Citations

Abstract Background. Autophagy maintains energy metabolism and enables tumor cell survival during cellular stress. Beclin 1 is involved in autophagosome formation during which membrane-bound LC3 binds to p62/sequestosome 1. We reported that ectopic p62 can promote 5-fluorouracil-induced cell death (JBC 286:40002-12, 2011). It is unknown whether these autophagy markers can predict clinical outcome. Materials & Methods. Beclin 1, LC3B, and p62 proteins were analyzed by immunohistochemistry in tissue microarrays containing stage II and III colon carcinomas (N=151) from patients receiving 5-FU or surgery alone. The mean weighted score (intensity x extent) was dichotomized for survival analysis using Cox models. Results. Increased levels of cytoplasmic LC3, Beclin 1, and p62 expression were detected in 21%, 14%, and 68% of cancers, respectively, and were not associated with tumor site, stage or grade. By multivariable analysis, increased LC3 was significantly associated with worse disease-free survival (DFS) (p=0.022), but not overall survival (OS) [p=0.069](Table). Beclin 1 was not prognostic, but increased p62 was significantly associated with improved DFS (p=0.038), but not OS (p= 0.176)[Table]. When restricted to 5-FU-treated patients, increased LC3 was associated with worse DFS [HR= 2.38 (1.19, 4.75, p=0.020), elevated Beclin 1 predicted worse OS [HR=2.83 (1.27, 6.34), p = 0.019], and p62 was not prognostic. Conclusion. LC3 and Beclin 1 overexpression were independently associated with adverse outcome. Increased p62 was associated with better outcome consistent with its ability to promote cell death. These data support strategies to inhibit autophagy for therapeutic advantage. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4112. doi:1538-7445.AM2012-4112

Abstract BACKGROUND. The mammalian target of rapamycin (mTOR) kinase forms two multiprotein complexes, mTORC1 and mTORC2, which regulate cell growth, cell survival, and autophagy. We determined whether mTOR inhibition can induce cytoprotective autophagy resulting in anti-cancer drug resistance, and whether inhibition of autophagy using the lysomotrophic agent, chloroquine, can restore chemosensitivity. EXPERIMENTAL METHODS. Inhibition of mTOR was accomplished in human colon cancer cells (HT-29, DLD1) using lentiviral shRNA or by AZD8055, a TORC1 and TORC2 kinase inhibitor. These cells were then incubated with 5-FU (0-100 µM) or CPT-11 (0-20 µM) alone and in combination with chloroquine. LC3I/II expression, phospho-histone H2A.X (a DNA damage marker), and caspase cleavage were analyzed by Western blotting using highly specific antibodies, and clonogenic survival assays were also performed. RESULTS. AZD8055 was shown to attenuate caspase-3 and -9 cleavage and to suppress the expression of phospho-histone H2A.X induced by 5-FU or CPT-11 in colon cancer cells. AZD8055 was also found to antagonize the ability of 5-FU to inhibit clonogenic survival. Similarly, shRNA knockdown of mTOR attenuated 5-FU-induced caspase cleavage and phospho-histone H2A.X expression. Inhibition of mTOR by AZD8055 or shRNA enhanced the conversion of the autophagosome-associated protein light chain 3 (LC3) from a cytosolic (LC3I) to a membrane-bound (LC3II) form, consistent with induction of autophagy. Inhibition of autophagy by chloroquine was shown to attenuate the ability of AZD8055 to suppress 5-FU-induced caspase cleavage. Consistent with this result, chloroquine sensitized DLD1 cells to AZD8055 and 5-FU in a clonogenic survival assay. CONCLUSION. Inhibition of mTOR can protect colon cancer cells from DNA damage induced by cytotoxic chemotherapy. This cytoprotection was due, in part, to induction of pro-survival autophagy since chloroquine was able to partially restore chemosensitivity. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5485. doi:10.1158/1538-7445.AM2011-5485

Abstract Mesenchymal Stem Cells (MSCs) have the ability to migrate specifically to tumours in vivo, and coupled with their capacity to bypass immune surveillance, are potentially attractive vehicles for tumor-targeted delivery of therapeutic agents. The aim of this study was to determine the potential of MSC-mediated expression of the sodium iodide symporter (NIS) for in vivo imaging and therapy of breast tumours. Expression of NIS allows cells to concentrate radionuclides including technetium-99m (Tc-99m) and 131-Iodine (131-I). This could potentially support imaging of NIS-expressing MSCs following engraftment, and 131-I therapy of surrounding tumour tissue based on the bystander effect of the radionuclide. Methods: Tumor bearing animals (MDA-MB-231 flank) were given an intravenous or intratumoral injection of NIS expressing human MSCs (MSC-NIS), followed by imaging 3, 7, 10 or 14 days later. Imaging was performed using a Bazooka SPECT γ-camera, following intraperitoneal injection of 2mCi/74MBq Tc-99m. Following imaging, animals were sacrificed and organs harvested for analysis of hNIS expression by RQ-PCR. A second group of animals received an intraperitoneal injection of 1mCi/37MBq 131-I or saline (controls) 14 days following intratumoral or intravenous injection of MSC-NIS, and tumour volume was tracked for 8 weeks. Results: Bazooka SPECT imaging of animals revealed uptake of tracer (Tc-99m) in the mouse thyroid/salivary glands and stomach, representing native NIS expression. Following injection of MSC-NIS, an image of animal intestines was also observed at D3, with a weak image of the tumour also visible. By Day14, uptake of tracer was visible at the site of the tumour with no significant image of non-target tissue observed. Expression of hNIS in harvested organs supported the imaging data, with hNIS detected in the intestines, heart, lungs and tumour at early timepoints. While NIS expression depleted in non-target tissues by D7, gene expression persisted at the tumour site. Based on imaging/biodistribution data, animals were given a therapeutic dose of 131-I 14 days following MSC-NIS injection to avoid toxicity to non-target organs. No adverse effect of injection of MSC-NIS or radionuclide was observed. Tracking of tumour volume revealed a significant reduction in tumour growth in animals that had received an intravenous injection of 131-I 14days following MSC-NIS injection (Mean ± SEM, 236 ± 62mm3 versus 665 ± 204 mm3 in controls). Following intratumoral injection of MSC-NIS + 131-I, although tumour size was reduced compared to controls (472 ± 80 mm3), the difference in volume was not significant. Conclusion: The ability to non-invasively track MSC migration and transgene expression in real time prior to therapy is a major advantage to this strategy. This promising data supports the viability of this approach as a novel therapy for metastatic breast cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5392. doi:10.1158/1538-7445.AM2011-5392

Abstract Pancreatic cancer is highly lethal with an estimated 5-year relative survival rate of less than 5%. A genome wide association study (GWAS) of pancreatic cancer (PanScan) conducted within the NCI Cohort Consortium and the Pancreatic Cancer Case-Control Consortium (PANC4) recently identified four common susceptibility loci for pancreatic cancer on chromosomes 1q32.1, 5p15.33, 9q34.2 and 13q22.1. To investigate how these germline variants influence pancreatic cancer risk, we are employing both genome wide and targeted approaches. For the former, we are characterizing the epigenome and transcriptome of pancreatic cell lines and tissue samples for genomic annotation and prioritization of methods to use in targeted approaches with the aim of directly linking pancreatic cancer risk variants to molecular phenotypes and mechanisms that underlie the association signals. As an example of our approach we will present results in a pancreatic cancer susceptibility locus on chromosome 1q32.2 marked by SNP (single nucleotide polymorphism) rs3790844. This SNP is located in the first intron of the NR5A2 gene that encodes a nuclear receptor of the Ftz-F1 family known to be involved in pancreatic development, cholesterol and bile acid homeostasis, steroidogenesis and cell proliferation. In ChIP-sequencing and MeDIP-chip experiments, we observed repressive histone modification marks (H3K27me3) and regions of DNA methylation in the NR5A2 gene locus in pancreatic cell lines whereas only a few weak positive histone modification marks were observed (H3K4me1 and H3K4me3). This pattern of epigenetic marks correlates well with low expression of NR5A2 in the same cell lines. Expression of the NR5A2 gene at the RNA level is high in normal pancreatic tissue samples as compared to stage I-III pancreatic adenocarcinoma, most islet cell and neuroendocrine tumors. Using immunohistochemistry, nuclear NR5A2 was detected in normal acinar cells and – at lower levels – in normal ductal cells. The protein was also expressed in low and high grade PanINs and in the majority of PDAC analyzed. Our analysis of the tissue specific expression of NR5A2 shows that it is mainly expressed in normal pancreas, colon, small intestine, liver and breast but detected at low levels or not at all in most other tissue types and cancer cell lines. Furthermore we have identified a novel NR5A2 transcript by rapid amplification of cDNA ends (RACE) and mapped a promoter in front of this transcript in the genomic region corresponding to first intron of the two annotated Refseq transcripts. Establishing a comprehensive catalog of epigenetic patterns coupled with targeted analysis to correlate risk variants to molecular phenotypes could provide functional plausibility for known pancreatic cancer association signals discovered by GWAS findings. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4965. doi:10.1158/1538-7445.AM2011-4965

Abstract —A rapidly emerging body of literature implicates a pivotal role for the Ca 2+ -calmodulin–dependent phosphatase calcineurin as a cellular target for a variety of Ca 2+ -dependent signaling pathways culminating in left ventricular hypertrophy (LVH). Most of the recent experimental support for this hypothesis is derived from in vitro studies or in vivo studies in transgenic mice expressing activated calcineurin or mutant sarcomeric proteins. The aim of the present study was to test whether calcineurin inhibitors, cyclosporin A (CsA) and FK 506, prevent pressure-overload LVH using 2 standard rat models: (1) the spontaneously hypertensive rat (SHR) and (2) aortic banding. The major new findings are 2-fold. First, in SHR, LVH (left ventricular weight to body weight ratio) was unaffected by a dose of CsA (5 mg · kg −1 · d −1 ) that was sufficient to raise blood pressure and to inhibit calcineurin-mediated transcriptional activation in skeletal muscle. Second, in rats with aortic banding, LVH was unaffected by FK 506 (0.3 mg · kg −1 · d −1 ) or even higher doses of CsA (10 and 20 mg · kg −1 · d −1 ) that were sufficient to inhibit 90% of total calcineurin phosphatase activity in the hypertrophied myocardium. In the latter experiments, CsA blocked neither the elevated left ventricular end-diastolic pressures, a measure of diastolic function, nor the induction in atrial natriuretic peptide mRNA in the hypertrophic ventricles. Thus, in numerous experiments, systemic administration of potent calcineurin inhibitors did not prevent the development of LVH in 2 classic models of pressure-overload hypertrophy. These results demonstrate that pressure-overload hypertrophy can arise through calcineurin-independent pathways.