Runx1/AML1 is a transcription factor implicated in tissue stem cell regulation and belongs to the small Runx family of cancer genes. In the hair follicle (HF), Runx1 epithelial deletion in morphogenesis impairs normal adult hair homeostasis (cycle) and blocks adult hair follicle stem cells (HFSCs) in quiescence. Here, we show that these effects are overcome later in adulthood. By deleting Runx1 after the end of morphogenesis, we demonstrate its direct role in promoting anagen onset and HFSC proliferation. Runx1 deletion resulted in cyclin-dependent kinase inhibitor Cdkn1a (p21) upregulation. Interfering with Runx1 function in cultured HFSCs impaired their proliferation and normal G 0 /G1 and G 1 /S cell cycle progression. The proliferation defect could be rescued by Runx1 readdition or by p21 deletion. Chemically induced skin tumorigenesis in mice turned on broad Runx1 expression in regions of the skin epithelium, papillomas, and squamous cell carcinomas. In addition, it revealed reduced rates of tumor formation in the absence of Runx1 that were accompanied by decreased epithelial levels of phospho-Stat3. Runx1 protein expression was similar in normal human and mouse hair cycles. We propose that Runx1 may act as a skin oncogene by directly promoting proliferation of the epithelial cells.

IntroductionTriple-negative breast cancer (TNBC) is aggressive and lacks targeted therapies. Phosphatidylinositide 3-kinase (PI3K)/mammalian target of rapamycin (mTOR) pathways are frequently activated in TNBC patient tumors at the genome, gene expression and protein levels, and mTOR inhibitors have been shown to inhibit growth in TNBC cell lines. We describe a panel of patient-derived xenografts representing multiple TNBC subtypes and use them to test preclinical drug efficacy of two mTOR inhibitors, sirolimus (rapamycin) and temsirolimus (CCI-779). MethodsWe generated a panel of seven patient-derived orthotopic xenografts from six primary TNBC tumors and one metastasis. Patient tumors and corresponding xenografts were compared by histology, immunohistochemistry, array comparative genomic hybridization (aCGH) and phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha (PIK3CA) sequencing; TNBC subtypes were determined. Using a previously published logistic regression approach, we generated a rapamycin response signature from Connectivity Map gene expression data and used it to predict rapamycin sensitivity in 1,401 human breast cancers of different intrinsic subtypes, prompting in vivo testing of mTOR inhibitors and doxorubicin in our TNBC xenografts.ResultsPatient-derived xenografts recapitulated histology, biomarker expression and global genomic features of patient tumors. Two primary tumors had PIK3CA coding mutations, and five of six primary tumors showed flanking intron single nucleotide polymorphisms (SNPs) with conservation of sequence variations between primary tumors and xenografts, even on subsequent xenograft passages. Gene expression profiling showed that our models represent at least four of six TNBC subtypes. The rapamycin response signature predicted sensitivity for 94% of basal-like breast cancers in a large dataset. Drug testing of mTOR inhibitors in our xenografts showed 77 to 99% growth inhibition, significantly more than doxorubicin; protein phosphorylation studies indicated constitutive activation of the mTOR pathway that decreased with treatment. However, no tumor was completely eradicated.ConclusionsA panel of patient-derived xenograft models covering a spectrum of TNBC subtypes was generated that histologically and genomically matched original patient tumors. Consistent with in silico predictions, mTOR inhibitor testing in our TNBC xenografts showed significant tumor growth inhibition in all, suggesting that mTOR inhibitors can be effective in TNBC, but will require use with additional therapies, warranting investigation of optimal drug combinations.

Osteosarcoma is an aggressive malignancy and represents the most frequent primary bone malignancy of dogs and humans. Prognostic factors reported for osteosarcoma include tumor size, presence of metastatic disease, and serum alkaline phosphatase (ALP) concentration at the time of diagnosis. To date, there have been no studies to determine whether the behavior of osteosarcoma cells differ based on serum ALP concentration. Here we report on the generation of six canine osteosarcoma cell lines from osteosarcoma-bearing dogs with differences in serum ALP concentration. To determine whether in vitro behavior differs between primary osteosarcoma cell lines generated from patients with normal or increased serum ALP assays were performed to evaluate proliferation, migration, invasion, and chemosensitivity. There were no significant differences in cell proliferation, migration, invasion, or chemosensitivity between cell lines associated normal or increased serum ALP concentration.

Osteosarcoma is an aggressive malignancy of the bone and an increase in serum alkaline phosphatase concentration has clinical prognostic value in both humans and canines. Increased serum alkaline phosphatase concentration at the time of diagnosis has been associated with poorer outcomes for osteosarcoma patients. The biology underlying this negative prognostic factor is poorly understood. Given that activation of the Wnt signaling pathway has been associated with alkaline phosphatase expression in osteoblasts, we hypothesized that the Wnt/β-catenin signaling pathway would be differentially activated in osteosarcoma tissue based on serum ALP status. Archived canine osteosarcoma samples and primary canine osteosarcoma cell lines were used to evaluate the status of Wnt/β-catenin signaling pathway activity through immunohistochemical staining, western immunoblot analyses, quantitative reverse-transcription polymerase chain reaction, and a Wnt-responsive promoter activity assay. We found no significant difference in β-catenin expression or activation between OSA populations differing in serum ALP concentration. Pathway activity was mildly increased in the primary OSA cell line generated from a patient with increased serum ALP compared to the normal serum ALP OSA cell line. Further investigation into the mechanisms underlying differences in serum ALP concentration is necessary to improve our understanding of the biological implications of this negative prognostic indicator.