Acquired PIK3CA amplification causes resistance to selective phosphoinositide 3-kinase inhibitors in breast cancer.

Agents targeting the PI3K/mTOR signaling axis have shown promise in early-phase clinical trials and are currently being studied in later stages of clinical development in multiple indications. Experience with other targeted agents suggests that clinical responses may be short-lived because of acquired resistance to therapy. Here, we report preclinical modeling of acquired resistance in a HER2-positive, PIK3CA mutant breast cancer cell line, KPL-4. We identified a heretofore-unreported mechanism of resistance, specifically high-level amplification of the mutant allele of PIK3CA, which resulted in a marked upregulation of PI3K signaling, enabling resistant cells to regain proliferative capacity at clinically relevant concentrations of the PI3K inhibitor, GDC-0941. We show that knockdown of the amplified PIK3CA mutant allele in these cells by small interfering RNA restored pathway signaling and sensitivity to PI3K inhibition at levels comparable to parental cells. These novel preclinical findings suggest that, in addition to assessment of other previously reported mechanisms of resistance, evaluation of PI3K copy number variation should be integrated into the exploratory analysis of biopsies obtained at disease progression.

EGO-1 is related to RNA-directed RNA polymerase and functions in germ-line development and RNA interference in C. elegans.

BACKGROUND: Cell-fate determination requires that cells choose between alternative developmental pathways. For example, germ cells in the nematode worm Caenorhabditis elegans choose between mitotic and meiotic division, and between oogenesis and spermatogenesis. Germ-line mitosis depends on a somatic signal that is mediated by a Notch-type signaling pathway. The ego-1 gene was originally identified on the basis of genetic interactions with the receptor in this pathway and was also shown to be required for oogenesis. Here, we provide more insight into the role of ego-1 in germ-line development. RESULTS: We have determined the ego-1 gene structure and the molecular basis of ego-1 alleles. Putative ego-1 null mutants had multiple, previously unreported defects in germ-line development. The ego-1 transcript was found predominantly in the germ line. The predicted EGO-1 protein was found to be related to the tomato RNA-directed RNA polymerase (RdRP) and to Neurospora crassa QDE-1, two proteins implicated in post-transcriptional gene silencing (PTGS). For a number of germ-line-expressed genes, ego-1 mutants were resistant to a form of PTGS called RNA interference. CONCLUSIONS: The ego-1 gene is the first example of a gene encoding an RdRP-related protein with an essential developmental function. The ego-1 gene is also required for a robust response to RNA interference by certain genes. Hence, a protein required for germ-line development in C. elegans may be a component of the RNA interference/PTGS machinery.

High-throughput isolation of Caenorhabditis elegans deletion mutants.

The nematode Caenorhabditis elegans is the first animal whose genome is completely sequenced, providing a rich source of gene information relevant to metazoan biology and human disease. This abundant sequence information permits a broad-based gene inactivation approach in C. elegans, in which chemically mutagenized nematode populations are screened by PCR for deletion mutations in a specific targeted gene. By handling mutagenized worm growth, genomic DNA templates, PCR screens, and mutant recovery all in 96-well microtiter plates, we have scaled up this approach to isolate deletion mutations in >100 genes to date. Four chemical mutagens, including ethyl methane sulfonate, ethlynitrosourea, diepoxyoctane, and ultraviolet-activated trimethylpsoralen, induced detectable deletions at comparable frequencies. The deletions averaged approximately 1400 bp in size when using a approximately 3 kb screening window. The vast majority of detected deletions removed portions of one or more exons, likely resulting in loss of gene function. This approach requires only the knowledge of a target gene sequence and a suitable mutagen, and thus provides a scalable systematic approach to gene inactivation for any organism that can be handled in high density arrays.