Kylie Mitchell , Sonnet Davis, Alexander Patent, and Arvind Ramanathan

1. Classes of Metabolites for Analysis Metabolic Action of Metformin in Caenorhabditis elegans . Application of High-Performance Liquid Chromatography-Mass Spectrometry Platform to Study Metabolism and Epigenetic Control of Metabolism Kylie Mitchell, Sonnet Davis, Alexander Patent, and Arvind Ramanathan Dominican University of California, and Buck Institute for Age Research, Novato, CA Introduction Epigenetic Control of Cancer Metabolism Naturally occurring small molecules (metabolites, signaling intermediates) are a critical component of the information flow in biology, along with DNA, RNA, and proteins. Metabolomics is an analytical approach that seeks to comprehensively analyze naturally occurring small molecules and quantify their dynamic changes in biological systems. In recent years metabolomics has begun to provide understanding of the metabolic basis of different diseases, such as heart disease, cancer, and diabetes. Our lab built a high performance liquid chromatography mass spectrometry (HPLC-MS) based metabolomics platform to analyze metabolites from mammalian cells, spent cellular medium, and model organisms such as C. elegans. We used C. elegans to elucidate the metabolic changes seen after treatment with Metformin; which is a known activator of the AMPK pathway. Cancer cells exhibit high levels of glycolysis producing large amounts of lactate; circumventing the mitochondrial pathway. This phenomenon is known as the Warburg effect. We hypothesize that cancer metabolism is epigenetically regulated. Epigenetics refers to inheritable traits that are not due to alterations in the primary DNA sequence. DNA methylation is an important epigenetic modification. DNA methylation mainly occurs in the CpG islands of the promoter region of genes. It is believed that during cancer development de novo DNA methyltransferases methylate tumor suppressing genes allowing cancer cells to proliferate uninhibited. There are two de novo DNMTs, DNMT3A, and DNMT3B. These DNMTs establish the pattern of methylation. We examined de novo DNMT mediated control of cellular metabolism, identifying global changes in metabolism, as well as differential sensitivity towards glycolytic and mitochondrial inhibitors. DNA methyltransferases affecting cancer metabolism Polar Metabolites Non-Polar Metabolites Figure 5: Based on preliminary evidence we hypothesized that cancer metabolism is epigenetically regulated by de novo DNMTs. A possible mechanism for this is via the a DNMT mediated transcriptional regulation of transcription factors (T.F.) involved in metabolism. PGC1α is a transcription factor, which is a master regulator of mitochondrial biogenesis, and energy homeostasis. PPARγ regulates fatty acid storage and glucose metabolism. HIF1α is important in oxygen homeostasis and becomes highly upregulated during oxygen deprivation, which is a known aspect of the Warburg Effect. Understanding the DNA methylation mediated regulation of these metabolic transcription factors could shed light on the epigenetic basis of the Warburg Effect. DNMT3A DNMT3B • Cancer Metabolism • PGC1α • HIF1α (Warburg Effect) • PPARγ • Other transcription factors? Cellular Metabolomics Profile Cancer Development / Warburg Effect Global Metabolism Changes after siRNA Mediated silencing of DNMT’s Specific Aims A. Figure 6: Global metabolism changes. After siRNA mediated silencing of DNMT’s in lung adenocarcinoma cells (A549) global metabolism changes can be seen. Figure 6A shows the de novo DNMTs lactic acid production. DNMT3A knockdown produces significantly higher levels of lactic acid. Figure 6B represents metabolites in the Specific Aim 1: Develop a High Performance Liquid Chromatography (HPLC) Mass Spectrometry based metabolomics platform. Specific Aim 2: Develop metabolomics ofC. elegans. 2A: Perform lifespan assay of ahr-1 null C. elegans in the presence of Metformin. 2B: Metabolic analysis on mass cultures of C. elegans treated with 50 mM Metformin. Specific Aim 3: Characterize the epigenetic control of metabolism 3A: Conduct a global analysis of metabolism changes after siRNA mediated silencing of DNMT’s. 3B: Identify potential transcription factors that are involved in epigenetic metabolism memory. Table 1: Ramanathan Metabolite Library. Metabolites have a wide range of chemical properties. Our database includes metabolites from different metabolic pathway, and chemical classes. *p ≤0.05 glycolytic pathway. DNMT1 knockdown demonstrates slight increases and decreases, similar to DNMT3B knockdown, the effect is more robust with DNMT3B samples. DNMT3A knockdown illustrates a significantly increased trend of metabolites. B. siNT siDNMT1 Lifespan of ahr-1 null C. elegans in presence of 50 mM Metformin *p ≤ 0.05 *p ≤ 0.05 *p ≤ 0.05 High-Performance Liquid Chromatography Mass-Spectrometry *p ≤ 0.005 *p ≤ 0.005 Agilent 6520 Accurate Mass Q-TOF *p ≤ 0.05 Figure 3: Lifespan of ahr-1 null worms in presence of 50 mM Metformin. Metformin significantly increased wild-type, as well as mutant worm lifespans. The lifespan extending affect of Metformin is still seen in ahr-1 null worms, leading us to believe ahr-1 is not Metformin’s mechanism of action. Transcription Factors Involved in Epigenetic Metabolism Memory A. B. Metabolomics Analysis of C. elegans treated with 50 mM Metformin B. A. A. Figure 1: Agilent 6520 Accurate Mass Quadrupole Time of Flight (Q-TOF) instrument. Metabolites are first separated using HPLC, then directly injected into the mass spectrometer via electrospray ionization, then metabolites fly into the flight tube. The time of flight of metabolites directly correlates to the mass of metabolites. The 6520 has a mass accuracy of 0.001 amu, reducing the number of potential chemical formulas. Overview of Experimental Design Cell Culture Figure 7: Transcription Factors involved in Epigenetic metabolism. PGC1α and HIF1α levels are significantly altered upon siRNA mediated silencing of DNMT’s. The de novo DNMT’s seem to be differentially regulating these transcription factors. • Human lung epithelial adenocarcinoma cell line (A549) Extraction 3:1:1 CHCl3:MeOH: H2O • Intracellular and extracellular metabolites • Polar and non-polar Metabolites Conclusions & Future Directions Garg, U., and Dasouki, M. (2006). Expanded newborn screening of inherited metabolic disorders by tandem mass spectrometry: clinical and laboratory aspects. Clinical biochemistry 39, 315-332. • Metformin’s exact mechanism of action is still unclear. Metformin causes global metabolism changes in C. elegans, the BCAA pathway could be working synergistically with other pathways. • DNA methylation is an important epigenetic modification. DNMT’s play a major role in cancer development, and metabolism. Understanding the relationship between these biological factors can shed light upon the mysterious Warburg Effect. • Further analysis of methylation levels in the promoter region of metabolic transcription factors can provide insight into how methylation controls metabolism. • A transcriptional profile of siRNA mediated knockdown of de novo DNMTs will be performed. Data Analysis Figure 4: Metabolomics with mass cultures of C. elegans: Day 1 adult worms were plated on 50 mM Metformin or FUdR plates. Worms were transferred to a new plate everyday for 5 days. At the end of the treatment metabolites were extracted. Figure 4A. is a subset of metabolites that were identified to be significantly altered after Metformin treatment. Figure 4B. A schematic of the Branched-Chain Amino Acid (BCAA) Degradation pathway.(Garg and Dasouki, 2006) Figure 4C. Additional metabolites from BCAA pathway were not found to be significantly altered. Our initial hypothesis was Metformin was working through this pathway, however since no additional metabolites were found significantly altered, analysis of additional pathways is necessary. • Quantitative and qualitative analysis • Statistical analysis • Biochemical pathway mapping LC-MS Analysis C. Targeted Profiling • “Hypothesis driven” • Database of “known” compounds Untargeted Profiling • “Discovery driven” • Identify all compound “known” and “unknown” • Liquid Chromatography (LC) • Mass Spectrometry (MS) www.agilent.com Acknowledgements Figure 2: Overview of the experimental design. Cells are grown to confluency, then intracellular and extracellular; polar and nonpolar metabolites are extracted. Samples are run on the HPLC-MS, then data analysis is conducted. Finally, pathway mapping is completed to put the dynamic changes into a biologically relevant context. I would like to acknowledge Dominican University of California and the Buck Institute for Research on Aging. I would also like to acknowledge Dr. Arvind Ramanathan, Dr. Sonnet Davis, Alexander Patent, Dr. Samiha Mateen, Nina Mahale, Matt Aguirre, the Lithgow lab, Gibson lab, and Benz lab.