AACR 2016: Abstracts 1-2696

By American Association for Cancer Research (AACR)

The AACR Annual Meeting is a must-attend event for cancer researchers and the broader cancer community. This year's theme, "Delivering Cures Through Cancer Science," reinforces the inextricable link between research and advances in patient care. The theme will be evident throughout the meeting as the latest, most exciting discoveries are presented in every area of cancer research. There will be a number of presentations that include exciting new data from cutting-edge clinical trials as well as companion presentations that spotlight the science behind the trials and implications for delivering improved care to patients. This book contains abstracts 1-2696 presented on April 17-18, 2016, at the AACR Annual Meeting.

• Language: English
• Publisher: CTI Meeting Technology
• Release date: Mar 28, 2016
• ISBN: 9781311286796

American Association for Cancer Research (AACR)

The AACR is the oldest and largest scientific organization in the world focused on every aspect of high-quality, innovative cancer research. Its reputation for scientific breadth and excellence attract the premier researchers in the field. The programs and services of the AACR foster the exchange of knowledge and new ideas among scientists dedicated to cancer research, provide training opportunities for the next generation of cancer researchers, and increase public understanding of cancer.

Book preview

Proceedings of the AACR

Volume 57 | April 2016

Part A: Abstracts 1-2696

TABLE OF CONTENTS

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Altered Cellular Signaling and Cancer Metabolomics

Altered Glucose Metabolism in Cancer

Functional Genomics and Genomics of Model Systems

Genomic Analysis of Cancers

Genomic Profiling of Cancers

Intratumor Heterogeneity and Resistance

Kinases and Phosphatases

Mitochondria, Autophagy, and Metabolic Vulnerabilities

EXPERIMENTAL AND MOLECULAR THERAPEUTICS:

Cellular Processes and Responses to Therapy

Combination Chemotherapy

Mechanisms of Drug Resistance 1

Novel Antitumor Agents

Novel Assays

PI3K/AKT Inhibitors

CLINICAL RESEARCH:

Biomarkers

Biomarkers for Genitourinary and Gynecological Cancers

Biomarkers for Melanoma and Uncommon Cancers

Circulating Biomarkers 1

Radiation Oncology

IMMUNOLOGY:

Genetic Determinants and Regulators of Cancer Immunity

Immune Modulating Agents 1

Therapeutic Antibodies

TUMOR BIOLOGY:

Drug Testing in Cell Lines and 3D Models

Human in Mouse Models

Mechanisms of Tumorigenesis in Animal Models of Cancer 1

Molecular Regulation of Tumor Invasion

Pro-Tumorigenic Microenvironment

Targeting the Microenvironment

BIOINFORMATICS AND SYSTEMS BIOLOGY:

Systems Biology

EPIDEMIOLOGY:

Genes and Function and Risk

PREVENTION RESEARCH:

Models and Mechanisms in Cancer Prevention

BIOINFORMATICS AND SYSTEMS BIOLOGY:

Novel and Integrative Analyses of Cancer Genome Data

CLINICAL RESEARCH:

Biomarkers to Direct Cancer Therapy

ENDOCRINOLOGY:

Molecular Pharmacology of Hormone-dependent Malignancies

EXPERIMENTAL AND MOLECULAR THERAPEUTICS:

Antibody-targeted Therapy

Approaches to Elucidating and Overcoming Drug Resistance

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Disordered Gene Regulation and Chromatin State in Malignant Transformation

Oncogene and Tumor Suppressor Function and Targeting

PREVENTION RESEARCH:

Highlights in Cancer Prevention Advances

TUMOR BIOLOGY:

Immunomodulation in Cancer

The Relevance of Stemness Properties in Cancer

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Cell-Cell Interaction and Tumor Microenvironment

Circulating RNAs, Epigenetics, and Novel Non-coding RNAs in Cancer

LncRNAs and Mechanisms in Cancer

Metabolic Reprogramming and Autophagy

Metabolic Reprogramming in Cancer

MicroRNAs as Biomarkers and Therapeutics

MicroRNAs as Oncogenes and Tumor Suppressors

Oncogene Function, Regulation, and Targeting

Oncogenes and Tumor Suppressor Genes and Pathways

EXPERIMENTAL AND MOLECULAR THERAPEUTICS:

Differentiation Therapy

Growth Factor Receptors and Surface Antigens as Therapeutic Targets

Novel Antitumor Agents and Epigenetics

Novel Molecular Targets

Regulation of Anticancer Drug Effects

CANCER CHEMISTRY:

Drug Delivery 1

Drug Delivery 2

CLINICAL RESEARCH:

Clinical Assay Development

Immune Response Monitoring: Clinical

Special Populations, Supportive Care, and Survivorship Research

IMMUNOLOGY:

Immune Microenvironment and Antitumor Immunity

Immune Modulating Agents and Therapeutic Antibodies

BIOINFORMATICS AND SYSTEMS BIOLOGY:

Applications of Bioinformatic Tools to Analyze Cancer Data

TUMOR BIOLOGY:

Biomarkers and Profiling of Metastasis

Cellular and Molecular Mediators of Metastasis

Epithelial-Mesenchymal Transition in Metastasis

Metastasis-Promoting and -Suppressing Genes

Radiation Science

Regulators of Epithelial-Mesenchymal Transition and Metastasis

Stemness Properties of Intestinal, Pancreatic, and Hepatic Cancer

EPIDEMIOLOGY:

Diet, Tobacco, Smoking, and Other Lifestyle Factors in Cancer Epidemiology

Race/Ethnicity and Disparities in Diagnosis, Treatment, and Outcomes

PREVENTION RESEARCH:

Prevention Clinical Research

ENDOCRINOLOGY:

Molecular Endocrinology of Hormone-dependent Malignancies

REGULATORY SCIENCE AND POLICY:

Regulatory Science and Policy

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Assessment of Gene Regulation in the Malignant Context

GTPase, RAF, and Growth Factor Pathways

MicroRNAs in Metastasis and Cancer

Non-coding RNAs and Mechanisms in Cancer

Profiling MicroRNA Expression in Cancer

Regulation of Chromatin State and Gene Expression

Transcriptional Regulation and Gene Expression

EXPERIMENTAL AND MOLECULAR THERAPEUTICS:

Clinical Pharmacology and Pharmacogenomics

Drug Delivery

Experimental Therapeutics

Mechanisms of Drug Resistance 2

New Drugs, Therapeutic Targets, and Treatment Approaches

CANCER CHEMISTRY:

High Throughput Screening and Natural Products

Nanotechnology and Drug Delivery

CLINICAL RESEARCH:

Adoptive Cell Therapy, Immune Checkpoints, and Vaccines

Biomarkers for Lung Cancer

Mechanisms of Response to Targeted Agents and Potential New Targets

IMMUNOLOGY:

Adoptive Cell Therapy

Immune Checkpoints 1

Therapeutic Antibodies and Vaccines

TUMOR BIOLOGY:

Clonal Heterogeneity and Evolution

Intratumor Heterogeneity and Treatment Responses

Pediatric Cancer Genomics, Genetics, and Epigenetics

Pediatric Cancer Molecular Pathways

Pediatric Cancer Targets, Models, Therapies, and Resistance

Stemness Properties and Therapeutic Targeting in Solid Tumors

Stemness Properties of Neuronal and Pediatric Tumors and New Approaches

EPIDEMIOLOGY:

Genetic Contributions to Cancer Epidemiology: Familial Cancers and GWAS

Risk Prediction, Screening, and Comparative Effectiveness Research

PREVENTION RESEARCH:

Targets, Markers, and Agents in Cancer Prevention

CLINICAL RESEARCH:

Genomic Landscapes

EPIDEMIOLOGY:

Biomarkers and Other Epidemiologic Factors in Cancer Prognosis

EXPERIMENTAL AND MOLECULAR THERAPEUTICS:

Hitting the Target Harder: Preclinical Development of Potent and Selective Inhibitors

IMMUNOLOGY:

Immunotherapy Trial Results and Correlates

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Epigenetic Alterations in Cancer

LncRNAs, MicroRNAs, and Non-coding RNAs in Cancer

Mechanisms and Vulnerabilities of Metabolic Reprogramming

TUMOR BIOLOGY:

Mechanism and Dynamics of Cancer Metastasis

Mechanisms of Tumorigenesis in Mouse Models of Human Cancer

New Molecular Advances in Pediatric Cancer

Advocates Poster Session 1

Sunday, April 17, 2016

MOLECULAR AND CELLULAR BIOLOGY / GENETICS:

Altered Cellular Signaling and Cancer Metabolomics

#2

Metabolomics analysis reveals distinct profiles of non-muscle invasive and muscle-invasive bladder cancer.

Divya Sahu,¹ Yair Lotan,² Bryan Wittmann,³ Bruce Neri,³ Donna Hansel¹. ¹UC San Diego, San Diego, CA; ²University of Texas Southwestern Medical Center, Dallas, TX; ³Metabolon Inc., Durham, NC.

Background: Bladder cancer affects >70,000 patients annually in the United States. Despite its high incidence, therapeutic options are limited in early or late stage. We wanted to identify key metabolic pathways that were altered in bladder cancer development and progression.

Experimental Design: We performed global metabolomics profiling of benign urothelium, high-grade non-muscle invasive bladder cancer and advanced, muscle-invasive bladder cancer using GC-MS and LC-MS platforms. This analysis was coupled with publicly available data on transcriptomics of key enzymes, to determine pathways that may be suitable for future therapeutics development.

Results: Categorical pathways globally dysregulated in cancer relative to benign urothelium included glucose, TCA cycle, lipid, amino acid and nucleotide pathways. Bladder cancers demonstrated Warburg metabolism, with elevated glucose utilization to

drive glycolysis and sorbitol pathway intermediates. Elevated late TCA cycle intermediates, coupled with higher levels of amino acids and dipeptides, suggest the possibility of anaplerotic activity in bladder cancer as a mechanism to sustain energy production. Medium and long chain fatty acids were produced at the expense of dicarboxylic fatty acids. Muscle-invasive bladder cancers showed enhanced use of COX and LOX metabolomics pathways and a possible role for inflammation in regulating NAD+ synthesis in muscle-invasive bladder cancer. Transcriptomic profiling validated that the majority of metabolomics pathway alterations corresponded to gene expression changes of enzymes responsible for metabolite production.

Conclusions: This study identifies multiple parallel metabolomics changes unique to non-muscle invasive and muscle-invasive bladder cancer that can be used to justify testing novel therapeutics targeting metabolic pathways in bladder cancer.

#3

Amino acid profiles indicate dependence on different metabolic pathways between leukemia subtypes.

Shannon R. Sweeney, Enrique Sentandreu, Stefano Tiziani. The University of Texas at Austin, Austin, TX.

Despite the growing body of evidence that the tumor microenvironment protects leukemia cells from chemotherapeutic stresses (1-3), the effect of many extracellular metabolites remains largely unknown. To explore the influence of extracellular metabolites on different leukemia subtypes, cells were treated for 24 hours in vitro with either a supplemental amino acid or amino acid derivative. From this initial screening, a subset of metabolites were chosen for metabolomics analysis. Mass spectrometry (UPLC-MS/MS) was performed on intracellular fractions to identify metabolic differences that resulted from supplementation. Metabolite profiles were also compared between leukemia cell types, namely AML, pre-B cell ALL, and T cell ALL. Of the metabolites tested, lysine and 4-hydroxyphenylpyruvate, an intermediate of tyrosine and phenylalanine metabolism, had the greatest impact on global amino acid profiles. In AML and T cell ALL cell lines, intracellular glutamate, glutamine, proline, and aspartate were increased relative to their respective controls. These amino acids can enter the tricarboxylic acid (TCA) cycle as either α-ketoglutarate or oxaloacetate, suggesting a central role of the TCA cycle in both AML and T cell ALL metabolism. Interestingly, these metabolites were not significantly increased in pre-B cell ALL, signifying the inverse it true for pre-B cells. This observation provides metabolomics evidence that is consistent with a previous study that reported downregulated expression of TCA cycle related genes in pre-B cell ALL (4). Our findings indicate that uptake and metabolism of amino acids and their derivatives is distinct for different leukemia types. Moreover, supplementation with a single metabolite can result in global changes in intracellular metabolite profiles, suggesting an influence not only as an energy substrate, but on overall metabolic pathway activity. Specifically, we conclude that the TCA cycle is more active in AML and T cell ALL and can be modulated by changing the extracellular environment, while pre-B cells are less sensitive to amino acid modulation.

(1) Meads MB, et al. Clin Cancer Res 2008;14(9):2519-2526.

(2) Ayala F, et al. Leukemia 2009;23:2233-2241.

(3) Konopleva M, et al. Drug Resist Updat 2009;12:103-113.

(4) Boag JM, et al. Leukemia 2006;20:1731-1737.

#4

Hyper activation of poly(ADP-ribose) polymerase 1 initiates large-scale metabolic changes in a cellular model of glioblastoma.

Anna M. Wilk,¹ Elise Fouquerel,² Bobbie Johnston,³ Samuel A.J. Trammell,⁴ Lindsay Schambeau,¹ Joel F. Andrews,¹ Lewis Pannell,¹ Sara J. Cooper,³ Charles Brenner,⁴ Robert W. Sobol¹. ¹Mitchell Cancer Institute, University of South Alabama, Mobile, AL; ²Department of Pharmacology & Chemical Biology, University of Pittsburgh School of Medicine, Pittsburgh, PA; ³HudsonAlpha Institute for Biotechnology, Huntsville, AL; ⁴Department of Biochemistry, Carver College of Medicine, University of Iowa, Iowa City, IA.

PARP1 is a key enzyme of the Base Excision Repair (BER) pathway, facilitating the repair of base damage and single-strand DNA breaks. Activated PARP1 synthesizes poly (ADP-ribose) (PAR), triggering chromatin de-condensation to facilitate recruitment of BER proteins to complete repair. PARP1 activation is attenuated upon successful repair of the DNA lesion. However, unrepaired DNA breaks lead to continuous PARP1 activation and cell death. The molecular mechanism underlying PARP1 activation induced cell death was recently revealed as independent from NAD+ depletion. We have shown that PARP1 activation and PAR synthesis affect glycolysis by directly inhibiting the glycolytic enzyme, hexokinase 1 (HK1). Following on these discoveries, we decided to investigate global metabolic changes triggered by hyperactivation of PARP1. For this study, we used gas chromatography mass spectrometry (GC-MS) to quantify over 150 cellular metabolites and Multiple-Reaction Monitoring Liquid Chromatography Mass Spectrometry (MRM LC-MS) to measure NAD+ metabolites. As a model, we tested glioblastoma cells overexpressing methylpurine DNA glycosylase (MPG) to enhance the PARP1-activation response to DNA damage induced by the alkylating agent MNNG. Simultaneously, to monitor independence from the DNA damaging effect of NAD+ depletion, we utilized an inhibitor of NAD+ biosynthesis, FK866. We found that PARP1 activation leads to a strong accumulation of glucose, likely as a secondary effect of HK1 inhibition. In addition, we observed a significant change in the level of other metabolites including an increase in inosine, inosine monophosphate (IMP), cytidine and uridine levels upon PARP1 activation, suggesting an indirect effect of PARP1 activation on purine and pyrimidine metabolism. Ongoing studies will use these global approaches to unravel the complete metabolic response of cancer cells to genotoxic treatment.

#5

Using stable isotope lipidomics to identify adipocyte-induced alterations in the acute lymphoblastic leukemia lipidome.

Jonathan Tucci,¹ Katy Margulis,² Cheng-Chih Hsu,³ Wesley Dixon,² Richard N. Zare,² Steven D. Mittelman¹. ¹Children's Hospital of Los Angeles, Los Angeles, CA; ²Stanford University, Stanford, CA; ³National Taiwan University, Taipei, Taiwan.

Obesity is associated with the development and progression of many cancers, including acute lymphoblastic leukemia (ALL). We have shown that ALL cells induce adipocyte lipolysis and take up adipocyte-derived free-fatty acids (FFAs). Here, we use targeted lipidomics with carbon-13 labeling to 1) determine which adipocyte-derived FFAs are taken up by ALL cells, and 2) identify the specific alterations in the ALL cell lipidomic profile caused by co-culture with adipocytes.

Mouse pre-adipocytes (3T3-L1) were differentiated into adipocytes in the presence of U¹³C-glucose to allow ¹³C incorporation into adipocyte lipids. ALL cells were then cultured alone or co-cultured over ¹³C-labeled or non-labeled adipocytes for 72 hours. Following co-culture, ALL cells were harvested and analyzed by nanospray desorption electrospray ionization mass spectrometry (nanoDESI-MS). Lipidomic spectra were analyzed to characterize uptake of adipocyte-derived FFAs identifying ¹³C-enrichment in ALL lipid moieties. The relative intensities of selected lipid peaks were compared between conditions, after normalization, to evaluate the effects of the presence of adipocytes on ALL lipidome.

Adipocytes differentiated in the presence of U¹³C-glucose incorporated ¹³C into numerous triglyceride moieties. In ALL cells co-cultured with labeled adipocytes, ¹³C enrichment was identified in FFA and phospholipids. Labeling of oleic acid primarily occurred in groups of two, suggesting incorporation of U¹³C-glucose into adipocyte FFAs through acetyl-CoA-mediated lipogenesis. Similarly, ¹³C enrichment of ALL cell phospholipids occurred in groups of two and three, mirroring adipocyte incorporation of U¹³C-glucose into phospholipids through both acetyl-CoA and glycerol. Moreover, ALL cell unsaturated FFAs had a greater ¹³C enrichment than saturated FFAs (Unsaturated FFAs: 14.1±5.5 vs. Saturated FFAs: 6.6±2.3, p=0.02; n=5). When co-cultured over adipocytes, ALL cells contained significantly more FFAs than ALL cells alone (Oleic Acid: 63.0±49.2 vs. 24.1±7.4, p=0.02; Stearic acid: 20.9±17.5 vs. 8.3±2.2, p=0.03; Palmitoleic Acid: 11.5±9.8 vs. 3.3±1.0, p=0.01; Palmitic Acid: 15.1±11.7 vs. 5.5±1.4, p=0.01; n=5).

Using stable-isotope lipidomics, we can effectively identify and quantify which ALL cell lipids are derived from nearby adipocytes. This methodology provides comprehensive insight into the cancer cell lipidome and can be extended to understand how genetics, the microenvironment, and treatment affect cancer lipid metabolism.

#6

Metabolomics of Gleason score in prostate tumor tissue and serum.

Kathryn Penney,¹ Svitlana Tyekucheva,² Massimo Loda². ¹Brigham and Women's Hospital/Harvard Medical School, Boston, MA; ²Dana-Farber Cancer Institute, Boston, MA.

Background: Gleason score is currently the best clinical predictor of prostate cancer specific-mortality. Identifying biomarkers associated with Gleason score can improve prediction and advance the understanding of the biology of aggressive disease. Using gene expression data, we previously determined that metabolism pathways, including pyrimidine, propanoate, and beta-alanine, were differentially expressed in high and low grade tumors. The goal of this study is to identify metabolites that differ in low and high Gleason score tumors and in blood from these patients.

Methods: Metabolic profiling was performed on Gleason score 6 (n=53) and Gleason score ≥8 (n=32) radical prostatectomy tumor tissue specimens, and on serum samples from the time of diagnosis (n=30 Gleason score 6 and n=19 Gleason score ≥8) from the DF/HCC Prostate Cancer SPORE Cohort. Samples were prepared for analysis by Metabolon, Inc. with their standard solvent extraction method to recover small molecules. Metabolites were identified and quantified by gas and liquid chromatography and mass spectrometry. Missing values were calculated using k-nearest neighbor imputation, and sample values were normalized across batches. The metabolite levels were compared across Gleason score 6 and Gleason score ≥8 tumors and serum using a Wilcoxon test.

Results: In the tumor tissue, 207 metabolites were identified. Of these, 30 were present in significantly different amounts in high and low Gleason score tumors, including previously reported citrate (p=0.002) and spermine (p=0.003). In a pathway analysis, the propanoate pathway was significantly different in high and low Gleason (p=0.03), but not the pyrimidine or beta-alanine pathways (p>0.05). In serum, 15 individual metabolites had significantly different levels, including uridine (p=0.008) and N-acetylserine (p=0.0001).

Conclusions: Metabolite levels do differ in high and low Gleason score prostate tumor tissue. The novel serum results suggest that metabolites in blood may serve as biomarkers of tumor differentiation. Our ongoing research will build prediction models, and will determine if the blood metabolites can indicate the presence of high-Gleason score tumor not detected at biopsy and could serve as biomarkers for monitoring disease progression in active surveillance patients.

#7

LKB1 deficiency is associated with a unique metabolic signature in Kras mutant non-small cell lung cancer (NSCLC).

Tingyu Liu,¹ Erin Sennott,¹ Zhengjie Zhong,¹ Mya Steadman,² Kelly Marsh,² Jonathan Hurov,² Cyril Benes,¹ Jeffrey A. Engelman¹. ¹Massachusetts General Hospital, Charlestown, MA; ²Agios Pharmaceuticals, Cambridge, MA.

LKB1 deficiency is found in approximately 10% of non-small cell lung cancer harboring Kras mutation. Loss of LKB1 has been associated with increased metastatic rates and decreased survival in patients. Currently, no effective therapy has been developed against this subtype of lung cancer. LKB1 can regulate multiple cellular activities. One of the most well-studied functions is its regulation on cell metabolism through AMPK signaling pathway. How LKB1 loss globally impacts NSCLC metabolism has not been well understood yet. Using isotope-labeled substrate tracing approach, we found an increase of specially-labeled intermediates in pentose phosphate pathway (PPP) that may suggest upregulation of PPP activity. In addition, we also performed an unbiased drug screen that showed higher sensitivity to an mTOR inhibitor in Kras and LKB1 double mutant lung cancer cells relative to cancer cells under other genetic background. These data suggest that loss of LKB1 may upregulate biosynthetic reactions in Kras mutant NSCLC and render these cells sensitive to inhibitors in biosynthesis. These findings may provide new insights in developing therapeutic agents targeting KrasMT/LKB1MT NSCLC.

#8

Global and targeted metabolomic profiling of colorectal cancer progression.

Beatriz Sanchez-Espiridion,¹ Lindsey White,² Lopa Mishra,³ Gottumukkala S. Raju,³ Scott Kopetz,⁴ Jian Gu,¹ Yuanquing Ye,¹ Xifeng Wu,¹ Dong Liang². ¹Department of Epidemiology, The University of Texas MD Anderson Cancer Center, Houston, TX; ²Department of Pharmaceutical Sciences, Texas Southern University, Houston, TX; ³Department of Gastroenterology, Hepatology & Nutrition, The University of Texas MD Anderson Cancer Center, Houston, TX; ⁴Department of GI Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX.

Background: Colorectal cancer (CRC) is one of the most prevalent and deadly cancers in the world. The development of improved and robust biomarkers to enable screening, surveillance, and early detection of CRC continues to be a challenge. Patients with colorectal adenoma are at higher risk of developing colon cancer; however, noninvasive methods to identify these patients are still on demand. The aim of this study was to identify biomarkers of CRC disease progression by using metabolomic profiling of human serum samples in a multistep approach.

Methods: We performed global metabolomic profiling on 30 human serum samples from patients with colorectal adenoma, 30 CRC patients and 30 healthy controls who were matched by age, gender and ethnicity. For validation, we measured the three top differentially expressed metabolites in an additional set of 50 adenoma, 50 CRC and 50 healthy controls.

Results: Global biochemical profiles of 404 metabolites were detected, with 301 metabolites remaining after quality control procedures. In discovery phase, 50 metabolites had differential levels between colorectal adenoma, CRC and controls (P for trend <0.05), with 19 metabolites showing increased levels in CRC and adenoma in comparison to controls and 31 metabolites with decreased levels. Further exploratory analyses of these metabolites showed a key role for metabolic pathways involving urea cycle, caffeine and galactose metabolism as associated with CRC progression. The top 3 differentially expressed metabolites (Xanthine, Hypoxanthine and D-mannose) were selected for validation. Consistent with the discovery phase, CRC cases and adenoma had lower levels of Xanthine than controls (mean ± SD; 9.95 ±0.92 mg/ml vs 10.63±0.97 mg/ml; P<0.001 and 9.87±0.75 vs 10.63±0.97 mg/ml; P<0.001). The same trend was observed for Hypoxanthine (mean ± SD; 10.72 ±0.62 mg/ml vs 12.29±1.60; P<0.001 and 10.71±0.61 vs 12.29±1.60 mg/ml; P <0.001) whereas higher levels of D-mannose where observed in both CRC cases and adenoma when compared to controls (3.32±0.58 mg/ml vs 2.32 ±0.90mg/ml; P<0.001 and 3.32±0.58 mg/ml vs 2.32 ±0.90mg/ml; P<0.001). Using the median value of controls as a cut-off point, 94% of the adenoma and CRC cases showed low levels of Xanthine ( Odds Ratio (OR)=10.47, 95% confidence interval (CI) = 2.63-41.63 for adenoma; OR=38.76 and 95% CI = 6.58-228.51 for CRC ) . For Hypoxanthine, 90% of the adenoma and CRC cases showed low levels (OR =6.50 and 95% CI = 1.98--21.24 for adenoma; OR=11.19 and 95% CI = 3.28-38.21 for CRC). For D-Mannose, all adenoma cases had high levels (OR is not available due to 0 count) and 92% of CRC cases (OR = 15.99, 95% Ci=4.07-62.88, P< 0.001) had high levels of D-Mannose, compared to 50% of controls having high levels of D-Mannose.

Conclusions: Our results suggest the potential utility of the identified metabolites as new valuable biomarkers for early detection of CRC.

#9

Metabolic profiling of bladder cancer cell lines reveals molecular alterations involved in methylation and novel epigenetic phenotype.

Rashmi Krishnapuram,¹ Franklin Gu,¹ Salil Kumar Bhowmik,¹ Suman Maity,¹ Mohan Manikkam,¹ Friedrich-Carl von Rundstedt,¹ Vasanta Putluri,¹ Yair Lotan,² Jonathan M. Levitt,¹ Seth P. Lerner,¹ Cristian Coarfa,¹ Arun Sreekumar,¹ Nagireddy Putlurip¹. ¹Baylor College of Medicine, Houston, TX; ²University of Texas Southwestern Medical Center, Dallas, TX.

Background: Bladder cancer (BCa) is primarily carcinogen driven cancer. Epidemiological studies indicate environmental factors play a major causative role compared to genetic factors. Xenobiotic metabolism is highly perturbed and precise mechanisms involved are poorly understood during BCa progression. We identified metabolic signature that can distinguish bladder cancer from controls and reveals major alterations in phaseI/II enzymes involved in xenobiotic metabolism and suggest a key role for epigenetic modifications.

Material and Methods:

In this study, we used mass spectrometry based metabolomics profiling coupled with enrichment-based bioprocess mapping to obtain insight into biochemical alterations in bladder cancer cell lines. We further validated related gene expression using real time quantitative PCR (qPCR) and proteins using western blotting.

Results:

In this study, we used high-throughput mass spectrometry to measure over 350 compounds in seven bladder cell lines, identifying 91 metabolites which exhibited significant changes in bladder Cancer. Most importantly, methylated, hydroxylated and acetylated metabolites are altered. Interestingly, S-Adenosyl methionine (SAM) is the most prominent pathway upregulated corroborated with our previous findings obtained using patient derived metabolomic data from two independent cohorts. Second, we observe many of phaseI/phaseII metabolic enzymes including aldehyde oxidase (AOX1), cytochrome P450 1A1 (CYP1A1), CYP1B1, Glutathione S-transferase T1 (GSTT1), Glutathione S-transferase M2 (GSTM2), N-acetyl transferase I NAT1 and NAT2 are transcriptionally repressed in BCa cell line compared to benign indicating the pivotal role of methylation in gene silencing. Interestingly, we observe differential expression of polycomb group of proteins (Pcg) associated with PRC2 and PRC1 complex. Specifically, histone-lysine N-methyl transferase (EZH2) protein, which is SAM dependent histone methyl transferase and concomitant 3meHK27 trimethylated histone K27, is highly expressed in metastatic UMUC3 BCa cell line further indicating prominence of epigenetic modifications.

Conclusion:

We present an integrative pathway analysis of a metabolic gene signature which has not been previously described in the context of bladder cancer cell lines. Further mechanistic analyses reveals prominent role for methylation status and associated epigenetic modifications being played in the transcriptional repression of key xenobiotic enzymes. Collectively, our novel findings provide an opportunity for development of efficient biomarker implications and epigenetic therapy targeting BCa progression.

#10

Silibinin exhibits anti-cachectic and anti-cancerous property by modulating metabolic properties of pancreatic cancer cells.

Surendra K. Shukla,¹ Aneesha Dasgupta,¹ Kamiya Mehla,¹ Venugopal Gunda,¹ Enza Vernucci,¹ Joshua Soucheck,¹ Gennifer Goode,¹ Ryan King,¹ Anusha Mishra,¹ Ibha Rai,¹ Sangeetha Natrajan,¹ Nina Chaika,¹ Fang Yu,² Pankaj K. Singh¹. ¹Eppley Institute for Cancer Research, UNMC, Omaha, NE; ²Department of Biostatistics, UNMC, Omaha, NE.

Cancer cachexia is a systemic syndrome characterized by progressive weight loss of the patient due to muscle wasting and fat depletion with or without anorexia. About 80% of pancreatic ductal adenocarcinoma (PDAC) patients exhibits cachectic phenotype and it significantly contributes in mortality and morbidity of the disease. In present study we have evaluated the effect of bioactive molecule silibinin on pancreatic cancer progression and cachectic properties by utilizing in vitro as well as in vivo models of PDAC. We observed that silibinin inhibits growth and induces apoptosis in multiple pancreatic cancer cell lines in a dose-dependent manner. We also observed silibinin-mediated reduction in the expression of key glycolytic genes and inhibition of glucose uptake and lactate secretion. By performing LC-MS/MS based metabolomics, we observed that silibinin treatment leads to global metabolic alterations in pancreatic cancer cells. Pancreatic cancer cells treated with silibinin exhibited reduced expression of c-MYC level, a key metabolic regulator. Furthermore, we observed that silibinin-mediated STAT3 inhibition leads to reduced c-MYC expression. Ectopic expression of constitutively active STAT3 significantly attenuated the effect of silibinin on c-MYC expression and metabolic phenotype of pancreatic cancer cells. Silibinin treatment also inhibited tumor growth and progression of cachexia. Silibinin treatment to tumor-bearing mice also lead to increased food intake, increased grip strength and body coordination. Overall, our results demonstrate that silibinin exhibits anti-cachectic and anti-cancerous properties by inducing metabolic reprogramming in pancreatic cancer cells.

#11

Metformin exerts differential metabolic effects in ovarian cancer cell lines.

Adnan Munkarah,¹ Laila Poisson,¹ Seongho Kim,² Jasdeep Chhina,¹ Shailendra Giri,¹ Ramandeep Rattan¹. ¹Henry Ford Hospital, Detroit, MI; ²Karmanos Cancer Institute, Detroit, MI.

Metformin is being actively repurposed for treatment of gynecologic malignancies including ovarian cancer, with various clinical trials underway. Metformin is known to alter the cancer cell metabolism, primarily by inhibiting oxidative phosphorylation and inducing glycolysis. Our aim was to investigate if metformin induces similar metabolic changes across ovarian cancer cells. Untargeted global metabolite assay, by ultra-high performance Liquid Chromatography and Gas Chromatography Mass Spectroscopy, was performed on A2780, C200, and SKOV3ip cell lines with and without metformin treatment (10mM for 48 hours). Per-metabolite comparisons were made across conditions. Interpretive analysis was performed using the KEGG molecular pathways (Kyoto Encyclopedia of Genes and Genomes) and the Ingenuity molecule library with a focus on energy pathways of glycolysis, oxidative phosphorylation and also other metabolic pathways. Additionally, glycolytic and mitochondrial respiration were measured using the Seahorse XFe Analyzer. Specific analysis of the glycolysis metabolites revealed that while glycolysis was increased by metformin in all the cells, the intracellular levels of glucose, lactose and pyruvate varied significantly across the cell lines and were differentially affected by metformin treatment. Metformin had little impact on the TCA cycle intermediates in the A2780 cells, which were significantly decreased in C200 and in contrast increased in SKOV3ip cells. Functional analysis showed the oxygen consumption rate to be significantly inhibited by metformin in all the three cell lines, while the increased fatty acid oxidation intermediates were observed across all the three cell lines albeit to a varying extent. Exploration of the global metabolite changes by metformin across the three cell lines revealed 57 common altered metabolites, of which 30 had consistent direction change, with 16 metabolite being up and 14 being downregulated by metformin treatment. Metabolite Set Enrichment analysis showed linolenic acid and methionine metabolism as most enriched in metabolites being increased by metformin, and RNA transcription and pyrimidine metabolism as most enriched in the metabolites downregulated by metformin treatment. Ingenuity analysis indicated cellular proliferation and signaling as the top common network pathway modulated by metformin. In conclusion metformin treatment had a significant and wide-spread effect on metabolism of ovarian cancer cell lines. While metformin resulted in certain consistent metabolic changes, it had cell line specific modulation on glycolysis and oxidative phosphorylation metabolites. These differential metabolic changes could indicate the degree of metformin response and suggest it to be context-dependent. Thus information about the cancer metabolism will aid in preclinical and clinical interpretation of metformin therapy in ovarian and other cancers.

#12

Combinatorial intervention with natural compounds induces key metabolic modulations for prostate cancer prevention and treatment.

Xiyuan Lu, Bo Wang, Achinto Saha, Enrique Sentandreu, Alessia Lodi, John DiGiovanni, Stefano Tiziani. University of Texas at Austin, Austin, TX.

Prostate cancer is one of the three most relevant cancer types in men, and it ranks second in death rate and first in newly diagnosed cancer cases according to the US cancer statistics 2015. In recent years, the impact of nutrition on cancer prevention has been increasingly recognized. Accordingly the study of natural compounds for cancer prevention and treatment has drawn attention, mainly for the low toxicity to normal tissues.

In this study, a natural compound library of 140 compounds was screened on cultured mouse prostate tumor cells from HiMyc mice (HMVP-2 cells), and ATP and reactive oxygen species (ROS) bioluminescence measurements were performed. Based on ATP and ROS results, three hits, ursolic acid (UA), curcumin (CURC) and resveratrol (RES) were selected for more in depth metabolomic and lipidomic analyses. High-resolution liquid chromatography mass spectrometry (HPLC-MS) and magnetic resonance spectroscopy (MRS) were applied for large scale untargeted metabolic and lipidomic profiling of intra and extracellular prostate cancer extracts after treatment with the three drugs and their combinations at different time points. We next performed an in vivo study on an allograft mouse model of prostate cancer by injecting HMVP-2 spheroids into FVB/N mice.

After 12 hours of treatment, 115 metabolites in KEGG database and 12 classes of lipids (664 features) were included in the study. Principal component analysis (PCA) showed that the combination of UA and CURC exerts the most profound metabolic perturbation in HMVP2 cells compared to the individual treatment or the combination of other selected hits. Further data mining showed that the CURC+UA had a synergistic effect on cell metabolism by altering metabolic pathways associated with alanine, aspartate, proline and glutamate metabolism. Moreover, key intermediates in glycerophospholipid and ceramide metabolism were highly perturbed in CURC +UA indicating a relevant response of lipid mechanism to treatment with the combination of these agents. The in vivo mouse model produces palpable tumors within 10-14 days post injection. Dietary administration of CURC+UA and UA+RES showed significant inhibition of tumor growth compared to the administration of the individual compounds, with CURC+UA yielding the most effective combination.

In summary, amongst the 140 screened compounds, CURC, UA and RES exerted the most prominent metabolic effects on prostate cancers cells, and the combined CURC+UA treatment showed a synergistic effect on cell metabolism and significantly affected key metabolic pathways active in mitochondria, most likely via lipid metabolisms.

#13

Serum caffeine metabolites and prostate cancer risk in the ATBC Study.

Shakira M. Nelson,¹ Alison M. Mondul,² Stephanie J. Weinstein,¹ Demetrius Albanes¹. ¹National Cancer Institute, Rockville, MD; ²University of Michigan School of Public Health, Ann Arbor, MI.

Background: Etiologic studies of prostate cancer have shown inconsistent associations with coffee consumption. We have previously shown increased risk with number of cups/day, with the strongest association for men consuming 4+ cups/day. The association was evident in the first 10 years of cohort follow-up, however, suggesting possible reverse causality. We conducted a serum metabolomic profiling study to biochemically re-evaluate our coffee findings and determine whether caffeine metabolites are similarly associated with prostate cancer risk.

Methods: A prospective, nested case-control analysis within the Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) Study cohort was conducted using 200 cases and 200 controls matched on age and date of baseline fasting serum collection. Sera were analyzed using ultrahigh performance liquid chromatography/mass spectroscopy (LC-MS) and gas chromatography/mass spectroscopy (GS-MS) (Metabolon, Inc.) which identified 626 known compounds, including 13 caffeine-related metabolites. Each metabolite was assessed using conditional logistic regression that estimated odds ratios (OR) and 95% confidence intervals (CI) of prostate cancer associated with a one standard deviation (1-S.D.) increment in metabolite signal strength. The analysis was also stratified by median time from baseline blood draw to prostate cancer diagnosis (<10 years vs. ≥10 years).

Results: Overall, caffeine metabolite associations with prostate cancer risk showed null results (e.g., caffeine OR 1.02, 95% CI 0.84-1.24, p=0.82; paraxanthine OR 0.99, 95% CI 0.83-1.20, p=0.96). Similar to our findings for coffee consumption, caffeine metabolites were positively associated with prostate cancer risk for men diagnosed within 10 years of blood collection. Theobromine and 1,7-dimethylurate showed the strongest signals (respectively, OR 1.69, 95% CI 1.13-2.54, p=0.01; OR 1.63, 95% CI 1.14-2.32, p=0.01). By contrast, in men diagnosed at least 10 years from baseline, caffeine metabolites were non-significantly inversely related to risk; e.g., 5-acetylamino-6-formylamino-3-methyluracil (AFMU), OR 0.77, 95% CI 0.56-1.06, p=0.10). Caffeine metabolites were positively correlated with both caffeine and coffee (median correlation coefficients, 0.78 and 0.20, respectively). Including coffee in the metabolite risk models led to little or no attenuation of the associations in either the early or later 10 year observation periods.

Conclusions: In this prospective analysis, serum caffeine metabolites were positively associated with risk of prostate cancer in men diagnosed within 10 years of blood collection. The findings provide biochemical corroboration of our previous findings for a positive coffee-prostate cancer risk association, and appear to support the possibility of reverse causality; i.e., men with subclinical tumors may increase their coffee consumption as a result of their underlying disease

#14

A switch in metabolic phenotype in vemurafenib-resistant melanoma cells to increased amino acid dependence serves as an alternate mechanism for survival.

Deeba N. Syed,¹ Rahul K. Lall,¹ Yong Wei Soon,¹ Ahmed Aljohani,¹ Changan Guo,¹ Feng Liu,² Frank L. Meyskens,² James M. Ntambi,¹ Hasan Mukhtar¹. ¹University of Wisconsin-Madison, Madison, WI; ²University of California, Irvine, CA.

Despite considerable progress in the understanding of melanoma, and an increase in overall survival, resistance ultimately develops in most patients treated with BRAF inhibitors. To study the molecular mechanism(s) involved in acquired resistance, we generated a Vemurafenib-resistant A2-1b cell line in the BRAFV600E mutated SK-Mel28 melanoma cell background. Viability assays established IC50 of parental SK-Mel28 cells as 0.2 µM, while IC50>12 µM was noted for Vemurafenib-resistant A2-1b cells. Utilizing a quantitative proteomic strategy, profiling of A2-1b versus SK-Mel28 cells resulted in positive identification of 1720 proteins (≤1% FDR). Analysis of this data set identified a subset of 13 proteins, including Enolase 2, Triose-phosphate isomerase and Glyceraldehyde-3-phosphate dehydrogenase with a role in glycolysis, gluconeogenesis and NADH repair, to be differentially regulated in resistant cells. Additional studies validated significantly decreased Enolase 1 and 2 protein expression, suppressed glucose uptake and reduced lactic acid levels in resistant A2-1b cells. The pentose phosphate pathway, in contrast to citric acid cycle intermediates, was unaffected while the latter was significantly suppressed in resistant cells. The energy status of cells was similar as evidenced by equivalent total adenylate, guanylate, NADH, NADP+ and NADPH levels. Flux experiments measuring extracelluar acidification and oxygen consumption rates demonstrated that resistant A2-1b cells are unable to efficiently utilize intrinsic or exogenous fatty acids to meet energy demands. However, A2-1b cells maintained in media deprived of arginine and lysine exhibited increased basal and maximal respiration indicative of an energy reserve that relies on amino acid metabolism. Finally, metabolomics studies showing an overall increase in the absolute concentrations of gluconeogenic and ketogenic amino acids along with significantly elevated S6Kinase expression signify a shift from glucose to amino acids as the major energy source, in resistant A2-1b cells. Our data are consistent with a model of prolonged treatment response in which de-repression of ERK-MAPK pathway activity upon emergence of resistance to Vemurafenib is associated with adaptive changes in cellular metabolism. Sensitivity to BRAF inhibitors in the context of melanoma may not be defined by a reliance on glycolysis for survival, and that switch in metabolic phenotype can serve as an alternate mechanism for cell survival.

#15

Metabolomic profiling of cell death in human lung cancer cells by a novel digitoxin analog.

Yogesh Kulkarni,¹ Neelam Azad,¹ Vivek Kaushik,¹ Juan Sebastian Yakisich,¹ Rajkumar Venkatadri,¹ Clayton Wright,¹ Yon Rojanasakul,² George O'Doherty,³ Anand Krishnan V Iyer¹. ¹Hampton University, Hampton, VA; ²West Virginia University, Morgantown, WV; ³Northeastern Univrsity, Boston, MA.

Background:

Digitoxin, a cardiac glycoside had shown considerable promise as an anti-cancer therapeutic. However, doses of digitoxin required to inhibit cancer cell proliferation lead to cardiotoxic side effects. We have previously shown that a novel analog of digitoxin, MonoD, activates apoptosis in human lung cancer cells at much lower concentrations as compared to digitoxin. We postulate that treatment with MonoD leads to distinct metabolic signatures on lung cancer cells as compared to digitoxin, which will help delineate the pathways that dictate MonoD action.

Experimental Approach:

NCI-H460 cells from identical passages were used for metabolite analysis using an optimized version of the methanol/water (80:20) protocol for metabolite extraction. The supernatant containing metabolites were suspended in 0.1% formic acid spiked with an internal standard for Q-Exactive analysis. The data was analyzed using SIEVETM 2.0 software with the component extraction algorithm that has been specifically designed for optimal data analysis in untargeted metabolomics experiments. Metabolites that were found to be significantly different between treatment groups were detected via Chemspider/HMDB database search.

Results:

We observed several metabolites along the prostaglandin synthesis pathway which were downregulated in MonoD treated lung cancer cells. Prostaglandins play a key role in the generation of the inflammatory response. Chronic inflammation has been shown to be a causative factor in a variety of cancers. Cyclooxygenase‐2 (COX-2) derived prostaglandin has been previously shown to promote tumor growth by binding its receptors and activating signaling pathways which control cell proliferation, migration, apoptosis, and angiogenesis. Using the metabolite profiling data, we posited that COX-2, a key enzyme involved in prostaglandin synthesis, in addition to NFκB, a key transcription factor of proinflammatory signaling pathway are downregulated with MonoD treatment. Immunoblotting confirmed the downregulation of COX-2 and NFκB in MonoD-treated cells.

Conclusion: Using metabolite profiling, we have elucidated an important mode of action for MonoD in lung cancer cells, and shown that MonoD exerts its anti-tumorigenic properties by downregulating key factors involved in inflammatory response.

#16

NAMPT/visfatin as a novel prognostic marker and therapeutic target in metastatic melanoma.

Valentina Audrito,¹ Davide Brusa,¹ Sofia La Vecchia,¹ Aureliano Stingi,¹ Francesca Mazzola,² Gianna Baroni,³ Francesco Neri,¹ Barbara Merelli,⁴ Salvatore Oliviero,¹ Daniela Massi,³ Nadia Raffaelli,² Mario Mandalà,⁴ Silvia Deaglio¹. ¹Human Genetics Fndn./Univ. of Turin, Turin, Italy; ²Università Politecnica delle Marche, Ancona, Italy; ³University of Florence, Florence, Italy; ⁴Papa Giovanni XXIII Hospital, Bergamo, Italy.

INTRODUCTION: Treatment of BRAF(V600E) mutant metastatic melanoma (MM) with BRAF inhibitors (BRAFi) can be highly effective. However, resistance to therapy develops rapidly, with paradoxical activation of the MAP kinase signalling. For this reason novel drug combinations are being tested.

Nicotinamide phosphoribosyltransferase (NAMPT) is a key enzyme in the biosynthesis of NAD, an adenine-based dinucleotide, essential cofactor in redox reactions and a substrate of key cellular enzymes. This enzyme may also be secreted in the extracellular space (eNAMPT), where it helps establish a cyto-protective microenvironment, favouring resistance to therapy. NAMPT inhibitors are being tested as cancer therapeutics.

The aim of this work is to determine function of NAMPT in MM and in BRAFi resistant patients.

METHODS: NAMPT expression and activity in MM samples was studied using biochemical, enzymatic, immunofluorescence, immunohistochemical and ELISA assays. RNAseq and functional/metabolic analysis were also performed.

RESULTS: After establishing BRAFi-resistant melanoma cell lines, we observed increased glycolysis and mitochondrial respiration compared to sensitive cells. In line with higher metabolic needs, higher levels of NAD were measured in resistant cells, suggesting that increased NAD synthesis supports the activated metabolic phenotype.

NAMPT levels in resistant cells were markedly elevated both at the transcriptional and protein levels, confirming that this is a critical enzyme in the NAD generation. Inhibition of the BRAF/MEK axis using different agents decreased NAMPT, suggesting a direct relationship between activation of the MAPK axis and NAMPT transcription.

Data using samples from MM patients (n=80) demonstrated that i) NAMPT is up-regulated in melanoma compared to normal melanocytes and ii) NAMPT expression sharply increased upon development of BRAFi resistance (p=0.001). Moreover, eNAMPT levels were markedly elevated in MM sera compared to healthy donors or to patients with localized melanomas. Furthermore, patients with eNAMPT levels >15 ng/ml were characterized by a shorter survival. Lastly, in our cohort of patients eNAMPT decreased sharply after the initiation of therapy with BRAFi, only to increase again in patients developing resistance.

Finally, pharmacological modulation of NAMPT affects melanoma progression and responses to BRAFi. Resistant cells treated with NAMPT inhibitors, FK866 and GMX1777, reduced their proliferation rate. Moreover, NAMPT inhibition led to i) NAD and ATP levels reduction, ii) mitochondrial respiration impairment and iii) decrease of ERK1/2 and NF-kB activation.

CONCLUSIONS: Together, these findings indicate that NAMPT is overexpressed during melanoma transformation and further increased upon paradoxical activation of the MAPK axis. They also provide preliminary data to test NAMPT inhibitors in combination with BRAFi in the treatment of patients with BRAF-mutated MM.

#17

Metabolic analysis of IDH mutant gliomas in Drosophila.

Michaela Brown, Jenna Buccetti, Marla Tipping. Providence College, Providence, RI.

Metabolic reprogramming is a common hallmark shared by nearly all proliferating cancer cells and has emerged as an exciting new direction in cancer research. Many signaling pathways have been implicated in mechanisms leading to the shift of metabolic programs in tumors, but more recently a small number of metabolic enzymes have also been identified in this process. Genes encoding the metabolic enzymes Isocitrate dehydrogenase 1 (IDH1) and 2 (IDH2) were found to be mutated in up to 70% of low-grade and medium grade gliomas, and in 15-20% of adult acute leukemia samples. These findings were the first to link the IDH gene to tumorigenesis. IDH1 and IDH2 function to irreversibly catalyze the oxidative decarboxylation of isocitrate to α-ketoglutarate (α-KG). Although these are important metabolic enzymes, little is known about the metabolic impact harboring mutant IDH protein has on cells. Our goal is to model the IDH mutant phenotype in Drosophila glial cells for further characterization of both metabolic status, and IDH enzymatic activity. To this end, we have generated Drosophila cell lines and transformant flies that express the most commonly identified IDH mutations under control of the UAS-Gal4 system. We have also investigated the metabolic flux of Drosophila cell lines expressing IDH mutant protein, and loss of IDH function, and glycolytic enzyme expression analysis. Currently, we are investigating the metabolic flux analysis (MFA) using labeled metabolites as well as comparing the protein interactions of the wild type to the mutant IDH cells. The results of these investigations will identify potential new targets for the treatment of aggressive gliomas at the level of cellular metabolism.

#18

Galactose impacts cell growth and redox status of glioblastoma multiforme U87 cell lines.

Jan C. Lumibao,¹ Vladimir Kolossov,² Matthew T. Leslie,¹ H R. Gaskins¹. ¹University of Illinois at Urbana-Champaign, Urbana, IL; ²Carl R. Woese Institute for Genomic Biology, Urbana, IL.

Glioblastoma multiforme (GBM) is the most common and malignant form of brain cancer in adults, with a median survival time of 12-15 months. Characterizing the metabolic alterations of GBM is a crucial step towards identifying potential targets for treatment, with mitochondria, the electron transport chain, and associated reactive oxygen species presenting lucrative targets for therapy. However, the Warburg effect, in which cells generate ATP mainly via glycolysis despite adequate oxygen and functional mitochondria, can render cancer cells resistant to antagonists of the electron transport chain. This prompts investigations into other nodes of oncogenic metabolism such as the epidermal growth factor receptor (EGFR)/phosphoinositide 3-kinase pathway which can regulate glucose metabolism. The present work defined the sensitivity of GBM U87MG, U87OE (over-expressed EGFR), and U87vIII (hyper-activated vIII mutated EGFR) cell lines to media supplemented with galactose versus conventional glucose concentrations. Cells were transfected with a mitochondrial-targeted Glrx-roGFP2 redox biosensor, a mutated GFP protein whose fluorescence reflects the local redox potential, and incubated in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with either 25 mM glucose (DMEM-Glu) or 10 mM galactose with and without 1 mM pyruvate (DMEM-Gal±Pyr). Glucose deprivation combined with galactose supplementation allows metabolic isolation of mitochondrial contributions to ATP generation because the glycolytic metabolism of galactose yields no net ATP, forcing cells to rely on increased mitochondrial respiration. Cells were exposed to differential media conditions and assessed via flow cytometry to monitor mitochondrial redox homeostasis after 24 h of media exposure. Wild type U87 cells exhibited a slight reduction upon glycolytic blockade, while mitochondria in the mutant EGFR U87OE and U87vIII cells exhibited extreme oxidation. The sulforhodamine B assay, which determines cell density based on a colorimetric measurement of cellular protein content, was performed to assess cell concentration in different media after 72 h of media exposure. Mutant EGFR U87OE and U87vIII cells experienced much greater growth inhibition and cell death compared to wild type U87MG cells. Thus, EGFR-activating mutations increase cellular dependence on glycolytic ATP generation for cell growth and redox homeostasis. Future work will build on these findings to characterize how the metabolic phenotypes and preferences of these cells are linked to their migratory behavior in an engineered 3D hydrogel environment.

#19

Neurofibromatosis type 1 (NF1) status determines sensitivity of soft tissue sarcoma and melanoma cell lines to glutaminase inhibitors.

Tahir N. Sheikh, Parag P. Patwardhan, Gary K. Schwartz. Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY.

Neurofibromatosis type 1 (NF1) is an autosomal dominant genetic syndrome caused by a mutation in or deletion of the NF1 gene. Mutations in the NF1 gene lead to the production of a nonfunctional version of neurofibromin that cannot regulate cell growth and division. As a result, tumors such as neurofibromas can form along nerves throughout the body. The NF1 phenotype is highly penetrant and occurs in 1 in 3,000 to 4,000 people worldwide. Individuals with an altered NF1 gene are at an increased risk of developing benign and/or malignant tumors.  NF1 has been shown to negatively regulate Ras activity. Ras-driven cancer cells have also been known to alter glucose and glutamine metabolism. In the present study, we evaluated the role played by NF1 status in determining the sensitivity of sarcoma and melanoma cell lines to glutaminase inhibition and its effect on Ras activity. We tested a panel of soft tissue sarcoma and melanoma cell lines that were either wild-type, null or mutant for NF1. Results from our in vitro proliferation assay showed that compared to wild-type NF1 (STS26T, LS141) cell lines, NF1 mutant (MPNST) and NF1 null (ST88 and MeWo) cell lines showed greater sensitivity to inhibition of proliferation by glutaminase inhibitors CB-839 and BPTES. Western blot analysis showed induction of apoptosis and down regulation of mTORC1 targets such as phospho-S6K and phospho-S6 by glutaminase inhibitors only in NF1 null and NF1 mutant but not wild-type NF1 cell lines. Gene silencing experiments showed that siRNA mediated knockdown of NF1 sensitizes LS141and STS26T cell lines to glutaminase inhibition. Conversely, overexpression of wild-type NF1 GRD (GAP related domain) in MeWo cell line resulted in decreased sensitivity to glutaminase inhibition when tested in a cell proliferation assay, thus, confirming the role played by NF1. Previous reports have shown that mutation or deletion of NF1 results in activation of Ras. In order to test the effect of glutaminase inhibition on Ras activity, we carried out Ras-GTP pull down assay following the treatment with glutaminase inhibitors in NF1 null and wild-type NF1 cell lines. Our results showed that glutaminase inhibition leads to down regulation of activated RAS in NF1 null but not wild-type NF1 cells. SiRNA mediated knockdown of NF1 followed by glutaminase inhibition in wild-type NF1 cell line (LS141) resulted in decreased Ras activity, further confirming our hypothesis. Results from patient derived MPNST tumor xenograft model showed a significant suppression of tumor volume when tumors were treated with glutaminase inhibitor compared to vehicle control. Taken together, our data strongly indicates that NF1 status determines the sensitivity of sarcoma and melanoma cell lines to glutaminase inhibition. Further research is warranted to explore glutaminase inhibition as potential therapy for patients with NF1 loss and/or mutation.

#20

Outlier overexpression of genes involved in fatty acid metabolism and processing define two aggressive prostate cancer phenotypes.

Sandra M. Gaston,¹ Kerry Dehimer,² James Kearns,² Kyle-Ravi Chinsky,³ Dennis Otali,⁴ Denise Oelschlager,⁴ Soroush Rais-Bahrami,⁴ Jeffrey W. Nix,⁴ Peter Kolettis,⁴ George W. Adams,⁵ William E. Grizzle⁴. ¹New England Baptist Hospital, Tufts Medical Center and Tufts Univ School of Medicine, Boston, MA; ²Tufts Medical Center, Boston, MA; ³Tufts University, Boston, MA; ⁴University of Alabama at Birmingham, Birmingham, AL; ⁵Urology Centers of Alabama, Birmingham, AL.

Major shifts in fatty acid metabolism are part of the malignant phenotype of many types of cancer, including prostate cancer. We have used a biopsy based approach to characterize high grade prostate cancers in both African American and European American patients. This approach allows us to analyze prostate cancers from patients whose disease is too advanced for radical prostatectomy and who are thus not represented in the study specimens available from conventional biorepositories. mRNA gene expression profiling of high grade prostate cancers from biopsies revealed cancer subtypes with very high (>10 fold) levels of overexpression of fatty acid binding protein 5 (FABP5) and fatty acid synthase (FASN). Interestingly, these prostate cancer subtypes showed an either-or phenotype, with super-overexpression of either FABP5 or FASN but not both. These findings suggest two outlier prostate cancer subtypes with striking differences in the pathways that drive a shift in tumor fatty acid metabolism; one is a fatty acid synthase (FASN) dominant phenotype and the other a previously unrecognized fatty acid binding protein (FABP5) dominant phenotype. In contrast, prostate cancers with lower levels of FABP5 and/or FASN overexpression (2-4 fold) often co-overexpressed both of these genes. Immunohistochemical analyses confirmed prostate cancer FABP5 and FASN overexpression; both showed strong cytoplasmic staining. FABP5 also showed robust nuclear staining consistent with its proposed role in regulating fatty acid mediated gene expression. At both the mRNA and protein levels, the high grade prostate cancers with the highest levels of FABP5 over-expression were more common in African Americans, while the highest levels FASN overexpression were more common in prostate cancers from European Americans. These findings may provide the basis for more effective dietary interventions and targeted therapies for African American and European American patients with these two different subtypes of high grade prostate cancer.

#21

The mTORC2 target Akt is regulated in response to glutamine metabolite levels.

Estela Jacinto,¹ Joseph Moloughney,¹ Peter K. Kim,¹ Nicole M. Vega-Cotto,¹ Chang-Chih Wu,¹ Thomas Lynch,¹ Sisi Zhang,² Matthew Adlam,¹ Sai Guntaka,¹ Po-Chien Chou,¹ Joshua D. Rabinowitz,² Guy Werlen¹. ¹Rutgers-RWJ Medical School, Piscataway, NJ; ²Princeton Univ., Princeton, NJ.

Highly proliferating cells are particularly dependent on glucose and glutamine for bioenergetics and to fuel biosynthesis of macromolecules. The signals that respond to the fluctuations of these nutrients and how they control metabolic pathways remain poorly understood. mTOR, as part of mTOR complex 1 (mTORC1), responds to amino acids and plays a central role in metabolism. On the other hand, little is known on how mTORC2, consisting of the core components mTOR, rictor, SIN1 and mLST8 is regulated and its metabolic functions. The phosphorylation of the mTORC2 substrate, Akt, is enhanced in cancer cells, suggesting that mTORC2 becomes deregulated during tumorigenesis. Here we found that the activity of mTORC2 is enhanced by diminishing glutamine-derived metabolites. mTORC2 activity is required by glutamine-requiring biosynthetic pathways such as the hexosamine biosynthetic pathway (HBP). Acute nutrient withdrawal augments Akt phosphorylation but does not affect GFAT1 expression. However, extreme starvation that eventually depletes intracellular glutamine metabolites inactivates mTORC2 and downregulates GFAT1 expression. Thus, while mTORC1 senses glutamine abundance to promote anabolism, mTORC2 responds to declining glutamine catabolites in order to restore metabolic homeostasis. Our findings uncover the role of mTORC2 in metabolic reprogramming and provide insights on more effective therapeutic strategies for glutamine-dependent tumors.

#22

Investigating the functional contribution of TANK binding kinase 1 to inflammation induced disease progression.

Victoria H. Burton, Rolf A. Brekken. UT Southwestern, Dallas, TX.

Initial stages of human pancreatic ductal adenocarcinoma (PDA) are commonly characterized by an activating mutation in K-RAS along with extensive immune cell infiltration. Direct inhibition of K-RAS through pharmacological means remains a challenge; however targeted inhibition of TANK Binding Kinase 1 (TBK1), a critical downstream effector of mutant active K-RAS is an attractive alternative. High levels of active TBK1 are associated with inflammatory disease and cancer progression. TBK1 and homolog, IKKe activate the immune response transcription factor NF-κB. In metabolically challenged mice, IKKe regulates energy balance by sustaining chronic, low-grade inflammation. We hypothesize that TBK1 signaling is also critical in metabolic regulation and is required for progression of inflammation-induced diseases such as PDA.

A kinase dead, Tbk1Δ/Δ mouse was used to determine the contribution of TBK1 to metabolic disease and PDA progression. Metabolic phenotyping experiments were performed with Tbk1Δ/Δ and Tbk1+/+ mice including a high fat diet weight study, body composition and metabolic chamber analyses. Tbk1+/+ and Tbk1Δ/Δ mice were crossed into a genetically engineered mouse model of PDA to determine the consequences of genetically removing Tbk1 on tumor development and overall survival.

Here we report that Tbk1Δ/Δ mice are significantly smaller, leaner and have less fat than Tbk1+/+ mice on high fat chow diets (HFD) for 14 weeks. Tbk1Δ/Δ mice are more active and have smaller and more abundant adipocytes relative to Tbk1+/+ mice. Additionally, Tbk1Δ/Δ mice are protected from HFD induced hypercholesterolemia and liver steatosis. White adipose tissue (WAT) from HFD fed Tbk1Δ/Δ mice exhibit an induction of brown fat gene expression suggesting that thermogenesis is a contributor to their healthier and more active phenotype. Macrophage infiltration is fairly low in WAT from Tbk1+/+ and Tbk1Δ/Δ mice on normal chow. However on HFD, WAT from Tbk1+/+ mice display a significant increase in macrophage marker expression compared to Tbk1Δ/Δ mice. These findings indicate that HFD fed Tbk1Δ/Δ mice are protected from classic phenotypes of metabolic syndrome.

In PDA, TBK1 is expressed and more active in human PDA cell lines relative to immortalized pancreatic epithelial lines and fibroblasts. Human PDA cell lines are sensitive to a small molecule inhibitor of TBK1 (compound II) in the low micromolar range. In a K-Ras driven genetic mouse model of PDA, TBK1 supports spontaneous pancreatic tumor growth as evidenced by smaller tumors in Tbk1Δ/Δ: PDA mice relative to Tbk1+/+: PDA mice. Our results suggest that TBK1 contributes to metabolic regulation and demonstrate the therapeutic potential of targeting TBK1 in pancreatic malignancies. Current work is focused on delineating the inflammatory and metabolic dysregulation in these animals and determining the precise mechanism by which TBK1 supports the progression of PDA.

#23

Caloric restriction slows tumor growth and metastases in both hormone-sensitive and hormone-resistant prostate cancers.

Robert S. Gitman,¹ Meredith LaRose,¹ Edouard J. Trabulsi,² Ajay Palagani,² Tiziana DeAngelis,¹ William K. Kelly,¹ Leonard G. Gomella,² Nicole L. Simone². ¹Thomas Jefferson University Hospital, philadelphia, PA; ²Thomas Jefferson University Hospital, Philadelphia, PA.

Introduction: The purpose of this study is to assess the effect of caloric restriction (CR) on prostate cancer (PCa) tumor burden, proliferation, and overall survival with and without radiation treatment (RT). Additionally, we determined if the beneficial effect of CR noted are due to a reduction in pro-inflammatory markers hypothesized to enhance PCa tumorgenesis.

Methods: To assess the effect of CR in vivo, 40 male 12 week-old NCRNu-M mice were injected with LNCaP or PC3 tumor cells. Once tumors were palpable, mice were randomly assigned to one of four treatment groups in cohorts of 10: ad libitum (AL) diet, 6.5Gy of radiation (RT), 30% reduction in caloric intake (CR), or CR+RT. Tumor growth and proliferation rate were measured 3 times per week via calipers and live bioluminescent imaging. Additionally, PCa tumor tissue were analyzed to determine if CR can exert its effect on tumor growth and metastases via IGF-1R signaling pathway.

Results: Adding CR to RT decreased tumor progression. In the PC3 murine model, when compared to AL, RT reduced tumor size by 22%, 77% with CR, and 80% with CR+RT. In the LNCaP murine model, compared with AL, CR reduced tumor size by 49% and a 55% reduction with CR+RT. Primary tumor formation began 9 weeks post tumor injection in the radiated cohort and was delayed to 15 weeks post tumor injection in the CR and CT+ RT cohorts. Additionally, time to metastasis was delayed with CR (86 days to metastasis in AL, 93 with RT, and 108 when combining CR+RT). Molecularly, CR decreased multiple members of the IGF-1R signaling pathway. The total IGF-1 and IGF-1R levels exhibited 50% reduction with CR and CR+RT but total INSR levels. CR also induced a significant decrease of pGSK3β and pAKT levels.

Conclusions: For the first time, we have shown that decreasing calories by 30% in both hormone-sensitive and hormone-refractory prostate cancer models, enhance the efficacy of radiation. Our experiments show that even CR alone and in combination with RT can improve PCa tumor proliferation rate, tumor burden, time to metastasis, and overall survival. We hypothesize that these changes are in part, due to the decrease of the IGF-1R signaling pathway. We propose that CR may be used as a novel therapeutic intervention to enhance outcomes of radiation treatment by altering the molecular profile of prostate tumors.

#24

Loss of ubiquitin-specific peptidase 18 (USP18) causes cold sensitive mice by destabilizing the critical regulator of thermogenesis: uncoupling protein-1 (UCP-1).

Xi Liu,¹ Yun Lu,² Lin Zheng,¹ David J. Sekula,¹ Sarah J. Freemantle,² Ethan Dmitrovsky¹. ¹University of Texas MD Anderson Cancer Center, Houston, TX; ²Geisel School of Medicine, Hanover, NH.

USP18 is the major ISG15 (Interferon-Stimulated Gene 15) deconjugase that removes ISG15 from substrate proteins. We recently reported that USP18 null mice spontaneously develop leiomyosarcomas. Intriguingly, these USP18 null mice are also markedly cold sensitive as compared to their wild-type littermates. When briefly exposed to cold stimuli, USP18 null mice significantly (P < 0.05) decreased their body temperature. In contrast, wild-type mice exposed to the same experimental conditions exhibited transient changes in their body temperature. We sought to elucidate the engaged mechanism. Expression profiles of critical thermogenic regulatory proteins were examined by immunoblot analyses of brown fat tissues of USP18 null mice. Strikingly, UCP-1 expression was substantially reduced in these tissues of USP18 null versus wild-type mice. To independently confirm that reduced UCP-1 expression was caused by loss of USP18, stable knock-down of USP18 was independently achieved in a panel of murine cell lines by use of transfected small hairpin RNAs (shRNAs). The obtained results were compared to that of transfected control vectors. USP18 down-regulation by different shRNAs markedly reduced UCP-1 protein as compared to controls. As expected, engineered gain of USP18 expression in the same murine cell lines stabilized UCP-1 protein. We explored if UCP-1 destabilization was due to its complex formation with the ubiquitin-like protein ISG15. This was the expected experimental outcome since USP18 is the distinct deconjugase that removes ISG15 from complexed proteins. Immunoprecipitation assays were performed to establish this direct association. These studies revealed a complex formed between ISG15 and UCP-1 protein. Thus, engineered loss of USP18 in mice led to cold sensitive mice from repression of UCP-1. As a consequence of its complex with ISG15, the thermogenic regulator UCP-1 was destabilized. Taken together, this study extends prior work by showing a previously unrecognized link between USP18 and regulation of thermogenesis. This defines a novel role for the USP18 protease in metabolism.

#25

SCAP links glucose to lipids for tumor growth.

Chunming Cheng, Feng Geng, Jeffrey Yunhua Guo, Xiaoning Wu, Xiang Cheng, Arnab Chakravarti, Deliang Guo. The Ohio State University Comprehensive Cancer Center, Columbus, OH.

Increased glucose consumption and elevated lipogenesis are co-occurred in malignancies. But the molecular mechanism of the link between glucose and lipid metabolism remains elusive. Our previous studies have uncovered that sterol regulatory element-binding protein (SREBP-1), an endoplasmic reticulum-bound transcription factor with a central role in lipid metabolism, is highly upregulated in glioblastoma (GBM), the most deadly brain tumor. In the current study, we found that SREBP-cleavage activating protein (SCAP), a key player regulating SREBP trafficking from ER to the Golgi and subsequent SREBP activation, links oncogenic signaling and glucose consumption to lipid metabolism. Our data showed that glucose is a critical activator for SREBP function via N-glycosylation of SCAP. N-glycosylation is a key signal to promote SCAP/SREBP complex move from ER to the Golgi and SREBP activation. Moreover, we found that EGFR/PI3K/Akt signaling activates SREBP-1 via upregulation of SCAP N-glycosylation and its protein levels through enhancing glucose uptake. Genetically silencing SCAP or mutation of the N-glycosylation sites on SCAP downregulates SREBP-1 activity and impairs GBM tumor growth. Taken together, our study revealed that glucose is a critical activator for SREBP-1 function and lipid metabolism, and SCAP integrates oncogenic signaling and fuel availability to SREBP-1-mediated lipogenesis for tumor growth. Our study also demonstrated that targeting SCAP N-glycosylation represents a promising therapeutic strategy to treat malignancies.

#26

Sensitive step in cholesterol biosynthesis reveals role for sterol metabolites in regulating growth of EGFR/KRAS-dependent tumors.

Linara Gabitova, Diana Restifo, Elizabeth Handorf, Kathy Q. Cai, Igor A. Astsaturov. Fox Chase Cancer Center, Philadelphia, PA.

We identified SC4MOL and NSDHL, two enzymes in the cholesterol pathway and their substrates, meiosis activating sterols (MAS), as critical regulators of receptor signaling and trafficking in normal development and in cancer. Oncogenic transformation by EGFR or RAS increases the demand for cholesterol, suggesting a possibility for metabolic interference. To test this idea in vivo, we ablated Nsdhl in adult keratinocytes expressing KRASG12D. Strikingly, Nsdhl inactivation antagonized the growth of skin tumors, while having little effect on normal skin. Loss of Nsdhl induced the expression of ATP-binding cassette (ABC) transporters ABCA1 and ABCG1, reduced the expression of low-density lipoprotein receptor (LDLR), decreased intracellular cholesterol and was dependent on the liver X receptor (LXR) α. Importantly, EGFR signaling opposed LXRα effects on cholesterol homeostasis, while an EGFR inhibitor synergized with LXRα agonists in killing cancer cells. Inhibition of SC4MOL or NSDHL, or activation of LXRα by sterol metabolites can be an effective strategy against carcinomas with activated EGFR-KRAS signaling.