Matt Strub and Tricia Braun to present at student seminar 2/11/2016
Matt Strub’s Abstract
Genomic Signature-Based Approaches to Drug Repositioning for ΔF508-CFTR Rescue
Cystic fibrosis (CF) is a lethal autosomal recessive disease caused by mutations in the CF transmembrane conductance regulator (CFTR) gene. The most common CFTR mutation, termed ΔF508, is a 3 base deletion resulting in the loss of a phenylalanine residue. This causes protein misfolding, resulting in proteosomal degradation. However, if CFTR-ΔF508 is allowed to traffick to the cell membrane, anion channel function may be partially restored. The McCray Lab previously reported that transfection with a miR-138 mimic or siRNA knockdown of SIN3A in primary cultures of CF airway epithelia increases CFTR mRNA and protein levels, and partially restores cAMP-stimulated Cl– conductance. The Connectivity Map (CMAP) is a catalog of gene expression profiles from cultured human cells treated with a variety of bioactive chemical compounds and has pattern-matching software to mine data. A CMAP query using previously generated gene expression signatures identified 27 candidate small molecules that mimicked miR-138 and DsiRNA SIN3A treatments. Our lab reported the identification of 4 small molecules that partially restored CFTR-ΔF508 function in primary CF airway epithelia, highlighting the utility of a genomic signature approach in drug discovery. Currently, the NIH is greatly expanding the CMAP dataset into the Library of Integrated Network-based Cellular Signatures (LINCS). Here we query the LINCS database using a meta-analysis of rescue signatures to identify candidate compounds for rescue of CFTR-ΔF508 function.
In collaboration with the Broad Institute, we used previously generated gene sets to iteratively query the Library of Integrated Network-based Cellular Signatures (LINCS). 125 candidate small molecules were selected for further testing. Functional screens performed in CFBE (ΔF508/ΔF508) cells identified 7/125 compounds that partially rescued ΔF508 function, as assessed by cAMP-activated Cl– conductance. Additional experiments performed to assess their activity in primary human CF epithelial cells confirmed the ability of these seven compounds to partially rescue ΔF508 function. Interestingly, some of these compounds showed significant cooperativity when administered with C18. We then obtained 70 congeners, derivatives, or related compounds of the seven validated LINCS hits and identified 18/70 compounds that increased cAMP-activated chloride conductance by at least 50% compared to a DMSO control. We also tested compounds in the presence of C18 and identified four compounds that increased conductance by at least 40% compared to a C18 control.
We recently used a meta-analytic approach to analyze multiple rescue signatures and relevant pathways to create a prioritized candidate drug list for functional screens. Transcriptomic profiles of RNAi (siRNA knockdown of SIN3A, SYVN1, and NEDD8, and miR-138 overexpression) and temperature (27°C for 24 hours, 40°C for 24 hours, and 27°C for 24 hours followed by 40°C for 24 hours) treatments were used to represent a rescue signature. Profiles of primary cells (both human and pig) from CF and healthy donors were used to represent a disease signature. The rescue and disease signatures were then analyzed together to create representative lists of up- and down-regulated genes. Meanwhile, gene sets were extracted from relevant curated pathways related to CFTR trafficking (from Thomson Reuters MetaCore). Gene sets from both strategies were then scored against all drug profiles contained in the CMAP and LINCS databases. The rescue signature- and disease signature-based scores were combined to form an overall score that was used to identify and prioritize candidate molecules. We have tested 115 compounds from the resulting meta-analysis-based candidate drug list and 36 compounds have been shown to increase cAMP-activated chloride conductance by at least 30% compared to DMSO. Additionally, 21 compounds in the presence of C18 increased conductance by at least 30% compared to a C18 control. Further analysis of preliminary hits will include validation in primary cells. We also have the ability to test the most efficacious drugs in our ΔF508/ΔF508 porcine model. Lastly, we are performing chemogenomic enrichment analysis using the results of all tested compounds to elucidate possible classification, structural, or transcriptomic similarities between efficacious drugs. Such analysis may help us to identify and prioritize additional candidates.
Tricia Braun’s Abstract
Genome-wide DNA methylation analysis of glucocorticoid treatment in human Blood and saliva and the correlation of DNA methylation between peripheral tissues and brain
Patricia Braun1, Yasunori Nagahama2, Marie Hafner1, Lauren O’Sullivan1, Melissa McKane1, Tanner D. Gardiner1, Andrew Grossbach2, Matthew A Howard III2, Hiroto Kawasaki2, James B. Potash1, Gen Shinozaki1
1Department of Psychiatry, University of Iowa Carver College of Medicine
2Department of Neurosurgery, University of Iowa Carver College of Medicine
Glucocorticoids help regulate the stress response, and an imbalance of glucocorticoids has been implicated in depression. Within mouse models, candidate genes have been shown to be differentially methylated in response to glucocorticoid treatment. Using the Infinium HumanMethylation 450K Array, which covers over 450,000 CpGs, we performed preliminary studies on genome-wide DNA methylation (DNAm) changes that occur within saliva samples from 10 subjects and blood samples from 6 subjects before and after treatment with dexamethasone, a corticosteroid, in the context of neurosurgery. Within saliva samples, average DNAm differences of >10% were observed for 84 CpGs. One CpG in a long intergenic non-coding RNA (LINC00871) attained near genome-wide level of significance (average DNAm: pre-dexamethasone 49%, post-dexamethasone 38%; p=5×10-7). Within blood samples, over 200 CpGs had >20% difference; however, none were statistically significant. These findings provide initial evidence for an influence of glucocorticoids on DNAm within humans. To understand the relevance DNAm in saliva and blood to brain DNAm, we also examined the correlation of DNAm between peripheral tissues and brain tissues. This is an essential issue for not only our analysis, but for the field more broadly. We obtained saliva, blood, and live brain tissue samples from 13 patients undergoing neurosurgery and analyzed genome-wide DNAm with the 450K Array. Blood and saliva showed a high degree of correlation for DNAm (r2=0.97), and saliva DNAm was revealed to be more similar to brain DNAm (r2=0.84) than blood (r2=0.81; p<1×10-4). As we have ongoing access to samples from neurosurgery patients, we will expand these studies to understand the extent to which peripheral tissues can be used as surrogate tissues for DNAm in the brain. Furthermore, we will collect samples from oral surgery patients given a higher dosage of glucocorticoids to more fully ascertain the global effects of glucocorticoids on DNAm.