Estrogen Receptor Beta (ERβ) Maintains Mitochondrial Network Regulating Invasiveness in an Obesity-Related Inflammation Condition in Breast Cancer

Obesity, a physiological situation where different proinflammatory cytokines and hormones are secreted, is a major risk factor for breast cancer. Mitochondrial functionality exhibits a relevant role in the tumorigenic potential of a cancer cell. In the present study, it has been examined the influence of an obesity-related inflammation ELIT treatment (17β-estradiol, leptin, IL-6, and TNFα), which aims to stimulate the hormonal conditions of a postmenopausal obese woman on the mitochondrial functionality and invasiveness of MCF7 and T47D breast cancer cell lines, which display a different ratio of both estrogen receptor isoforms, ERα and ERβ. The results showed a decrease in mitochondrial functionality, with an increase in oxidative stress and invasiveness and motility, in the MCF7 cell line (high ERα/ERβ ratio) compared to a maintained status in the T47D cell line (low ERα/ERβ ratio) after ELIT treatment. In addition, breast cancer biopsies were analyzed, showing that breast tumors of obese patients present a high positive correlation between IL-6 receptor and ERβ and have an increased expression of cytokines, antioxidant enzymes, and mitochondrial biogenesis and dynamics genes. Altogether, giving special importance to ERβ in the pathology of obese patients with breast cancer is necessary, approaching to personalized medicine.


Introduction
Cancer, a pathology characterized by the excessive proliferation of tumor cells, is the second cause of death in the world, responsible for approximately 9.6 million deaths in 2018. In addition, breast cancer is a disease causing more than half a million deaths annually, projecting an increase in its incidence to a total of 3.2 million new cases per year in 2050 [1,2]. Breast cancer is a heterogeneous and multifactorial disease that involves both genetic predisposition, lifestyle, and environmental factors. In fact, it has been estimated that about 20% of breast cancer cases around the world are attributed to modifiable risk factors, which include obesity [1].
Obesity is a chronic metabolic disease characterized by excess fat accumulation in the body, and its prevalence has increased markedly in the last two decades in the most developed countries [3]. In obesity conditions, there is an alteration of the adipocyte secretome, which can lead to an imbalance of secreted adipokines, affecting processes as cell proliferation, invasive growth, apoptosis, angiogenesis, and metastasis in tumor cells

Wound Healing Assay
Cells were seeded in six-well plates at a density of 1 × 10 6 cells/well for T47D cell line and 8.5 × 10 5 cells/well for MCF7 cell line. Wound healing assay was performed as previously described by Torrens-Mas et al. [35]. The area of the scratch was measured using the MRI Wound Healing Tool macro for ImageJ software.
Mitochondrial Roundness was analyzed in ImageJ software with Mito-Morphology Macro designed by Ruben K. Dagda at the University of Pittsburgh (2010). This macro is currently maintained and supported by grants NIH/NINDS R01NS105783-01 grant and by NIH/NIGMS R25 1R25-OD023795-01.

Seahorse Metabolic Analyzer
Real-time oxygen consumption rates (OCRs) were determined for MCF7 and T47D cells using the Seahorse Extracellular Flux (XFe96) analyzer (Seahorse Bioscience, North Billerica, MA, USA). Cells were seeded at a density of 4.8 × 10 3 cells/well for MCF7 and 9.600 cells/well for T47D into XFe96 well cell culture plates and incubated overnight to allow attachment at 37 • C in 5% CO 2 . After 24 h, cells were incubated with vehicle or ELIT treatment. After 48 h of incubation, cells were maintained in 200 µL/well of XF assay media at 37 • C, in a non-CO 2 incubator for 1 h. During the incubation time, mitochondrial complex inhibitors (1 µM oligomycin, 2 µM FCCP, 0.5 µM rotenone, and 0.5 µM antimycin A) were preloaded for OCR measurements, in XF assay media into the injection ports in the XFe96 sensor cartridge.

Human Samples
Human breast cancer biopsies were obtained from 33 women, ages between 45-90 years. Samples of these patients were obtained from the Biological Specimen Bank of Son Llàtzer Hospital and as specified by and with the necessary permission granted from the Balearic Island Bioethics Committee. Tumor samples were collected immediately after tumor removal and were frozen in isopentane for analysis as described by Sastre-Serra et al. [38]. Written informed consent was obtained from the patients before surgery. All the patients presented an invasive ductal carcinoma (ER-positive, PR-negative, and HER2-negative, as determined by immunohistochemistry) and were classified in normal weight (nw), overweight (ow), and obese (o) by BMI (kg/m 2 ).

Statistical Analysis
The statistical analyses were performed with the Statistical Programme for the Social Sciences software for Windows (SPSS, version 27.0; SPSS Inc, Chicago, IL, USA). Data are presented as mean ± standard error of the mean (SEM). The statistical differences in cell lines between vehicle-and ELIT-treated cells were analyzed using a Student's t-test with statistical significance was set at p < 0.05 (*). The statistical differences in human samples were analyzed using Pearson's correlation with statistical significance was set at p < 0.01 (**) and p < 0.05 (*).

Obesity-Related Inflammation Treatment Increases Inflammation-Related Genes Expression in Breast Cancer Cell Lines
To confirm the obesity-related inflammation treatment effectivity, mRNA expression of main inflammatory genes was determined. As shown in Figure 1, inflammatory genes expression (IL6, IL6R, CXCL8, PTGS2, TNF, and STAT3) showed a statistically significant increase in both MCF7 and T47D cell lines after 24 h of ELIT (17β-estradiol (10 nM), leptin (100 ng/mL), interleukin-6 (50 ng/mL), and TNFα (10 ng/mL)) treatment. In contrast, PPARG anti-inflammatory gene had decreased expression in the MCF7 cell line, and TGFB anti-inflammatory gene showed a statistically significant increase in the T47D cell line.  Table 2, ELIT treatment, an increase ROS production in both MCF7 and T47D cell lines was seen. Superoxide anion levels increased more than 10 times in MCF7 cells (1177%) and 5 times in T47D cells (526%) after ELIT treatment, as shown in Table 2, and H 2 O 2 production increased in both cell lines (+100% in MCF7 and +34% in T47D cells). However, cardiolipin content, as an indicator of mitochondrial inner membrane quantity, decreased in MCF7 treated cells. Moreover, as shown in Table 2, oxidative damage increased (+45%) in MCF7 cell line after ELIT treatment, but not in the T47D cell line. A representative blot of 4-HNE detection is shown in Supplementary Materials Figure S2. The mRNA expression of antioxidant enzymes was analyzed ( Figure 2) in both MCF7 and T47D cell lines after 24 h ELIT treatment. As shown in Figure 2, SOD2 (mitochondrial superoxide dismutase) showed a high increase in both cell lines; in addition, a decrease in SOD1 (copper/zinc superoxide dismutase) was observed. However, ELIT-treated MCF7 cells showed a general decrease in catalase (CAT) and glutathione reductase (GSR). Nevertheless, in the T47D cell line, ELIT treatment increased GSR and nuclear factor erythroid 2-related factor 2 (NFE2L2) expression, as shown in Figure 2.

As shown in
Main antioxidant protein expression levels were determined (Table 3 and Supplementary Materials) after 48 h ELIT treatment and resulted in a high increase in SOD2 in both MCF7 and T47D cell lines (more than 10 times in MCF7 and 30 times in T47D). Moreover, SOD1 showed a statistically significant increase in the T47D cell line after treatment. As shown in Table 3, CAT and GSR protein levels decrease only in the MCF7 cell line after ELIT treatment (−30% and −51%, respectively). Representative detection bands are shown in Supplementary Materials Figures S3-S6.

Mitochondrial Biogenesis and Functionality Are Reduced in Breast Cancer Cell Lines after ELIT Treatment with a High ERα/ERβ Ratio
Mitochondrial biogenesis genes were also checked, as shown in Figure 3. ELIT treatment decreased the expression of almost all mitochondrial biogenesis genes analyzed in MCF7 cell line. Nevertheless, in the T47D cell line, not only did the mRNA expression not decrease but also, especially high levels of estrogen-related receptor alpha (ESRRA) and Twinkle (TWNK) were found in ELIT-treated cells versus non-treated cells. Moreover, uncoupling proteins 2 and 5 (UCP2 and SLC25A14, respectively) mRNA expression was analyzed, and a decrease in UCP2 expression in both cell lines was accompanied by a decrease in uncoupling protein 5 expression in the MCF7 cell line.
Mitochondrial biogenesis master regulator PPARGC1A protein levels increased after ELIT treatment in the MCF7 cell line. However, to check the effect of ELIT treatment on the mitochondrial respiratory chain (OXPHOS), protein levels of these OXPHOS complexes were analyzed. As seen in Table 4  analyzed, and a decrease in UCP2 expression in both cell lines was accompani decrease in uncoupling protein 5 expression in the MCF7 cell line. Mitochondrial biogenesis master regulator PPARGC1A protein levels increas ELIT treatment in the MCF7 cell line. However, to check the effect of ELIT treatmen mitochondrial respiratory chain (OXPHOS), protein levels of these OXPHOS complex analyzed. As seen in Table 4     To further investigate mitochondrial function, the oxygen consumption rate (OCR) was determined. As shown in Figure 4a,c, basal OCR, Maximal respiratory capacity, ATPlinked respiration, and proton leak were statistically significantly lower in the MCF7 cell line after ELIT 48 h treatment. These changes were not observed in the T47D cell line (Figure 4b,d). Moreover, reserve capacity, calculated as basal minus maximal respiratory capacity rates, showing an increase in both cell lines, though this increase was higher in the T47D than in the MCF7 cell line. Basal respiration: initial rate-antimycin A + rotenone rate; ATP-Linked Respiration: maximal respiratory capacity: FCCP rate-antimycin A+ rotenone rate; initial rate-oligomycin rate; Reserve capacity: FCCP rate-initial rate; proton leak: oligomycin rate-antimycin A+ rotenone rate. Values are expressed as means ± SEM (n = 5). * Statistically significant difference between treated and vehicle-treated cells (Student's t-test, p < 0.05).

Obesity-Related Inflammation Treatment Affects Mitochondrial Dynamics and Mitochondrial Network in Breast Cancer Cell Lines
As shown in Figure 5, mitochondrial dynamics genes expression was decreased in MCF7 cells after 24 h ELIT treatment, but in T47D treated cells, this situation was not observed. In fact, almost all the genes showed an increase after treatment in this cell line. The expression of mitochondrial fusion-related genes MFN1, MFN2, OMA1, OPA1 showed decreased levels in the MCF7 cell line, whereas, in the T47D cell line, ELIT treatment increased expression of MFN1, MFN2, and OMA1. Fission-related mitochondrial genes presented a different pattern expression between cell lines, with a statistically significant increase in FIS1 mRNA expression in the T47D cell line after ELIT treatment. On the one hand, mitochondrial networking was modified after 24 h-ELIT treatment in both MCF7 and T47D cell lines. However, as shown in Figure 6a, the MTG signal was higher in the MCF7 cell line after treatment, but LTR intensity was also high. These changes were not observed in T47D cell lines after ELIT treatment. On the other hand, lysosomes distribution in MCF7 cell lines, whereas more equally distributed dispersed through the cytoplasm after 24 h-ELIT treatment, whereas in T47D cell line this distribution showed seemed to be more perinuclear. Index of elongation, calculated as the average circularity of mitochondria of confocal microscopy images, as shown in Figure 6b. As seen, the MCF7 cell line after 24 h ELIT treatment showed more circular mitochondria than in control cells. These morphological modifications were not seen in the T47D cell line with treatment.

Obesity-Related Inflammation Treatment Increase Invasiveness in Breast Cancer Cell Lines with a High ERα/ERβ Ratio
As shown in Figure 7a, Cadherin E (CDH1) expression was decreased in MCF7 cells after 24 h ELIT treatment, but Matrix Metalloproteinase 9 (MMP9) expression increased three times with respect to vehicle-treated cells. In contrast, CDH1 expression increased in the T47D cell line after treatment. In addition to these results, a wound-healing assay was performed. As shown in Figure 7b and in Supplementary Materials Figure S9, after 24 h, ELIT-treated MCF7 cells were able to better close the wound, leaving 80% of the initial scratch open, while the control cells were not able to close the wound. In the T47D cell line, there were no differences between control and ELIT-treated cells.

Estrogen Receptor Ratio Is Modified by ELIT Treatment in Breast Cancer Cell Lines
To observe the effects of obesity-related inflammation treatment over estrogen receptors alpha (ERα), beta (ERβ), and GPER in MCF7 and T47D breast cancer cell lines, mRNA expression was analyzed. As shown in Figure 8, estrogen receptor alpha (ESR1) mRNA expression was decreased in both cell lines after 24 h-ELIT treatment. However, estrogen receptor beta (ESR2) and GPER (GPER1) mRNA expression only shown a decrease in the MCF7 cell line, maintaining its expression in the T47D cell line.  Table 5 shows the correlation with Pearson correlation values between the IL6R mRNA expression and studied genes in different BMI situations: normal weight (nw), overweight (ow), and obese (o). CXCR8, CXCL8, TNF, SLC25A14, NRF1, NFE2L2, PPARGC1A, and FIS1 were significantly and positively correlated with IL6R expression in the nw group. CXCR8, TNF, SLC25A14, NRF1, NFE2L2, and PPARGC1A were significantly and positively correlated with IL6R expression in the ow group. CXCR8, CXCL8, TNF, PTGS2, ESR2, GPX1, SOD1, SLC25A14, NRF1, NFE2L2, PPARGC1A, SIRT1, FIS1, and OMA1 were significantly and positively correlated with IL6R expression in the ow group. Data represent Pearson correlations and significance (unilateral) in normal weight (n = 9). overweight (n = 8) and obesity (n = 16) groups. * (p < 0.05) ** (p < 0.01) significant difference.

IL6R in Breast Tumors Correlates with Inflammation, Mitochondrial Biogenesis, and Oxidative Stress Markers in Different BMI Situations
Moreover, ESR1 and ESR2 gene expression correlations are shown in Table 6a,b, respectively. On the one hand, IL6, CXCL8, SOD1, and SSBP1 were significantly and negatively correlated with ESR1 expression; instead, SIRT3 was significantly and positively correlated. On the other hand, IL6R, CXCR8, GPX1, SLC25A14, NRF1, PPARGC1A, SIRT1, and OMA1 were significantly and positively correlated with ESR2 expression.

Discussion
In this study, the effects of obesity related-inflammation on mitochondrial functionality in breast cancer cell lines and breast tumors, focusing on estrogen receptors ratio, were analyzed. Moreover, invasiveness in this situation was also analyzed. We demonstrated that ERβ, in an inflammatory and obesity condition, maintains mitochondrial functionality and avoids invasiveness in breast cancer cell lines. Moreover, we found a strong correlation between interleukin-6 receptor gene expression and inflammation, mitochondrial functionality, and oxidative stress markers, as well as with estrogen receptor beta, in breast cancer human samples in different BMI situations.
Obesity stimulates the adipose tissues to release inflammatory mediators such as tumor necrosis factor α and interleukin 6, predisposing them to a proinflammatory state and oxidative stress [14,39]. These signals also stimulate the release of inflammatory mediators by breast cancer cells, creating an autocrine feedback loop [40]. To start the study, we confirmed the effects of treatment on inflammatory genes expression. ELIT treatment (17β-estradiol (10 nM), Leptin (100 ng/mL), IL6 (50 ng/mL), and TNFα (10 ng/mL)), simulates circulating hormonal conditions in a postmenopausal obese woman and inducing a high increase in proinflammatory expression genes. Furthermore, it is worth noting that, in breast tumors in different BMI situations, inflammatory genes expression was positively correlated with interleukin-6 receptor gene expression. These results are in concordance with other studies where proinflammatory markers are studied in postmenopausal breast cancer patients [32].
Previous studies in our group and other research groups have shown the independent effects of 17β-estradiol or leptin on oxidative stress and mitochondrial biogenesis and dynamics [9,[41][42][43][44], but never in an associated-inflammation situation. It is well known that ERα is the predominant estrogen receptor found in the MCF7 cell line, and it responds to estrogens by increasing proliferation, while, if ERβ is overexpressed in these cells, the proliferative effect of estrogens is inhibited [9]. Therefore, the response to estrogens in breast cancer not only depends on the concentration of estrogens in the cellular environment but also depends on the ERα/ERβ ratio presented by cells [9,43].
In this study, MCF7 and T47D cell lines have been treated with an ELIT inflammatory cocktail with the aim of generates an inflammation situation (IL6 and TNFα) in the presence of Leptin and 17β-estradiol levels, simulating the physiological condition of postmenopausal obese women. In this way, an increase in reactive oxygen species has been observed in both cell lines; however, in the T47D cell line, a lower increase in the production of hydrogen peroxide and levels of superoxide anion was observed. These data make more sense when studying the oxidative damage present in both cell lines, where a significant increase was only observed in the MCF7 cell line. On the other hand, it seems that the T47D treated cell line does not present differences compared to the control group, despite presenting higher levels of H 2 O 2 and superoxide anion, as similarly described by some authors [41,45,46].
Oxidative stress is generated in breast cancer, and in many other pathologies, in two main ways. The first one, as observed in the high ERα/ERβ ratio MCF7 cell line after ELIt treatment, is a decrease in the expression of the antioxidant enzymes. It was observed that all antioxidant enzymes had decreased their gene or protein expression in the MCF7 cell line, except the SOD2 enzyme that had increased expression in both MCF7 and T47D cell lines. It has been described that SOD2 plays a role as a free radical detector, increasing gene and/or protein expression in order to alleviate oxidative damage [47]. Moreover, NFE2L2, a transcription factor that controls mainly antioxidant enzymes expression, and glutathione reductase, which recycles glutathione, was also increased, avoiding a high increase in free radical levels in low ERα/ERβ ratio T47D cell line. In addition, it should be noted that those tumors with a high correlation between IL6R and ESR2 gene expression presented an increase in antioxidant enzymes, thus could palliate levels of free radicals, which can ultimately diminish oxidative damage.
The other pathway that increases oxidative stress is the poor maintenance of a functional mitochondrial pool [48]. To achieve good maintenance, two highly coordinated processes, such as mitochondrial biogenesis and dynamics, are very important [49,50]. In the MCF7 cell line, where ERα predominates, the protein levels of the OXPHOS system are decreased after ELIT treatment, as well as the amount of inner mitochondrial membrane (cardiolipin levels), required for the activity of complexes I, III, and IV, and plays an important role in mitochondrial biogenesis [51,52]. Likewise, both mitochondrial biogenesis and mitochondrial dynamics are diminished by ELIT treatment in the MCF7 cell line. Therefore, neither mitochondria are generated, nor are those that do not function properly eliminated; thus, the production of free radicals and, consequently, the oxidative damage is increased. Mitochondrial dysfunction has been postulated as one of the hallmarks of cancer, increasing free radicals and decrease energetic efficiency [19]. If this situation is accompanied by low levels of antioxidant enzymes, oxidative damage will be higher, as observed in the MCF7 cell line.
However, the T47D cell line, with high levels of ERβ, had mitochondrial dynamics elevated, maintaining a more functional pool of mitochondria. In the case of mitochondrial biogenesis, it should be noted that there are two genes that presented a very significant elevation, estrogen-related receptor alpha (ESRRA) and Twinkle (TWNK). ESRRA belongs to a superfamily of nuclear receptors independent of estrogen activation. Its expression is induced and activated by PGC1α, and the two factors together are capable of binding to response elements and promoting the initiation and elongation of the transcription of metabolic and mitochondrial target genes [53]. Twinkle is the mitochondrial helicase encoded by nuclear DNA that acts during mtDNA replication, and its overexpression has been shown to be related to increased mitochondrial biogenesis [54]. In fact, ESRRA should be considered as a possible factor responsible for the difference in the rate of mitochondrial biogenesis between both MCF7 and T47D cell lines in an obese-related inflammation situation. As shown in the results, the increased expression of the nuclear receptor ESRRA in the T47D cell line could be responsible for the maintenance of the mitochondrial pool in the cell. Furthermore, it has been observed that the STAT3 signaling pathway, activated by leptin and IL6, upregulates the expression of ESRRA [55].
Taking into account that patients with tumors with low ERα/ERβ ratios have a worse response to chemical agents oxidative damage inductors [56], we observed that breast tumors with a high correlation between IL6R and ESR2 in obese patients had genes related to mitochondrial biogenesis, and dynamics increased gene expression, which could be an attempt to increase the number of mitochondria due to the inflammatory state and oxidative stress generated over a long period of time, such happens in other situations like aging [57,58], but maintaining a functional mitochondrial pool. In addition, it is worthy to note that both interleukin 6 and 8 gene expression correlated negatively with estrogen receptor alpha gene expression, whereas both interleukin 6 and 8 receptors correlated positively with estrogen receptor beta gene expression. These results remark the importance of more studies are needed in order to better understand estrogen receptors subtypes' role in breast cancer.
In addition to mitochondrial biogenesis and dynamics, our study reveals an oxygen consumption rate decreased in the MCF7 cell line that supports all the results commented above. It is worth noting that ELIT treatment decreases basal respiration, maximal respiratory capacity, ATP-linked respiration as much as proton leak in the MCF7 cell line. Proton leak decrease is in concordance with low uncoupling proteins mRNA expression levels found in the MCF7 cell line after ELIT treatment. UCPs, which promote proton leak across the inner mitochondrial membrane, have emerged as essential regulators of mitochondrial membrane potential, respiratory activity, and ROS generation [59]. As seen in our results, the mitochondrial network was modified in the MCF7 cell line after treatment. Mitochondria appear more fragmented and circular than fused in the ELIT-treated MCF7 cell line but not in the T47D cell line. In fact, it seems that estrogen receptor beta could maintain mitochondrial network as well as oxygen consumption rate. Moreover, the T47D cell line showed an increase in the reserve capacity of mitochondria, a fact that supports good mitochondrial pool maintenance [48,60,61].
Oxidative stress, inflammation, and poor mitochondrial network observed in MCF7 cell lines after ELIT treatment leads to a worse situation, and some authors have described a relationship between this situation and metastasis [62]. Our results are in concordance with this idea due to increased motility and modification of invasiveness markers presented in the MCF7 cell line but not in the T47D cell line after treatment. In fact, the T47D cell line, with a better mitochondrial profile, including biogenesis, dynamics, and mitochondrial network, showed high levels of Cadherin-E that could simulate an epithelial-like phenotype, as suggested by some authors [63,64].
As mentioned above, the MCF7 cell line has a higher ERα/ERβ ratio, and the T47D cell line has a lower ratio. ELIT-treatment decreases estrogen receptor alpha mRNA expression in both MCF7 and T47D cell lines, as shown in results. However, estrogen receptor beta mRNA expression only decreases in the MCF7 cell line but not in the T47D cell line. This fact supports that all the results showed in our study could be through estrogen receptor beta T47D cell lines maintenance, giving to this receptor subtype the protective role that previously some studies have been described [9,39].

Conclusions
The presence of estrogen receptor beta allows maintaining a more functional mitochondrial pool, with active mitochondrial biogenesis and dynamics, which means less production of reactive oxygen species and better mitochondrial metabolism in an obesityrelated inflammation condition. In addition, antioxidant enzymes are active, preventing oxidative damage and, at least in part, invasiveness.
This study could be important to remark the importance of estrogen receptor beta and mitochondria as an important organelle in the development and prognosis of breast cancer in obese patients. Likewise, more studies are necessary in order to clarify the estrogen receptor beta mechanism in breast cancer and establish it as a clinical biomarker, as is the estrogen receptor alpha.