The role of immunosuppressive cells in metastasis

National Institute On Aging

Investigators

Linked publications & trials

Abstract

Here we have hypothesized that breast cancer cells utilize a migratory pathway of immune cells, such as CCR4-expressing T cells, to metastasize into inflamed lungs producing TARC and MDC. Interestingly, primary tumor growing at distant site induced production of TARC/CCL17 (a chemokine ligand for CCR4) in lungs of mice presumably to facilitate recruitment of CCR4-expressing tumor cells. Thus, to demonstrate this, we have asked question whether 4T1 tumor cells express and utilize CCR4 for lung metastasis. However, our flow cytometry analysis has only detected a weak expression of CCR4 on the surface of 4T1 cells, suggesting that a proportion of the cells may indeed express the receptor. In support, the cells were also able to chemotax in vitro to TARC, indicating that the receptor is functionally active. To confirm this, 4T1 cells were treated with CCR4-targeting chemotoxin, a novel formulation developed by us to specifically kill cells through unique chemokine receptors. As a result, the treatment generated resistant to chemotoxin cells (designated 4T1wt-PE and 4T1.2-PE). Unlike parental 4T1 tumor cells, 4T1wt-PE and 4T1.2-PE cells did not chemotax to TARC and could not metastasize in to lungs. Our data have clearly demonstrated that CCR4 was expressed on proportion of 4T1 tumor cells, and only CCR4-expressing cells metastasized into lungs. However and surprisingly, this inherent capability alone was not sufficient to establish lung metastasis. Lung metastasis also required an active participation of CCR4+ Tregs. We have found that Tregs facilitated lung metastasis by regulating or even killing NK cells. This presumably explains our finding that NK cell counts were significantly reduced in peripheral blood (PB) of both tumor-bearing mice and human patients with an advanced stage IV breast cancer. Taken together, we propose that CCR4 is a lung metastasis Zip code utilized for dissemination of tumors together with their protector immune cells, Tregs. Strategies that abrogate any part of this process, such as targeted inactivation of CCR4+ cells or direct depletion of Tregs, would be expected to improve the outcome of the disease. This in turn would activate both antitumor innate and adaptive immune responses through activation of NK cells and cytolytic T cells. Indeed, the treatment of tumor -bearing mice with TARC-chemotoxin significantly reduced lung metastasis of 4T1 cells through depletion of Tregs and metastasizing tumors and activating NK cells.[unreadable] [unreadable] Our data clearly indicate the important role of Tregs in regulation of antitumor responses. Despite significant efforts, practically very little is known about mechanisms of suppressive activity exerted by Tregs. Recently, we have demonstrated that Tregs in human PBL consisted of at least two distinct subsets, memory-type CCR4+Tregs and nave-type CCR4- Tregs (Baatar et al., 2007). While freshly isolated CCR4+Tregs presumably represent natural Tregs and appear to be primed to readily suppress T cell proliferation, CCR4-Tregs require TCR-mediated activation to render them fully active. It was reported by others that activated Tregs could kill target cells, including CD4+Tcells, CD8+ T cells, utilizing both GZ-A and GZ-B. However, we did not detect GZ-A expression by either Treg subsets regardless of anti-CD3/CD28 or IL-2 stimulation. In our hands, GZ-B was not detected in non-activated T cells, and Tregs do not kill target T cells. In addition, upon activation, GZ-B was exclusively expressed in CCR4- T cells (both Tregs and non-Treg cells), but not in CCR4+Tregs, and did not correlate with their suppressive activities. Thus, GZ-B may not be a primary mechanism of regulation, although we cannot rule out its utilization by activated CCR4-Tregs, for example, when used at significantly higher cell concentrations. On the other hand, we have found that Tregs regulate T cell proliferation through a cell contact-dependent process involving FasL/Fas signaling. Thus, we have first hypothesized that Tregs may also regulate NK cells utilizing FasL/Fas signaling. Our study indicates that the process is independent of FasL/Fas signaling, as Tregs retained ability to regulate NK cells in the presence of neutralizing Fas antibodies. In contrast, we have found that Tregs express and utilize lectin-type proteins, such as beta-galactoside-binding protein (bGBP). bGBP and its dimeric lectin form Galectin-1 are immunosuppressive proteins expressed by activated immune cells, such as T cells, B cells and macrophages. Interestingly, bGBP actively participated in regulation of CD8+ T cells. It was used by Tregs to regulate TCR signaling of CD8+ T cells without induction of cell death, although bGBP is known to be cytotoxic for activated T cells. We have demonstrated that the mechanism of this process is in the capacity of bGBP to induce limited (non-processive) TCR signaling in target T cells; as it only activates Zap70, but not downstream molecules, such as ERK, Ras and PI3K. Although non-processive TCR signaling can lead to anergy, bGBP does not induce anergy of CD8+ T cells. This presumably allows Tregs to transiently prevent activation of CD8+ T cells by self-antigens, while keeping responses to xenogeneic antigens unaffected, indicating the important biological role of bGBP in the maintenance and control of peripheral tolerance.