Obesity-Altered Adipose Stem Cells Promote ER⁺ Breast Cancer Metastasis through Estrogen Independent Pathways

Abstract

Adipose stem cells (ASCs) play an essential role in tumor microenvironments. These cells are altered by obesity (obASCs) and previous studies have shown that obASCs secrete higher levels of leptin. Increased leptin, which upregulates estrogen receptor alpha (ERα) and aromatase, enhances estrogen bioavailability and signaling in estrogen receptor positive (ER⁺) breast cancer (BC) tumor growth and metastasis. In this study, we evaluate the effect of obASCs on ER⁺BC outside of the ERα signaling axis using breast cancer models with constitutively active ERα resulting from clinically relevant mutations (Y537S and D538G). We found that while obASCs promote tumor growth and proliferation, it occurs mostly through abrogated estrogen signaling when BC has constitutive ER activity. However, obASCs have a similar promotion of metastasis irrespective of ER status, demonstrating that obASC promotion of metastasis may not be completely estrogen dependent. We found that obASCs upregulate two genes in both ER wild type (WT) and ER mutant (MUT) BC: SERPINE1 and ABCB1. This study demonstrates that obASCs promote metastasis in ER WT and MUT xenografts and an ER MUT patient derived xenograft (PDX) model. However, obASCs promote tumor growth only in ER WT xenografts.

Keywords: adipose stem cells; breast cancer; estrogen receptor; metastasis; obesity.

Figure 1

obASCs promote metastasis but not tumor growth of constitutively active ERα xenograft models—MCF7-Y537S and WHIM20 PDX (A) Tumor volume was tracked over time. (Day of injection = Day 0). There is no change in tumor volume when BC was implanted in the presence of lnASCs or obASCs compared to BC alone except lnASCs compared to control WHIM20 tumor volume at day 60 (# −ln vs. ctrl p < 0.05). Caliper measurements were taken every three to four days until the tumor volume reached 750–1000 mm³. Values reported are the mean (n = 5 mice/group). Data were analyzed using two-way analysis of variance (ANOVA) and a Bonferroni post-test. (B) Area of the lung occupied by metastasis (metastatic index) was evaluated at the endpoint. Groups, where BC was implanted with obASCs, had higher levels of metastasis compared to BC alone or grown with lnASCs. Data were analyzed using one-way ANOVA and Tukey post-test. Bars, ± SEM. * p < 0.05, ** p < 0.01. (C) Circulating tumor cells were analyzed in animals harboring patient-derived xenograft (WHIM20) at endpoint using flow cytometry. There was no change in human (HLA1⁺) cells across groups; however, analysis of circulating tumor cells enriched for the cancer stem cell marker CD44⁺CD24⁻ was increased in PDX+obASCs compared to PDX alone. Data were analyzed using one-way ANOVA and Tukey post-test and no significant difference was found. Bars, ± SEM.

Figure 2

In an estrogen-depleted environment, obASCs promote proliferation and migration of ER WT and ER MUT BCCs in vitro. (A) Conditioned media collected from ASCs after 24 h promotes proliferation of ER WT (MCF7) and ER MUT with constitutively active ERα (MCF7-Y537S and PDX-derived) cells. Data were analyzed using two-way ANOVA and Bonferroni post-test. *** p < 0.001 (B) obASCs promotes increased migration of MCF7, MCF7-Y537S and PDX derived cells (WHIM43) through a 0.4 um membrane. Scale bar represents 100 μm. Data were analyzed using one-way ANOVA and Tukey post-test. Bars, ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001. Values reported are the mean of three independent experiments each performed in triplicate.

Figure 3

PCR array demonstrates that obASCs upregulate two common genes in ER WT and ER MUT cells. PCR array demonstrates that obASCs upregulate above a cutoff of 2x: 22 BC related genes in ER WT (MCF7) cells compared to 5 genes ER MUT (WHIM43) and downregulated 9 BC related genes in ER WT compared to 4 in ER MUT. Two genes were upregulated in both ER WT and ER MUT: SERPINE1 and ABCB1. PCR arrays were conducted (n = 1) for each condition.