Modeling Malignant Bone Disease Using Microgravity Bioreactors, Genetically Engineered Cell Lines, Patient-Derived Tumor Xenografts and Stem Cells

Transcriptome data from mesenchymal stem cells cultured in osteogenic conditions on monolayers of microcarriers coated with collagen I or extracellular matrix derived from osteogenically enhanced human mesenchymal stem cells (hereafter referred to as hMatrix) Two million GFP-labeled hMSCs were attached to collagen I- or ECM-coated beads with a combined growth area of 50 cm2. The loaded beads were then suspended in 10 mL of CCM and transferred into the RWV culture system. After 48 hours of equilibration, the CCM was removed and replaced with OEM so as to generate OEhMSCs. The cultures were incubated for 8 days with changes of media every 2 days. Triplicate cultures for collagen-I coated and ECM coated beads were performed. After 8 days, the cell-laden beads were recovered by brief centrifugation, washed in PBS and subjected to total RNA purification. Resultant total RNA yields ranged from 4.5-13.5 ug at a concentration of 150-450 ug mL-1 with OD260/280 ratios ranging from between 1.85-2.0. Thereafter, sample preparation and data acquisition was performed by the UT Southwestern Genomics and Microarray Core Facility. RNA integrity was confirmed by analysis using an Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA) and RNA integrity scores were 10 in each case. Biotin-UTP labeled antisense copy RNA (cRNA) was generated from 200 ng of total RNA using a commercially available kit (Illumina TotalPrep RNA Amplification Kit, Life Technologies, Carlsbad, CA). Labeled cRNA was hybridized to a HumanHT-v4.0 Expression BeadChip (Illumina, San Diego, CA) and visualized with biotin-Cy3 (Amersham, Piscataway, NJ, USA). Chips were read on an Illumina Hiscan scanner and analyzed according to standard manufacturer’s protocols using GenomeStudio version 3 (Illumina). Background correction, quality control, and quantile normalization were performed in accordance with Illumina standard operating procedures. Mean normalized fluorescent intensities and standard deviations were calculated for each transcript using biological triplicates for each condition. Data for a given transcript were excluded if the standard deviations exceeded 0.25 of the mean. Linear fold changes were calculated between type I collagen and ECM coatings using mean intensity values and lists were compiled of those genes that were upregulated by 2 fold or higher.