Macrophages Upregulate Estrogen Receptor Expression in the Model of Obesity-Associated Breast Carcinoma

Breast cancer (BC) and obesity are two heterogeneous conditions with a tremendous impact on health. BC is the most commonly diagnosed neoplasm and the leading cause of cancer-related mortality among women, and the prevalence of obesity in women worldwide reaches pandemic proportions. Obesity is a significant risk factor for both incidence and worse prognosis in estrogen receptor positive (ER+) BC. Yet, the mechanisms underlying the association between excess adiposity and increased risk/therapy resistance/poorer outcome of ER+, but not ER−negative (ER−), BC are not fully understood. Tumor-promoting action of obesity, predominantly in ER + BC patients, is often attributed to the augmented production of estrogen in ‘obese’ adipose tissue. However, in addition to the estrogen production, expression levels of ER represent a key determinant in hormone-driven breast tumorigenesis and therapy response. Here, utilizing in vitro and in vivo models of BC, we show that macrophages, whose adverse activation by obesogenic substances is fueled by heparanase (extracellular matrix-degrading enzyme), are capable of upregulating ER expression in tumor cells, in the setting of obesity-associated BC. These findings underscore a previously unknown mechanism through which interplay between cellular/extracellular elements of obesity-associated BC microenvironment influences estrogen sensitivity—a critical component in hormone-related cancer progression and resistance to therapy.


Introduction
Breast carcinoma (BC) is the most commonly diagnosed cancer type and the leading cause of cancer-related mortality among women [1]. BC is a highly heterogeneous disease [2,3]. The widely employed classification, referred to as PAM50, classifies BC into five molecular intrinsic subtypes: luminal A, luminal B, HER2-enriched, basal-like, and normal [3,4]. Over 75% of BCs express estrogen receptor α (ER) in >1% of the tumor cells by IHC, and overlap with luminal A and B subtypes [2]. ERα acts as a key driver of BC development, progression and dissemination [5][6][7][8][9][10][11]. Hormonal therapies for ERα-positive (ER+) BC target ERα either directly, by selective ER modulators and down-regulators, or indirectly, by abolishing estrogen production by inhibitors of aromatase, the rate-limiting enzyme in estrogen biosynthesis. Hence, ER expression level (governed by multiple and poorly-understood mechanisms [6][7][8]) is an important determinant in BC tumorigenesis, as well as in the tumor response to hormonal therapy (reviewed in [7,12]).

Clinical Data Analysis
Breast tumor tissue specimens and clinical data from 58 ER+ breast carcinoma female patients were available from the Sharett Oncology Institute, Hadassah Medical Center, Jerusalem. A summary of clinical and pathological characteristics of the study population is shown in Table 1. The use of these data and formalin-fixed, paraffin-embedded breast carcinoma tissues in research was approved by the Human Subjects Research Ethics Committee of the Hadassah Medical Center. Tissue microarray construction was performed as previously described [30,53]. Immunodetection of ERα was performed as in [30,53], using ERα monoclonal antibody NCL-L-ER 6F11 (Novocastra, Newcastle, UK). ERα positivity and staining intensity (scored as weak (=1), moderate (=2), or strong (=3)) was determined in accordance with [54,55] by an expert pathologist (B.M.). Immunodetection of heparanase was performed using polyclonal rabbit anti-heparanase antibody (733) directed against a synthetic peptide ( 158 KKFKNSTYRSSSVD 171 ) corresponding to the N-terminus of the 50-kDa subunit of the heparanase enzyme [30,53]. The antibody was diluted 1:100 in 10% goat serum in PBS. Control slides were incubated with 10% goat serum alone. Color was developed as described in [30,53], slides were visualized with a Zeiss axioscope microscope and manually read by an expert pathologist (B.M.). To define tumor as heparanase-positive, a cut-off point of 25% immuno-stained tumor cells was chosen, based on an initial overview of the cases, in order to improve signal-to-noise ratios. Cut-off was chosen before any attempt at correlating heparanase expression with the obese status of the patients.
For tumor irradiation studies tumor-bearing obese and lean mice were anesthetized 7 days before sacrifice and their tumors irradiated conformally to 5Gy, using surface brachytherapy, essentially as described in [57]. Briefly, radiation was delivered to the tumor-bearing mammary gland by using a brachytherapy after-loader (I 192 Nucletron mi-croSelectron HDR, Veenendaal, The Netherlands); the brachytherapy sleeve was positioned over the tumor (above a 0.5 cm width silicon bolus). The dose distribution calculations were based on CT simulation; the dose of 4.5Gy was calculated to 1.0 cm isodose line, in order to achieve 5Gy to 90% of the tumor volume, while relatively protecting the surrounding normal tissues. The prescribed dose was confirmed by film dosimetry. Variation of dose inside the tumors was estimated to be within ±15% of the prescribed dose. Tumor growth was monitored immediately prior to irradiation and at the indicated times after irradiation. All experiments were performed in accordance with the Hebrew University Institutional Animal Care and Use Committee.

Immunoblotting
Cell and tumor tissue lysates were homogenized in lysis buffer containing 0.6% SDS, 1 mM Tris-HCl, pH 7.5, supplemented with a mixture of protease inhibitors (Roche) and phosphatase inhibitors (Thermo Scientific, Waltham, MA, USA). Equal protein aliquots were subjected to SDS-PAGE (10% acrylamide) under reducing conditions and proteins were transferred to a polyvinylidene difluoride membrane (Millipore, Darmstadt, Germany). Membranes were blocked with 3% BSA for 1 h at room temperature and probed with the appropriate antibody, followed by horseradish peroxidase-conjugated secondary antibody (KPL, Milford, MA, USA) and a chemi-luminescent substrate (Biological Industries, Beit-Haemek, Israel). Band intensity was quantified by densitometry analysis using ImageJ software.

Statistical Analysis
The results are presented as the mean ±SD unless otherwise stated. p values ≤ 0.05 were considered statistically significant. Statistical analysis of in vitro and in vivo data was performed by unpaired Student's t-test. Pearson Chi-Square test was applied to analyze the relationship between heparanase and estrogen receptor alpha expression of breast carcinoma patients, using SPSS software (SPSS Inc., Chicago, IL, USA). All statistical tests were two-sided.

Increased ERα Expression Levels in Obesity-Associated E0771 Murine Breast Carcinoma
Tumor-promoting action of obesity in ER+ breast tumors is often explained by enhanced estrogen signaling, owing to induction of aromatase expression in "obese" adipose tissue and augmented extragonadal synthesis of estrogen [30,41,42,58]. However, increase in estrogen signaling may also occur due to the upregulation of ER protein levels in BC cells. To test the impact of obesity on ER expression in BC, we applied an immunocompetent orthotopic mouse model of obesity-associated breast cancer, based on murine BC cell line Cells 2022, 11, 2844 6 of 18 E0771, growing in female syngeneic C57BL/6J mice with high fat diet (HFD)-induced obesity [30,34,59]. E0771 cells express relatively low levels of ERα and are considered as a luminal B subtype of BC [60][61][62]. HFD-fed C57BL6 mice represent a reliable model of diet-induced obesity and related pathological conditions, i.e., inflammation [63], FA accumulation [64], and obesity-accelerated tumor growth [34,59,65] Importantly, female (as opposed to male) C57BL/6J mice are protected against HFD/obesity-induced hyperinsulinemia/hyperglycemia [66], therefore allowing scrutinization of the specific contribution of obesity-related inflammatory events to BC progression.
In agreement with earlier reports [30,34,59,65], following 12 weeks on HFD, the experimental mice became obese, as evidenced by a 38% increase in their body weight, compared to control diet-fed lean mice (p < 0.001). E0771 cells were then implanted orthotopically in obese and lean mice, as described in Methods, and tumor growth was monitored for 14 days. Consistent with previous findings [30,34,59,65], growth of E0771 tumors was augmented in obese, as compared to lean, mice (on day 14 tumors from obese mice were 2.4-fold larger than tumors from lean mice, p < 0.004). As expected, the extent of Mφ infiltration, the hallmark of obesity-accelerated breast tumor growth [33,34], was increased 1.9-fold (p < 0.03) in tumors grown in obese versus lean mice (Supplementary Figure S1A,B). Additionally, the obese state attenuated the sensitivity of E0771 tumors to radiotherapy-backbone treatment modality in ER + BC (Supplementary Figure S1C). Given the role of ERα in growth and radio-resistance of BC [5][6][7][8], we examined ERα expression levels in tumors derived from obese versus lean mice. Immunoblot and qPCR analysis of the tumor tissue revealed that expression of the full length 66-kDa ERα protein and mRNA was significantly increased in E0771 tumors growing in obese, as compared to lean, mice ( Figure 1A-C). Of note, there was no difference in the levels of 46-kDa ER isoform (known to antagonize the full-length ER−mediated responses in BC [67,68]) between tumors growing in obese versus lean mice. The band intensity was quantified using ImageJ software; intensity ratio for ERα /GAPDH is shown, n ≤ 4 mice per condition. (C) qRT-PCR was used to determine the levels of ERα mRNA in tumor tissue samples collected from lean and obese mice. Error bars represent ± SD. * p < 0.04; ** p < 0.001.

Macrophages Stimulated by Obese Milieu Components Mediate Upregulation of ERα in BC
Cells.
Emerging evidence indicates that tumor-infiltrating immunocytes (including Mϕ) are capable of inducing ERα expression in hormone-responsive tumor types [43,69] via cytokine-mediated mechanism. Based on these observations, along with the notion that Mϕ represent the principal pro-cancerous immunocyte population implicated in obesityaccelerated BC tumorigenesis [35,42], and the well-documented increase in Mϕ infiltration in E0771 tumors under obese conditions (Refs. [30,33,34] and Supplementary Figure S1B), we sought to investigate whether Mϕ are capable of modulating ER The band intensity was quantified using ImageJ software; intensity ratio for ERα/GAPDH is shown, n ≤ 4 mice per condition. (C) qRT-PCR was used to determine the levels of ERα mRNA in tumor tissue samples collected from lean and obese mice. Error bars represent ± SD. * p < 0.04; ** p < 0.001.

Macrophages Stimulated by Obese Milieu Components Mediate Upregulation of ERα in BC Cells
Emerging evidence indicates that tumor-infiltrating immunocytes (including Mφ) are capable of inducing ERα expression in hormone-responsive tumor types [43,69] via cytokine-mediated mechanism. Based on these observations, along with the notion that Mφ represent the principal pro-cancerous immunocyte population implicated in obesity-Cells 2022, 11, 2844 7 of 18 accelerated BC tumorigenesis [35,42], and the well-documented increase in Mφ infiltration in E0771 tumors under obese conditions (References [30,33,34] and Supplementary Figure S1B), we sought to investigate whether Mφ are capable of modulating ER expression in the setting of obesity-associated BC. To mimic conditions occurring in breast tumorbearing obese patients/experimental animals (i.e., adverse activation of Mφ [35,41] due to the presence of increased concentrations of sFA in circulation [36]), murine and human (i.e., E0771, MCF7) ER + BC cell lines were incubated with medium conditioned by primary Mφ (isolated as described in Methods), either unstimulated or stimulated by SFA (i.e., palmitate, dominant SFA present in fat depots/circulation of female patients [70] and responsible for adverse Mφ activation under obese conditions [35,37,39,71]. As shown in Figure 2, SFA-stimulated Mφ-conditioned medium markedly upregulated ERα protein ( Figure 2A,B,D,E), as well as mRNA expression ( Figure 2C,F) of ESR1 gene, encoding for ERα, in E0771 and MCF7 cells. The effect of SFA-stimulated Mφ on ER expression was evident even in the 4T1 murine BC cell line (Supplementary Figure S2A), which is often considered ER−negative, but in fact expresses extremely low levels of ERα [72]. Similar effects were observed in MCF7 cells incubated with a medium conditioned by SFA-stimulated human Mφ differentiated from promonocytic cell line THP-1 [73] (Supplementary Figure S2B). Since Mϕ activation by SFA under the obese state involves a TLR4-dependent mechanism [35,37,39,71], we next tested the effect of TLR4 inhibition on the ability of SFAstimulated Mϕ to induce ER expression. As shown in Figure 3A  Since Mφ activation by SFA under the obese state involves a TLR4-dependent mechanism [35,37,39,71], we next tested the effect of TLR4 inhibition on the ability of SFAstimulated Mφ to induce ER expression. As shown in Figure 3A,B, specific TLR4 inhibitor TAK-242 abolished the ability of SFA-stimulated Mφ to induce ERα expression in E0771 (A) and MCF7 (B) cells, in agreement with TLR4-activating properties of SFA [37][38][39].
E0771 and MCF7 cells, while pretreatment of Mϕ by TAK-242 prior to LPS stimulation abolished this upregulation ( Figure 3C,D), further validating involvement of TLR4mediated signaling. Since TLR4 could also be expressed by BC cells per se, to exclude the possible effect of carry-over of TLR4 ligand via the Mϕ-conditioned medium, we compared levels of ERα protein in untreated BC cells (E0771, MCF-7) versus those incubated with medium conditioned by LPS (0.1 ng/mL)-stimulated Mϕ, or those incubated with standard medium containing the same concentration of LPS. As shown in Supplementary Figure S3, ERα protein levels were significantly increased in BC cells treated with medium conditioned by LPS-stimulated Mϕ, as compared to either untreated cells or cells treated directly with LPS, confirming macrophage specific function. . ERα protein levels were determined by immunoblotting (top) and quantified using ImageJ software (bottom). The intensity ratio for ERα /GAPDH is shown. The data are representative of 3 independent experiments; error bars represent ± SD. * p < 0.007; ** p < 0.0009. Of note, the canonic ligand of TLR4 lipopolysaccharide [LPS] per se is a component of the systemic obese milieu, since it is chronically present (at extremely low levels, typically ranging between 0.01 to 0.2 ng/mL) in the circulation of obese individuals/experimental animals [74][75][76]. This phenomenon, known as 'metabolic endotoxemia' [74][75][76], occurs due to impaired intestinal permeability under obese conditions [74][75][76][77][78]. Interestingly, we found that medium conditioned by Mφ stimulated by LPS at a concentration reflecting metabolic endotoxemia (i.e., 0.1 ng/mL) also upregulated ER in E0771 and MCF7 cells, while pretreatment of Mφ by TAK-242 prior to LPS stimulation abolished this upregulation ( Figure 3C,D), further validating involvement of TLR4-mediated signaling. Since TLR4 could also be expressed by BC cells per se, to exclude the possible effect of carry-over of TLR4 ligand via the Mφ-conditioned medium, we compared levels of ERα protein in untreated BC cells (E0771, MCF-7) versus those incubated with medium conditioned by LPS (0.1 ng/mL)-stimulated Mφ, or those incubated with standard medium containing the same concentration of LPS. As shown in Supplementary Figure S3, ERα protein levels were significantly increased in BC cells treated with medium conditioned by LPS-stimulated Mφ, as compared to either untreated cells or cells treated directly with LPS, confirming macrophage specific function.

Macrophages Stimulated by SFA Increase the Sensitivity of BC Cells to Estrogen
To verify the functional significance of ERα upregulation, depicted in Figure 2, we next assessed the effect of medium conditioned by SFA-stimulated Mφ on the proliferative response of MCF7 cells to estrogen. For this purpose, cells were first depleted of estrogen for 2 weeks (as detailed in Methods) and then treated with 17β estradiol in the presence or absence of medium conditioned by either unstimulated or SFA-stimulated Mφ ( Figure 4). As expected, 17β estradiol increased proliferation of MCF7 under all conditions. However, the presence of medium conditioned by SFA-stimulated Mφ (but not vehicle-stimulated Mφ) significantly augmented proliferative response to estradiol (Figure 4), consistent with the upregulation of ERα levels ( Figure 2). Importantly, stimulated Mφ also enhanced MCF7 proliferation under conditions of estrogen depletion (i.e., without addition of exogenous estrogen- Figure 4, white bars). In vitro estrogen depletion procedure ensures that only extremely low residual estrogen levels are present in the medium. In addition, this procedure mimics depletion of estrogen that occurs in BC patients treated with aromatase inhibitors. In this context, it is worthy to note that enhanced MCF7 proliferation, triggered by stimulated (as compared to unstimulated) Mφ, was observed even in the presence of residual levels of estrogen ( Figure 4). These data suggest that in the setting of obesity, upregulation of ER levels in BC cells by SFA stimulated Mφ may enable their proliferative response even to extremely low levels of estrogen, thus providing a molecular/cellular explanation for the observation that obese patients are more likely to experience ER+BC recurrence after treatment with aromatase inhibitors [23] and generally contributing to the worse response to hormonal therapy in BC [13,19,79]. Collectively, these results confirmed the ability of obese milieu component-activated Mφ to increase the sensitivity of BC cells to estrogen through upregulation of ERα.

Mφ-Mediated Augmentation of ER Expression in BC Is Dependent on Heparanase, the Endoglycosidase Enzyme Essential for Mφ Reactivity
Heparanase enzyme is preferentially expressed in obesity-associated breast tumors in clinical/experimental settings [30] and associated with worse prognosis in ER+, but not ER− BC [52]. Heparanase is the only known mammalian endoglycosidase capable of cleaving heparan sulfate (HS) at the cell surface and ECM. Of note, intact extracellular HS was shown to inhibit TLR4 signaling and Mφ activation, while enzymatic cleavage of HS relieves this inhibition [80]. Furthermore, soluble fragments released through enzymatic cleavage of HS stimulate TLR4 signaling [80][81][82][83][84]. Thus, heparanase enzyme is causally involved in Mφ activation by TLR4 ligands [45][46][47][48]50], including FA [30,49]. In light of the above data, we next investigated the role of heparanase in Mφ-mediated induction of ER under obese conditions. estrogen, thus providing a molecular/cellular explanation for the observation that obese patients are more likely to experience ER+BC recurrence after treatment with aromatase inhibitors [23] and generally contributing to the worse response to hormonal therapy in BC [13,19,79]. Collectively, these results confirmed the ability of obese milieu componentactivated Mϕ to increase the sensitivity of BC cells to estrogen through upregulation of ERα.  First, utilizing the obesity-associated E0771 BC model in heparanase-null Hpse-KO mice, we found that, unlike in wild-type (wt) animals, the obese state failed to upregulate ERα expression in BC tumors under conditions of heparanase deficiency ( Figure 5A-C). Of note, heparanase deficiency has been previously shown by Hermano et al., to abolish BC-promoting action of obesity in vivo [30]. Data from these experiments were used in the present study for comparative purposes. It should be noted that heparanase deficiency did not affect growth of E0771 tumors in lean mice (as no statistically significant difference was detected between volume of tumors derived from wt lean (105 mm 3 ) versus Hpse-KO lean (154 mm 3 ) mice on day 14, p > 0.06). In agreement with the in vivo findings ( Figure 5), in vitro Mφ derived from heparanase-null mice (KO-Mφ), differently from Mφ derived from wt mice (wt-Mφ), failed to upregulate ERα in murine (E0771, Figure 6A) and human (T47D, Figure 6B) BC cell lines following SFA stimulation. On the other hand, in vitro treatment with recombinant heparanase enzyme prior to stimulation by SFA (mimicking overexpression of the enzyme in obesity-associated breast tumors [30]) resulted in a 4-fold increased ability of SFA-stimulated Mφ to upregulate ERα expression in E0771 and MCF7 cells (Supplementary Figure S4). Corroborating the above findings in the clinical setting, immunohistochemical analysis of ER + BC tissue specimens derived from 58 patients revealed that heparanase positivity significantly correlated with higher levels of ERα expression (quantified as in [54] in the tumor tissue): a nearly 2-fold higher proportion of high ERα expression was detected in heparanase-positive versus heparanase-negative tumors (73% vs. 40%, chi-square test p = 0.0135, Figure 7).      Ref. [53]. The intensity of ERα staining was scored as weak (1), moderate (2), or strong (3), as described in Methods; tumors with staining score 1 were categorized as "low ER" (black bars); tumors with staining score 2 and 3 as "high ER" (grey bars). Chi-squared analysis was then used to assess the relationship between heparanase positivity and high versus low ER levels. Significant correlation between expression of heparanase and high ER levels was noted:an almost 2-fold higher proportion of high ER expression was detected in heparanase-positive versus heparanase-negative tumors (73% vs 40%, chi-square test * p = 0.0135).

Discussion
Breast tumors remain a major cause of morbidity/ mortality worldwide, despite remarkable progress in the diagnosis/endocrine therapy of ER+BC [28]. Approximately 30% of patients with hormone-responsive BC develop resistance. Moreover, the alarming global increase in the prevalence of obesity [85,86], a well-characterized risk factor for both incidence and worse prognosis of ER+/luminal BC subtypes, threatens to limit the progress in the management of this disease, highlighting the need for further in-depth characterization of obesity related ER-driven tumorigenesis.
The diverse spectrum of obesity effects in mammary tumorigenesis, as well as association between obesity and ER+ tumors predominantly in postmenopausal patients [13,14,26,27] (in which circulating levels of the hormone drop after cessation of ovarian estrogen production), are far from being fully characterized [41,42]. The ER+BC promoting action of obesity, mainly in postmenopausal women, has been attributed to increase in circulating [87] and adipose tissue levels of estrogen, as a result of upregulation of aromatase enzyme in fat depots of obese patients [13,16,18,30,41,42]. Yet, postmenopausal circulating levels of the hormone in these patients appear to be insufficient to activate signaling through ERα expressed in the uterine tissue [88] (as endometrial thickening does not occur in the majority of the cases [42]). This notion implies that, in addition to Figure 7. Sections of human ER + BC tissue samples (n = 58) were processed for immunohistochemistry with anti-ERα and anti-heparanase (Hpse) antibodies, as described in Methods. Inset: representative images of heparanase-positive (top) and heparanase-negative (bottom) breast tumor tissue specimens (invasive ductal carcinoma), original magnification ×200. To define tumor as Hpsepositive, a cutoff point of 25% immuno-stained tumor cells was used, as in Ref. [53]. The intensity of ERα staining was scored as weak (1), moderate (2), or strong (3), as described in Methods; tumors with staining score 1 were categorized as "low ER" (black bars); tumors with staining score 2 and 3 as "high ER" (grey bars). Chi-squared analysis was then used to assess the relationship between heparanase positivity and high versus low ER levels. Significant correlation between expression of heparanase and high ER levels was noted:an almost 2-fold higher proportion of high ER expression was detected in heparanase-positive versus heparanase-negative tumors (73% vs. 40%, chi-square test * p = 0.0135).
Of note, although heparanase is expressed by both Mφ [46][47][48][49] and BC cells [52] numerous observations have repeatedly indicated that in the setting of breast tumors, carcinoma cells per se appear to represent the major cellular source of the enzyme (reviewed in [52]). Given the above notion, the secreted nature of heparanase, and the role of extracellular heparan sulfate in modulation of TLR signaling ( [80][81][82]), it is plausible that in the BC under obese conditions the excessive enzyme overexpressed by the tumor cells is a main contributor to adverse activation of macrophages. Collectively, these data highlight a novel mechanism through which interplay between elements of the obesity-associated BC microenvironment, both cellular (i.e., Mφ, upregulating ERα expression; carcinoma cells, supplying heparanase) and extracellular (obese milieu components, HS), influences estrogen sensitivity-a critical component in hormone-related cancer progression and resistance to endocrine therapy.

Discussion
Breast tumors remain a major cause of morbidity/mortality worldwide, despite remarkable progress in the diagnosis/endocrine therapy of ER + BC [28]. Approximately 30% of patients with hormone-responsive BC develop resistance. Moreover, the alarming global increase in the prevalence of obesity [85,86], a well-characterized risk factor for both incidence and worse prognosis of ER+/luminal BC subtypes, threatens to limit the progress in the management of this disease, highlighting the need for further in-depth characterization of obesity related ER−driven tumorigenesis.
The diverse spectrum of obesity effects in mammary tumorigenesis, as well as association between obesity and ER+ tumors predominantly in postmenopausal patients [13,14,26,27] (in which circulating levels of the hormone drop after cessation of ovarian estrogen production), are far from being fully characterized [41,42]. The ER+BC promoting action of obesity, mainly in postmenopausal women, has been attributed to increase in circulating [87] and adipose tissue levels of estrogen, as a result of upregulation of aromatase enzyme in fat depots of obese patients [13,16,18,30,41,42]. Yet, postmenopausal circulating levels of the hormone in these patients appear to be insufficient to activate signaling through ERα expressed in the uterine tissue [88] (as endometrial thickening does not occur in the majority of the cases [42]). This notion implies that, in addition to increased estrogen levels, the extent of ERα expression may serve as an important determinant of estrogen-driven BC progression under obese conditions.
Here we showed that tumor-associated macrophages (TAMs) can augment estrogen signaling in BC cells through upregulation of ERα expression. We found that the obese state upregulated ERα expression in a murine model of obesity-associated BC (Figure 1), and demonstrated contribution of Mφ, adversely-activated by obesogenic substances (e.g., SFA) through TLR4-dependent mechanism, to ERα upregulation in mouse/human BC cells (Figures 2 and 3; Supplementary Figure S2). We also showed that Mφ-mediated increase of ERα levels in BC cells influenced estrogen sensitivity ( Figure 4)-a critical component in hormone-related cancer progression and response to endocrine therapy [7]. Indeed, ERα is the key driver of breast tumorigenesis in~75% of cases, and, at the same time, higher ERα load is often considered as predicting better prognosis in the general population of systemically-treated ER + BC patients, which is typically explained by the ability to effectively target ERα by existing hormonal therapy. Noteworthy is the fact that a recent study based on a cohort of >30,000 BC patients found no clear evidence for an association between higher ER load and better outcome [89]. In this respect, our present findings suggest that in the sub-population of obese postmenopausal ER+BC patients (especially in the face of increased estrogen production in fat depots [13,16,20,29,32]), upregulation of ERα expression by the tumor may promote BC progression and render it less sensitive to standard therapeutic approaches (i.e., estrogen deprivation (Figure 4), and ionizing radiation (Supplementary Figure S1C)).
Mφ infiltrating the tumor tissue or adjacent adipose tissue have been shown to contribute to several important features of malignant growth, and correlate with poor prognostic signs (i.e., high tumor grade, high mitotic activity, and metastasis [17,[90][91][92]). Yet, the emerging role of Mφ (and other tumor-residing immunocytes) in the regulation of steroid receptor expression in hormone-responsive tumors (i.e., through secretion of cytokines capable of inducing ERS1 gene expression) only recently came to be appreciated [43,44,69]. When considered in the context of hormone-dependent tumors, our findings suggest a bidirectional action of Mφ in obesity-associated breast tumorigenesis, through which TAMs augment estrogen signaling in BC via upregulation of ERα (present study), while adipose tissue Mφ contribute to this augmentation through an increase in estrogen levels (via induction of aromatase expression in stromal cells of 'obese' adipose tissue [16,30,41,42]). Interesting to note, Mφ, per se, reportedly express ER and estrogen can affect some of the macrophage functions; yet, the role of ER signaling in the control of the Mφ reactivity remains poorly understood, and both findings supporting the activating and suppressing effects of the hormone on Mφ have been published [93].
Taken together, the above findings suggest beneficial effect(s) of approaches aimed at manipulating Mφ in BC, especially when taken together with previous reports attesting to Mφ being an attractive target in mammary tumorigenesis, in general, and obesity-accelerated BC progression, in particular [16,30,33,34,94]. Yet, initial clinical studies (utilizing various agents decreasing Mφ content) demonstrated mixed results [95,96]. The likely explanation for the somewhat limited benefit of the direct Mφ targeting approach is provided by the dynamic nature and highly heterogeneous phenotypes of Mφ [92,97]. This notion is particularly relevant in the context of obesity: a recent report revealed that 'obese' visceral adipose tissue contains seven distinct Mφ populations with potentially different functions/cytokine profiles [98]. Given this complexity, our findings pointing to the ability of heparanase to dictate the ER−inducing phenotype of Mφ in BC (Figures 5 and 6), and, therefore, control one of the fundamental features of hormone-dependent breast tumorigenesis, namely, the level of ER expression, suggest that modulation of the enzyme activity (rather than targeting Mφ per se [95]) may offer an appealing alternative approach to uncouple obesity and breast cancer progression in a rapidly growing population of obese patients. Of note, a recent study (based on microarray datasets comprising >7000 BC patients, as well as prospective data from the BIG 2-98 repository [52]) demonstrated that overexpression of heparanase in BC tumors was associated with worse outcome/increased risk of recurrence in luminal (ER+) subtypes, but did not affect outcome/risk in non-luminal (ER−) subtypes [52]. This pattern of association is essentially similar to one reported for obesity, per se, which correlates with worse prognosis in BC patients bearing luminal tumors, while no such association was found in non-luminal tumors [13,14,26,27], and further supports the contribution of the enzyme to BC-promoting effects of obesity through Mφ-mediated upregulation of ER.
Collectively, the data presented here reveal previously unknown crosstalk occurring between tumor, adipose and immune compartments in hormone-responsive obesity-associated BC. Moreover, as techniques to effectively screen for the Mφ infiltration/heparanase expression in tumor tissue become available, and heparanase-targeting approaches are pre-clinically and clinically developed [99], our findings provide a basis for further studies aimed at manipulating estrogen signaling and suppressing breast cancer-promoting consequences of excess adiposity in an increasingly obese population.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cells11182844/s1, Figure S1: Effect of HFD-induced obesity on tumor growth, radiotherapy resistance and macrophage infiltration in murine breast carcinoma model; Figure S2: 4T1 murine BC cells; Figure S3: E0771 (left) and MCF7 (right) cells were either remained untreated (Cont.) or incubated (24 h, 37 • C) with LPS or medium conditioned by macrophages stimulated by LPS at concentration 0.1 ng/mL (Mφ·LPS). ERα protein levels were determined by immunoblotting (top) and quantified using ImageJ software (bottom). Intensity ratio for ERα/GAPDH is shown. The data are representative of at least 3 independent experiments; error bars represent ±SD. * p < 0.007 Figure S4: Heparanase augments ability of sFA-stimulated macrophages to upregulate ERα expression in BC cells.  Data Availability Statement: All data generated or analyzed during this study are included in this article.

Conflicts of Interest:
The authors declare no potential conflicts of interest.