New mechanistic insights of integrin β1 in breast cancer bone colonization.

Bone metastasis is a frequent and life-threatening complication of breast cancer. The molecular mechanisms supporting the establishment of breast cancer cells in the skeleton are still not fully understood, which may be attributed to the lack of suitable models that interrogate interactions between human breast cancer cells and the bone microenvironment. Although it is well-known that integrins mediate adhesion of malignant cells to bone extracellular matrix, their role during bone colonization remains unclear. Here, the role of β1 integrins in bone colonization was investigated using tissue-engineered humanized in vitro and in vivo bone models. In vitro, bone-metastatic breast cancer cells with suppressed integrin β1 expression showed reduced attachment, spreading, and migration within human bone matrix compared to control cells. Cell proliferation in vitro was not affected by β1 integrin knockdown, yet tumor growth in vivo within humanized bone microenvironments was significantly inhibited upon β1 integrin suppression, as revealed by quantitative in/ex vivo fluorescence imaging and histological analysis. Tumor cells invaded bone marrow spaces in the humanized bone and formed osteolytic lesions; osteoclastic bone resorption was, however, not reduced by β1 integrin knockdown. Taken together, we demonstrate that β1 integrins have a pivotal role in bone colonization using unique tissue-engineered humanized bone models.

Cell morphology and attachment assays BC cells were seeded at a density of 2500 cells/cm 2 (for morphology assessment) or 10000 cells/cm 2 (for attachment assessment) on TCP or hOBM. Cell spreading area and shape factor were analyzed after 24 hours of culture. The shape factor is a quantity used to describe the circularity of the cells and is calculated as 4π[area]/[perimeter] 2 . A value of 1 indicates a perfect circle and a value approaching 0 indicates an increasingly elongated shape. Cells were stained by immunofluorescence with 10 µg/ml human fibronectin-specific antibody (HFN 7.1, DSHB), followed by 10 µg/ml secondary fluorescently labeled antibody (anti-mouse Alexa 633, Cell signaling), 0.8 U/ml rhodamine-conjugated phalloidin and 5 µg/ml 4',6-diamidino-2-phenylindole (DAPI) and imaged using a Leica TCS SP5 confocal laser scanning microscope (Leica Microsystems, Mannheim, Germany) as previously described [1,2].
Quantification of spreading area and shape factor was performed using ImageJ (National Institutes of Health, Bethesda, Maryland, USA). To quantify attachment, cell layers were rinsed three times with PBS after an attachment period of 30 minutes, before plates were frozen at -80°C. Quantification of attached cells was performed by measuring the DNA content in each well using a Quant-iT™ PicoGreen assay (Life Technologies) according to the manufacturer's instructions. Three independent experiments were performed with at least three replicates for each condition.
Proliferation assays BC cells were seeded at a density of 2500 cells/cm 2 on TCP or hOBM. Cell proliferation was assessed by an Alamar Blue (Life Technologies) assay. At each time point, BC cells were incubated for 2 hours in culture medium containing 8% (v/v) Alamar Blue reagent. Alamar Blue added to cell-free wells served as a background controls. Supernatants were then transferred to a 96-well black plate (Corning Incorporated, Corning, NY, USA) and fluorescence signals (excitation 544 nm, emission 590 nm) were detected using a POLARstar OPTIMA plate reader (BMG LABTECH, Offenburg, Germany). Three independent experiments were performed with at least six replicates for each condition.
BC cell proliferation and colony formation in 3D hydrogels BC cells were embedded in 5% (w/v) gelatin methacrylate (gelMA) at a density of 7000 cells per 50 µl and gel crosslinking was performed under ultra-violet light exposure for 10 minutes, as previously described [3]. Cell proliferation was assessed over time by an Alamar Blue assay (Life Technologies). Three independent experiments were performed with at least 6 samples for each condition. After 2 weeks, cell-loaded hydrogels were fixed in 4% paraformaldehyde for 15 minutes, stained with rhodamine-conjugated phalloidin and 4',6diamidino-2-phenylindole (DAPI) and imaged using a Leica TCS SP5 confocal laser scanning microscope (Leica Microsystems, Mannheim, Germany) as previously described [3]. BC colony size was quantified using ImageJ.
Migration assays BC cells were seeded at a density of 15000 cells/ cm 2 on hOBM grown in 6-well plates and imaged with a Leica AF6000 LX wide-field fluorescence microscope (Leica Microsystems, Wetzlar, Germany) at 37°C and 5% CO 2 . Brightfield and fluorescence images were recorded in sequence every 16 minutes for a total period of 24 hours. Image stacks were analyzed in Image J as described previously [1] and the effective distance (start-to-end distance), total distance, instantaneous speed (average frame-to-frame speed) and directionality (effective distance/total distance) of cell migration were calculated. A minimum of 96 individual cells from a total of 12 different movies were analyzed for each condition.

qRT-PCR
For qRT-PCR analysis of tumor specimens, tissue homogenization was performed using 2 mm glass beads (Sigma-Aldrich) in a Mini-BeadBeater (Biospec Products), followed by RNA extraction using Trizol (Life Technologies) according to the manufacturer's instructions.  Table 2. qRT-PCR was performed on a 7900HT Fast Real-Time PCR System (Life Technologies) using a 384-well block module, as previously described [1]. Relative gene expression was calculated as 2^(ct geometric mean of housekeeping genes -ct gene of interest ). Ki67-positive cells in IHC stains. Quantification was performed using the ImageJ cell counter plug-in or the ImmunoRatio application. For the ImmunoRatio analysis representative images from the IHC stains and the corresponding pseudo-colored images with the segmented staining components are shown. Data are represented as mean ± standard error.