Resurrecting the Dead: Dendritic Cells Cross-Present Apoptotic Cells
A novel pathway for the activation of tumor-specific killer T cells

Matthew L. Albert

"To examine the causes of life, we must first have recourse to death."
Mary Shelley, Frankenstein

Like most Sunday nights, Ms. S, a 53-year-old woman, got into bed with a good book. However, the next morning, she awoke feeling dizzy. Over the next 4 days, new symptoms came on with startling rapidity. Her speech became slurred. She became unable to walk without falling over, forcing her to sit in chairs, until she became unable even to sit up. The dizziness worsened and she experienced a constant feeling of nausea. What she did not know was that her neurologic symptoms were the result of her immune system attacking a tiny cluster of cancer cells originating from her breast.

Immunologists have puzzled, since the discovery of B cells and T cells, over whether it was possible to direct an immune response against tumors. The discovery of a rare set of neurologic degenerations, such as the cerebellar degeneration experienced by Ms. S, has offered the possibility for exploring naturally occurring tumor immunity. It is known that the autoimmune attack against Purkinje neurons in Ms. S, was initiated by an immune system response targeting her cancer (1). As is often the case in medicine, answers to seemingly obscure questions can open new windows on generally important issues. Studying the cause of Paraneoplastic ("alongside a cancer") Neuronal ("brain") Degenerations (PNDs) has provided a wealth of information about tumor immunology. My thesis work focused on understanding the pathogenesis of the PNDs and demonstrated that killer T cells are involved in targeting human tumors. Additionally, we described a mechanism for antigen presentation that allows killer T cells to be activated to tumor-specific proteins.

To study naturally occurring tumor immunity in PND patients, we first had to find a mechanism for the capture and presention of tumor antigen by antigen-presenting cells (APCs). It was previously known that cytotoxic T lymphocytes (CTLs) are important components of the adaptive immune response, responsible for destroying virally infected cells and considered critical for the eradication of cells undergoing malignant transformation (2). To become an effector cell and perform these tasks, CTLs must first be activated by APC expressing MHC I/peptide complexes. Classically, only antigen that is synthesized endogenously was thought to access the MHC I presentation pathway (3). This belief raised the following question: How are CTLs reactive to tumor antigens activated de novo, for while tumor cells express MHC I/peptide complex ("signal 1"), they lack costimulatory molecules ("signal 2") required for the activation of naïve T cells (3, 4).

One answer to this question was offered by the in vivo phenomenon of "cross-priming." Originally described by Dr. Michael Bevan in 1976 (5), this indirect pathway for generating CTLs to exogenous antigen was disregarded as artifact by others in the field. However, subsequent in vivo data established this pathway as critical for activating CTLs specific for tumor and viral antigens not expressed by hematopoetic cells (6, 7). Still, a physiologically relevant mechanism remained elusive.

We defined such a mechanism: (i) identifying two important components to this exogenous pathway, and (ii) revealing a novel signaling pathway responsible for antigen capture. We first identified the APC required for antigen cross-presentation, and the mechanism by which such antigen is acquired. We found that dendritic cells (DCs) efficiently present antigen derived from apoptotic cells, thus allowing for the induction of MHC I-restricted antigen-specific CTLs (8). Interestingly, only antigenic- material derived from apoptotic, but not from necrotic cells, can charge DCs in this manner. We have documented that immature DCs phagocytose the apoptotic cells and upon maturation, they cross-present antigen derived from the apoptotic corpse (8, 9). Moreover, we demonstrated that DCs are unique in their ability to cross-present antigen from apoptotic cells. While macrophages efficiently engulf apoptotic cells, they degrade, rather than cross-present the internalized antigen (see figure).

Dendritic cells utilize avb5 and CD36 for the phagocytosis and cross-presentation of antigen derived from apoptotic cells. Dendritic cells (DCs) phagocytose apoptotic cells and cross-present antigens derived from the dying cells to stimulate virus- and tumor-specific CD8+ T cells. Phagocytosis is restricted to the immature stage of DCs and is accompanied by the expression of the avb5 integrin and CD36. Although macrophages (Mp) efficiently phagocytose apoptotic cells via many receptors, including avb3 and CD36, they do not cross-present antigens to CD8+ T cells.

This striking distinction between the two APCs with respect to their handling of internalized apoptotic material was correlated with their use of distinct receptors for engagement and internalization of apoptotic cells. Specifically, we found that DCs employ the a_nb₅ integrin for recognition of dying cells (9). This finding distinguished DCs from macrophages, which lack the a_nb₅ receptor, and instead employ a_nb₃ for phagocytosis (10). We next mapped downstream signaling molecules responsible for a_nb₅-mediated phagocytosis, demonstrating a role for the CrkII/Dock180/Rac1 signaling complex in phagosome formation (11). In addition to defining integrin-receptor signaling, these results constituted evidence that the CED-2/CED-5/CED-10 complex defined in Caenorhabditis elegans (12) is functionally analogous to the CrkII/Dock180/Rac1 molecular complex in mammalian cells. By linking the a_nb₅ receptor to this molecular switch, we revealed an evolutionarily conserved signaling pathway that is responsible for recognition and internalization of apoptotic cells by both professional and nonprofessional phagocytes. Subsequent work has illustrated that while DCs utilize Dock180, macrophages employ a distinct isoform of this scaffolding protein. Furthermore, these distinct signaling cascades seem to result in the unique trafficking of internalized material--to an antigen-processing endosome in DCs and to the lysosome in macrophages--thus offering an explanation for the distinct fate of apoptotic material captured by these two cell types (13).

While defining the apoptosis-dependant pathway for T cell activation, we returned to the bedside of Ms. S and other patients with PCD. Seeking confirmation of the physiologic relevance of this pathway, we examined the peripheral blood of patients for evidence of cellular immunity in their disease pathogenesis and, indeed, we observed cdr2-specific CTLs in five out of five patients with PCD (14). Additionally, we demonstrated that cross-presentation of apoptotic tumor cells expressing the PCD antigen, resulted in activation of potent tumor-specific CTLs. Based on our studies, we propose a physiologically relevant mechanism whereby immature dendritic cells may engulf apoptotic cells via the a_nb₅ integrin, generate MHC I/peptide complexes, and induce a CTL response specific for tissue-restricted antigens (e.g., tumor proteins). The identification of the cellular and molecular events responsible for antigen cross-presentation has already resulted in new therapeutic approaches aimed at activating the immune system against systemic tumors.

References

1. R. B. Darnell, Proc. Natl. Acad. Sci. U.S.A 93, 4529 (1996).
2. M. T. Heemels, H. Ploegh, Annu. Rev. Biochem. 64, 463 (1995).
3. E. Pamer, P. Cresswell, Annu. Rev. Immunol. 16, 323 (1998).
4. J. Banchereau, R. M. Steinman, Nature 392, 245 (1998).
5. M. J. Bevan, J. Exp. Med. 143, 1283 (1976).
6. F. R. Carbone, M. J. Bevan, J. Exp. Med. 171, 377 (1990).
7. A. Y. Huang et al., Science 264, 961 (1994).
8. M. L. Albert, B. Sauter, N. Bhardwaj, Nature 392, 86 (1998).
9. M. L. Albert et al., J. Exp. Med. 188, 1359 (1998).
10. J. Savill, I. Dransfield, N. Hogg, C. Haslett, Nature 343, 170 (1990).
11. M. L. Albert, J. I. Kim, R. B. Birge, Nature Cell Biol. 2, 899 (2000).
12. P. W. Reddien, H. R. Horvitz, Nature Cell Biol. 2, 131 (2000).
13. M. L. Albert, in Dendritic Cells: Biology and Clinical Applications, M. Lotze, Ed. (Academic Press, in press).
14. M. L. Albert et al., Nature Med. 4, 1321 (1998).

To Advertise   |   Find Products