Role of MIP1a in Anti-Tumor Immunity and Therapy
By Maja Maric, PhD
Specific immune responses are usually preceded by local inflammation. During an anti-tumor immune response, T cells and antigen presenting cells (APCs), in addition to other leukocytes, are recruited into the tumor. It is clear that T cell effector response requires inflammation, which brings leukocytes into contact with tumor cells. What causes the selective leukocyte recruitment into solid tumor? The molecular basis for this selective recruitment in tumor is still being explored, but a family of proteins called chemokines are emerging as major players in these processes. Therefore, our ability to manipulate the recruitment of a specific subset of leukocytes into tumor may enhance the anti-tumor immune response.
It is widely accepted that the most important effector cells in tumor immunity are activated T cells, especially CD8+ subset, because most somatic tissues express MHC class I but not class II molecules. However, their arrival into the tumor is usually preceded by cells of innate immunity like macrophages, neutrophils, and natural killer (NK) cells, which also have the ability to kill tumor cells. How do all these immune cell subsets find their way to a site on the tumor in an ordered sequence? Our immune system is equipped with a "911 system" that brings all potential help to the area that needs the help most. A very important part of this "911 system" is made of a family of proteins called chemokines. Among the four known families, a-, b-, g-, and d-family, potentially the most interesting family for tumor immunotherapy is b-family, whose members show ability to selectively attract monocytes, eosinophils, and lymphocytes. One of the most explored and interesting members of this group is macrophage inflammatory protein 1a (MIP1a), known also as LD78. MIP1a is involved in inducing chemotaxis and inflammatory responses, and in the homeostatic control of stem cell proliferation. It is produced by a variety of cells: macrophages, neutrophils, T cells (preferentially CD8+ subset), fibroblasts, endothelial cells, B cells, and NK cells.1,2 MIP1a acts through interaction with its receptor, CCR5 (also a co-receptor for HIV entry). This chemokine has been detected at locally increased levels in various pathological states from injuries to autoimmune disorders like rheumatoid arthritis, tumors, infection with intracellular parasites, or viruses.3-6
It has been shown that MIP1a can selectively recruit CD8+ T cells and macrophages, and that MIP1a-/- mice have significantly reduced inflammatory response to influenza virus infection.7-9 In vivo studies using mouse plasmacytoma with the defined tumor antigen (P1A) as tumor model have shown that local expression of MIP1a results in strong inflammation of leukocytes in tumors and leads also to the induction of strong anti-tumor immune response.10
When plasmacytoma was manipulated to express different levels of MIP1a, increased infiltration of APCs (macrophages, B cells, and dendritic cells) was detected without alteration of their composition in infiltrate. When tumor-infiltrating leukocytes (TILs) were isolated from MIP1a secreting tumor and used in cytotoxic assay, TILs from MIP1a secreting tumor were 3- to 30-fold more efficient in lysis of target cells than TILs from the tumor that did not express MIP1a. However, despite the drastic difference in CTL response there was no difference in the rate of tumor growth between tumors secreting MIP1a and tumors that were not secreting MIP1a. This result suggests that the production of cytotoxic lymphocytes within tumor is not sufficient to cause tumor rejection. In ex vivo CTL assay targets expressing costimulatory molecule B7 were more efficiently lysed than targets without costimulatory molecule. This leads us to the question of whether costimulatory molecules B7-1 and B7-2, expressed by host APCs, can play a role in the induction of CTLs.
In the same study, two groups of mice were injected with MIP1a-secreting tumors and treated with either PBS or anti-B7-1 and anti-B7-2 monoclonal antibodies, and cytotoxic activity was measured at different time points. In mice treated with anti-B7 antibodies, tumor antigen-specific cytotoxic activity was significantly abrogated while the control group developed significant cytotoxicity. These experiments demonstrate that local/tumor expression of MIP1a can induce strong CTL response against tumor antigen without further in vitro restimulation. The fact that in vivo tumors are not rejected could be explained by the requirement for B7 costimulatory molecules directly on tumor cells and/or still undefined tumor factors (related or unrelated to MIP1a) with ability to suppress anti-tumor CTL activity. This is plausible, because tumor antigens are, in most cases, self-antigens and self-reactive T cells would be potentially harmful for host organisms. If both antigen and costimulatory molecules are required at the effector phase of T cells, it suggests that tumor CTLs are not fully activated.
Cell Cycle Arrest and MIPIa
Another potentially useful function of MIP1a in tumor immunotherapy at a more systemic level is its ability induce cell cycle arrest in immature hematopoietic progenitors, and it could, therefore, be used to reduce the hematologic toxicity of cell cycle active therapy. Recently, a genetically engineered analog of MIP1a, BB-10010, has been used in several studies on mice and in clinical trial.11-13 This variant of MIP1a has single amino acid substitution of Asp26>Ala that confers a reduced tendency of this molecule to form large, less active polymers at physiologic pH and ionic strength.14 This variant of MIP1a has been tested in several studies for potential protective effects in chemotherapy-induced neutropenia, which is a major dose-limiting factor in chemotherapy. Most chemotherapeutic reagents are active against proliferating cells by interfering with DNA replication and mitosis. BB-10010 shows potential protective abilities because it reduces accumulated hematopoietic stem cell damage following repeated non-cell cycle specific cytotoxic insults.11 In another study, it also reduced toxicity of three different cytotoxic drugs: cyclophosphamide, 5-fluorouracil, and cytosine arabinoside.12 In a phase I study with cancer patients and healthy volunteers, doses of BB-10010 from 10-300 mg/kg (given IV) were well tolerated, although it caused acute, short-lived monocytopenia.13 It remains to be seen whether further clinical studies will confirm usefulness of MIP1a variant BB-10010 in chemotherapy.
Animal and human studies with MIP1a suggest that this chemokine and/or its analog BB-10010 may have two important roles in the fight against cancer. When expressed locally in tumor tissue, it attracts effector cells and acts as an additional costimulatory factor in activation of CD8+ T cells. Also, when distributed systemically it may protect hematopoietic stem cells from damage from chemotherapy. The whole picture is more complicated because in animal studies described here, the sole action of MIP1a is not sufficient to induce such a vigorous cytotoxic response in vivo that it would be expected to cause immediate tumor rejection. Rather, when working in conjunction with other costimulatory molecules like B7, its action is more effective. It should also be kept in mind that chemokines in general bind promiscuously to a family of receptors and that such complicated interplay of many factors is easy to throw out of balance and difficult to control. By focusing on one chemokine, we cannot exclude the possible roles of others that bind and compete for the same receptor.
Further studies are necessary to explore more details and additional factors that may alter or control the role of MIP1a as a locally expressed factor in the tumor immunotherapy. As a protective agent from negative effects of chemotherapy, BB-10010 is certainly promising, and perhaps more agents like that should be defined while the scientific community searches for less toxic ways of fighting the cancer. (Dr. Maric is a Post-Doctoral Fellow, Immunobiology Section, Yale School of Medicine, New Haven, CT.)
1. Cook DN, Smithies O, Strieter RM, et al. CD8+ T cells are a biologically relevant source of MIP1a in vivo. J Immunol 1999;162:5423-5428.
2. Krzysiek R, Lefevre EA, Zou W, et al. Antigen receptor engagement selectively induces MIP1a and MIP1b chemokine production in human B cells. J Immunol 1999;162:4455-4463.
3. Hatano Y, Kasama K, Iwabuchi H, et al. MIP1a expression by synovial fluid neutrophils in rheumatoid arthritis. Ann Rheum Dis 1999;58:297-302.
4. Maric M, Chen L, Sherry B, et al. A mechanism for selective recruitment of CD8 T cells into B7-1-transfected plasmacytoma: Role of MIP1a. J Immunol 1997;159:360-368.
5. Bliss SK, Marshall AJ, Zhang Y, et al. Human polymorphonuclear leukocytes produce IL-12, TNF-alpha, and the chemokines MIP1a and MIP1b in response to Toxoplasma gondii antigens. J Immunol 1999;162: 7369-7375.
6. McColl SR, Roberge CJ, Larochelle B, et al. EBV induces the production and release of IL8 and MIP1a in human neutrophils. J Immunol 1997;159:6164.
7. Taub DD, Konlon K, Lloyd AR, et al. Preferential migration of activated CD4 and CD8 T cells in response to MIP1a and MIP1a. Science 1993;260: 355-358.
8. Schall TJ, Bacon K, Camp RD, et al. Human MIP1a andMIP1b chemokines attract distinct populations of lymphocytes. J Exp Med 1993;177:1821-1826.
9. Cook DN, Beck MA, Coffman TM, et al. Requirement of MIP1a for an inflammatory response to viral infection. Science 1995;269:1583-1585.
10. Maric M, Liu Y. Strong cytotoxic T lymphocyte responses to a MIP1a-expressing tumor: Linkage between inflammation and specific immunity. Cancer Res 1999;59:5549-5553.
11. Lord BI, Marshall E, Woolford LB, et al. BB-10010/MIP1a in vivo maintains haemopoietic recovery following repeated cycles of sublethal irradiation. Br J Cancer 1996;74:1017-1022.
12. Gilmore GL, DePasquale DK, Shaduck RK. Protective effects of BB-10010 treatment on chemotherapy-induced neutropenia in mice. Exp Hematol 1999;27:195-202.
13. Marshall E, Howell AH, Powles R, et al. Clinical effects of human MIP1a (LD78) administration to humans: A phase I study in cancer patients and normal healthy volunteers with the genetically engineered variant, BB-10010. Eur J Cancer 1998;34:1023-1029.
14. Hunter MG, Bawden L, Brotherton D, et al. BB-10010: An active variant of human MIP1a with improved pharmaceutical properties. Blood 1995;86:440-448.