity with T cells. In addition, monocytes represent an abundant source of DC precursors during physiological stress. Another subset of DC, plasmacytoid DC (pDC) originate from a lymphoid progenitor cell in lymphoid organs. By contrast, follicular DC (FDC) are probably not of hematopoietic origin, despite similar morphology and function to the abovementioned subsets of DC. FDC are APC of the B cell follicles in lymph nodes and central players in humoral immunity.
After application of danger signals, DC express several different types of membrane molecules that determine their phenotypic and functional characteristics:
- DC display a high surface density of antigenpresenting molecules, such as CD1a, major histocompatibility complex (MHC) class I and class II molecules. The level of expression of these molecules is 10to 100-fold higher compared to other APC (e.g., B cells).
- In addition, mature DC have high expression levels of costimulatory and adhesion molecules: CD40, ICAM-1/CD54, ICAM-3/ CD50, LFA-3/CD58, B7-1/CD80, and B7-2/ CD86. Binding of these molecules with their respective receptors on T cells results in T cell activation and subsequently stimulates the expression of cytokines, cytokine receptors, and genes for cell survival.
- Several members of the integrin family are expressed by DC. Cadherins contribute to the generation of cell contacts and selectins are important for the motility of DC.
- DC also express pathogen-recognition receptors, e.g., DEC-205, a macrophage-mannose receptor capable of binding bacterial carbohydrates and toll-like receptors (TLR), recognizing a variety of pathogen-αssociated molecular patterns (PAMP), such as carbohydrates, nucleic acids, peptidoglycans, and lipoteichoic acids.
- Cytokine and chemokine receptors are also important for DC function, since growth, differentiation, and migration of DC as well as antigen processing and presentation are tightly regulated by cytokines and/or chemokines.
Table 1. Different subsets of dendritic cells
|CD34+ hematopoietic stem cell|
|Myeloid progenitor cell||Lymphoid progenitor cell|
|Monocyte-derived DC||Langerhans cells||Interstitial DC||Plasmacytoid DC|
The widespread distribution of DC and their expression of a variety of membrane molecules underline their sentinel function: they patrol the body to capture invading pathogens and certain malignant cells in order to induce efficient antimicrobial or anti-tumor T cell responses. In their in vivo steady-state condition, immature DC are specialized in capturing antigens, i.e., they efficiently take up pathogens, apoptotic cells, and antigens from the environment by phagocytosis, macropinocytosis, or endocytosis. However, immature DC remain tissue resident, expressing only small amounts of (MHC) class II and of costimulatory molecules, which leads to T cell unresponsiveness. After encounter of a “danger” signal (e.g., TLR ligand) immature DC mature and migrate to the secondary lymphoid organs. Mature DC are considered to be immunogenic, mainly due to the marked upregulation of MHC class II and costimulatory molecules. This maturation step is believed to be a crucial event to regulate DC function and makes DC potent inducers of T cell immunity.
Dendritic cell-based immunotherapy
Despite our immune system’s function to protect us from malignant cells, tumor cells grow undisturbed and, unless treated, are fatal to the host. The reasons for the failure to eliminate tumor burden in a majority of patients can be the consequence of different tumor escape mechanisms. For example, tumor-derived inhibitory factors (e.g., PD-L1/2 IL-10 and/or TGF-β) or tumor cell-induced T regulatory cells (Treg) might be involved in downregulating or altering immune function. The goal of cancer immunotherapy is to resolve or circumvent these problems and generate tumor-specific immune responses. It is important to realize that immunotherapies will likely only be successful after reducing tumor mass via primary therapies: surgery and radioand/or chemotherapy, i.e., in a minimal residual disease (MRD) setting.
Because of their pivotal immune-stimulatory capacity and their ability to activate naЁive tumorspecific T cells, DC-based cancer vaccines could have important applications in the future treatment of cancer. For this, it was necessary to cultivate DC with high yields. Several cultivation protocols were developed for in vitro generation of DC. First, DC can be differentiated from CD34+ hematopoietic progenitor cells using granulocyte-monocyte colony stimulating factor (GM-CSF), tumor necrosis factor (TNF-α), stem cell factor (SCF), interleukin (IL)-3, and interleukin-6. Second, DC can be generated starting from monocytes using GM-CSF and Interleukin-4. Finally, DC can be directly harvested from the peripheral blood of a patient, where they reside at low percentages (0.1 %).
Next, cultivated DC can be loaded with the tumor antigen of importance in different ways:
- DC can be grown in vitro in the presence of tumor-αssociated antigens (TAA). This technique is called peptide pulsing and results in direct binding of the immunodominant epitope on an empty MHC class I molecule on the DC membrane. This circumvents the need for antigen uptake and processing and ensures the stimulation of tumor-specific cellmediated cytotoxicity. However, the number of known TAA is still restricted and highly dependent on the human leukocyte antigen (HLA) haplotype of the patient.
- DC can also be fused with the patient’s tumor cells in vitro or pulsed with tumor cell lysates. The former method combines sustained tumor antigen expression with the antigenpresenting and immunostimulatory capacities of DC. DC-tumor cell hybrids will also stimulate an active antitumoral immune response.
- Tumor antigen can also be loaded on DC using plasmid DNA transfection or viral vector mediated gene transfer. The former method results in only low transfection efficiencies. On the other hand, viral transduction, for example by using adenoviral or lentiviral, vectors is very effective with regard to transfection efficiency. However, the immunogenic character of the viral vector itself is a serious disadvantage. In both cases, DC will transcribe and process the tumor antigen. This will result in a cytotoxic immune response, necessary for immunological defense against cancer cells.
- It is also possible to transfect DC using in vitro transcribed mRNA coding for tumor antigens or total tumor RNA. It has been shown that electroporation of RNA is the most effective nonviral transfection method for DC (Nonviral Vectors for Cancer Therapy). mRNA is brought directly into the cytoplasm and the cell’s metabolism will translate mRNA into proteins, which can be presented onto MHC class I molecules after processing. This will guarantee a specific cell-mediated antitumoral immune response.
In a clinical context, in vitro cultured and activated DC loaded with appropriate tumor antigens could be administered to cancer patients in a therapeutic setting (active specific immunotherapy). The aimed generation of anti-tumor immunity, mediated by DC, could be of importance for both treatment (as adjuvant to conventional therapies) and to prevent relapse in an MRD setting. On the other hand, tumor antigen-loaded DC can also be used for the ex vivo generation of tumorspecific cytotoxic T lymphocytes (CTL) in an autologous system. These tumor-specific CTL can, in their turn, be administered to the patient to exert a direct cytotoxic effect on the patient’s cancer cells (passive or adoptive immunotherapy).
The impact of a DC-based cancer vaccine is clear: an antigen-specific anti-tumor vaccine would influence both morbidity and mortality of various cancers. Currently, several phase I–II or III clinical trials using TAA-loaded DC are ongoing worldwide in order to stimulate the patient’s immune system against tumor antigens. A number of these trials demonstrated some clinical and immunological responses (as evidenced by T cell proliferation, IFN-g ELISPOT, and delayed type hypersensitivity [DTH] reaction) without any significant toxicity. However, despite the presence of expanded antigen-specific T cells in patients after vaccination, only a minor population of these patients showed a beneficial biologically relevant clinical response, i.e., tumor regression and increased disease-free survival. Clinical trials using DC have shown moderate, success. To date, the combination of a targeted therapy exploiting the capacity of DC. To Stimulate the patient’s own immune system against cancer with so-called immune checkpoint inhibitors is being examined in ongoing and future trials in order to eliminate tumor burden in patients.
Dendritic cells in cancers
DC can also infiltrate human tumors where they are involved in the induction of anti-tumor immune responses. It is likely that the establishment of tumor-specific immune responses depends on the migratory capacity of DC from the tumor microenvironment to the draining lymph nodes, where tumor antigen presentation to T cells takes place. Moreover, by their expression of costimulatory molecules and several cytokines, such as IFN-α and IL-12, DC also mediate T cell survival by preventing T cell apoptosis.
In addition, mature DC have been reported to cause direct lysis, apoptosis, as well as cell cycle arrest of cancer cells through the secretion of soluble factors. As a consequence, the presence of a high number of DC in the tumoral or peritumoral area, as well as in the draining lymph nodes of various human tumors, has been shown to correlate with patients’ survival and a better prognosis. Decreased numbers or dysfunction (e.g., decreased expression of costimulatory molecules) of DC is reported in poor-prognosis tumors. Furthermore, tumor cells can secrete certain factors (e.g., IL-10 and TGF-β) that counteract DC maturation and migration and thus actively contribute to DC dysfunction.
Occasionally, neoplasms of accessory immune cells (antigen-presenting cells, dendritic cells) can occur. These are primarily found in lymph nodes and extranodal lymphoid tissues (lymph node interdigitating cell sarcoma), but are also reported from other sites such as the skin (Langerhans Cell Histiocytosis). The incidence of dendritic cell tumors is very rare: until now, only a few dozens of cases have been reported in literature.
Banchereau J, Steinman RM (1998) Dendritic cells and the control of immunity. Nature 392:245–252
Gilboa E (2007) DC-based cancer vaccines. J Clin Invest 117:1195–1203
Lotze MT, Thomson AW (2001) Dendritic cells, 2nd edn. Academic, London
Ponsaerts P, Van Tendeloo VF, Berneman ZN (2003) Cancer immunotherapy using RNA-loaded dendritic cells. Clin Exp Immunol 134:378–384
Van Tendeloo VF, Van Broeckhoven C, Berneman ZW (2001) Gene-based cancer vaccines: an ex vivo approach. Leukemia 15:545–558