MPA alters metabolic phenotype of endometrial cancer-associated fibroblasts from obese women via IRS2 signaling

Abstract

Obese women have a higher risk of developing endometrial cancer (EC) than lean women. Besides affecting EC progression, obesity also affects sensitivity of patients to treatment including medroxprogesterone acetate (MPA). Obese women have a lower response to MPA with an increased risk for tumor recurrence. While MPA inhibits the growth of normal fibroblasts, human endometrial cancer-associated fibroblasts (CAFs) were reported to be less responsive to MPA. However, it is still unknown how CAFs from obese women respond to progesterone. CAFs from the EC tissues of obese (CO) and non-obese (CN) women were established as primary cell models. MPA increased cell proliferation and downregulated stromal differentiation genes, including BMP2 in CO than in CN. Induction of IRS2 (a BMP2 regulator) mRNA expression by MPA led to activation of glucose metabolism in CO, with evidence of greater mRNA levels of GLUT6, GAPDH, PKM2, LDHA, and increased in GAPDH enzymatic activity. Concomitantly, MPA increased the mRNA expression of a fatty acid transporter, CD36 and lipid droplet formation in CO. MPA-mediated increase in glucose metabolism genes in CO was reversed with a progesterone receptor inhibitor, mifepristone (RU486), leading to a decreased proliferation. Our data suggests that PR signaling is aberrantly activated by MPA in CAFs isolated from endometrial tissues of obese women, leading to activation of IRS2 and glucose metabolism, which may lead to lower response and sensitivity to progesterone in obese women.

Fig 1. MPA induced cell proliferation in CAFs of obese women by decreasing the expression of stromal differentiation genes.

(A) CAFs from non-obese (CN) and obese (CO) women were treated with 10 and 1000 nM MPA for 24 h, before measuring their proliferation using BrdU cell proliferation assay. The data was normalized to vehicle and the means of the data were shown. (B) The heatmap shows the expression of stromal differentiation genes. The bottom of the heatmap is labeled with cells CN1-4 and CO1-4, while the row is labeled with genes. CN and CO were treated with 10 nM MPA for 4 hours, before RNA isolation for quantitative real-time PCR analysis of stromal differentiation genes. The data was normalized to GAPDH followed by the value from cells treated with the vehicle. (C) The mean expression values of CN and CO from (B) were shown. Bar graphs represent mean ± SEM. **P<0.01, ***P<0.001, comparing MPA to vehicle. #P<0.05, ##P<0.01, comparing CO to CN. Data shown are representative of three independent experiments.

Fig 2. MPA induced IRS2 and increased glucose metabolism in CAFs of obese women.

(A) Diagram highlighting genes involved in glycolytic and fatty acid metabolism. (B) The heatmap shows the expression of metabolic genes in CN and CO after 6 h treatment of MPA. The bottom of the heatmap is labeled with cells CN1-4 and CO1-4, while the row is labeled with genes. (C) The mean expression values of IRS2 and glycolytic genes for CN1-4 and CO1-4. The data was normalized to housekeeping gene 18S rRNA. (D) GAPDH enzymatic activity in CN1-4 and CO1-4 were measured with GAPDH assay kit 24 h post MPA treatment, and normalized to the protein concentration of the samples. V: vehicle, M: MPA; *P<0.05, **P<0.01, comparing MPA treatment to vehicle for CN1-4 and CO1-4. Data represent the mean ± SEM., and are representative of three independent experiments.

Fig 3. MPA increased expression of CD36 and lipid droplet formation in CAFs of obese women.

(A) The heatmap shows the expression of metabolic genes in CN and CO after 6 h treatment of MPA, and the mean expression values of genes involved in fatty acid metabolism in CN1-4 and CO1-4. The data was normalized to housekeeping gene 18S rRNA. (B) Fatty acid oxidation (FAO) activity in CN1-4 and CO1-4 was measured with FAO assay kit 24 h post MPA treatment, and normalized to the protein concentration of the samples. Data shown are mean ± SEM of duplicates. (C) Lipid droplet formation was measured in CN (2, 4) and CO (1, 4) after treatment with 10 nM MPA for 24 h. The coverslips were stained with Oil Red O dye and counterstained with haematoxylin, before analyzing with light microscopy at 40X magnification. Scale, 200 μM. Images were analyzed using ImageJ software. Representative images were shown for each CN and CO. (D) Bar graphs showing quantification of lipid droplet formation. Data shown are the mean ± SEM of three different spots in a coverslip. V: vehicle, M: MPA; **P<0.01, ***P<0.001, comparing MPA treatment to vehicle. #P<0.05, ##P<0.01 comparing CO to CN.

Fig 4. Activation of genes related to glucose and fatty acid metabolism is progesterone receptor-dependent.

Gene expression of metabolic genes IRS2, GLUT6, GAPDH and CD36 in CN1-4 and CO1-4 after treatment with 10 nM MPA and 10 μM RU486 (A-D) for 6 h. (E) IRS2 protein level in CN and CO 24 h after treatment in vehicle (V), 10 nM MPA (M), and MPA+10 μM RU486 (M+RU) treated cells, evaluated using Western blotting. (F) CN and CO were treated with 10 nM MPA, in combination with PR inhibitor, RU486 for 24 h, before measuring their proliferation using BrdU cell proliferation assay. The data was normalized to vehicle and the means of the data were shown. Bar graphs represent mean ± SEM. NS: non-significant. *P<0.05, **P<0.01, ***P<0.001. Data shown are representative of three independent experiments.

Fig 5. MPA-induced IRS2 mediates activation of genes related to glucose and fatty acid metabolism.

Gene expression of GLUT6, GAPDH and CD36 in CN1-4 and CO1-4 after treatment with 10 nM MPA and 1 μM NT157 (A-C) for 6 h. (D) CN and CO were treated with 10 nM MPA, in combination with IRS2 inhibitor, NT157 for 24 h, before measuring their proliferation using BrdU cell proliferation assay. The data was normalized to vehicle and the means of the data were shown. Bar graphs represent mean ± SEM. NS: non-significant. *P<0.05, **P<0.01. Data shown are representative of three independent experiments.