An Unrecognized Role Of Ccl5 In Triple Negative Breast Tumor Construction
Elizabeth Brett, Matthias Sauter, Hans-Guenther Machens, Dominik Duscher.
Technical University of Munich, Munich, Germany.
PURPOSE: CCL5 (RANTES) has long been identified as a driver for triple negative breast cancer (TNBC). Numerous studies demonstrate that silencing CCL5 decreases invasion, metastasis, and migration. Similarly, the phenomenon of linearized collagen extending from the tumor body, has been well identified. We hypothesize there is an unrecognized role of CCL5 in depositing linear collagen type VI, identifiable by a juxtacrine relationship between triple negative breast cancer cells and healthy cells.
METHODS: Group i (MDA-MB-231, ASCs, and fibroblasts in juxtacrine contact), group ii (ASCs and fibroblasts juxtacrine, MDA-MB-231 in paracrine contact through a 0.2m semipermeable membrane), and group iii (ASCs and fibroblasts in juxtacrine contact) were set up and cultured in normal conditions for 7 days. During this time ECM was deposited, followed by decellularization using NH4OH and Triton-X. ECMs were analyzed using brightfield images, immunohistochemistry, SEM, AFM, and mass spectroscopy. An 86-cytokine panel (Ray Biotech) screened cytokines in the media during ECM deposition (day 5 of culture). Cell response to ECM was measured by reseeding MDA-MB-231, and cultured under normal conditions. Immunohistochemistry on human breast tumor sections targeting collagen VI and CCL5 was performed. As an intervention study, monoclonal antibodies against human CCL5, and recombinant human CCL5 were used to block and supply CCL5 in vitro respectively.
RESULTS: Co-culture where all three cell types had juxtacrine contact created ECM linear in format (per SEM and AFM), high in collagen VI (per mass spectroscopy), and was produced under high levels of CCL5 (per cytokine panel). Human breast tumor samples show positive staining for collagen VI and CCL5. These data are in contrast to the healthy cell co-culture, which had rough ECM ultrastructure, lower collagen VI, and was produced in low CCL5 concentration. When CCL5 was blocked in vitro during matrix deposition, the linear matrix formation was disrupted, and collagen VI production was significantly decreased. Furthermore, reseeded cells on the matrix formed without CCL5 had less integrin 1 compared to the matrix formed with unlimited CCL5.
CONCLUSION: The TNBC co-culture of MDA-MB-231, ASCs, and fibroblasts in juxtacrine contact, reproducibly builds a matrix which high in collagen type VI, linear in format, encourages formation of road-like structures by reseeded cells, and induces high integrin B1 expression of the reseeded cells. This is a yet undescribed role of CCL5, which may help explain the abundant literature on the oncologic benefits of its inhibition.
Back to 2020 Abstracts