Increase in fatty acids and flotillins upon resveratrol treatment of human breast cancer cells

Flotillin-1 and flotillin-2 are highly conserved proteins that localize into cholesterol-rich microdomains in cellular membranes. Flotillins are closely related to the occurrence and development of various types of human cancers. Flotillin-1 is highly expressed in breast cancer, and the high expression level of flotillin-1 is significantly correlated with poorer patient survival. Here we studied the relationship between the formation of lipid rafts and the expression of flotillins and lipids in human breast cancer cells. We used the polyphenol compound resveratrol to alter the structure and function of the plasma membrane. Our data revealed an increase in fatty acids in MCF-7 and MDA-MB-231 cells upon resveratrol treatment. Interestingly, we also found an increase in the expression of both flotillin-1 and flotillin-2 in breast tumor cells after treatment. Resveratrol also induced changes in the pattern of flotillin distribution among detergent-resistant lipid rafts fractions in both cell lines and induced the nuclear translocation of flotillin-2. Since resveratrol has been pointed out as a putative cancer therapy agent, our results could have an impact on the understanding of the effects of resveratrol in tumor cells.


Results
Analysis of cell viability after resveratrol treatment. Here our aim was to study the effects of the lipid raft-inducer compound, resveratrol, in the expression of flotillins in human breast cancer cells. First, we investigated the effects of resveratrol treatment on the viability, using the MTT assay, of two human breast tumor cells (MCF-7 and MDA-MB-231) and in the non-tumor mouse breast epithelial cell line 4T1. We analyzed the viability of the cells after 24 hours of treatment using different concentrations of resveratrol (10,25,50,100,200 and 300 µM). No significant changes in cell viability was observed in non-tumor 4T1 cells (Fig. 1a).
Isolation of lipid rafts. To investigate the effects of resveratrol on lipid rafts, we decided to isolate lipid rafts from the two human breast tumor cell lines (untreated and resveratrol-treated cells) using a discontinuous sucrose density gradient centrifugation in the presence of Triton X-100 at 4 °C. Lipid rafts are insoluble in the nonionic detergent Triton X-100. At the end of the isolation procedure, it was possible to observe the formation of an intact lipid raft fraction floating on the top of the centrifuge tube in all the experiments (Fig. 2a). Twelve fractions were collected from each isolation procedure and the percentage of sucrose was determined in each one. We confirmed the formation of a sucrose gradient ranging from 5% to 50% in all four experimental conditions: untreated and resveratrol-treated MCF-7 and MDA-MB-231 cells (Fig. 2b-e).

Resveratrol induces changes in the pattern of flotillin distribution among lipid rafts fractions.
Next, we decided to analyze the presence of flotillin proteins in the 12 fractions obtained in the lipid raft isolation since flotillins are considered lipid raft markers. Upon treatment of cells with resveratrol, lipid rafts were separated by sucrose gradient and then examined by dot blot analysis to assess the presence of flotillin-1 in both raft and non-raft fractions. Changes in the distribution of flotillin among the fractions were observed when comparing untreated with resveratrol-treated cells (Fig. 3). In resveratrol-treated cells, flotillins were found in fractions containing lower concentrations of sucrose as compared to control cells (Fig. 3). The results shown in Fig. 3 demonstrate that resveratrol treatment of MCF-7 and MDA-MB-231 cells resulted in a significant redistribution of the cellular flotillin-1 from the cytosol and plasma membrane (fractions 7-9) to the lipid raft microdomains (fractions 3-6).

Resveratrol induces an increase in the expression of flotillins in breast cancer cells.
Since we detected changes in the distribution of flotillins among rafts and non-rafts fractions after treatment with resveratrol (Figs 2 and 3), we decided to analyze the expression of flotillin proteins in untreated and resveratrol-treated cells using immunofluorescence microscopy. MDA-MB-231 cells were treated with resveratrol, fixed and labeled with anti-flotillin-2 antibodies. Immunofluorescence microcopy analysis revealed a 10-fold increase in the expression of flotillin-2 in treated cells as compared to control (Fig. 4). It is important to point out that treatment of MDA-MB-231 cells with 200 uM resveratrol induced a 32% decrease in cell viability (Fig. 1c), and nevertheless we found a 10-fold increase in the labeling of flotillin-2 in these cells.
To further study the subcellular localization of flotillin-2 in control and resveratrol-treated cells, we triple-labelled MDA-MB-231 cells with Bodipy-ceramide (a Golgi specific fluorescent probe), flotillin-2 and DAPI and analyzed the cells under a Spinning Disk confocal fluorescence microscope (Fig. 5). Interestingly, we found that flotillin-2 colocalizes with Golgi membranes in control MDA-MB-231 cells (DMSO) but translocate to the nuclei after resveratrol treatment (Fig. 5).
We also investigated the expression of flotillins at both the mRNA and protein levels in MCF-7 and MDA-MB-231 cells after treatment with resveratrol. The results of qPCR showed that resveratrol treatment induces an increase in the expression of flotillin-1 and -2 ( Fig. 6), with the only exception of the expression flotillin-2 protein in MCF-7 cells, which showed no significant alterations. Western blots showed no significant changes in the expression of both flotillin-1 and -2, at the protein level, in MCF-7 and MDA-MB-231 cells after resveratrol treatment (Fig. 7). Interestingly, a slight increase in the levels of both flotillin-1 and -2 can be observed in MDA-MB-231 cells after resveratrol treatment, but not in MCF-7 cells (Fig. 7).
Isolation of neutral lipids. Then we decided to investigate whether the changes in flotillins expression and in lipid rafts observed after resveratrol treatment were accompanied by changes in neutral lipids. Neutral lipids were isolated from the same initial number of MCF-7 and MDA-MB-231 cells (10 7 cells) and we quantified the amounts of fatty acids (FA), triacylglycerol (TG), cholesterol ester (CHOE), and cholesterol (CHO) in MCF-7 and MDA-MB-231 untreated and treated with resveratrol (Fig. 8). Interestingly, we found a significant increase in the total amount of fatty acids (FA) in both cell lines (MCF-7 and MDA-MB-231) after resveratrol treatment: an 188% increase in MCF-7 cells and an 435% increase in MDA-MB-231 cells (Fig. 8). of resveratrol, or 0.1% DMSO (control), for 24 h at 37 °C. They were then washed and maintained with MTT (0.5 mg/mL) for 3 h at 37 °C for formazan crystals formation and the associated optical density was analyzed at 570 and 650 nm. The results were expressed as mean ± SD, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 significantly different (p < 0.05) from the values corresponding to the control, using One-way ANOVA, followed by multiple comparison test Dunnett. (2019) 9:13960 | https://doi.org/10.1038/s41598-019-50416-5 www.nature.com/scientificreports www.nature.com/scientificreports/

Discussion
Our study analyzed the effects of resveratrol treatment in the expression and distribution of flotillins and neutral lipids in human breast cancer cells. We found that resveratrol induced: (I) a cytotoxic effect in a dose-dependent manner; (ii) an increase in the expression of flotillin-1 and -2; (iii) an increase in the total amount of fatty acids; and (iv) a redistribution of flotillin-1 from the cytosol and plasma membrane to lipid raft microdomains and the nuclear translocation of flotillin-2.
In relation to the cytotoxic effects, we found that concentrations of resveratrol above 50 μM induced a significant decrease in cell viability in MCF-7 and in MDA-MB-231 cells, although MDA-MB-231 cells were found to be more resistance to treatment. These results show that MCF-7 cells are more sensitive to resveratrol than MDA-MB-231 cells, which are in accordance with a previous work 12 . MDA-MB-231 is a triple-negative type of human breast cancer (TNBC) cell line and consider to be a highly aggressive type of cancer cell, whereas MCF-7 expresses estrogen receptor and is a less aggressive cancer cell.
The increased sensitivity in cell viability of MCF-7 cells to resveratrol suggests that resveratrol action could be mediated, at least in part, by the estrogen receptor found in these cells. More studies are necessary to investigate the molecular details of the interaction between resveratrol and membrane receptors in MCF-7 cells. Importantly, no significant changes in cell viability was observed in the non-tumor breast cell line 4T1 reinforcing the idea that breast tumor cells are more sensitive to resveratrol than non-tumor breast cells.
Interestingly, we found an increase in fatty acids and in flotillin-1 and -2 in MCF-7 and MDA-MB-231 cells upon resveratrol treatment. Flotillins and fatty acids are both essential components for the formation of lipid rafts and therefore our results are in accordance with a previous work showing that resveratrol induces phase separation and formation of liquid-ordered domains in bilayer structures 7 . Favoring this hypothesis, we found a significant increase in the total amount of fatty acids, but not in cholesteryl ester and triacylglycerol, in both cell lines (MCF-7 and MDA-MB-231) after resveratrol treatment. It has been shown that lipid rafts are rich in phospholipids and cholesterol, whereas contain small amounts of cholesteryl ester and triacylglycerol 13 .
Our data also showed that resveratrol induced changes in the pattern of flotillin distribution among detergent-resistant lipid rafts fractions in both cell lines. Resveratrol treatment of MCF-7 and MDA-MB-231 cells resulted in a significant redistribution of the cellular flotillin-1 from the cytosol and plasma membrane to lipid raft microdomains.  of genes related to cell survival and cell proliferation, which are hallmarks of cancer cells. In accordance with our results and hypothesis, the nuclear translocation of flotillins has been previously correlated with mitogenic activity in the human prostate cancer cell line PC-3 14 .
Curiously, only few papers describe the presence of flotillin-2 within the nuclei of cancer cells, and therefore nuclear flotillin-2 cannot be described as a characteristic of all cancer cell types or specific to breast tumor cancer cells. Further studies are necessary to unravel the role of nuclear flotillin-2 in different cancer cell types.
Finally, we found flotillin-2 within vesicle-like compartments after resveratrol treatment. These results suggest that upon resveratrol treatment flotillins are upregulated and they accumulate in intracellular compartments before being directed to the nucleus or to lipid rafts at the plasma membrane. It has been shown by different groups that flotillins display a dynamic cellular localization and they can be found in four major locations in the cells: plasma membrane, nucleus, endosomal structures and/or intracellular vesicles 15 . The molecular regulation of the dynamic changes in the cellular distribution of flotillins after resveratrol treatment needs to be further studied.
Triple-negative MDA-MB-231 cells exhibit a high level of resistance to chemotherapeutic drugs 16,17 . Chemoresistance has a range of causes, which includes the following: highest expression of phosphorylated protein kinase D isoform 2 (PKD2), activity of ATP-binding cassette (ABC) transporters, such as P-glycoprotein1 (MDR1 and ABCB1), ABCG2 and CCL20, which function as drug efflux pumps 16,18 .
One main question that arises from our work is how resveratrol induces flotillin expression. The expression of both flotillins involves transcriptional regulation and can be modulated by different transcription factors 19 . Flotillins are transcriptional targets of the extracellular signal-regulated kinases (ERK1/2) and the transcription factor Egr1 and serum response factor (SRF), both of which are ERK1/2 downstream targets, were identified as positive regulators of flotillin expression. Interestingly, it has been shown that ERK1/2 accumulates in lipid rafts on resveratrol exposure and could enter the cell by endocytosis via lipid rafts activating ERK1/2 downstream www.nature.com/scientificreports www.nature.com/scientificreports/ signaling pathways 11 . Here, we hypothesize that the expression of flotillin genes could be activated by ERK1/2 signaling pathway after resveratrol treatment.
Since resveratrol has been pointed out as a putative cancer therapy agent, our results could have an impact in the understanding of the effects of resveratrol in different human breast tumor cells. Cell viability assay. Cell viability was determined using 3-(4,5-dimethyl-2-thiazyl)-2,5-diphenyl-2Htetrazolium bromide (MTT) reagent (Sigma-Aldrich, USA). Briefly, cells were plated at an initial density of 2.0 × 10 5 cells per well in 24-well plates and incubated for 24 h at 37 °C under 5% CO 2 . Cultures were treated with resveratrol (10 to 300 μM) for 24 h. Cells treated with 0.1% DMSO were used as control. After 24 h of treatment, the supernatant of each well was removed, and cells were washed twice with PBS. Then, 500 µL MTT solution  www.nature.com/scientificreports www.nature.com/scientificreports/ (5.0 mg/mL in PBS) were added and cells were incubated for 3 h at 37 °C, 5% CO 2 . The resultant formazan crystals were dissolved in isopropanol (500 µL) and absorbance intensities were measured in a microplate reader (FlexStation® 3 Reader, USA) at 570 nm and 650 nm 20 . All experiments were performed in triplicates, and the relative cell viability (%) was expressed as a percentage relative to the control cells.

Detergent-resistant membrane purification and Dot blot.
Cells were washed in phosphate buffered saline (PBS, pH 7.2), and TNE buffer (100 mM Tris, 150 mM NaCl, and 0.2 mM EDTA; pH 7.5) was added together with 2 µL protease inhibitor cocktail 10 times diluted (Sigma-Aldrich, USA). Aliquots of 1 × 10 8 cells in TNE buffer were carefully disrupted on ice with a Branson Sonifier Cell disruptor 250 (Analog Ultrasonic Homogenizer, USA) operating at 10% of total amplitude, with 3 cycles of 10 s and with 1 s rest between cycles, yielding a total homogenate. This homogenate was incubated for 20 min at 4 °C with 2% Triton X-100 in TNE buffer. After incubation, the sample was mixed 1:1 with 80% sucrose (w/v) in TNE buffer and transferred to a Beckman SW41 centrifuge tube (Beckman Coulter Inc, USA). This mixture was first overlaid with 30% sucrose, followed by 5% sucrose, and centrifuged at 38000 rpm for 20 h at 4 °C. Twelve 1-ml fractions could be sequentially collected from the top. Anti-flotillin-1 dot blotting -0.5 mL from each gradient fraction was adsorbed onto a nitrocellulose filter using a dot-blot apparatus (GE Healthcare Life Sciences Corp., UK). Membranes were blocked for 2 h in Tris-buffered saline containing 1% Tween 20 (TBS-T) and 5% nonfat milk and incubated for 2 h with the primary antibody (1:1000), then washed with TBS-T and incubated with a peroxidase-conjugated secondary antibody (1:1000) for 2 h. Antigen-antibody complex formed were visualized with the enhanced chemiluminescence (ECL) kit (Amersham, UK) using Kodak X-Omat films (Eastman Kodak Comp., USA). Images were analyzed using IMAGE J 1.50i software (NIH, USA) and results were expressed as arbitrary units.  www.nature.com/scientificreports www.nature.com/scientificreports/ with Alexa Fluor-conjugated secondary antibodies for 1 h at 37 °C, and nuclei were labeled with DAPI (0.1 μg/mL in 0.9% NaCl). Some cells were triple-labelled with Bodipy-ceramide (Thermo Fisher Scientific, USA), flotillin-2 and DAPI. Bodipy-ceramide was used for labeling Golgi with a fluorescent probe. Cells were examined either with an Axiovert 100 microscope (Carl Zeiss, Germany) or with a Spinning Disk confocal microscope (Olympus, Japan). Image processing was performed using Fiji software (based on ImageJ, http://imageJ.nih.gov/ij/). qPCR. Real-time quantitative polymerase chain reaction (qPCR) was performed in untreated and resveratrol-treated cells. mRNA from cells were isolated using Trizol (Invitrogen, USA) reagent, according to the manufacturer's instructions, and quantified using a Nanodrop spectrophotometer. Two micrograms of total RNA were used as a template for cDNA synthesis, using the High Capacity cDNA Reverse Transcription kit (Life Technologies, USA). PowerUpTM SYBR® Green Master Mix (Applied Biosystems, USA), was used to quantify mRNA expression levels, with actin as an endogenous control. Real-time reactions were performed in triplicates using a StepOne Real-Time PCR Systems (Applied Biosystems, USA). Relative quantification was performed using the Delta-Delta Ct method. Primer sequences were as follows: Actin_F: GTCCGCGATATCAAGGAAAA Actin_R: GTGTTGGCGTACAGGTCCTT Flotillin-1_F: CTCCACCCCACCTCAACTTATTTA Flotillin-1_R: TCCAGCCCATCCCTCAGTCT Flotillin-2_F: CCCCAGATTGCTGCCAAA Flotillin-2_R: TCCA-CTGAGGACCACAATCTCA Lipid extraction. After treatment with resveratrol, 1.0 × 10 7 cells were washed twice with PBS (Phosphate-buffered saline) (pH 7.5). The lipids were extracted as described previously by Bligh and Dyer 21 , methanol:chloroform:water (2:1:0.8 v/v) solution was added to the cells, they were, then, vortexed for 1 hour and centrifuged at 4 °C for 20 min at 3000 rpm. The lipid-containing supernatants were separated from the pellet. This process was performed twice, and the supernatants were united. The supernatants were added to water:chloroform (1:1) and vortexed for 30 s. The samples were centrifuged (3000 rpm/30 min at 4 °C), then the lipid phase was separated, and the solvent was evaporated by an N 2 stream. Lipids extracts were resuspended in chloroform for analysis by TLC and HPTLC.
Neutral lipids analysis. Lipid extracts used for neutral lipids were analyzed by one-dimensional thin layer chromatography (TLC) and high performance thin layer chromatography (HPTLC) on Silica plate (Merck, Germany). After applying the samples on the plate, the run was performed with n-hexane:ethylether:acetic acid (60:40:1 for TLC and 30:20:05 for HPTLC; v/v) 22 . Cholesterol (CHO), cholesterol ester (CHOE), triacylglycerol (TG) and fatty acid (FA) (Sigma-Aldrich, USA) standards were running in parallel to the samples to enable identifications and quantification of the bands. The lipids were visualized using a charring reagent after heating at 200 °C for 5 min 23 . After that, the chromatography plates were digitalized and analyzed using Image J 1.50i software (NIH, USA). The results were expressed as µg/µL, calculated by the fraction of pixels presented by each band relative to the respectively standard lipids.