BSA 2016 Symposium Abstracts

THE ω-HYDROXYLASE ACTIVITY OF CYTOCHROME P450 4A11 IS ATTENUATED IN THE PRESENCE OF HYDROGEN PEROXIDE: A POSSIBLE INSIGHT TO BLOOD PRESSURE CONTROL

Matt Albertolle, Donghak Kim, F. Peter Guengerich
Cytochrome P450 (CYP) 4A11 is a ω-hydroxylase that catalyzes the oxidation of arachidonic acid (AA) to 20-hydroxyeicosatetraenoic acid (20-HETE). 20-HETE can be pro-hypertensive by constricting microvasculature or anti-hypertensive by inhibiting salt reabsorption in the kidney. Thus, 20-HETE and CYP4A11 have important roles in blood pressure regulation. CYP4A11 activity is attenuated in the presence of hydrogen peroxide, an important secondary signaling molecule. This leads to the central hypothesis that oxidation of CYP4A11 alters regulation and enzymatic synthesis of hypertensive mediators that contribute to pro- and/or anti-hypertensive phenotypes. The results indicate that thiol redox status alters the enzymatic activity of CYP4A11, a biological control mechanism that mediates 20-HETE production. 

 

IDENTIFICATION OF A SUBSTRATE RECOGNITION DOMAIN IN THE REPLICATION STRESS RESPONSE PROTEIN ZINC FINGER RAN-BINDING DOMAIN CONTAINING PROTEIN 3 (ZRANB3)

Akosua Badu-Nkansah, Aaron Mason, Brandt Eichman, David Cortez
DNA damage and other forms of replication stress can cause replication forks to stall. Replication stress response proteins stabilize and resolve stalled forks by mechanisms that include fork remodeling to facilitate repair or bypass of damaged templates. Several enzymes including SMARCAL1, HLTF, and ZRANB3 catalyze these reactions. SMARCAL1 and HLTF utilize structurally distinct accessory domains attached to an ATPase motor domain to facilitate DNA binding and catalysis of fork remodeling reactions. Here we describe a substrate recognition domain within ZRANB3 that is needed for it to recognize forked DNA structures, hydrolyze ATP, catalyze fork remodeling, and act as a structure-specific endonuclease. Thus, substrate recognition domains are a common feature of fork remodeling, SNF2-family, DNA-dependent ATPases, and our study provides further mechanistic understanding of how these enzymes maintain genome integrity during DNA replication. 

 

METHODS FOR MONITORING DNA METHYLATION AND CHROMATIN DYNAMICS AT GENE REGULATORY ELEMENTS

Kelly Barnett, Bob Chen, Emily Hodges
DNA methylation is thought to be a critical component of the gene regulatory mechanisms that guide developmental cascades and stabilize cell fate decisions. In a typical mammalian genome, cytosine methylation is the most common, default state, with 70-80% of all CpG sites modified in non-tumorigenic settings; yet a large number of discrete domains, hypomethylated regions (HMR), escape this modification. Previous work (Hodges et al.) identified a number of highly cell-type specific intergenic and intronic HMRs (iHMRs). These iHMRs have been found to overlap chromatin marks associated with enhancers, and be variable in methylation state across multiple cell types and differentiation states. We propose techniques utilizing ATACseq (Assay for Transposase Accessible Chromatin sequencing) coupled with BSseq (bisulfite sequencing) to identify putative enhancers genome wide as well as elucidate the timing of demethylation and concomitant activation of these iHMRs. In tandem, we propose techniques to validate these iHMRs as functional enhancers using ATACseq coupled with STARRseq (self transcribing active regulatory region sequencing). We would like to apply these techniques in cell models of hormone receptor stimulation (MCF7 estradiol signaling) and cell differentiation (human embryonic stem cells differentiated to neural progenitor cells). 

 

INVESTIGATING THE ROLE OF THE P63 SAM DOMAIN IN PROTEIN-PROTEIN INTERACTIONS AND TRANSCRIPTIONAL REGULATION

J. Scott Beeler, Timothy Shaver, Jennifer Pietenpol
p63, a member of the p53 family of transcription factors, regulates signaling pathways that play essential roles during development and tumorigenesis. Germline missense mutations in the p63 sterile alpha motif (SAM) domain are causative for ankyloblepharon-ectodermal dysplasia-clefting (AEC) syndrome, which is characterized by varying ectodermal dysplasia, limb malformations, and orofacial clefting. The biological role of the p63 SAM domain is not well understood. In model organisms, SAM domains have been shown to mediate important protein-protein interactions. My research tests the hypothesis that the p63 SAM domain is required to regulate key protein-protein interactions involved in transcriptional regulation of p63 target genes required during development. Using the CRISPR/Cas system, I engineered isogenic cell lines with p63 SAM domain mutations and used the cell models in proteomic-based experiments to identify candidate SAM domain-interacting proteins. In parallel, I performed ChIP-seq and RNA-seq to determine the role and functional significance of the p63 SAM domain in p63-mediated transcriptional regulation. Understanding the function of the p63 SAM domain will provide insight into the roles of p63 in epidermal differentiation and may also be useful for understanding the mechanism underlying AEC syndrome and the function of the closely related p73 SAM domain. 
 

IMPLICATIONS OF R4-OXO-2-NONENAL-DERIVED HISTONE ADDUCTS IN MEDIATING GENE EXPRESSION

Jeannie M. Camarillo, James J. Galligan, Lawrence J. Marnett
Histone modifications play a critical role in the maintenance of chromatin structure and the regulation of gene expression. For example, Lys acetylation to histone tails results in chromatin remodeling to allow transcriptional activation of targeted genes. Recent work has identified core histones as targets for adduction by the oxidative stress-derived lipid electrophile, 4-oxo-2-nonenal (4-ONE). Here, we investigate the potential implications of these modifications on gene expression. Using an alkyne analog of 4-ONE, a4-ONE, we employed click chemistry to selectively tag modified proteins from cells treated with electrophile. Our data reveal that a4-ONE is a long-lived histone modification, present up to 24 hours after treatment. Using pharmacological inhibitors of histone acetylation and deacetylation, we observe alterations in the acetylation of core histones when cells are co-treated with a4-ONE. These data suggest that a4-ONE modifications occur at physiologically relevant sites that regulate chromatin structure and gene expression. Utilizing a combination of chromatin immunoprecipitation and click chemistry, we can selectively pull out DNA associated with adducted histones, further suggesting a putative role for these modifications in the regulation of gene expression. Together, our data suggest that histone adducts may play a role in chromatin structure and the mediation of gene expression. 

 

IDENTIFICATION OF NOVEL INOSITOL 1,3,4,5,6-PENTAKISPHOSPHATE BINDING PROTEINS

Bradley Clarke, Aidong Qi, John York
Inositol Phosphates (IPs) are small stereo-chemically distinct signaling molecules produced by a set of enzymes that are conserved throughout eukaryotes. Over forty of the possible 729 IP derivatives have been found in cells, yet only a few of these chemical codes have well described functions. Thus it remains an important question how the cell is able to decode these signals and elicit a biological response. We have developed a new strategy to identify novel inositol 1,3,4,5,6-pentakisphosphate (IP5) binding proteins. By fractionating and screening yeast cellular extracts with this novel approach we have found several candidates IP5 binding proteins. This work may provide insights into how IP5 is decoded by cellular machinery and how loss of the kinase that produces IP5, Ipk2/IPMK, leads to defects in organismal development and adaptive responses to nutrient changes. 

 

THE ROLE OF P73 IN HORMONALLY-REGULATED TISSUES

Gabriela Santos Guasch, Clayton Marshall, Jennifer Pietenpol
The p53 family of proteins are sequence-specific transcription factors that are required for appropriate control of cell cycle, DNA repair, apoptosis, development, and cell differentiation. Unlike p53, p73 is rarely mutated in human cancers and its role in epithelial cell growth and differentiation is not well understood. One focus of research in the Pietenpol Laboratory is the function of p73 in normal tissue development and during tumorigenesis. Mice that lack all p73 isoforms show developmental defects including hippocampal dysgenesis, hydrocephalus, inflammation, sterility, and abnormalities in pheromone sensory pathways. In a p73-knockout mouse model generated in the Pietenpol Laboratory, we observe a significant decrease in lobular alveolar budding in the mammary gland and in circulating progesterone levels, and an increased number of degenerative follicles in the ovaries. Mammary gland branching is regulated by integrated cell autonomous and non-autonomous stimuli including stromal interactions, circulating hormones and the estrous cycle. One research aim is to identify the mechanism by which p73 regulates endocrine signaling pathways to promote proper oocyte maturation, ovulation and subsequent lobular alveolar budding in the mammary gland. Using p73 ChIP-seq, performed on isolated murine granulosa cells, we are discovering target genes bound directly by p73 in the ovarian cells. In parallel, RNA-seq is being performed on the isolated granulosa cells and integrated with the ChIP-seq data to determine p73-regulated transcriptional networks. Results from the experiments will be presented and provide mechanistic insight to the role of p73 in oocyte maturation, ovulation and lobular alveolar budding in the mammary gland.

 

INVESTIGATING A NOVEL ROLE OF NUCLEAR PHOSPHOINOSITIDE REMODELING ENZYMES IN GENE EXPRESSION

Bree Hamann, Raymond Blind
Loss of function mutations of the lipid phosphatase PTEN are found in many human tumors, explaining the tremendous effort currently being put forth to inhibit the lipid kinases that antagonize PTEN function in cancer. The role of PTEN at the plasma membrane is well known, but PTEN function in the nucleus is far less well studied. Most of the nuclear functions of PTEN identified thus far are phosphatase-independent, making those functions extraordinarily difficult to target using drugs. If an enzyme-dependent function for PTEN could be identified in the nucleus, inhibitors of the kinases that antagonize nuclear PTEN might be extraordinarily effective in treating PTEN-dependent cancers, particularly cancers resistant to current PI3-Kinase inhibitors. We discovered that a little-studied inositol lipid kinase called IPMK (Inositol Polyphosphate MultiKinase) antagonizes a phosphatase-dependent function of nuclear PTEN: both IPMK and PTEN remodel phosphoinositides bound to the NR5A-class of nuclear receptor transcription factors, resulting in dramatic changes in the transcriptional programs mediated by NR5As. This proposal will demonstrate that IPMK and PTEN similarly remodel other phosphoinositide-bound transcriptional factors throughout the human genome, introducing a new paradigm for PTEN function and identifying new "low hanging" kinase targets for anti-cancer therapies. 

 

DRUG-LINKED OLIGONUCLEOTIDES SPECIFICALLY INDUCE HUMAN TOPOISOMERASE II-MEDIATED DNA CLEAVAGE

Lorena Infante, Sabine Fenner, Ben Bax, Neil Osheroff
Topoisomerases are ubiquitous nuclear enzymes that remove knots and tangles and regulate the supercoiling of DNA. They are essential for replication, transcription, recombination, and chromosome segregation. Type II topoisomerases transiently cleave both strands of the DNA as part of their catalytic cycle. They cleave the DNA backbone while using a catalytic tyrosine to covalently stabilize the break, pass another DNA helix through the break, and restore the DNA to its original condition. Drugs that stabilize the topoisomerase II-DNA cleavage complexes are called topoisomerase poisons, as they convert the enzyme into a cellular toxin by increasing the probability of permanent double-stranded DNA breaks. The Osheroff lab is currently collaborating with GlaxoSmithKline to improve topoisomerase II poison targeting and reduce off-target effects when treating cancers. The approach utilizes drug-linked oligonucleotides (DLO), oligonucleotides of a specific sequence that are covalently bound to a molecule of etoposide, which is a topoisomerase II poison. The hypothesis is that the DLOs will direct topoisomerase II to cleave at specific sites, interfering with replication and translation. Ultimately, the goal is to design DLOs personalized to specific cancer cell mutations in patients. We tested a 50-nucleotide DLO that matches the sequence of the human PML gene. We monitored the levels of topoisomerase II-mediated DNA cleavage when the enzyme was incubated with either a wild type oligomer (with the same PML sequence as the DLO) plus etoposide, or a DLO with no extra drug added. The DLO induces more than twice as much cleavage as when 500 μM etoposide is added to the wild type oligomer. The DLOs also inhibit the religation of DNA and stabilize the cleaved DNA-enzyme complex to the same extent as etoposide. Results also demonstrate that DLO precursors do not induce topoisomerase II-mediated cleavage, suggesting that the effect of the DLO is mediated by the linked drug and not by any other feature of the oligomer. Current data indicate that DLOs act as topoisomerase II poisons and that it is possible to direct topoisomerase II cleavage to a specific site, potentially reducing off-target cleavage. 

 

ANGIOTENSIN II AT1 RECPETOR POTENTIATES PROSTAGLADIN E2 VASOCONSTRICTOR EFFECTS

Maria Kraemer, Fred Lamb, Richard Breyer
Prostaglandins are key modulators of blood pressure and arterial tone. Prostaglandin E2 (PGE2), is a prostanoid that has vasodepressor effects; however, under certain circumstances PGE2 can induce vasopressor responses. Recent reports demonstrated that sub-threshold concentrations of vasoconstrictors augment PGE2-mediated constriction in rat femoral arteries. However, whether angiotensin II (Ang II) could affect PGE2-mediated contraction is not known. Using a wire myograph, we demonstrated that PGE2 had no significant effect on mouse femoral arterial rings at doses up to 1 μM. However, priming of arterial rings with 1 nM Ang II potentiated PGE2-evoked constriction in a concentration dependent manner (Area Under the Curve, AUCuntreated 1.784 ± 0.353, AUCAng II 23.27± 9.820, P<0.05). We tested femoral arteries from EP1, EP2, and EP3 receptor knockout mice. Only the EP3-/- arteries were unable to respond to PGE2 after Ang II priming (figure below). Pretreatment of arterial rings with 1 μM losartan, an angiotensin receptor antagonist, blocked PGE2-induced constrictor effects primed with Ang II (% of KCl, Ang II 21.72 ± 5.296, Ang II + losartan 3.025 ± 1.046, n=3). We have determined that re-addition of extracellular Ca2+ to a Ca2+-free artery restores PGE2-induced contractions (n=5) and that the Rho-kinase inhibitor Y-27632 blocks contraction (n=3). Taken together these data are consistent with angiotensin AT1 and prostaglandin EP3 receptors mediating a synergistic Rho-kinase-dependent contractile response. We are continuing to investigate the relationship between Ang II and PGE2 to determine the physiological relevance this may have in modulating blood pressure. 

 

STRUCTURAL ANALYSIS OF KCNE1 TRANSMEMBRANE MUTANT YIELDING KCNE3-LIKE FUNCTION

Cheryl Law, Charles Sanders
The KCNE family contains five single transmembrane-spanning proteins that modulate the voltage-gated potassium channel, KCNQ1 and provide functional diversity to the KCNQ1 channel. For example, KCNE1 modulates the KCNQ1 channel by slowing its activation and increasing its conductance. The KCNE3 protein also increases the conductance, but induces constitutive activation of the channel. In 2001, Melman et al made a series of KCNE1 and KCNE3 chimeras, and identified the transmembrane region as key to distinct functions of KCNE1 and KCNE3. They showed that by swapping three transmembrane residues, T71V72G73, of KCNE3 for KCNE1, F57T58L59, to create a KCNE1/3 chimera they could create a current trace similar to KCNE3 and eliminate KCNE1's ability to delay opening. Thus, KCNE1 with TVG from KCNE3 yields a constitutively active channel without a delay in opening. We hypothesize that this is due to changes in flexibility of the transmembrane-helix of KCNE1 when TVG triple mutation is present. By using NMR spectroscopy, biochemical studies, and computational docking, we aim to look at structural and conformational differences between KCNE1 and KCNE1TVG. We have expressed and purified KCNE1TVG for NMR studies and collected 2D-NMR spectra using a TROSY-based pulse sequence. Partial backbone assignments of KCNE1TVG have been determined by aligning and transfer assignments of the WT KCNE1 previous determined in our lab. Further 3D experiments have been carried out to complete assignments. Further NMR studied will be done to structurally determine KCNE1TVG. Computational models of KCNE1TVG/KCNQ1 in the open and closed state will be created. These models, along with working models for KCNQ1/KCNE1 will provide insight into how KCNE1TVG interacts with the channel differently and similarly than both KCNE1 and KCNE3. This will provide insight into how KCNE1TVG interacts with the channel differently and similarly than both KCNE1 and KCNE3. 

 

THE ROLE OF P73 IN HORMONALLY-REGULATED TISSUES

Clayton Marshall, Deborah Mays, Scott Beeler, Jennifer Rosenbluth, Kelli Boyd, Gabriela Santos, Timothy Shaver, Lucy Tang, Qi Liu, Yu Shyr, Bryan Venters, Mark Magnuson, Jennifer Pietenpol
Utilizing a mouse model generated in our laboratory that targets exons 7-9 of p73, we discovered that p73 is required for multiciliated cells (MCCs) development, is required for MCC differentiation, and directly regulates transcriptional modulators of multiciliogenesis. Loss of ciliary biogenesis provides a unifying mechanism for many phenotypes observed in p73 knockout mice including hydrocephalus, hippocampal dysgenesis, sterility and chronic inflammation/infection of lung, middle ear and sinus. We further focused on the pulmonary epithelium to determine the role p73 plays in ciliary biogenesis. We found p73 is expressed in MCCs as well as co-expressed with p63 in a subset of basal, tracheal epithelial cells, suggesting that p73 'marks' these cells for MCC differentiation. Through in situ p63/p73 ChIP-seq of the murine trachea, we identified genomic binding sites in proximity to genes that regulate MCC differentiation, from cell cycle arrest (Cdkn1a) and amplification of centrioles (Myb) to apical docking of centrioles with components that make up the axoneme [Foxj1 and Traf3ip1]. By combining our ChIP-seq data with RNA-seq of tracheal epithelial cells, we found evidence for p73-dependent, direct and indirect transcriptional regulation of a broad network of cilia-associated genes. In sum, p73 is essential for MCC differentiation, functions as a critical regulator of a transcriptome required for MCC differentiation and, like p63, has an essential role in development of tissues. 

 

TARGETNIG IMMUNO CHECKPOINT PATHWAYS BY SMALL MOLECULES

Evan Perry, Feng Wang, Bin Zhao, Qi Sun, Tyson Rietz, Amyn Murji, Jason Phan, Olejniczak Ed, Taekyu Lee, Stephen Fesik
T lymphocytes (T cells) of the immune system are able to recognize tumor antigens by their T cell receptor (TCR) and are capable of killing tumor cells. However initial attempts to stimulate the immune response against cancer were largely unsuccessful. Two co-receptors (CTLA-4 and PD-1) expressed on the T cell surface have been found to negatively regulate T cell activity when bound to their ligands at the stages of T cell activation (CTLA-4) and at tumor cell recognition (PD-1). By blocking these T cell checkpoints by antibody-based therapeutics, patients in clinical trails showed significant tumor regression and long term durability. Antibody therapeutics are now FDA approved for both PD-1 and CTLA-4, but to date no small molecule inhibitors yet exist for these pathways. We hypothesize that a potent small molecule inhibitor of these checkpoint pathways will have more preferred pharmacokinetic properties over antibody therapeutics allowing for more control over the dosing regime potentially leading to an alterative therapeutic that is more cost effective and better suited for combination therapies. Our lab is currently screening our fragment library against CTLA-4, PD-1 and its ligand PD-L1 by NMR screening methodology to access the druggability of these targets and find small fragments that bind to advance to lead optimization by structure based design. 

 

DIFFERENTIATING THE FUNCTION OF SNF2 FAMILY DNA TRANSLOCASES

Lisa A. Poole, Runxiang Zhao, David Cortez
Each round of cell cycle requires complete and faithful replication of the DNA to yield two identical copies of the genome. Replication can be hindered by several sources of replication stress including inherent characteristics of the DNA that impede rapid completion of this essential process. To ensure accurate replication, the genomic stress response (GSR) exists to regulate cell cycle checkpoints, stabilize stalled replication forks, and address the source of replication stress. The SNF2 family of DNA translocases is a member of the GSR that stabilizes and restarts stalled replication forks. This family includes three enzymes - SMARCAL1, ZRANB3, and HLTF - that utilize ATP to catalyze DNA remodeling in vitro. Additionally, our lab has shown that all three enzymes are found at DNA replication forks in the absence of added stress agents. Despite the similarities observed between these three enzymes in vitro, it is known that these enzymes do not have redundant functions in cells as depletion of individual proteins results different cell phenotypes. I have previously shown that SMARCAL1, and not HLTF or ZRANB3, is needed to promote replication through telomeres, a difficult to replicate region of the genome. However, it is known that SMARCAL1 also functions in other regions of the genome during DNA replication. I propose that the difference in knockdown phenotypes between these enzymes is due to their function in separate genomic contexts during DNA replication. 

 

POST-TRANSCRIPTIONAL GENE REGULATION IN INNATE IMMUNITY

Katie Rothamel, Manny Ascano
A proper molecular response to the detection of pathogens requires dynamic gene expression and the ability to manage the new transcription program of cytokines such as Interferons (IFNs) such that cells are able to survive and/or to initiate an inflammatory process. Previous work in the field has focused on the transcriptional changes that occur during cellular stress, yet little is known about how these RNAs are regulated during pathogenic infections or in the activation of the cell's immune response. RNA-binding proteins (RBPs), the main players of post-transcriptional gene regulation (PTGR), control the mRNA transcriptional landscape and mRNA translation during dynamic gene expression. Specifically, we hypothesize that RBPs regulate key mRNAs during innate immune activation, and allow for rapid changes in their translation and preventing the translation of unnecessary mRNA. RBPs can regulate gene expression post-transcriptionally by affecting RNA stability; thereby affecting how much mRNA is ultimately used for protein translation. The transcriptional induction of IFN-β and interferon stimulatory genes (ISGs) is a hallmark of immune stimulation and activation. Our long-term goal is to characterize critical mRNA-protein interaction networks during immune cell activation, which will enhance our molecular understanding of the inflammatory process, and contribute towards the development of novel immunotherapies. To test our hypothesis, we will employ cross-linking methods in conjunction with affinity purification and immunoprecipitation of mRNA-proteins complexes. This will allow us to how and what particular RBPs bind to mRNA transcripts during immune activation, and if there are changes in mRNA modifications- such as methylation- during post-transcription.

 

GENOME WIDE POL II DYNAMICS UPON PHARMACOLOGICAL INHIBITION OF CDKS IN T(8;21) AML

Shilpa Sampathi, Pankaj Acharya, Yue Zhao, Jing Wang, Kristy Stengel, Qi Liu, Michael Savona, Scott Hiebert
The t(8;21) translocation fuses the DNA binding domain of RUNX1, a master regulator of hematopoiesis to the ETO/MTG, a transcriptional co-repressor. The resultant fusion protein interferes with RUNX1 functions. The incidence of occurrence of the t(8;21) in AML is approximately 4%-12% making it the most frequently occurring chromosomal translocation. The AML1-ETO fusion protein exerts a dominant negative effect on activation of RUNX1 gene targets. Although there are several advances made within the past decade in understanding the pathogenesis of AML1-ETO, there is still both a need and urgency to understand the mechanistic events in systems that express the fusion protein. This dependency also makes t(8;21) an attractive model to explore druggable vulnerabilities for potential therapeutic approaches. Towards this end, by using yeast-two-hybrid screens our lab has identified several components of the transcriptional super elongation complex as associating with the MTG family members hinting at the role of these proteins in transcriptional elongation. Furthermore, we validated one such interaction using co-immunoprecipitation of MTG proteins with AFF4, a scaffolding member of the SEC complex. Interestingly, MTGs also associate with the positive transcription elongation factor B complex (pTEFB = CDK9 + cyclinT1) of the SEC. Also, CDK9 mediated transcriptional elongation control was essential for MYC driven hepatocellular carcinoma. Therefore, we hypothesized that CDK9 inhibition by pharmacological approach to target the critical transcriptional elongation control would provide mechanistic information towards better therapy. We used several CDK inhibitors targeting CDK9 including classical CDK inhibitors such Flavopiridol on Kasumi-1 cells expressing the fusion protein AML1-ETO. The cells were sensitive to all the inhibitors tested. These inhibitors triggered both cell cycle proliferation defects and apoptosis perhaps via a DNA damage response in as little as 6 hrs of treatment. To delineate the transcriptional regulation upon inhibition of CDKs, we performed high resolution analysis of genome wide RNA polymerases by precision global run-on transcription sequencing (PRO-Seq). As a goal towards understanding the early events, PRO-Seq was done after 1 hr of treatment with these inhibitors. Initial analysis showed that about 95% of the genes showed accumulation of paused RNA Pol II near the transcriptional start site of promoters. This method also identified regulatory elements such as enhancers being affected by CDK inhibition (e.g. MYC super enhancer). Corresponding RNA-Seq analysis revealed several transcripts being deregulated such as MYC, MCL1 etc. Most interestingly, THZ1, a highly selective covalent inhibitor of CDK7 (a component of TFIIH), caused the reverse effect and stimulated transcriptional elongation, while also causing cell death. These results suggest that employing CDK9 inhibition can open up to new avenues of therapy in AML. 

 

MECHANISMS OF RESISTANCE TO PI3K INHIBITION IN TRIPLE NEGATIVE BREAST CANCER

Johanna Schafer, Brian Lehmann, Luojia Tang, Jennifer Pietenpol
Since the phosphatidylinositide 3-kinase (PI3K) pathway is frequently deregulated in human tumors and primarily involved in growth, survival, and proliferation, it has become a major target for cancer drug discovery efforts. In collaboration with Dr. Vandana Abramson, the laboratory is currently investigating a β-sparing, pan-PI3K inhibitor, GDC-0032, in combination with enzalutamide, in a Phase Ib/II clinical trial at the Vanderbilt Ingram Cancer Center (VICCBRE1374) for patients with triple negative breast cancer (TNBC) that express the androgen receptor (AR). Due to the prevalence of acquired resistance to other targeted therapies in previous trials, I am generating cell line models of acquired resistance to PI3K inhibition (PI3Ki) through continuous culture of select TNBC cell lines in the presence of increasing concentrations of PI3Ki over time. By sequencing RNA from the first TNBC cell line to acquire resistance, CAL51, I discovered differential expression of MYC family isoforms between the parental cell line (CAL51Parental: c-MYC) and the newly generated PI3Ki-resistant cell lines (CAL51GDC-0941R, CAL51GDC-0032R, and CAL51BYL719R: N-Myc). In analyzing clonal populations developed from CAL51Parental and CAL51GDC-0032R, I observed that only clones expressing MYCN were resistant to PI3Ki. These clonal cell lines were then subjected to the NCI library of FDA-approved oncology drugs as well as several inhibitors under clinical development to target N-Myc. Not only did drugs targeting N-Myc rank among the highest in the screen for differential effects on viability between CAL51Parental and CAL51GDC-0032R, but all CAL51GDC-0032R clones tested had a substantial decrease in viability upon N-Myc siRNA-mediated knockdown. Furthermore, additional TNBC cell lines resistant to PI3Ki are displaying MYC expression no longer regulated by the PI3K pathway. Based on similar methods by investigators studying other targeted therapies, I hypothesize that this mechanism of resistance will mirror those identified in residual and/or recurrent tumor of patients receiving PI3Ki. 

 

BYPASS AND MISINCORPORATION OF DNA-PEPTIDE CROSS-LINKS BY DNA POLYMERASES

Carl Sedgeman, Kevin Johnson, F. Peter Guengerich
The DNA of all living organisms is constantly exposed to chemicals that can cause diverse harmful lesions, one of which is DNA-peptide cross-links. The formation of DNA-peptide cross-links (DPCs) has been shown to occur from the conjunction of bis-electrophiles to DNA bases and concomitantly nuclear proteins such as glutathione and O6-alkylguanine DNA alkyltransferase (AGT). These conjugates have been discovered in cells and can induce both DNA mutations and cell death. Our hypothesis is that select translesion DNA polymerases are capable of replicating past DNA-peptide cross-links. A corollary of our central hypothesis is that AGT cross-linked to DNA causes mutations after being partially degraded by proteases. For the investigation of the bypass and miscoding ability of DPCs by translesion DNA polymerases, DNA replication assays have been performed in duplex DNA substrates containing an N6dA-DPC formed by glutathione and peptides from the active site of AGT (3-15 aa). Results indicate that Pols kappa and iota have low processivity past these adducts compared to Pol eta and Dpo4. However, preliminary results suggest that the bypass by Pol eta and Dpo4 have high levels of misincorporation. 

 

MECHANISM OF RIBONUCLEOTIDE INCORPORTATION BY HUMAN DNA POLYMERASE ETA

Yan Su, Martin Egli, F. Peter Guengerich
Ribonucleotides and 2 ́-deoxyribonucleotides are the basic units for RNA and DNA, respectively, and the only difference is the extra 2 ́-OH group on the ribonucleotide sugar. Cellular rNTP concentrations are much higher than those of dNTP. When copying DNA, DNA polymerases not only select the base of the incoming dNTP to form a Watson-Crick pair with the template base but also distinguish the sugar moiety. Some DNA polymerases use a steric gate residue to prevent rNTP incorporation by creating a clash with the 2 ́-OH group. Y-family human DNA polymerase eta (hpol eta) is of interest owing to its spacious active site (especially in the major groove) and tolerance of DNA lesions. Here, we show that hpol eta maintains base selectivity when incorporating rNTPs opposite undamaged DNA and the DNA lesions 7,8-dihydro-8-oxo-2 ́-deoxyguanosine (8-oxodG) and cyclobutane pyrimidine dimer (CPD) but with rates that are 1000-fold lower than for inserting the corresponding dNTPs. X-ray crystal structures show that hpol eta scaffolds the incoming rNTP to pair with the template base (dG) or 8-oxodG with a significant propeller twist. As a result, the 2 ́-OH group avoids a clash with the steric gate, Phe-18, but the distance between primer end and Pα of the incoming rNTP increases by 1 Å, elevating the energy barrier and slowing polymerization compared with dNTP. In addition, Tyr-92 was identified as a second line of defense to maintain the position of Phe-18. This is the first crystal structure of a DNA polymerase with an incoming rNTP opposite a DNA lesion. 

 

MICE LACKING THE PHAGOCYTIC RECEPTOR JEDI1 DEVELOP AUTOIMMUNITY AND PRURITUS

Alexandra J. Trevisan, Chelsea S. Sullivan, Francis E. Hickman, Ashley Wilhelm, Jillian P Rhoads, Kristy R. Stengel, Scott W. Hiebert, Amy S. Major, Kevin P.M. Currie, Bruce D. Carter
Programmed cell death is essential for proper organism development, tissue homeostasis, and preservation of cell integrity. The oft overlooked final stage of apoptosis is non-immunogenic clearance of cellular corpses through the process of phagocytosis. Failure to efficiently engulf dead cells is one presumed cause of autoimmune disease such as systemic lupus erythematosus (SLE). In the developing nervous system, 50% of initially born neuroblasts (neuron precursors) eventually apoptosis. Our lab recently discovered that satellite glial cells (SGCs) are responsible for phagocytosis of dying primary afferent sensory neurons in the dorsal root ganglia (DRG) and that the mammalian homolog of the C. elegans engulfment receptor CED1, JEDI1(PEAR1/MEGF12) is important for this process. We used jedi1-/- mice to test the hypothesis that JEDI1 is an engulfment receptor important for the clearance of dead neurons in vivo. The preliminary data demonstrate that jedi1-/- mice develop canonical symptoms of SLE such as serum autoantibodies, proteinurea, and accumulation of antibodies in the kidney. We have shown by RNA that macrophages, professional phagocytic cells, express Jedi1 and hypothesize that lack of Jedi1 may diminish macrophage activity, leading to an accumulation of dead cells in vivo and subsequent autoimmune disease. Surprisingly, jedi1-/- animals also develop pruritus, which manifests as open wounds in aged mice. Electrophysiological evidence shows that jedi1-/- sensory neurons are hyperactivated, which could lead to increased scratching behavior. We propose that lack of Jedi1 in the DRG leads to decreased SGC engulfment of dead neurons, causing glial activation. SGC activation has been shown to increase sensation of pain, and because pain and itch share highly similar neuronal circuitry pathways, we suspect that a similar mechanism elevates itch in jedi1-/- animals. 

 

THE REGULATION AND FUNCTION OF DUAL-DOMAIN INOSITOL PHOSPHATE KINASE/PYROPHOSPHATE VIP

Shidong Wang, John York
Inositol pyrophosphates (PP-IPs, also known as IP7 and IP8) are important and evolutionarily conserved signaling molecules that are synthesized by two classes of kinases: IP6Ks and VIPs that phosphorylate the D-5 and D-1 positions respectively. Remarkably, in addition to the N-terminal 1-kinase domain, the VIP proteins also possess a pyrophosphatase domain that functions to remove the D-1 beta-phosphate. Thus, VIP proteins function to control levels of 1-IP7 and 1,5-IP8 through its dual-functional enzyme activities. The goal of my thesis is to understand the biochemical and functional roles of VIP kinase and pyrophosphatase activities. I propose the hypothesis that to ensure proper temporal production of PP-IPs signaling molecules and to avoid a futile consumption of ATP, the two opposing activities of VIP proteins are tightly regulated. Previous studies in S. cerevisiae reveal an example of potential activity regulation and define a phosphate starvation stress transcriptional regulation pathway sensitive to levels of 1-IP7. This pathway provides a framework to study kinase/pyrophosphatase activity changes under high and low phosphate conditions. In addition to post-translational regulation, I speculated that association with other proteins may also alter VIP activity. My preliminary studies using affinity purification strategy in human cells identified that VIP2 associates with 14-3-3 protein through its C-terminal region. I seek to determine if this association regulates VIP2 kinase/pyrophosphatase activity balance, subcellular localization or protein stability. The long-term goal of my project is to achieve better understanding of inositol pyrophosphate signaling and the enzyme regulation of the dual-functional class of VIP proteins. 

 

INVESTIGATING THE METABOLISM OF MAJOR PEROXIDATION DNA ADDUCT, M1DG, IN THE GENOME

Orette R. Wauchope, William N. Beavers, James J. Galligan, Michelle M Mitchener, Phillip J. Kingsley, Lawrence J. Marnett
Chronic inflammation results in increased production of reactive oxygen species (ROS), which can oxidize cellular molecules including lipids and DNA. Our laboratory has shown that 3-(2-deoxy-β-D-erythro-pentofuranosyl)pyrimido[1,2-α]purin-10(3H)-one (M1dG) is the most abundant DNA adduct formed from the lipid peroxidation product, malondialdehyde, or the DNA peroxidation product, base propenal. M1dG is mutagenic in bacterial and mammalian cells and is repaired via the nucleotide excision repair system. Here, we report that M1dG levels in intact nuclear DNA were increased from basal levels of 1 adduct per 108 nucleotides to 2 adducts per 106 nucleotides following adenine propenal treatment of RKO, HEK293 or HepG2 cells. We also found that M1dG in genomic DNA was oxidized in a time-dependent fashion to a single product, 6-oxo-M1dG, (to ~ 5 adducts per 107 nucleotides) and that this oxidation correlated with a decline in M1dG levels. Investigations in RAW264.7 macrophages indicate the presence of high basal levels of M1dG (1 adduct per 106 nucleotides) and the endogenous formation of 6-oxo-M1dG. Interestingly, in all cell lines investigated, 6-oxo-M1dG was not observed in mitochondrial DNA. Basal levels of M1dG in the mitochondria were found to be approximately two orders of magnitude higher (~ 1 adduct per 106 nucleotides) than those observed in nuclear DNA and were increased with electrophilic stimulation. Further studies, in all the investigated cell lines, indicated a correlation between M1dG levels and oxidative stress in the mitochondrion. This is the first report of M1dG in mitochondrial DNA in intact cells and it has significant implications for understanding the role of inflammation in DNA damage, mutagenesis and repair.

 

PROCESSIVITY OF HUMAN CYTOCHROME P450 17A1 AND ROLE OF CYTOCHOME b5

Eric Gonzalez and F. Peter Guengerich

The human cytochrome P450 (P450) enzyme family is vital in the steroid biosynthetic pathway. Although steroid hormones are essential for normal development and reproductive functions, the abolition of these molecules has become a major therapeutic goal for the treatment of hormone-driven cancers. Human P450 17A1 is one of the steroid-metabolizing human enzymes and has been targeted in prostate cancer therapy due to its function in androgen formation. Inhibition of P450 17A1 has proven difficult because the enzyme catalyzes two sequential reactions, and the intermediate products (17-OH steroids) are critical for the production of glucocorticoids. The initial reaction catalyzed by P450 17A1 is hydroxylation of pregnenolone and progesterone into 17α-OH pregnenolone and 17α-OH progesterone, respectively, followed by a 17,20-lyase reaction generating dehydropeiandrosterone and androstenedione, respectively. The development of a reaction-specific P450 17A1 inhibitor is a current objective in the pharmaceutical industry, but none have been approved. The primary objective of this project was to clarify whether P450 17A1 is (a) distributive and dissociates the intermediate 17α-OH product or (b) processive and proceeds directly to the cleavage reaction. One hypothesis is that P450 17A1 is primarily distributive but becomes processive when associated with cytochrome b5 (b5), a recognized stimulator of the lyase reaction. Recombinant enzymes have been used to conduct detailed kinetic analyses of individual steps in the P450 17A1 mechanism. First, the steady-state parameters for binding and catalysis were obtained. Association of P450 17A1 with its substrates and products is very tight, all having spectrally-derived Kd values <500 nM. Further, the steady-state incubation assays revealed a significant catalytic efficiency increase in both the cleavage (7- & 26-fold) and hydroxylation (14- & 4-fold) reactions when b5 was included. The koff rates for the substrates and products, in the presence and absence of b5, have been estimated using an inhibitor-trapping assay, but only small differences were observed. Single-turnover incubations did not show significant b5-mediated changes in the rate of progesterone metabolism but did show an increased rate in the 17-hydroxylation of pregnenolone (0.27 to 0.43 s-1) when supplemented with b5. However, the rates for 17α-OH pregnenolone and 17α-OH progesterone lyase reactions were increased 10- and 4-fold, respectively, when b5 was added under single-turnover conditions. These results have been combined with data from pulse-chase experiments, which utilize radiolabeled substrates, to further model the P450 17A1 enzyme mechanism, with and without b5, utilizing KinTek Explorer® software. In conclusion, P450 17A1 has a distributive mechanism that is enhanced by b5, but the combination does not generate a completely processive system.

 

GUANIDINE INDUCED UNFOLDING OF TWO MANGANESE-DEPENDENT SUPEROXIDE DISMUTASES FROM STAPHYLOCOCCUS AUREUS

Saffron Little, Sashari Pinnace, Kofi Amoah, Steven Damo

The rapidly increasing antibiotic resistance of bacteria poses a threat to global public health. In particular, the pathogenic bacterium, Staphylococcus aureus, causes health risks that range from mild infections to diseases that can be fatal. Understanding the molecular mechanisms of S. aureus virulence will allow for the development of new and possibly more effective therapeutics. This project focuses on the role of superoxide dismutase (SOD), an essential enzyme that protects bacteria from oxidative stress by catalyzing the conversion of superoxide to hydrogen peroxide and oxygen. Unique to all Staphylococci, S. aureus expresses two manganese dependent Sods, SodA and SodM, both of which are critical virulence factors. SodA and SodM share >80% sequence identity, yet, little is known about their mechanism of action. Our overarching hypothesis is that understanding the structure-function relationships of these enzymes will give insights into bacterial pathogenesis that can be exploited for the design of new antimicrobial agents. SodA and SodM were recombinantly expressed in E. coli and purified using nickel affinity column chromatography. We determined the high-resolution crystal structure of SodM to 2.0 Å but were unable to deduce any structural variations that could be attributed functional differences. To assess any differences in conformational dynamics between SodA and SodM, the thermodynamic stability of each protein was determined by equilibrium unfolding experiments. We measured the intrinsic tryptophan fluorescence as a function of guanidine hydrochloride concentration. Preliminary results demonstrate that SodM is more stable than SodA by ~1 kcal/mol. These studies demonstrate the feasibility of obtaining a complete biophysical characterization of SodA and SodM which will allow for a detailed comparative analysis of their function.

 

HISTONES ARE TARGETS FOR MODIFICATION BY GLUCOSE-DERIVED METHYLGLYOXAL

Galligan JJ, Mitchener MM, Wang T, Wauchope OR, Rose KL, Spiegel DA, Marnett LJ.
Diabetic nephropathy (DN) is the major cause of morbidity and mortality in diabetic patients. The sustained hyperglycemia associated with diabetes results in substantial oxidative stress and inflammation, leading to the generation of the reactive α-oxoaldehyde, methylglyoxal (MGO). MGO has been shown to play an integral role in DN pathogenesis and its effects are often seen in DN patients, despite glycemic control, via a phenomenon called metabolic memory. Metabolic memory is hypothesized to result from alterations in histone posttranslational modifications. Here, we describe histones as targets for modification by MGO in a cell culture model of hyperglycemia. Utilizing novel LC-MS/MS methodologies developed in our laboratory, histones were purified from cultured HEK293 cells treated with low (5mM) or high (25mM) glucose. Following exhaustive proteolytic digestion, absolute concentrations of histone posttranslational modifications were quantified by LC-MS/MS. These analyses revealed both Lys and Arg-derived MGO adducts, with the Arg hydroimidazalone (MG-H) adducts being the most prevalent. To validate that these adducts were indeed derived from cellular MGO, glyoxalase 1 was knocked down, resulting in substantial increases in both cellular MGO and MG-H adducts. Sites of MGO modification were identified via high-resolution mass spectrometry, which revealed Lys and Arg adducts on all four-core histones. We hypothesize that MGO adduction of histones offers a novel link between inflammation, hyperglycemia, and DN pathogenesis. The effects of these modifications on cellular homeostasis are the subject of current investigation. 

 

CHARACTERIZATION OF “MYCOBACTERIUM GYRASE INHIBITORS” (MGIS): A NOVEL CLASS OF GYRASE POISONS

Elizabeth Gibson, Timothy Blower, Monica Cacho, James Burger, Neil Osheroff
Tuberculosis is a leading cause of mortality worldwide. The current standard treatment is the RIPE regimen: rifampin, isoniazid, pyrazinamide, and ethambutol. Because resistance is developing against these drugs and the second-line treatment, fluoroquinolones (FQs), there is an urgent need for the development of novel antitubercular drugs to combat resistance. Therefore, we characterized a new class of "Mycobacterium Gyrase Inhibitors" (MGIs) that display activity against Mycobacterium tuberculosis (Mtb). Three MGIs obtained from GlaxoSmithKline, GSK000, GSK325, and GSK126, displayed activity against Mtb gyrase, with GSK000 being the most efficacious. In marked contrast to FQs (which induce double-stranded DNA breaks), MGIs induce gyrase-mediated single-stranded DNA breaks, even at high drug concentrations or long cleavage time courses.  While increasing single-stranded cleavage, MGIs appear to suppress double-stranded DNA breaks. Like FQs, MGIs act by inhibiting religation of cleaved DNA. GSK000 displays a strong preference for Mtb gyrase over Bacillus anthracis topoisomerase IV or gyrase and Escherichia coli topoisomerase IV, suggesting specificity for Mtb. Finally, MGIs retain activity against common FQ resistant mutant gyrase enzymes (GyrA A90V, D94H, and D94G) and displayed no significant activity against recombinant human topoisomerase IIα. Our results suggest that MGIs are a novel class of gyrase poisons that have potential as antitubercular drugs. Supported by NIH grants GM033944 and GM007628. 

 

FRAGMENT-BASED DISCOVERY OF RAGE INHIBITORS

Benjamin Gilston, Alex Waterson, Walter Chazin
Activation of the cell-surface receptor RAGE (the Receptor for Advanced Glycation Endproducts) contributes to the inflammatory response through the NFB signaling pathway. Overactivation of RAGE is specifically associated with atherosclerosis, hypertension, renal damage, and cataracts in diabetic patients. We are using a fragment-based approach to develop highly targeted small molecule inhibitors of RAGE. An initial screen of a 14,000 member, highly curated library of molecular fragments will be performed using the Structure Activity Relationship by NMR (SAR by NMR) strategy. This is a highly efficient approach to identify linked-fragment inhibitor compounds that bind to RAGE precisely at the ligand binding site. In order to accomplish this screen we optimized our production of RAGE protein using two important criteria: determine the construct that produces the largest yield of pure protein, while simultaneously evaluating which has the highest quality 2D NMR spectra. Promising preliminary data have been obtained in the form of hits in the chemically diverse mixtures of molecular fragments that target key recognition sites on RAGE. Our observation that mixtures of fragments can target multiple binding pockets of the RAGE protein suggest that a multivalent inhibitor of RAGE could be produced. Our ultimate goal is to discover small molecule inhibitors that disrupt RAGE as a means to explore the therapeutic potential for slowing the degenerative effects associated with diabetes. 

 

LOSS OF JEDI-1 RESULTS IN INCREASED PROLIFERATION OF NEURAL PRECURSOR CELLS IN VITRO

F. Edward Hickman, Elizabeth Crummy, Bruce Carter
While most neurogenesis in the developing mouse is complete by post-natal day 0 (p0), proliferation of stem cells continues in both the ventricular-subventricular zone (V-SVZ) and subgranular zone of the dentate gyrus of the hippocampus (SGZ). Our lab has identified Jedi-1 as an engulfment receptor necessary for clearance of apoptotic sensory neurons in the dorsal root ganglia that present during normal development. Its role in the central nervous system, however, is still not understood. Our analysis of jedi-1-/- mice have shown deletion of Jedi-1 expression in the brain and in neural precursor cell populations compared to wild type controls. Interestingly, jedi-1-/- neural precursor cells derived from the V-SVZ showed a two-fold increase in proliferation observed by BrdU incorporation. Most proliferating cells were Nestin+, a common marker of early NPCs. Preliminary data also show that jedi-1-/- NPCs do not have any deficits in engulfment capability. Future experiments will determine the mechanism by which Jedi-1 negatively affects the proliferation of NPCs. 
 

CYTOCHROME P450 27C1 CATALYZES THE DESATURATION OF RETINOL TO 3,4-DEHYDRORETINOL

Kevin M. Johnson, Valerie M. Kramlinger, Rina Fujiwara, Leslie D. Nagy, F. Peter Guengerich
Although cytochromes P450 (P450) are extensively studied, there are still a number of enzymes from this superfamily that have no assigned function. Until recently, P450 27C1 was one of these "orphan P450s". We have identified that, in zebrafish, this enzyme is responsible for the desaturation of retinol (vitamin A1) to 3,4-dehydroretinol (vitamin A2) and are currently investigating this mechanism. In vertebrates, such as migrating salmon and bullfrog, this desaturation is triggered in response to changes in the light environment. The presence of 3,4-dehydroretinol leads to spectral tuning of the vertebrate photoreceptors allowing vision to shift toward the near infrared range. Human P450 27C1 can also catalyze this reaction in vitro and steady-state kinetic parameters indicate it does so with unusually high catalytic efficiency. The function of P450 27C1 in humans is currently unknown, but several studies suggest diverse roles for this gene outside of the eye. P450 27C1 mRNA is expressed in a number of issues, including liver, kidney, and spleen. Thus, various human issues are being analyzed to quantify P450 27C1 and assess activity. 

 

ncRNA AND EPIGENETICS: THE REGULATION OF A HISTONE DEMETHYLASE BY STRUCTURED RNAS

William Martin, Alexander Hirschi, Nicholas Reiter
Lysine specific demethylase 1 (LSD1) is an essential epigenetic regulator which is responsible for the demethylation of histone 3 lysine 4/9 (H3K4 and H3K9). LSD1 plays an important role in the differentiation of ESCs and is a driver of various cancers. Furthermore, the non-coding RNAs TERRA and HOTAIR are known to recruit LSD1 to certain regions of chromatin. We have demonstrated that LSD1 specifically binds G-quadruplex RNAs and that these RNAs can modulate the activity of LSD1 in vitro. Current efforts seek to determine the binding mechanism and the global role of RNA in LSD1 regulation. 

 

DIFFERENTIAL LIPID AND PROTEIN DISTRIBUTIONS IN NORMAL AND TYPE I DIABETIC HUMAN PANCREATA REVEALED BY TISSUE IMAGING MASS SPECTROMETRY

Boone Prentice, Rachana Haliyur, Nathaniel Hart, Audra Judd, Radhika Armandala, Marcela Brissova, Jeffery Spraggins, Jeremy Norris, Alvin Powers, Richard Caprioli
Hormone-secreting pancreatic islet cells play key roles in glucose homeostasis and derangement of these cells leads to altered metabolic states such as type 1 diabetes (T1D). However, our understanding of the processes leading to islet β-cell death/dysfunction has been limited by the small size of pancreatic islets and the lack of experimental approaches to fully interrogate these groups of cells. Although alterations in lipid metabolism are known to cause disruptions in cell structure, signaling, and energy homeostasis, the roles of lipids in pancreatic development and diabetes-associated loss of function remain unclear. In concert with immunohistochemistry (IHC), imaging mass spectrometry (IMS) offers the ability to measure more completely the array of molecular species present in a cell population. By overlaying IHC images onto scanned microscope images of serial tissue sections, targeted high spatial resolution (20 μm), high mass resolution Fourier transform ion cyclotron resonance (FTICR) IMS measurements were performed on portions of the tissue rich in islets. In these experiments, age-matched control tissues of normal (e.g., 19 year old male) and T1D (e.g., 20 year old male with 7 years of T1D) samples were placed on the same slide and analyzed in the same experimental run, allowing us to discover distinctive lipid profiles. For example, the spatial localization and level of fatty acid unsaturation of several phosphatidylcholine molecules varied in the exocrine and endocrine pancreas [e.g., PC(16:0/18:2) vs. PC(16:0/18:1)]. Additionally, arachidonic acid-containing phosphatidylinositol molecules appear elevated in insulin-positive cells [e.g., PI(18:0/20:4) vs. PI(18:0/22:4)]. Several sulfatides were differentially expressed between islets and acinar tissue (e.g., C22:0-OH vs. C22:1-OH). This work aims to uncover new molecular understanding of islet and exocrine cell lipid synthesis, expression, and metabolism. Protein analysis by IMS was also performed and compared to IHC as a proof of principle experiment to confirm that insulin and glucagon IMS signals co-register with insulin and glucagon IHC fluorescence images. Additionally, other protein signals showed interesting differential expression between the normal and T1D tissues. For example, protein signals at m/z 1936.28 and m/z 3020.51 were elevated in the islets of the normal tissue while m/z 4111.98 and m/z 9970.90 were elevated in T1D islets. Combined with our access to a unique set of human pancreatic tissue samples, we are seeking to use the molecular specificity and spatial information afforded by IMS in order to gain new insights into pancreatic pathophysiology through understandings of islet lipid and protein expression. 

 

SYSTEMS ANALYSIS OF SKELETAL MUSCLES FROM MYOTONIC DYSTROPHY PATIENTS AND HEALTHY INDIVIDUALS REVEALS KEY DIFFERENTIALLY REGULATED PROTEINS AND PATHWAYS
Parimal Samir
, Nripesh Prasad, Kristen Hoek, Shawn Levy, Andrew J. Link
Myotonic dystrophy, a form of muscular dsystrophy, is an autosomal dominant multi-systemic disorder caused by the expansion of nucleotide repeats. There are two types of myotonic dystrophy. Myotonic dystrophy type 1 (DM1) is cause by a CTG trinucleotide repeat expansion in the 3' untranslated region of dystrophia myotonica-protein kinase (DMPK) gene. Myotonic dystrophy type 2 (DM2) is caused by a CCTG tetranucleotide repeat expansion in the first intron of zinc finger 9 (ZNF9) gene. The expression of the repeat expansions in both cases lead to the nuclear accumulation of RNA granules, which sequester the RNA processing factors. This RNA toxicity is thought to be the cause of the disease symptoms. However, the effect of the repeat expansions on the proteome is poorly understood. To address this, we quantified the proteomes of the skeletal muscles of myotonic dystrophy patients and healthy volunteers to identify differentially regulated proteins. We used iTRAQ labeling followed by liquid chromatography tandem mass spectrometry for protein quantitation. Skeletal muscles from 5 healthy volunteers, 7 DM1 patients and 6 DM2 patients were used in this study. We quantitated 3575 proteins across all the samples. We used a one way ANOVA, with the Benjamini Hochberg procedure for controlling false discovery rate in multiple comparisons, to identify differentially regulated proteins. We found identified 30 proteins to be upregulated and 4 proteins to be downregulated in both DM1 and DM2. We found 154 proteins to be upregulated and 218 proteins to downregulated uniquely in DM1 patients. Pathway analysis of these proteins revealed biochemical pathways that appear to be affected by the repeat expansions. 

 

ELUCIDATION OF BOTTLENECKS IN THE CELLULAR QUALITY CONTROL PATHWAY OF PERIPHERAL MYELIN PROTEIN 22

Jonathan P. Schlebach, Lara Scott, Alan Tang, R. Luke Wiseman, Charles R. Sanders
Mutations in the sequence or copy number of the gene encoding the integral membrane protein peripheral myelin protein 22 (PMP22) are the most common cause of Charcot-Marie-Tooth Disease. The majority of these mutations result in the accumulation of misfolded PMP22 in the secretory pathway and a concomitant decrease in the yield of mature protein at the plasma membrane. We recently found that the extent of mutant PMP22 misfolding is proportional to the degree of dysmyelination exhibited by patients carrying these mutations, which suggests that strategies to restore PMP22 folding and trafficking may be useful for therapeutic intervention. To identify key modulators of PMP22 quality control, we used flow cytometry to quantitatively assess the effects of selective induction of two transcription activators of the unfolded protein response (ATF6 and XBP1) on the trafficking efficiency of PMP22. Initial results suggest that ATF6 activation selectively retains and degrades destabilized PMP22 variants. Alternatively, we find that XBP1 induction increases the trafficking efficiency of PMP22. Unlike ATF6, XBP1 upregulates the expression of lectin chaperones that carryout glycosylation-mediated quality control in the ER. To determine whether the effect of XBP1 is critically dependent upon the glycosylation state, we characterized the cellular trafficking of a non-glycosylated PMP22 variant (N41Q). Strikingly, our results show that removal of the glycan doubles the trafficking efficiency of PMP22, which suggests that glycan recognition constitutes a limiting interaction in the quality control pathway. Furthermore, we find that XBP1 induction has no effect on the trafficking of N41Q PMP22, which suggests that lectin chaperones play a critical role in XBP1-mediated PMP22 rescue. Together our preliminary findings suggest glycan recognition constitutes a key bottleneck in the cellular trafficking of PMP22 and suggest new pharmacological strategies to correct its pathogenic misfolding. 

 

DIVERSE, BIOLOGICALLY RELEVANT, AND TARGETABLE GENE REARRANGEMENTS IN TRIPLE-NEGATIVE BREAST CANCER AND OTHER MALIGNANCIES 

Tim Shaver, Brian D. Lehmann, J. Scott Beeler, Chung-I Li, Zhu Li, Hailing Jin, Thomas P. Stricker, Yu Shyr, Jennifer A. Pietenpol
Triple-negative breast cancer (TNBC) and other molecularly heterogeneous malignancies present a significant clinical challenge due to a lack of high-frequency "driver" alterations amenable to therapeutic intervention. These cancers often exhibit genomic instability, resulting in chromosomal rearrangements that impact the structure and expression of protein-coding genes. However, identification of these rearrangements remains technically challenging. Using a newly developed approach that quantitatively predicts gene rearrangements in tumor-derived genetic material, we identified and characterized a novel oncogenic kinase fusion and discovered a clinical occurrence and cell line model of an additional, clinically targetable fusion in TNBC. Expanding our analysis to other malignancies, we identified a diverse array of novel and known hybrid transcripts, including rearrangements between non-coding regions and an array of clinically relevant genes involved in tumor proliferation, survival, and immune evasion. The over 1000 genetic alterations we identified highlight the importance of considering non-coding gene rearrangement partners, and the targetable gene fusions identified in TNBC demonstrate the need to advance gene fusion detection for molecularly heterogeneous cancers. 

 

HDAC3 IS REQUIRED FOR B CELL DEVELOPMENT AND REGULATES TRANSCRIPTIONAL PROGRAMS DRIVING TERMINAL B CELL DIFFERENTIATION

Kristy Stengel, Srividya Bhaskara, Yue Zhao, Qi Liu, Scott Hiebert
Hdac3 is a targeted by HDAC inhibitors, which are FDA approved for the treatment of cutaneous T cell lymphoma (CTCL) and multiple myeloma and currently being assessed for efficacy in other cancer types, including B cell lymphomas. Deletion of Hdac3 early in hematopoiesis revealed its requirement for the formation of the earliest lymphoid progenitor cells, preventing further examination of the role of Hdac3 in B cell function. Therefore, in order to address the role of Hdac3 in B cells, we crossed conditional Hdac3 animals to Mb1-cre mice to drive recombination at the Hdac3 locus in early B cell progenitors. Deletion of Hdac3 early in B cell development prevents formation of mature B cells due to defects in VDJ recombination. Furthermore, transmission electron microscopy and micrococcal nuclease digestion revealed global changes in chromatin structure in B cell progenitors lacking Hdac3, suggesting that the deregulation of chromatin structure likely contributes to the observed defects. While Mb1-cre-mediated deletion of Hdac3 results in a complete loss of mature B cells from peripheral lymphoid tissues, Hdac3 is recruited by the proto-oncogene, Bcl6, a critical regulator of B cell function, suggesting that Hdac3 may play a role in mature B cells as well. Therefore, in order to define the role of Hdac3 in mature B cell functions, we used a Cd19-cre transgene to specifically delete Hdac3 later in B cell development. While Cd19-cre-driven loss of Hdac3 had no effect on B cell maturation within the bone marrow, Hdac3-/- spleens showed significant alterations in germinal center (GC) structure. Additionally, in vitro stimulation of Hdac3-deficient B cells showed intact BCR signaling, yet a failure to drive plasmablast formation and expansion. These in vitro studies were consistent with the observed reduction of plasma cell numbers in the knockout mice. Further analysis of the Hdac3-/- germinal centers revealed an accumulation of light zone centrocytes, which give rise to fully differentiated plasma cells and memory B cells. Finally, RNA-Seq analysis of Hdac3-/- GC B cells revealed an up-regulation of plasma cell transcription factors (Prdm1, Irf4, Xbp1) without a concomitant reduction of germinal center-associated transcription factors (Bcl6, Pax5, Irf8, Bach2). In all, these studies demonstrate that loss of Hdac3 from B cell progenitors is associated with changes in global chromatin structure and defective VDJ recombination, while the loss of Hdac3 from more mature B cells results in the deregulation of transcriptional programs required for the differentiation of activated B cells into plasma cells.