inos inhibitor Whole exome sequencing of tumor

Whole-exome sequencing of tumor DNA from GBM 5 identified a total of 32 SNVs (Table S3). A subset of these SNVs was selected for single-cell analysis based on putative gene function and variant allele fractions encompassing high, low, and intermediate frequencies. The genomic targets selected included SNVs in KCNH5, PLCB2, GDF5, TRMT5, TP53, and PALB2, and CNAs in CDKN2A, TP53, and EGFR. Single-cell analysis for the simultaneous presence of six SNVs and three CNAs was carried out on flow-sorted neurosphere and xenograft tumor inos inhibitor from GBM 5 (a representative heatmap of the qPCR data from the BioMark HD is given in Figure S3). A comparison of the clonal phylogeny and subclonal architecture of neurosphere and xenograft cells is shown in Figure 4. Homozygous CDKN2A deletion, gain of EGFR (up to four copies), and KCN5, PLCB2, GDF5, and TRMT5 mutations all occurred early and were present in all subclones of the neurospheres (Figure 4A). Loss of one TP53 wild-type allele occurred after EGFR amplification of more than four copies. Heterozygous TP53 and PALB2 mutations occurred after further EGFR amplification. According to the chromosome 7 copy number as assessed by FISH (Figure S4) and single-cell data (not shown), EGFR gain was uncoupled from chromosome copy number at three or four copies of chromosome 7. Subsequently, there was an increasing gain of EGFR, consistent with the formation of extrachromosomal double minutes. All of the cells in the secondary mouse xenograft possessed all of the mutations, including heterozygous TP53 and PALB2 mutations, and most likely derived from two of the most evolved subclones in the neurospheres (being present in only 3.3% and 4%, respectively; Figure 4B). The subclone with one copy of TP53 mutant and more than ten copies of EGFR evolved further in the xenograft cells by acquiring two mutated copies of TP53, and all subclones evolved to show high-level amplification of EGFR (>100 copies) (Figure 4B). We used secondary transplantation as a more stringent measure of stem cell renewal (Dick et al., 1997). In five cases (GBM 1, GBM 5, GBM 8, GBM 9, and GBM 11), we observed a statistically shorter time to tumor formation in the secondary transplant than in the primary xenograft tumor. This pattern of evolution is consistent with the typical pattern of disease progression seen in patients and would be consistent with the genetically more evolved subclonal structure observed in the neurospheres of GBM 5, GBM 8, and GBM 11, and with the presence of TP53 mutations in GBM 5 and GBM 8. Clones with EGFR amplification consistently read out after serial transplantation, and usually further evolved with an incremental gain of more copies of EGFR. Other investigators have demonstrated a mosaic pattern of growth factor amplification in GBM tumors, with EGFR, MET, and PDGFRA gain occurring in distinct populations of cells (Snuderl et al., 2011; Szerlip et al., 2012). In the present study, there was one case (GBM 11) with subclones in the neurosphere cells that showed concurrent PDGFRA and EGFR gain in the same cell, as well as subclones with only EGFR gain (Figure 2). However, only the subclones with high-level EGFR amplification repopulated the mouse xenograft; none of the subclones with PDGFRA gain were present. These observations reveal the dynamic complexity of subclonal interactions in GBM and provide deeper insight into the role of PDGFRA. We previously showed that amplification of PDGFRA occurs in the midphase of GBM evolution (Sottoriva et al., 2013) rather than as a primary driver event. Evidence suggests that tumor-propagating clones may arise from a common precursor, with key early events including genetic alterations in EGFR, CDKN2A/B, and TP53 (Goodenberger and Jenkins, 2012; Snuderl et al., 2011; Sottoriva et al., 2013). These observations are supported by data from glioma susceptibility studies that revealed prominent roles for alterations in EGFR, CDKN2A, and TP53 in glioma evolution (Andersson et al., 2010; Egan et al., 2012; Shete et al., 2009; Stacey et al., 2011; Wrensch et al., 2009).