Identification of lysyl oxidase as an adipocyte-secreted mediator that promotes a partial mesenchymal-to-epithelial transition in MDA-MB-231 cells.

Aim: Breast cancer (BC) is the most common cancer in women worldwide, where adiposity has been linked to BC morbidity. In general, obese premenopausal women diagnosed with triple-negative BC (TNBC) tend to have larger tumours with more metastases, particularly to the bone marrow, and worse prognosis. Previous work using a 3-dimensional (3D) co-culture system consisting of TNBC cells, adipocytes and the laminin-rich extracellular matrix (ECM) trademarked as Matrigel, demonstrated that adipocytes and adipocyte-derived conditioned media (CM) caused a partial mesenchymal-to-epithelial transition (MET). Given that MET has been associated with secondary tumour formation, this study sought to identify molecular mediators responsible for this phenotypic change. Methods: Adipocytes were cultured with and without Matrigel, where semi-quantitative proteomics was used to identify proteins whose presence in the CM was induced or enhanced by Matrigel, which were referred to as adipocyte-secreted ECM-induced proteins (AEPs). The AEPs identified were assessed for association with prognosis in published proteomic datasets and prior literature. Of these, 4 were evaluated by the reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and enzyme-linked immunosorbent assay (ELISA), followed by a functional and MET marker analysis of 1 AEP on MDA-MB-231 cells grown on Matrigel or as monolayers. Results: The 4 AEPs showed a positive correlation between protein expression and poor prognosis. RT-qPCR analysis reported no significant change in AEPs mRNA expression. However, lysyl oxidase (LOX) was increased in CM of ECM-exposed adipocytes. Recombinant LOX (rLOX) caused the mesenchymal MDA-MB-231 TNBC cells to form less branched 3D structures and reduced the expression of vimentin. Conclusions: The data suggest that adipocyte-secreted LOX changes the mesenchymal phenotype of BC cells in a manner that could promote secondary tumour formation, particularly at sites high in adipocytes such as the bone marrow. Future efforts should focus on determining whether targeting LOX could reduce BC metastasis in obese individuals.

Three-Dimensional Co-Culture Method for Studying Interactions Between Adipocytes, Extracellular Matrix, and Cancer Cells.

Three-dimensional (D) culture models are increasingly becoming the model of choice for studying different biological phenomena such as cell-cell interaction, drug resistance, and gene expression. These models include extracellular matrix (ECM) proteins that better model the in vivo conditions as it allows cells to have both cell-cell and cell-ECM contacts. In the context of the tumor microenvironment, there are additional types of cells present in addition to the ECM. Thus, an intermediate between 2D cell culture and in vivo mouse models can be desired to interrogate the interactions between multiple cell types under the influence of the ECM. Here we describe a 3D co-culture technique for studying breast cancer-adipocyte interactions. This technique could easily be modified to analyze interactions between other cancer cell types and different fibroblast-like cells.

Adipose Tissue from Lean and Obese Mice Induces a Mesenchymal to Epithelial Transition-Like Effect in Triple Negative Breast Cancers Cells Grown in 3-Dimensional Culture.

Breast cancer is the second leading cause of cancer-related mortality among women globally with obesity being one risk factor. Obese breast cancer patients have at least a 30% increased risk of death from breast cancer compared to non-obese breast cancer patients because they present with larger tumors and generally have increased rates of metastasis. Moreover, obese breast cancer patients respond more poorly to treatment compared to non-obese patients, particularly pre-menopausal women diagnosed with triple negative breast cancer (TNBC). To help understand the molecular mechanisms underlying the increased metastasis associated with obesity, we previously established a three-dimensional culture system that permits the co-culture of adipocytes and TNBC cells in a manner that mimics an in vivo milieu. Using this system, we demonstrate that white adipose tissue from both lean and obese mice can induce a partial mesenchymal-to-epithelial transition (MET). Triple negative breast cancer cells adopt an epithelial morphology and have an increased expression of some epithelial markers, but they maintain the expression of mesenchymal markers, furnishing the breast cancer cells with hybrid properties that are associated with more aggressive tumors. Thus, these data suggest that adipose tissue has the potential to promote secondary tumor formation in lean and obese women. Further work is needed to determine if targeting the partial MET induced by adipose tissue could reduce metastasis.

Adipocytes in the Tumour Microenvironment.

Adipose tissue contribution to body mass ranges from 6% in male athletes to over 25% in obese men and over 30% in obese women. Crosstalk between adipocytes and cancer cells that exist in close proximity can lead to changes in the function and phenotype of both cell types. These interactions actively alter the tumour microenvironment (TME). Obesity is one of the major risk factors for multiple types of cancer, including breast cancer. In obesity, the increase in both size and number of adipocytes leads to instability of the TME, as well as increased hypoxia within the TME, which further enhances tumour invasion and metastasis. In this chapter, we will discuss the diverse aspects of adipocytes and adipocyte-derived factors that affect the TME as well as tumour progression and metastasis. In addition, we discuss how obesity affects the TME. We focus primarily on breast cancer but discuss what is known in other cancer types when relevant. We finish by discussing the studies needed to further understand these complex interactions.

A Novel 3-Dimensional Co-culture Method Reveals a Partial Mesenchymal to Epithelial Transition in Breast Cancer Cells Induced by Adipocytes.

Cancer metastases are accountable for almost 90% of all human cancer related deaths including from breast cancer (BC). Adipocytes can alter the tumor microenvironment, which can promote metastasis by inducing an epithelial-to-mesenchymal transition (EMT) in BC cells. However, the role of adipocytes during the mesenchymal-to-epithelial transition (MET), that can be important in metastasis, is not clear. To understand the effect of adipocytes on the BC progression, there is a requirement for a better in vitro 3-dimensional (3D) co-culture system that mimics the breast tissue and allows for more accurate analysis of EMT and MET. We developed a co-culture system to analyze the relationship of BC cells grown in a 3D culture with adipocytes. We found that adipocytes and adipocyte-derived conditioned media, but not pre-adipocytes, caused the mesenchymal MDA-MB-231 and Hs578t cells to form significantly more epithelial-like structures when compared to the typical stellate colonies formed in control 3D cultures. SUM159 cells and MCF7 cells had a less dramatic shift as they normally have more epithelial-like structure in 3D culture. Biomarker expression analysis revealed that adipocytes only induced a partial MET with proliferation unaffected. In addition, adipocytes had reduced lipid droplet size when co-cultured with BC cells. Thus, we found that physical interaction with adipocytes and ECM changes the mesenchymal phenotype of BC cells in a manner that could promote secondary tumor formation.

CD24 is required for regulating gene expression, but not glucose uptake, during adipogenesis.

Adipose tissue dysfunction in obesity and lipodystrophy results in major health complications such as heart disease and stroke, and is associated with an increased risk of some cancers. We have previously found that the cell surface receptor CD24 regulates adipogenesis as measured by lipid accumulation and gene expression in mature adipocytes. How CD24 regulates these processes remains unknown. To begin answering this question, we first determined that CD24 does not affect glucose uptake in differentiating adipocytes in vitro. We then examined changes in global gene expression via DNA microarray in 3T3-L1 adipocytes with siRNA-mediated knock-down of CD24 expression. We found that CD24 expression is necessary for upregulation of up to 134 genes. We validated the CD24-mediated regulation of 4 of these genes during in vitro adipogenesis of 3T3-L1 and primary cells isolated from the inguinal white adipose tissue depots of CD24 knockout mice. Surprisingly, we found that only 1 of these genes was also regulated by CD24 in cells from the epididymal depot. Overall, these data suggest that CD24 is necessary for select gene expression in a depot-specific manner during adipogenesis in vitro. These findings could help elucidate the mechanisms regulating lipid accumulation in adipocytes thereby aiding in the development of novel treatment strategies for obesity and lipodystophy.

Collagen overlays can inhibit leptin and adiponectin secretion but not lipid accumulation in adipocytes.

BACKGROUND: White adipose tissue (WAT) is essential for energy storage as well as being an active endocrine organ. The secretion of adipokines by adipocytes can affect whole body metabolism, appetite, and contribute to overall health. WAT is comprised of lipid-laden mature adipocytes, as well as immune cells, endothelial cells, pre-adipocytes, and adipose-derived stem cells. In addition, the presence of extracellular matrix (ECM) proteins in WAT can actively influence adipocyte differentiation, growth, and function. Type I collagen is an abundant fibrous ECM protein in WAT that is secreted by developing adipocytes. However, the extent and overall effect of Type I collagen on adipokine secretion in mature adipocytes when added exogenously has not been established. METHODS: We characterized the effects of Type I collagen overlays prepared using two different buffers on adipocyte physiology and function when added at different times during differentiation. In addition, we compared the effect of collagen overlays when adipocytes were cultured on two different tissue culture plastics that have different adherent capabilities. Triglyceride accumulation was analyzed to measure adipocyte physiology, and leptin and adiponectin secretion was determined to analyze effects on adipokine secretion. RESULTS: We found that collagen overlays, particularly when added during the early differentiation stage, impaired adipokine secretion from mature adipocytes. Collagen prepared using PBS had a greater suppression of leptin than adiponectin while collagen prepared using HANKS buffer suppressed the secretion of both adipokines. The use of CellBind plates further suppressed leptin secretion. Triglyceride accumulation was not substantially impacted with any of the collagen overlays. DISCUSSION: Adipokine secretion can be selectively altered by collagen overlays. Thus, it is feasible to selectively manipulate the secretion of adipokines by adipocytes in vitro by altering the composition or timing of collagen overlays. The use of this technique could be applied to studies of adipokine function and secretion in vitro as well as having potential therapeutic implications to specifically alter adipocyte functionality in vivo.

Analysis of the structure, evolution, and expression of CD24, an important regulator of cell fate.

BACKGROUND: CD24 is a small, glycophosphatidylinositol-anchored cell surface receptor, expressed in a variety of cells types and tissues. CD24 gene and protein expression is highly dynamic in response to cellular differentiation and stimulation in a cell-specific manner. Furthermore, CD24 interacts with a diverse collection of ligands, including cell adhesion molecules such as P-selectin, and the immune-associated siglec family of transmembrane proteins. While much is known regarding the biological roles of CD24 in regulating cell survival, death and differentiation, little is known about the evolution and organization of CD24 across species or the relationship between CD24 expression and its known ligands. RESULTS: We analyzed the organization and evolution of the CD24 gene from 56 mammalian, avian and reptilian species. We further examined the mRNA expression of CD24 and its known ligands in Mus musculus in immune cells, immunologically privileged tissues, developing brain, and developing and regenerating liver. CD24 arose prior to the reptilian-avian divergence and is conserved across many mammalian species, although we were unable to identify CD24 in marsupials or monotremes. The CD24 genomic structure is diverse between and within species, with varying numbers of exons, introns, and the presence of untranslated regions. Of note, we found no obvious criteria distinguishing CD24 genes from those annotated as CD24-like. The expression of CD24 is similarly complex, with immune cells showing dynamic changes in mRNA levels during development, while immunologically privileged and developing tissues show a high, static expression level that decreases in mature tissues. Furthermore, the expression of CD24 correlated with some but not all of its known ligands in a tissues-specific manner, suggesting that novel ligands have yet to be identified and that cell-specific ligand expression can influence CD24 function. CONCLUSIONS: We find that CD24 arose prior to the divergence of reptiles, birds and mammals. Furthermore, the most highly conserved areas of the protein are the amino acids which can be glycosylated. We also find that CD24 expression is highly tissue-specific and in many cases, not well conserved with known CD24 ligands, suggesting yet-unknown CD24-ligand interactions. Together, these data are a valuable resource for furthering studies in CD24 biology.

Repression of CD24 surface protein expression by oncogenic Ras is relieved by inhibition of Raf but not MEK or PI3K.

CD24 is a dynamically regulated cell surface protein. High expression of CD24 leads to progression of lung, prostrate, colon, and pancreatic cancers, among others. In contrast, low expression of CD24 leads to cell proliferation and metastasis of breast cancer stem cells (BCSCs). Activating mutations in Ras are found in 30% of all human cancers. Oncogenic Ras constitutively stimulates the Raf, PI3K, and Ral GDS signaling pathways, leading to cellular transformation. Previous studies have shown that expression of oncogenic Ras in breast cancer cells generates CD24(-) cells from CD24(+) cells. However, the molecular mechanisms involved in the generation of CD24(-) cells were not determined. Here, we demonstrate that oncogenic Ras (RasV12) expression suppresses CD24 mRNA, protein, and promoter levels when expressed in NIH/3T3 cells. Furthermore, activation of only the Raf pathway was sufficient to downregulate CD24 mRNA and protein expression to levels similar to those seen in with RasV12 expression. In contrast, activation of the PI3K pathway downregulated mRNA expression with a partial effect on protein expression whereas activation of the RalGDS pathway only partially affected protein expression. Surprisingly, inhibition of MEK with U0126 only partially restored CD24 mRNA expression but not surface protein expression. In contrast, inhibition of Raf with sorafenib did not restore CD24 mRNA expression but significantly increased the proportion of RasV12 cells expressing CD24. Therefore, the Raf pathway is the major repressor of CD24 mRNA and protein expression, with PI3K also able to substantially inhibit CD24 expression. Moreover, these data indicate that the levels of CD24 mRNA and surface protein are independently regulated. Although inhibition of Raf by sorafenib only partially restored CD24 expression, sorafenib should still be considered as a potential therapeutic strategy to alter CD24 expression in CD24(-) cells, such as BCSCs.

Dynamic upregulation of CD24 in pre-adipocytes promotes adipogenesis.

The development of mature adipocytes from pre-adipocytes is a highly regulated process. CD24 is a glycophosphatidylinositol-linked cell surface receptor that has been identified as a critical cell surface marker for identifying pre-adipocytes that are able to reconstitute white adipose tissue (WAT) in vivo. Here, we examined the role and regulation of CD24 during adipogenesis in vitro. We found that CD24 mRNA and protein expression is upregulated early during adipogenesis in the 3T3-L1 pre-adipocytes and in murine primary pre-adipocytes isolated from subcutaneous and visceral WAT, followed by downregulation in mature adipocytes. CD24 mRNA expression was found to be dependent on increased transcription due to increased promoter activity in response to activation of a pre-existing transcriptional regulator. Furthermore, either intracellular cAMP or dexamethasone were sufficient to increase expression in pre-adipocytes, while both additively increased CD24 expression. Preventing the increase in CD24 expression, by siRNA-mediated knock-down, resulted in fewer mature lipid-laden adipocytes and decreased expression of mature adipogenic genes. Therefore, conditions experienced during adipogenesis in vitro are sufficient to increase CD24 expression, which is necessary for differentiation. Overall, we conclude that the dynamic upregulation of CD24 actively promotes adipogenesis in vitro.