The role of hypertrophic adipocytes in breast cancer progression

Introduction:: Obese breast cancer patients have a 1.8-fold increased risk of mortality compared to normal weight women and over 40% of adult women in the U.S. are obese. Previous research has identified that obesity-associated changes of fibroblasts, immune cells, and the interstitial extracellular matrix (ECM) of white adipose tissue (WAT) affect this process. However, how obesity-associated changes of adipocytes, the primary cell type composing WAT, affect breast cancer malignancy remains less clear.



Adipose tissue expands in obese individuals through a combination of adipocyte hyperplasia (increase in number) and hypertrophy (increase in size). In particular, adipocyte hypertrophy has been shown to correlate with breast cancer progression, but the underlying mechanisms remain unclear as primary adipocytes are buoyant and post-mitotic, making them difficult to study in culture. Moreover, limited methods currently exist that allow characterizing the biochemical and biophysical properties of adipocytes as a function of their size. For these reasons, prior studies with tumor cells often leveraged preadipocytes differentiated in vitro. However, these cells fail to recapitulate key aspects of mature adipocytes including the ability to become hypertrophic.



Here, we present novel methods to separate primary adipocytes by size to characterize and culture these sorted populations ex vivo. Using these methods, we investigate how the altered biochemical and biophysical properties of hypertrophic adipocytes impact breast cancer cell behavior and disease progression via ex vivo coculture studies and computational simulations of tumor invasion.


Materials and Methods:: For studies with murine tissue, WAT was isolated from lean or obese (genetic or diet-induced) female C57BL/6 mice. For studies with human tissue, deidentified WAT from women undergoing reduction mammoplasties at Weill Cornell Medicine was used. Isolated WAT was minced, incubated in collagenase, and passed through cell strainers to obtain a cell suspension. The resulting cells were washed and separated by buoyancy to enrich for adipocytes. For experiments requiring size-sorted cells, adipocytes were further stratified via differential buoyancy using a separatory funnel.



For coculture studies, packed adipocytes were plated on the bottom of tissue culture inserts and then inverted into wells pre-seeded with tumor cells. Studies were performed for 72 hours before breast cancer cells were collected and reseeded for endpoint analysis, including changes in lipid content (BODIPY staining), proliferation (EdU incorporation), and migration (Boyden chamber assay).



For nanoindentation studies, primary adipocytes were attached to glass coverslips functionalized with Cell-Tak. Coverslips were then used for image-guided mechanical characterization via an Optics11 Life Chiaro Nanoindenter. For all experiments, a probe with a radius of 8.5 μm and spring constant of 0.032 N/m was used.



For computational studies, experimental data was used to parameterize discrete element method (DEM) simulations of invasion which model adipocytes as deformable polyhedra with shape-based energy functions that include cytoplasm incompressibility, membrane surface tension, substrate adhesion, and volume exclusion. Tumor cells were modeled as motile polyhedra with cell-cell adhesions and a migration strategy that uses cell shape changes.


Results, Conclusions, and Discussions:: Genetic and diet-induced models of obesity in mice (Fig. 1A) recapitulate the hypertrophic expansion of WAT seen in humans associated with greater body mass index (Fig. 1B). In addition to increased cell volume, hypertrophic adipocytes isolated from ob/ob or HFD mice are stiffer than non-hypertrophic cells isolated from donor-matched controls (Fig. 1C), which restricts tumor cell migration in simulations of invasion (Fig. 1D). These hypertrophic cells also demonstrate cortical actin remodeling into dense, bundled fibers (Fig. 1E) and bulk sequencing data from reduction mammoplasty patients indicate genes related to actin remodeling (ACTR2, ACTR3, PFN1) correlate with average adipocyte diameter (Fig. 1F). Collectively, these results suggest that obesity prompts hypertrophic expansion of a subset of adipocytes in WAT and that these hypertrophic cells become stiffer (potentially mediated by actin remodeling), which mechanically limits tumor invasion.



By leveraging their differential buoyancy, adipocytes from the same donor can be sorted into small (non-hypertrophic) and large (hypertrophic) populations (Fig. 1G). Coculture with large adipocytes promoted MDA-MB-231 EdU incorporation and migration to a greater extent than coculture with small adipocytes from the same donor (Fig. 1H). These differences may be explained by greater extracellular vesicle secretion by large adipocytes (Fig. 1I), as determined via nanoparticle tracking analysis, which can transfer neutral lipid to tumor cells for metabolic use (Fig. 1J). Collectively, these results suggest that altered paracrine factors secreted by hypertrophic adipocytes promote the proliferation and motility of cocultured breast cancer cells to a greater extent than factors secreted by non-hypertrophic cells.



Using the tools presented here, we investigated the properties of primary adipocytes whose characterization has historically been limited by insufficient protocols to manipulate these cells ex vivo. We found that hypertrophic adipocytes, which are more common in obesity, become stiffer and physically restrict simulated tumor invasion. Conversely, these cells secrete altered biochemical factors which promote breast cancer cell proliferation and migration when compared to non-hypertrophic adipocytes from the same donor. Continued studies based on these findings will help determine whether mammary adipocyte size could serve as a valuable prognostic biomarker for breast cancer patients.


Acknowledgements (Optional): :

References (Optional): :