Growth curve Two sets of WT

Growth curve — Two sets of WT and Ptp4a3-KO Go 6976 were independently compared using the same number of starting cells (1.2×106) on day in a 24-well plate seeded with feeder cells. Wells were analyzed by flow cytometry in triplicate on days 1, 4, 8 and 11 to determine the absolute number of EpCAM+ cells in a defined volume using the MACSQuant™ cell analyzer. Triplicates were averaged and plotted against time on a line graph to reflect growth rate. Quantitative RT-PCR — Gene expression levels were assayed by quantitative RT-PCR. Total RNA was isolated from tumor cells and tissues with TRIzol reagent (Invitrogen) and resuspended in nuclease-free water. First strand synthesis was performed with the iScript cDNA synthesis kit (Bio-rad) using 1μg of total RNA. Gene expression was assayed with 2xSYBR Green Mastermix (Bio-rad) with real-time detection performed with the iCycler thermocycler (Bio-Rad). PCR was performed by incubating at 95°C for 5min, followed by 39cycles of 95°C for 30s, 58°C for 30s, 72°C for 1min. The Ct values for determining relative gene expression were normalized to β−actin as an endogenous control. The following primer sets were used for each target gene: Ptp4a3 (F-CGGGATGAAGTACGAGGACG, R-GCGTGTGTGGGTCTTTGAAC), Muc2 (F-AGTCTGCTCTGTGAAGTGCC, R-GGCAAACACAGTCCTTGCAG), β-actin (F-TCATGAAGTGTGACGTTGACATCCGT, R-CCTAGAAGCATTTGCGGTGCACGATG). Immunohistochemistry — Tissues were first isolated and rinsed in ice-cold phosphate buffered saline (PBS). Samples were immediately submerged in 10% neutral buffered formalin, incubated at room temperature overnight, and dehydrated through an ethanol gradient. Fixed tissues were then embedded in paraffin and sectioned onto glass slides. Sections were deparaffinized and rehydrated prior to staining. For chamber slides, slides were rinsed in PBS, fixed in 4% paraformaldehyde, rinsed again and frozen prior to staining. For both tissue and chamber slides, samples were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich, MO) for 10min, blocked with 5% Bovine Serum Albumin (BSA) for 30min, and stained with hematoxylin & eosin or primary antibody against Mucin2 (MUC2) (Santa Cruz, TX) in PBS containing 0.5% BSA (Fisher Scientific, PA) for one hour followed by anti-rabbit Alexa Fluor secondary (Life Technologies, NY) for 30min followed by counterstain with Hoechst 33342. Images were taken using an IX71 inverted microscope (Olympus, PA). Subcutaneous tumor growth — Subcutaneous injection of tumor cells allowed assessment of tumorigenic potential after in vitro expansion. Expanded tumor cells (passage seven) were washed, trypsinized, resuspended in cold HBSS/Matrigel (BD Biosciences, CA) 1:1 and stored on ice. Nude mice were injected with either 5×105 (n=3) or 1×106 (n=3) cultured tumor cells subcutaneously. Mice were injected in both flanks with WT (left) and Ptp4a3-KO (right) tumor cells and monitored for tumor formation. Mice were sacrificed at 18weeks post-injection of cells or the point when tumors reached 20mm in diameter. Statistics — Assays for quantitative RT-PCR and cell cycle analysis were quantified and analyzed statistically by two-tailed T-test, and the growth curve was analyzed by repeated measures ANOVA. In both cases, significance was defined as p<0.05.
Results and discussion
Conclusions In this report we have adapted a powerful technique used to study human solid tumor-initiating cells for use in a mouse model system. Using a clonogenic assay, we showed that the colony forming activity of both primary and expanded Ptp4a3-KO tumor cells is lower than their WT tumor cell counterparts. Colonies are formed by cells with self-renewal potential, therefore, the decrease in frequency of colony-forming tumor cells from Ptp4a3-KO mice is likely due to a defect in self-renewal potential. Furthermore, previous reports have demonstrated that increased CFU frequency is associated with tumorigenicity and metastatic potential (Huang et al., 2012; Chen et al., 2013). While we observed that both expanded WT and Ptp4a3-KO tumor cells shared the characteristics of tumor-initiating cells, with nearly identical expression of cell surface markers assayed, Ptp4a3-KO tumor cells were significantly deficient in clonogenicity and tumorigenesis.