Utilization of adipocyte-derived lipids and enhanced intracellular trafficking of fatty acids contribute to breast cancer progression

Abstract

Background: To determine whether adipocyte-derived lipids could be transferred into breast cancer cells and investigate the underlying mechanisms of subsequent lipolysis and fatty acid trafficking in breast cancer cells.

Methods: A Transwell co-culture system was used in which human breast cancer cells were cultured in the absence or presence of differentiated murine 3 T3-L1 adipocytes. Migration/invasion and proliferation abilities were compared between breast cancer cells that were cultivated alone and those co-cultivated with mature adipocytes. The ability of lipolysis in breast cancer cells were measured, as well as the expression of the rate-limiting lipase ATGL and fatty acid transporter FABP5. ATGL and FABP5 were then ablated to investigate their impact on the aggressiveness of breast cancer cells that were surrounded by adipocytes. Further, immunohistochemistry was performed to detect differential expression of ATGL and FABP5 in breast cancer tissue sections.

Results: The migration and invasion abilities of cancer cells were significantly enhanced after co-culture with adipocytes, accompanied by elevated lipolysis and expression of ATGL and FABP5. Abrogation of ATGL and FABP5 sharply attenuated the malignancy of co-cultivated breast cancer cells. However, this phenomenon was not observed if a lipid emulsion was added to the culture medium to substitute for adipocytes. Furthermore, epithelial-mesenchymal transaction was induced in co-cultivated breast cancer cells. That may partially due to the stimulation of PPARβ/δ and MAPK, which was resulted from upregulation of FABP5. As evidenced by immunohistochemistry, ATGL and FABP5 also had higher expression levels at the invasive front of the breast tumor, in where the adipocytes abound, compared to the central area in tissue specimens.

Conclusions: Lipid originating from tumor-surrounding adipocytes could be transferred into breast cancer cells. Adipocyte-cancer cell crosstalk rather than lipids alone induced upregulation of lipases and fatty acid transport protein in cancer cells to utilize stored lipids for tumor progression. The increased expression of the key lipase ATGL and intracellular fatty acid trafficking protein FABP5 played crucial roles in this process via fueling or signaling.

Keywords: ATGL; Adipocyte; Aggressiveness; Breast cancer; Crosstalk; FABP5.

Fig. 1

Lipid transfer during co-culture and co-cultivated breast cancer cells increased aggressiveness. a Lipid accumulation in cancer cells shown by Bodipy staining (lipids in green and nuclei in blue; scale bar, 50 μm), NC, non-co-culture; Coc, co-culture. b TG content in SK-BR-3 (left) and SUM159PT (right) cells cultured alone (NC) or with mature adipocytes (Coc) for 3 days. c Non-co-cultivated (NC) and co-cultivated (Coc) SK-BR-3 (left) and SUM159PT (right) cells migrated/invaded to the outer surface of the Transwell chamber; the migration times were 24 h and 6 h, respectively. Five fields were randomly taken for each chamber, and the representative migration images are shown. d Comparison of the proliferative ability of non-co-cultivated (NC) and co-cultivated (Coc) SUM159PT cells. Fold change comparison of the cell optical density during 48 h of SUM159PT cell non-co-cultivation (NC) or co-cultivation (Coc) with mature adipocytes for 3 days (n = 3). Representative results from at least three independent experiments are shown. *p < 0.05, **p < 0.01

Fig. 2

Co-culture with adipocytes induces ATGL-dependent lipolysis in breast cancer cells. a Expression of ATGL in mammary epithelial cell lines using western blotting. b After co-culture with mature adipocytes for 3 days, expression of ATGL in SK-BR-3 and SUM159PT cells was obviously increased. c Removing adipocytes and depriving serum from the medium led to lipolysis in cancer cells, and the TG content in tumor cells was detected at 0, 4 and 24 h after release, while free glycerol in the medium was detected at 4 and 24 h. d Down-regulation of ATGL in SUM159PT cells by shRAN plasmid transfection and validation of the effects using western blotting. Proliferative ability was compared between shCtrl and shATGL stably expressed SUM159PT cells (left panel). Transwell assays conducted to elucidate the impact of ATGL blockade on the migration ability of cancer cells. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, “ns” stands for not significant

Fig. 3

Elevated expression of FABP5 induce EMT inbreast cancer cells. a Expression of FABP5 in different mammary epithelial cell lines (left panel) and in non-co-cultivated (NC) and co-cultivated (Coc) breast cancer cells (SUM159PT and SK-BR-3, right panel). b Detection of E-cadherin in the SK-BR-3 breast cancer cells by immunofluorescence analysis (upper panel) and some molecules that may contribute to the enhance malignancy of breast cancers by western blot (lower panel). c Representative immunohistochemical staining of FABP5 in normal breast and tumor tissues

Fig. 4

In the presence of adipocytes, accelerated FA trafficking in breast cancer cells was mediated by FABP5. a After ablation of FABP5 using an siRNA kit, cells were harvested 48 h after transfection for western blotting to the validate knockdown effectiveness. b Impact on the proliferative ability of breast cancer cell lines resulting from FABP5knockdown. c Altered migration ability resulting from FABP5 deletion in non-cultivated (NC) and co-cultivated (Coc) cancer cells. d TG storage in breast cancer cells, with FABP5 expressed or ablated, cultivated in the absence or presence of adipocytes. *p < 0.05, **p < 0.01, ****p < 0.0001, “ns” stands for not significant

Fig. 5

Exogenous lipids in the absence of adipocytes could not promote the malignant behavior of cancer cells. Mature adipocytes were replaced by 0.02% Intralipid in 3 day co-culture with SUM159PT breast cancer cells. a Bodipy staining was performed on SUM159PT cells grown alone (NC) and in medium supplemented with Intralipid. (lipids are shown in green, and nuclei are shown in blue; scale bar, 50 μm). b The TG content in the two cell populations (NC vs. Intralipid) was measured. c-d Proliferation and migration assays were employed to compare the malignant properties of non-treated cells (NC) and Intralipid-treated cells. e SUM159PT cells were not treated (NC) or treated with Intralipid for 3 days and then released similar to co-culture. Cells were collected at 0, 4 and 24 h for TG measurement, and medium was collected at 4 and 24 h for free glycerol measurement. f Differential expression of lipases, pAMPK/AMPK and FABP5 in breast cancer cells (SUM159PT) grown in Intralipid or with adipocytes. ***p < 0.001, “ns” means not significant