PDF file, 87KB, Differentially Expressed Genes by miR-708 Status in Never-Smoker Lung Adenocarcinoma.

PDF file, 147KB, Validation of miR-708 Overexpression by Real-Time qPCR in Lung Squamous cell carcinoma. Non-paired mostly smoker Squamous cell carcinomas and normal FFPE tissues from the National Cancer Institute cohort were examined. Sample numbers (n) and tissue types are indicated on each graph. P values are as shown by unpaired 2-tailed t test.

PDF file, 90KB, Endogenous expression level of miR-708 in various cell lines. Real-time quantitative polymerase chain reaction analyses were performed as described in the "Methods" section.

PDF file, 154KB, TMEM88 is a putative target of miR-708 by MicroCosm database. The target site of miR-708 is located in the 3'UTR, 286 to 309 bases from downstream of the stop codon sequence of TMEM88. Asterisks indicate the binding site.

PDF file, 69KB, Clinical Information of the three lung adenocarcinoma Cohorts.

PDF file, 58KB, Potential Target Genes of miR-708.

Lung cancer occurs in never-smokers. Epigenetic changes in lung cancer potentially represent important diagnostic, prognostic, and therapeutic targets. We compared DNA methylation profiles of 28 adenocarcinomas of the lungs of never-smokers with paired adjacent nonmalignant lung tissue. We correlated differential methylation changes with gene expression changes from the same 28 sample pairs. Using principal component analysis, we observed a distinct separation in methylation profiles between tumor and adjacent nonmalignant lung tissue. Tumors were generally hypomethylated compared with adjacent nonmalignant tissue. Of 1,906 CpG sites differentially methylated between tumor and nonmalignant tissue, 1,198 were within classically defined CpG islands where tumors were hypermethylated compared with nonmalignant tissue. A total of 708 sites were outside CpG islands where tumors were hypomethylated compared with nonmalignant tissue. There were significant differences in expression of 351 genes (23%) of the 1,522 genes matched to the differentially methylated CpG sites. Genes that were not significantly differentially expressed and were hypermethylated within CpG sites were enriched for homeobox genes. These results suggest that the methylation profiles of lung adenocarcinomas of never-smokers and adjacent nonmalignant lung tissue are significantly different. Despite the differential methylation of homeobox genes, no significant changes in expression of these genes were detected.

Background: Irritable Bowel Syndrome (IBS) has been associated with changes in the rectosigmoid mucosal expression of immune (TNFSF15, TLRs) and nonimmune (tight junctions, mucus and serotonergic) factors. In a recent microarray based study on Campylobacter jejuni post-infectious IBS (PI-IBS), patients were found to have an increased rectal mucosal expression of CCL11, CCL13, Calpain 8 (pro-inflammatory), GABRE and decreased expression of NR1D1, GPR161. Aim: To determine changes in colonic mucosal RNA expression among patients with C. jejuni PI-IBS in comparison to healthy volunteers using whole transcriptome sequencing. Methods: Sigmoid colonic biopsies were obtained from 5 PI-IBS patients (3 females) and 7 age-matched healthy volunteers (5 females). Biopsies were preserved in RNALater at -80°C. Total RNA was isolated using the RNeasy Plus Mini Kit (Qiagen, Valencia, CA). All samples had RNA Integrity Numbers >7.0. RNA library preparation was performed using the Illumina TruSeq RNA Sample Prep v2 (San Diego, CA) and sequencing was performed as paired-end 51 base reads on an Illumina HiSeq 2000 with 3 samples/ lane using TruSeq SBS Sequencing Kit Version 3. Base calling was performed using Illumina's RTA version 1.12.4.2. Analysis (alignment statistics, in-depth quality control, gene and exon expression, fusion transcripts, and single nucleotide variants) was done using MAP-RSeq v1.2.1.3. Differential expression between samples was done using edgeR algorithm. Results were also compared to our recently published transcriptomic data from rectosigmoid biopsies of IBS-D patients (Am J Physiol 306: G1089-G1098, 2014). Results: Overall, mRNA expressions of 20 genes were changed in patients with C. jejuni PI-IBS as compared to healthy volunteers (P values obtained using false detection rate (FDR) threshold of 5% ranged from 10-5 to 10-9) (Table 1). The genes that stood out as potentially relevant to pathophysiology of PI-IBS included: neurotransmitter-related genes (↑neuropeptide Y receptor type 2, ↑Galanin, ↓nitric oxide synthase 2 (inducible), and ↓indoleamine 2, 3-dioxygenase 1), proteases (↑Secretory leukocyte peptidase inhibitor, ↓Matrix metallopeptidase 3), chemokines (↑chemokine C-C motif ligand 18, ↓chemokine C-X-C motif ligand 10 and 11), carbohydratemetabolism (↑aldolase B, ↑sucrase-isomaltase), ion channels and transporters (↑bestrophin 4, ↑calbindin 2, ↑solute carrier family 3). With the exception of aldolase B expression, there was no overlap with the IBS-D dataset. Conclusion: Transcriptome sequencing of sigmoid colonic mucosa in C. jejuni PI-IBS shows distinct transcriptomic differences from healthy controls and from non-post-infectious IBS-D. These data provide the rationale to explore the role of mucosal factors in the pathophysiology of C. jejuni PI-IBS. Table 1: Gene expression identified by edgeR showing relative expression in C. jejuni PIIBS compared to healthy volunteers

OBJECTIVE:To investigate transcriptional regulation of late-onset Alzheimer’s disease (LOAD) GWAS risk loci genes in the human brain. BACKGROUND:We previously identified genetic associations of brain levels of genes at the LOAD risk GWAS loci ABCA7 , BIN1, CLU, CR1 and MS4A4A with cisSNPs, including some of the top AD risk variants. Recently, 11 additional risk loci were identified through a large meta-analysis of multiple LOAD GWAS. In this study, our first aim is to investigate the association of brain levels of these 11 novel LOAD risk loci genes with cis SNPs. Our second aim is to evaluate the potential role of alternative splicing in these gene expression associations. DESIGN/METHODS:We have already obtained gene expression measures of ~24,000 transcripts in two brain regions (temporal cortex and cerebellum) for ~ 200 AD subjects and ~200 subjects with other pathologies using Illumina’s WG-DASL assay. LOAD GWAS SNPs at the 11 novel loci will be analyzed for association with expression levels of the corresponding genes in cis , using linear regression adjusting for appropriate covariates. To evaluate alternative splicing, we will use NanoString nCounter™ technology to measure levels of all known transcripts of the candidate genes at all LOAD GWAS loci in ~350 brain RNA samples. RESULTS:We have previously demonstrated the utility of this approach and anticipate that there exist additional LOAD genes, brain levels of which are likewise influenced by SNPs. We postulate that some of these SNPs also confer LOAD risk and some of the expression associations are due to alternative splicing of exons. CONCLUSIONS:Many genes at the LOAD GWAS loci associate with cis SNPs, some of which are the top LOAD risk SNPs. This suggests that many LOAD GWAS variants confer risk by transcriptional regulation of genes at these loci. Dissection of this transcriptional regulation is expected to have implications in the understanding of LOAD pathophysiology. Study Supported by: R01 AG032990, P50 AG016574 Disclosure: Dr. Kachadoorian has nothing to disclose. Dr. Allen has nothing to disclose. Dr. Karhade has nothing to disclose. Dr. Karhade has nothing to disclose. Dr. Zou has nothing to disclose. Dr. Chai has nothing to disclose. Dr. Younkin has nothing to disclose. Dr. Crook has nothing to disclose. Dr. Pankratz has received research support from Abbott Laboratories, Inc. Dr. Carrasquillo has nothing to disclose. Dr. Nair has nothing to disclose. Dr. Middha has nothing to disclose. Dr. Maharjan has nothing to disclose. Dr. Nguyen has nothing to disclose. Dr. Ma has nothing to disclose. Dr. Malphrus has nothing to disclose. Dr. Lincoln has nothing to disclose. Dr. Bisceglio has nothing to disclose. Dr. Kolbert has nothing to disclose. Dr. Jen has nothing to disclose. Dr. Petersen has received personal compensation for activities with Pfizer, Inc., and Janssen Alzheimer9s Immunotherapy. Dr. Petersen has received royalty payments from Oxford University Press. Dr. Graff-Radford has received personal compensation for activities with Codman as a member of a scientific advisory board. Dr. Graff-Radford has received personal compensation in an editorial capacity for The Neurologist. Dr. Graff-Radford has received research support from Janssen, Pfizer Inc., Medivation, Forest Laboratories Inc., and Allon. Dr. Younkin has nothing to disclose. Dr. Dickson has received personal compensation for activities with Neotope, Inc. as a consultant. Dr. Taner has nothing to disclose.

Summary Based upon genetic analysis, decorin is an exciting pharmacologic agent of potential anti-fibrogenic effect on arthrofibrosis in our animal model. Introduction While the pathophysiology of arthrofibrosis is not fully understood, some anti-fibrotic molecules such as decorin could potentially be used for the prevention or treatment of joint stiffness. The goal of this study was to determine whether intra-articular administration of decorin influences the expression of genes involved in the fibrotic cascade ultimately leading to less contracture in an animal model. Material and Methods Eighteen rabbits had their right knees operated on to form contractures. The left knees served as controls. The 6 right limbs in the experimental group (Group 1) received four 500 ug/ml intra-articular injections of decorin over 8 days starting at 8 week, for a total of 2 mg. The 6 right limbs in the first control group (Group 2) received four intra-articular injections of bovine serum albumin (BSA) over 8 days starting at 8 weeks as well. The 6 six right limbs in the second control group (Group 3) received no injections. The contracted limbs of rabbits in Group 1 were biomechanically and genetically compared to the contracted limbs of rabbits in Groups 2 and 3 with the use of a calibrated joint measuring device and custom microarray, respectively. Results There was no statistical difference in the flexion contracture angles between those right limbs that received intra-articular decorin versus those that received intra-articular BSA (66° vs. 69°; p = 0.41). Likewise, there was no statistical difference between those right limbs that received intra-articular decorin as opposed to those who had no injection (66° vs. 72°; p = 0.27). The lack of significance remained when the control left limbs were taken into account (p > 0.40). When compared to bovine serum albumin (BSA), decorin led to a statistically significant increase in the mRNA expression of 5 genes: substance P, neuropeptide γ, and neurokinin A, cyclin E2, and MMP-9 (p Conclusions In this model, when administered intra-articularly at 8 weeks, 2 mg of decorin had no significant effect on joint contractures. However, our genetic analysis revealed a significant alteration in the expression of several fibrotic genes. Further studies investigating the route of administration, dosing, frequency, and timing are required before definitive conclusions may be drawn on the effects of decorin on joint contractures.

OBJECTIVE: To use existing microarray data to identify genes that are differentially expressed in AD vs. non-AD brains. BACKGROUND: Despite the success of recent LOAD GWAS studies, much of the heritability of LOAD remains unexplained. We hypothesize that evaluation of brain gene expression provides an additional avenue for identification of novel LOAD genes and pathways that may be potential drug targets. We previously performed a gene expression GWAS that assessed mRNA levels of ~24,000 transcripts in two brain regions (temporal cortex and cerebellum) for ~ 200 AD subjects and ~200 subjects with other pathologies We have now performed transcript profiling analysis comparing gene expression levels between AD and non-AD subjects in both brain regions and conducted pathway analysis on these data. In addition, we are collecting whole genome CpG methylation data (methylome) on a subset of these subjects for future evaluation. DESIGN/METHODS: Gene expression levels were measured by the Illumina HT-12 V4 Expression BeadChip arrays using the Whole-Genome DASL HT assay and appropriate data quality control was implemented. Transcript profiling analysis was carried out in R using linear regression with appropriate covariates. Gene pathway analysis was executed using MetaCore for the two brain regions separately. RESULTS: Following QC ~17,000 gene expression measures were robustly detected in each brain region. Transcript profiling analysis identified 743 targets in the cerebellum and 2839 targets in the temporal cortex (un-corrected p-value <0.01) that were selected for pathway analysis. In the temporal cortex several significant pathways and GO Processes were identified including but not limited to oxidative phosphorylation and lipid metabolism. CONCLUSIONS: Transcript profiling and pathway analysis of brain gene expression data has identified several interesting targets for further exploration of molecular pathways involved in AD. The addition of methylome data will further enhance our understanding of the role of gene expression and control in this disease. Study Supported by: R01 AG032990, P50 AG016574, Mayo Clinic Center for Individualized Medicine Epigenomics Grant. Disclosure: Dr. Allen has nothing to disclose. Dr. Serie has nothing to disclose. Dr. Walsh has nothing to disclose. Dr. Zhifu has nothing to disclose. Dr. Baheti has nothing to disclose. Dr. Zou has nothing to disclose. Dr. Chai has nothing to disclose. Dr. Younkin has nothing to disclose. Dr. Crook has nothing to disclose. Dr. Pankratz has received research support from Abbott Laboratories, Inc. Dr. Carrasquillo has nothing to disclose. Dr. Nair has nothing to disclose. Dr. Middha has nothing to disclose. Dr. Maharjan has nothing to disclose. Dr. Nguyen has nothing to disclose. Dr. Ma has nothing to disclose. Dr. Malphrus has nothing to disclose. Dr. Lincoln has nothing to disclose. Dr. Bisceglio has nothing to disclose. Dr. Kolbert has nothing to disclose. Dr. Jen has nothing to disclose. Dr. Petersen has received personal compensation for activities with Pfizer, Inc., and Janssen Alzheimer9s Immunotherapy. Dr. Petersen has received royalty payments from Oxford University Press. Dr. Graff-Radford has received personal compensation for activities with Codman as a member of a scientific advisory board. Dr. Graff-Radford has received personal compensation in an editorial capacity for The Neurologist. Dr. Graff-Radford has received research support from Janssen, Pfizer Inc., Medivation, Forest Laboratories Inc., and Allon. Dr. Dickson has received personal compensation for activities with Neotope, Inc. as a consultant. Dr. Younkin has nothing to disclose. Dr. Asmann has nothing to disclose. Dr. Taner has nothing to disclose.

OBJECTIVE: To use existing microarray data to identify genes that are differentially expressed in AD vs. non-AD brains. BACKGROUND: Despite the success of recent LOAD GWAS studies, much of the heritability of LOAD remains unexplained. We hypothesize that evaluation of brain gene expression provides an additional avenue for identification of novel LOAD genes and pathways that may be potential drug targets. We previously performed a gene expression GWAS that assessed mRNA levels of ~24,000 transcripts in two brain regions (temporal cortex and cerebellum) for ~ 200 AD subjects and ~200 subjects with other pathologies We have now performed transcript profiling analysis comparing gene expression levels between AD and non-AD subjects in both brain regions and conducted pathway analysis on these data. In addition, we are collecting whole genome CpG methylation data (methylome) on a subset of these subjects for future evaluation. DESIGN/METHODS: Gene expression levels were measured by the Illumina HT-12 V4 Expression BeadChip arrays using the Whole-Genome DASL HT assay and appropriate data quality control was implemented. Transcript profiling analysis was carried out in R using linear regression with appropriate covariates. Gene pathway analysis was executed using MetaCore for the two brain regions separately. RESULTS: Following QC ~17,000 gene expression measures were robustly detected in each brain region. Transcript profiling analysis identified 743 targets in the cerebellum and 2839 targets in the temporal cortex (un-corrected p-value <0.01) that were selected for pathway analysis. In the temporal cortex several significant pathways and GO Processes were identified including but not limited to oxidative phosphorylation and lipid metabolism. CONCLUSIONS: Transcript profiling and pathway analysis of brain gene expression data has identified several interesting targets for further exploration of molecular pathways involved in AD. The addition of methylome data will further enhance our understanding of the role of gene expression and control in this disease. Study Supported by: R01 AG032990, P50 AG016574, Mayo Clinic Center for Individualized Medicine Epigenomics Grant.

OBJECTIVE:To investigate transcriptional regulation of late-onset Alzheimer's disease (LOAD) GWAS risk loci genes in the human brain. BACKGROUND:We previously identified genetic associations of brain levels of genes at the LOAD risk GWAS loci ABCA7, BIN1, CLU, CR1 and MS4A4A with cisSNPs, including some of the top AD risk variants. Recently, 11 additional risk loci were identified through a large meta-analysis of multiple LOAD GWAS. In this study, our first aim is to investigate the association of brain levels of these 11 novel LOAD risk loci genes with cisSNPs. Our second aim is to evaluate the potential role of alternative splicing in these gene expression associations. DESIGN/METHODS:We have already obtained gene expression measures of ~24,000 transcripts in two brain regions (temporal cortex and cerebellum) for ~ 200 AD subjects and ~200 subjects with other pathologies using Illumina's WG-DASL assay. LOAD GWAS SNPs at the 11 novel loci will be analyzed for association with expression levels of the corresponding genes in cis, using linear regression adjusting for appropriate covariates. To evaluate alternative splicing, we will use NanoString nCounter™ technology to measure levels of all known transcripts of the candidate genes at all LOAD GWAS loci in ~350 brain RNA samples. RESULTS:We have previously demonstrated the utility of this approach and anticipate that there exist additional LOAD genes, brain levels of which are likewise influenced by SNPs. We postulate that some of these SNPs also confer LOAD risk and some of the expression associations are due to alternative splicing of exons. CONCLUSIONS:Many genes at the LOAD GWAS loci associate with cisSNPs, some of which are the top LOAD risk SNPs. This suggests that many LOAD GWAS variants confer risk by transcriptional regulation of genes at these loci. Dissection of this transcriptional regulation is expected to have implications in the understanding of LOAD pathophysiology. Study Supported by: R01 AG032990, P50 AG016574

MAPT encodes for tau, the predominant component of neurofibrillary tangles that are neuropathological hallmarks of Alzheimer’s disease (AD). Genetic association of MAPT variants with late-onset AD (LOAD) risk has been inconsistent, although insufficient power and incomplete assessment of MAPT haplotypes may account for this.

OBJECTIVE: To evaluate MAPT subhaplotypes for association with risk for LOAD and MAPT brain expression levels.

Intracellular neurofibrillary tangles composed predominantly of tau protein are a classic neuropathological feature of AD. The MAPT gene (Chr17) encodes for tau, however the evidence for genetic involvement of this locus in LOAD etiology has been inconsistent. This may be because previous studies have generally used relatively small numbers of LOAD cases and controls, and/or have not investigated haplotypic variability at the locus in depth. We sought to examine well-established MAPT sub-haplotype-tagging variants in our large LOAD case-control series to determine their effect on risk for LOAD. We also hypothesized that some of the MAPT variants may confer disease risk by influencing brain expression levels of MAPT, based on our expression GWAS (eGWAS) and previous reports by others. We genotyped six SNPs which tag the most common MAPT locus haplotypes (frequencies >1%), in three Caucasian LOAD case-control series (N-cases=2052; N-controls=3,406). We measured MAPT expression levels in the cerebellum (N=197) and temporal cortex (N=202) of autopsied AD subjects as part of our eGWAS. The six SNPs and the estimated haplotypes were tested for: (I) association with LOAD risk using logistic regression and (II) gene expression of MAPT using linear regression; all analysis included appropriate covariates. The H2 haplotype was significantly associated with decreased risk of LOAD (OR = 0.80, p=4.10E-04), and decreased MAPT expression in both brain regions (beta=-0.16 to -0.49, p=3.4E-03 to 8.70E-33). The most common H1 sub-haplotype (H1b) had nominally significant association with increased risk of LOAD (OR=1.15, p=0.046) and a global test for the 19 haplotypes identified in this study was significant (p = 0.0123). Interestingly we do not find significant association for the H1c haplotype that has previously been implicated in LOAD (p = 0.277). Our results strongly suggest that MAPT is a LOAD gene that has regulatory variants which confer LOAD risk by influencing its brain expression. Additional studies are needed to identify the precise mechanism of disease attributable to genetic variation at this locus.

MicroRNAs play a role in regulating diverse biological processes and have considerable utility as molecular markers for diagnosis and monitoring of human disease.Several technologies are available commercially for measuring microRNA expression.However, cross-platform comparisons do not necessarily correlate well, making it difficult to determine which platform most closely represents the true microRNA expression level in a tissue.To address this issue, we have analyzed RNA derived from cell lines, as well as fresh frozen and formalin-fixed paraffin embedded tissues, using Affymetrix, Agilent, and Illumina microRNA arrays, NanoString counting, and Illumina Next Generation Sequencing.We compared the performance within-and between the different platforms, and then verified these results with those of quantitative PCR data.Our results demonstrate that the within-platform reproducibility for each method is consistently high and although the gene expression profiles from each platform show unique traits, comparison of genes that were commonly detectable showed that detection of microRNA transcripts was similar across multiple platforms.

Technical advancements in quantitative PCR (qPCR) instrumentation have made it possible to perform gene expression measurements using small sample input to support both basic and clinical research studies. As part of the strategic goals to assess new technologies and identify protocols that best fit the needs of the Mayo Clinic, we compared the Fluidigm BioMark system with standard Applied Biosystems (AB) instrumentation for mRNA and miRNA gene expression measurements. We also examined the performance of the BioMark system when using very low-input RNA. We evaluated a set of control samples using the same TaqMan assays with both systems. We observed that the BioMark-generated data routinely yields Ct values approximately 10 cycles lower than those obtained with AB instrumentation. The correlations between the two platforms were high (r = 0.96) for both mRNA and miRNA expression experiments. For miRNA expression, a similarly high correlation was observed between fresh frozen and formalin-fixed paraffin embedded (FFPE) samples. In an effort to accommodate our customer needs, we also evaluated the performance of the BioMark for evaluating gene expression in very low-input samples. Using six standard TaqMan control assays (having high, medium and low expression levels), we observed that high quality RNA samples as low as 10pg achieved linear amplification across four different pre-amplification cycles (10, 14, 18 and 22). At 10pg total RNA input, low-expression control assay IPO8 demonstrated a correlation of r = .999 among the four pre-amplification cycles. This linearity was also observed at higher RNA input levels, up to 10ng. The only control assay that did not perform in a linear fashion across all input amounts and all pre-amplification cycles was 18S ribosomal RNA. The highest correlation observed for 18S was r = 0.801, and this supports the vendor suggestion that 18S is not the best control assay option.