Effects of adiponectin on breast cancer cell growth and signaling

Abstract

Obesity is a risk factor for postmenopausal breast cancer. Adiponectin/Acrp30 is lower in obese individuals and may be negatively regulating breast cancer growth. Here we determined that five breast cancer cell lines, MDA-MB-231, MDA-MB-361, MCF-7, T47D, and SK-BR-3, expressed one or both of the Acrp30 receptors. In addition, we found that the addition of Acrp30 to MCF-7, T47D, and SK-BR-3 cell lines inhibited growth. Oestrogen receptor (ER) positive MCF-7 and T47D cells were inhibited at lower Acrp30 concentrations than ER-negative SK-BR-3 cells. Growth inhibition may be related to apoptosis since PARP cleavage was increased by Acrp30 in the ER-positive cell lines. To investigate the role of ER in the response of breast cancer cells to Acrp30, we established the MDA-ERalpha7 cell line by insertion of ER-alpha into ER-alpha-negative MDA-MB-231 cells. This line readily formed tumours in athymic mice and was responsive to oestradiol in vivo. In vitro, MDA-ERalpha7 cells were growth inhibited by globular Acrp30 while the parental cells were not. This inhibition appeared to be due to blockage of JNK2 signalling. These results provide information on how obesity may influence breast cancer cell proliferation and establish a new model to examine interactions between ER and Acrp30.

Figure 1

Growth curves for breast cancer cell lines in response to increasing concentrations of Acrp30. (A) MCF-7, (B) T47-D, (C) MDA-MB-361, (D) SK-BR-3 and (E) MDA-MB-231 cells. The concentration of Acrp30 in ng ml⁻¹ is shown below each graph. Each point represents three or more wells. ^*Indicates significantly different from 0 ng ml⁻¹ as defined by ANOVA p=0.0002, Dunnett's multiple comparison post-test P<0.01.

Figure 2

Western analysis of protein regulation by Acrp30. (A) The individual cell lines are shown above their two corresponding lanes on the gels. Minus symbols above the lanes indicate extracts from untreated cells and plus symbols above the lanes indicate extracts from cells treated with 1 μg ml⁻¹ Acrp30 for 24 h. The proteins detected are shown to the left of the figure. Abbreviations are AdipoR1=adiponectin receptor 1; AdipoR2=adiponectin receptor 2; Ob-Rb=leptin receptor isoform b; PCNA=proliferating cell nuclear antigen and PARP=Poly (ADP-ribose) polymerase. (B) Densitometry of cleaved caspase 8 (solid bars) and cleaved PARP (checkered bars) normalised to b-actin. The x axis shows the different cell lines tested without Acrp30 (−) or with Acrp30 (+). The y axis is percent where 100% is the value of each cell line without Acrp30. Note: cleaved PARP was not detected in SK-BR-3 and MDA-MB-231 cells under these conditions.

Figure 3

Growth of MDA-wt and clones expressing ERα in response to oestradiol in vivo. Rag1 mice were injected with the cell lines shown along the x axis in the presence (E) or absence (C) of oestradiol pellets. Mice were intact for experiment 1 and ovariectomised in experiment 2. Tumours were excised, weighed and measured at the end of the experiment. The average volume in millimetres of the tumours are shown in (A) and (C) and the average weight in milligrams are shown in (B) and (D) along the y axis for experiments one and two respectively. The. volume for each individual tumour was computed according to the formula length ((longest dimension) × width squared (widest point at right angle to length) × 0.52). The different groups are shown along the x axis. Bars represent standard error of the mean. An ‘a’ above the bars shows groups that are statically different than all other groups as determined by ANOVA with Newman–Keul's multiple comparison post-test (P<0.01).

Figure 4

Growth of MDA-wt and MDA-ERα7 cells in response to gAcrp30 and Acrp30 in vitro. The MDA-wt (checkered bars) and MDA-ERα7 (filled bars) cells are shown with cell growth as a percent along the y axis. Cells in only serum-free L-15 were considered to be 100%. The concentrations of gAcrp30 (A) or Acrp30 (B) are shown along the x axis. The assays were performed in triplicate and repeated a total of three separate times. Bars represent standard error of the mean of the three different experiments. An ‘a’ corresponds to significantly different from the 0 control of the same cell line as determined by ANOVA with the Dunnett's post-test and ‘b’ corresponds to a significant difference between the two cell lines at the same concentration as determined by ANOVA with the Newman–Keul's test.

Figure 5

Alteration of signalling pathways by Acrp30 and gAcrp30. MDA-wt and MDA-ERα7 cells were treated with 10 μg ml⁻¹ of gAcrp30 or full-length Acrp30 for 24 h in serum-free media. After pretreatment 5% FCS was added to the cultures indicated for 15 min (A) or 24 h (B). The different antibodies used are shown along the side and the cell lines with treatments are shown above the lanes. In (C) (MDA-wt) and (D) (MDA-ERα7) densitometry is shown of (A) that has been normalised to β-actin. The x axis is the different proteins and the y axis is percent where the unstimulated control is set to 100%. Solid bars are untreated (lanes 1 and 5), diagonal slashes are 5% FCS (lanes 2 and 6), checkered bars are 5% FCS with Acrp30 pretreatment (lanes 3 and 7) and horizontal bars are 5% FCS with gAcrp30 pretreatment (lanes 4 and 8).