Application of the S-score algorithm for analysis of oligonucleotide microarrays

Differential potassium channel gene regulation in BXD mice reveals novel targets for pharmacogenetic therapies to reduce heavy alcohol drinking

Alcohol (ethanol) dependence is a chronic relapsing brain disorder partially influenced by genetics and characterized by an inability to regulate harmful levels of drinking. Emerging evidence has linked genes that encode K_V7, K_IR, and K_Ca2 K⁺ channels with variation in alcohol-related behaviors in rodents and humans. This led us to experimentally test relations between K⁺ channel genes and escalation of drinking in a chronic-intermittent ethanol (CIE) exposure model of dependence in BXD recombinant inbred strains of mice. Transcript levels for K⁺ channel genes in the prefrontal cortex (PFC) and nucleus accumbens (NAc) covary with voluntary ethanol drinking in a non-dependent cohort. Transcripts that encode K_V7 channels covary negatively with drinking in non-dependent BXD strains. Using a pharmacological approach to validate the genetic findings, C57BL/6J mice were allowed intermittent access to ethanol to establish baseline consumption before they were treated with retigabine, an FDA-approved K_V7 channel positive modulator. Systemic administration significantly reduced drinking, and consistent with previous evidence, retigabine was more effective at reducing voluntary consumption in high-drinking than low-drinking subjects. We evaluated the specific K⁺ channel genes that were most sensitive to CIE exposure and identified a gene subset in the NAc and PFC that were dysregulated in the alcohol-dependent BXD cohort. CIE-induced modulation of nine genes in the NAc and six genes in the PFC covaried well with the changes in drinking induced by ethanol dependence. Here we identified novel candidate genes in the NAc and PFC that are regulated by ethanol dependence and correlate with voluntary drinking in non-dependent and dependent BXD mice. The findings that Kcnq expression correlates with drinking and that retigabine reduces consumption suggest that K_V7 channels could be pharmacogenetic targets to treat individuals with alcohol addiction.

The allostatic impact of chronic ethanol on gene expression: A genetic analysis of chronic intermittent ethanol treatment in the BXD cohort

The transition from acute to chronic ethanol exposure leads to lasting behavioral and physiological changes such as increased consumption, dependence, and withdrawal. Changes in brain gene expression are hypothesized to underlie these adaptive responses to ethanol. Previous studies on acute ethanol identified genetic variation in brain gene expression networks and behavioral responses to ethanol across the BXD panel of recombinant inbred mice. In this work, we have performed the first joint genetic and genomic analysis of transcriptome shifts in response to chronic intermittent ethanol (CIE) by vapor chamber exposure in a BXD cohort. CIE treatment is known to produce significant and sustained changes in ethanol consumption with repeated cycles of ethanol vapor. Using Affymetrix microarray analysis of prefrontal cortex (PFC) and nucleus accumbens (NAC) RNA, we compared CIE expression responses to those seen following acute ethanol treatment, and to voluntary ethanol consumption. Gene expression changes in PFC and NAC after CIE overlapped significantly across brain regions and with previously published expression following acute ethanol. Genes highly modulated by CIE were enriched for specific biological processes including synaptic transmission, neuron ensheathment, intracellular signaling, and neuronal projection development. Expression quantitative trait locus (eQTL) analyses identified genomic loci associated with ethanol-induced transcriptional changes with largely distinct loci identified between brain regions. Correlating CIE-regulated genes to ethanol consumption data identified specific genes highly associated with variation in the increase in drinking seen with repeated cycles of CIE. In particular, multiple myelin-related genes were identified. Furthermore, genetic variance in or near dynamin3 (Dnm3) on Chr1 at ∼164 Mb may have a major regulatory role in CIE-responsive gene expression. Dnm3 expression correlates significantly with ethanol consumption, is contained in a highly ranked functional group of CIE-regulated genes in the NAC, and has a cis-eQTL within a genomic region linked with multiple CIE-responsive genes.

Genetic variation within the Chrna7 gene modulates nicotine reward-like phenotypes in mice

Mortality from tobacco smoking remains the leading cause of preventable death in the world, yet current cessation therapies are only modestly successful, suggesting new molecular targets are needed. Genetic analysis of gene expression and behavior identified Chrna7 as potentially modulating nicotine place conditioning in the BXD panel of inbred mice. We used gene targeting and pharmacological tools to confirm the role of Chrna7 in nicotine conditioned place preference (CPP). To identify molecular events downstream of Chrna7 that may modulate nicotine preference, we performed microarray analysis of α7 knock-out (KO) and wild-type (WT) nucleus accumbens (NAc) tissue, followed by confirmation with quantitative polymerase chain reaction (PCR) and immunoblotting. In the BXD panel, we found a putative cis expression quantitative trait loci (eQTL) for Chrna7 in NAc that correlated inversely to nicotine CPP. We observed that gain-of-function α7 mice did not display nicotine preference at any dose tested, whereas conversely, α7 KO mice demonstrated nicotine place preference at a dose below that routinely required to produce preference. In B6 mice, the α7 nicotinic acetylcholine receptor (nAChR)-selective agonist, PHA-543613, dose-dependently blocked nicotine CPP, which was restored using the α7 nAChR-selective antagonist, methyllycaconitine citrate (MLA). Our genomic studies implicated a messenger RNA (mRNA) co-expression network regulated by Chrna7 in NAc. Mice lacking Chrna7 demonstrate increased insulin signaling in the NAc, which may modulate nicotine place preference. Our studies provide novel targets for future work on development of more effective therapeutic approaches to counteract the rewarding properties of nicotine for smoking cessation.

Age dependence of lung mesenchymal stromal cell dynamics following pneumonectomy

Aging is a critical determinant of regenerative capacity in many organ systems, but it remains unresolved in the lung. This study examines murine lung cell dynamics during age-dependent lung regeneration. Proliferation of lung progenitor cells (EpCAM(neg)/Sca-1(high) lung mesenchymal stromal cells - LMSCs, EpCAM(pos)/Sca-1(low) epithelial progenitor cells, proSP-C(pos) alveolar type II epithelial cells - AECII, and CD31(pos) - endothelial cells) was tracked to day 3 or 7 after pneumonectomy (PNX) or SHAM surgery in 3, 9, and 17 month mice. In 3 month mice, post-PNX LMSC proliferation peaked early (3 days), with 50%-80% more BrdU-positive cells than the other cell types, which peaked later (4-7 days). In older mice (9 and 17 month), abundance and post-PNX proliferation of LMSCs at day 3 were reduced (40%-80%). In both young and old mice, LMSCs were isolated and compared phenotypically with whole lung non-LMSCs. Donor age had no qualitative effect on the phenotype (LMSC vs. non-LMSC), with increased expression of CD90/Thy1, CD105/Eng, CD106/Vcam, CD146/Mcam, and Pdgfrα, and up-regulation of mRNA encoding Fap, Eln, Col1a1, Col3a1, Aldh1a3, Arhgef25, Dner, Fgfr1, and Midkine. However, compared with LMSCs isolated from young mice, LMSCs from older mice exhibited reduced mRNA expression of retinoic acid (Aldh1a3, Rbp4), Fgf/Wnt (Fgfr1, Sfrp1, Wnt2, and Ctnnb1), and elastogenesis (Col1a1, Eln, Fbn1, and Sdc2) pathway genes. Isolated LMSCs from older mice also demonstrated lower colony-forming units (-67%), growth potential (-60% by day 7), ALDH activity (-49%), and telomerase activity (-47%). Therefore, age is associated with declining proliferative potential and regenerative functions of LMSCs in the lung.

Using expression genetics to study the neurobiology of ethanol and alcoholism

Recent simultaneous progress in human and animal model genetics and the advent of microarray whole genome expression profiling have produced prodigious data sets on genetic loci, potential candidate genes, and differential gene expression related to alcoholism and ethanol behaviors. Validated target genes or gene networks functioning in alcoholism are still of meager proportions. Genetical genomics, which combines genetic analysis of both traditional phenotypes and whole genome expression data, offers a potential methodology for characterizing brain gene networks functioning in alcoholism. This chapter will describe concepts, approaches, and recent findings in the field of genetical genomics as it applies to alcohol research.

Global gene expression patterns in the post-pneumonectomy lung of adult mice

Background

Adult mice have a remarkable capacity to regenerate functional alveoli following either lung resection or injury that exceeds the regenerative capacity observed in larger adult mammals. The molecular basis for this unique capability in mice is largely unknown. We examined the transcriptomic responses to single lung pneumonectomy in adult mice in order to elucidate prospective molecular signaling mechanisms used in this species during lung regeneration.

Methods

Unilateral left pneumonectomy or sham thoracotomy was performed under general anesthesia (n = 8 mice per group for each of the four time points). Total RNA was isolated from the remaining lung tissue at four time points post-surgery (6 hours, 1 day, 3 days, 7 days) and analyzed using microarray technology.

Results

The observed transcriptomic patterns revealed mesenchymal cell signaling, including up-regulation of genes previously associated with activated fibroblasts (Tnfrsf12a, Tnc, Eln, Col3A1), as well as modulation of Igf1-mediated signaling. The data set also revealed early down-regulation of pro-inflammatory cytokine transcripts and up-regulation of genes involved in T cell development/function, but few similarities to transcriptomic patterns observed during embryonic or post-natal lung development. Immunohistochemical analysis suggests that early fibroblast but not myofibroblast proliferation is important during lung regeneration and may explain the preponderance of mesenchymal-associated genes that are over-expressed in this model. This again appears to differ from embryonic alveologenesis.

Conclusion

These data suggest that modulation of mesenchymal cell transcriptome patterns and proliferation of S100A4 positive mesenchymal cells, as well as modulation of pro-inflammatory transcriptome patterns, are important during post-pneumonectomy lung regeneration in adult mice.

Microarray analysis of ethanol-induced changes in gene expression

DNA microarray studies offer a robust method for nonbiased analysis of whole genome messenger ribonucleic acid expression patterns. A growing number of studies have applied this experimental approach to studies on ethanol either in cell culture of animal models of ethanol exposure or self-administration. Expression profiling has identified novel gene networks responding to ethanol or differing across animal strains with differing responses to ethanol. Recent studies have shown benefit for meta-analysis of microarray data across different laboratories. Gene network analysis offers unique opportunities for understanding the molecular mechanisms of ethanol responses, toxicity and addiction. Eventually, such work may generate novel targets for future pharmacotherapy. To fully capitalize on the prom ise alluded to above, particularly in regard to meta-analysis of microarray data, it is critical that high quality standards are followed in the generation and analysis of microarray studies. This chapter will discuss experience of our laboratory in performing and analyzing microarray studies on ethanol, focusing discussion mainly on short oligonucleotide microarrays (Affymetrix). However, the general principals of technique and analysis that are discussed have broad applicability to other types of microarray platforms and experimental designs.

Transcriptional profiling of C57 and DBA strains of mice in the absence and presence of morphine

Background

The mouse C57BL/6 (C57) and DBA/2J (DBA) inbred strains differ substantially in many aspects of their response to drugs of abuse. The development of microarray analyses represents a genome-wide method for measuring differences across strains, focusing on expression differences. In the current study, we carried out microarray analysis in C57 and DBA mice in the nucleus accumbens of drug-naïve and morphine-treated animals.

Results

We identified mRNAs with altered expression between the two strains. We validated the mRNA expression changes of several such mRNAs, including Gnb1, which has been observed to be regulated by several drugs of abuse. In addition, we validated alterations in the enzyme activity of one mRNA product, catechol-O-methyltransferase (Comt). Data mining of expression and behavioral data indicates that both Gnb1 and Comt expression correlate with aspects of drug response in C57/DBA recombinant inbred strains. Pathway analysis was carried out to identify pathways showing significant alterations as a result of treatment and/or due to strain differences. These analyses identified axon guidance genes, particularly the semaphorins, as showing altered expression in the presence of morphine, and plasticity genes as showing altered expression across strains. Pathway analysis of genes showing strain by treatment interaction suggest that the phosphatidylinositol signaling pathway may represent an important difference between the strains as related to morphine exposure.

Conclusion

mRNAs with differing expression between the two strains could potentially contribute to strain-specific responses to drugs of abuse. One such mRNA is Comt and we hypothesize that altered expression of Comt may represent a potential mechanism for regulating the effect of, and response to, multiple substances of abuse. Similarly, a role for Gnb1 in responses to multiple drugs of abuse is supported by expression data from our study and from other studies. Finally, the data support a role for semaphorin signaling in morphine effects, and indicate that altered expression of genes involved in phosphatidylinositol signaling and plasticity might also affect the altered drug responses in the two strains.

RhoA, encoding a Rho GTPase, is associated with smoking initiation

We used microarray analysis of acute nicotine responses in mouse brain to choose rationale candidates for human association studies on tobacco smoking and nicotine dependence (ND). Microarray studies on the time-course of acute response to nicotine in mouse brain identified 95 genes regulated in ventral tegmental area. Among these, 30 genes were part of a gene network, with functions relevant to neural plasticity. On this basis and their known roles in drug abuse or synaptic plasticity, we chose the genes RhoA and Ywhag as candidates for human association studies. A synteny search identified human orthologs and we investigated their role in tobacco smoking and ND in a human case-control association study. We genotyped five and three single nucleotide polymorphisms from the RhoA and Ywhag genes, respectively. Both single marker and haplotype analyses were negative for the Ywhag gene. For the RhoA gene, rs2878298 showed highly significant genotypic association with both smoking initiation (SI) and ND (P = 0.00005 for SI and P = 0.0007 for ND). In the allelic analyses, rs2878298 was only significant for SI. In the multimarker haplotype analyses, significant association with SI was found for the RhoA gene (empirical global P values ranged from 9 x 10(-5) to 10(-5)). In all multimarker combinations analyzed, with or without inclusion of the single most significant marker rs2878298, identical risk and protective haplotypes were identified. Our results indicated that the RhoA gene is likely involved in initiation of tobacco smoking and ND. Replication and future model system studies will be needed to validate the role of RhoA gene in SI and ND.

DNA microarray and proteomic strategies for understanding alcohol action

This article summarizes the proceedings of a symposium presented at the 2005 annual meeting of the Research Society on Alcoholism in Santa Barbara, California. The organizer was James M. Sikela, and he and Michael F. Miles were chairs. The presentations were (1) Genomewide Surveys of Gene Copy Number Variation in Human and Mouse: Implications for the Genetics of Alcohol Action, by James M. Sikela; (2) Regional Differences in the Regulation of Brain Gene Expression: Relevance to the Detection of Genes Associated with Alcohol-Related Traits, by Robert Hitzemann; (3) Identification of Ethanol Quantitative Trait Loci Candidate Genes by Expression Profiling in Inbred Long Sleep/Inbred Short Sleep Congenic Mice, by Robnet T. Kerns; and (4) Quantitative Proteomic Analysis of AC7-Modified Mice, by Kathleen J. Grant.

SScore: an R package for detecting differential gene expression without gene expression summaries

SUMMARY

SScore is an R package that facilitates the comparison of gene expression between Affymetrix GeneChips using the S-score algorithm. The S-score algorithm uses probe level data directly to assess differences in gene expression, without requiring a preliminary separate step of probe set expression summary estimation. Therefore, the algorithm avoids introduction of error associated with the expression summary estimation process and has been demonstrated to improve the accuracy of identifying differentially expressed genes. The S-score produces accurate results even when few or no replicates are available.

AVAILABILITY

The R package SScore is available from Bioconductor at http://www.bioconductor.org

Empirical validation of the S-Score algorithm in the analysis of gene expression data

Background

Current methods of analyzing Affymetrix GeneChip^® microarray data require the estimation of probe set expression summaries, followed by application of statistical tests to determine which genes are differentially expressed. The S-Score algorithm described by Zhang and colleagues is an alternative method that allows tests of hypotheses directly from probe level data. It is based on an error model in which the detected signal is proportional to the probe pair signal for highly expressed genes, but approaches a background level (rather than 0) for genes with low levels of expression. This model is used to calculate relative change in probe pair intensities that converts probe signals into multiple measurements with equalized errors, which are summed over a probe set to form the S-Score. Assuming no expression differences between chips, the S-Score follows a standard normal distribution, allowing direct tests of hypotheses to be made. Using spike-in and dilution datasets, we validated the S-Score method against comparisons of gene expression utilizing the more recently developed methods RMA, dChip, and MAS5.

Results

The S-score showed excellent sensitivity and specificity in detecting low-level gene expression changes. Rank ordering of S-Score values more accurately reflected known fold-change values compared to other algorithms.

Conclusion

The S-score method, utilizing probe level data directly, offers significant advantages over comparisons using only probe set expression summaries.

Tumor-derived p53 mutants induce NF-kappaB2 gene expression

Overexpression of mutant p53 is a common theme in tumors, suggesting a selective pressure for p53 mutation in cancer development and progression. To determine how mutant p53 expression may lead to survival advantage in human cancer cells, we generated stable cell lines expressing p53 mutants p53-R175H, -R273H, and -D281G by use of p53-null human H1299 (lung carcinoma) cells. Compared to vector-transfected cells, H1299 cells expressing mutant p53 showed a survival advantage when treated with etoposide, a common chemotherapeutic agent; however, cells expressing the transactivation-deficient triple mutant p53-D281G (L22Q/W23S) had significantly lower resistance to etoposide. Gene expression profiling of cells expressing transcriptionally active mutant p53 proteins revealed the striking pattern that all three p53 mutants induced expression of approximately 100 genes involved in cell growth, survival, and adhesion. The gene NF-kappaB2 is a prominent member of this group, whose overexpression in H1299 cells also leads to chemoresistance. Treatment of H1299 cells expressing p53-R175H with small interfering RNA specific for NF-kappaB2 made these cells more sensitive to etoposide. We have also observed activation of the NF-kappaB2 pathway in mutant p53-expressing cells. Thus, one possible pathway through which mutants of p53 may induce loss of drug sensitivity is via the NF-kappaB2 pathway.

The search for candidate genes of alcoholism: evidence from expression profiling studies

Alcoholism is the outcome of complex interactions between the environment and multiple gene loci, which may encode pre-existing susceptibility, or contribute to the neuroadaptations underlying the process of developing dependence. Because of this, the prospect of simultaneous, genome wide, high-throughput analysis of gene expression allowed by microarray technology has met with great expectations. The hope has been that new insights into pathogenesis of substance disorders will rapidly be gained, leading to identification of novel treatment targets. The usefulness of this approach as a discovery tool in addiction research will be critically reviewed here. In this article, we describe the evolution of our experimental approaches, from first generation Affymetrix expression arrays to present high-density arrays, and from the use of original Affymetrix software to more advanced analysis of the probe signal, and different statistical approaches to creating candidate gene lists. Further, we address some methodological issues critical to the study of brain samples by microarray technology. We also summarize findings from several expression profiling experiments involving different animal models of alcoholism. The accumulation of expression data from different animal models allows mining the database for patterns of overlap. Such second level analysis depends on the generation of uniform and reliable datasets.

Using in vitro models for expression profiling studies on ethanol and drugs of abuse

The use of expression profiling with microarrays offers great potential for studying the mechanisms of action of drugs of abuse. Studies with the intact nervous system seem likely to be most relevant to understanding the mechanisms of drug abuse-related behaviours. However, the use of expression profiling with in vitro culture models offers significant advantages for identifying details of cellular signalling actions and toxicity for drugs of abuse. This study discusses general issues of the use of microarrays and cell culture models for studies on drugs of abuse. Specific results from existing studies are also discussed, providing clear examples of relevance for in vitro studies on ethanol, nicotine, opiates, cannabinoids and hallucinogens such as LSD. In addition to providing details on signalling mechanisms relevant to the neurobiology of drugs of abuse, microarray studies on a variety of cell culture systems have also provided important information on mechanisms of cellular/organ toxicity with drugs of abuse. Efforts to integrate genomic studies on drugs of abuse with both in vivo and in vitro models offer the potential for novel mechanistic rigor and physiological relevance.

Alcohol Effects on Central Nervous System Gene Expression in Genetic Animal Models

This article summarizes the proceedings of a symposium presented at the 2004 annual meeting of the Research Society on Alcoholism in Vancouver, British Columbia, Canada. The organizers and chairs were William J. McBride and Michael F. Miles. The presentations were (1) Molecular Triangulation on Gene Expression Patterns in Behavioral Responses to Acute Ethanol, by Robnet T. Kerns; (2) Gene Expression in Limbic Regions After Ethanol Self-Infusion Into the Posterior Ventral Tegmental Area, by Zachary A. Rodd; (3) Microarray Analysis of CNS Limbic Regions of Inbred Alcohol-Preferring and -Nonpreferring rats and Effects of Alcohol Drinking, by Wendy N. Strother and Howard J. Edenberg; and (4) Microarray Analysis of Mouse Lines Selected for Chronic Ethanol Withdrawal Severity: The Convergence of Basal, Ethanol Regulated, and Proximity to Ethanol Quantitative Trait Loci to Identify Candidate Genes, by Joel G. Hashimoto and Kristine M. Wiren.

Modulation of gene expression by tumor-derived p53 mutants

p53 mutants with a single amino acid substitution are overexpressed in a majority of human cancers containing a p53 mutation. Overexpression of the mutant protein suggests that there is a selection pressure on the cell indicative of an active functional role for mutant p53. Indeed, H1299 cells expressing mutant p53-R175H, p53-R273H or p53-D281G grow at a faster rate compared with a control cell line. Using p53-specific small interfering RNA, we show that the growth rate of mutant p53-expressing cells decreases as mutant p53 level decreases, demonstrating that the increased cellular growth is dependent on p53 expression. Increased growth rate is not observed for H1299 cell clones expressing mutant p53-D281G (L22Q/W23S), which has been shown to be defective in transactivation in transient transcriptional assays. This shows that the increased growth rate imparted by mutant p53 in H1299 cells requires the transactivation function of mutant p53. By performing microarray hybridization analyses, we show that constitutive expression of three common p53 mutants (p53-R175H, p53-R273H, and p53-D281G) in H1299 human lung carcinoma cells evokes regulation of a common set of genes, a significant number of which are involved in cell growth regulation. Predictably, H1299 cells expressing p53-D281G (L22Q/W23S) are defective in up-regulating a number of these genes. The differences in expression profiles induced by individual p53 mutants in the cells may be representative of the p53 mutants and how they can affect gene expression resulting in the observed "gain of function" phenotypes (i.e., increased growth rate, decreased sensitivity to chemotherapeutic agents, and so forth).

Evaluation of gene expression profile of keratinocytes in response to JP-8 jet fuel

The skin is the principal barrier against any environmental insult. Therefore, there is a high risk for a large number of military and civilian personnel exposed to jet fuel JP-8 to suffer percutaneous absorption of this fuel. This paper reports the use of cDNA microarray to identify the gene expression profile in normal human epidermal keratinocytes exposed to JP-8 for 24-h and 7-day periods. The effects of JP-8 exposure on keratinocytes at these two different periods induced a set of genes with altered expression in response to this type of insult. Microarray data were visualized using a novel algorithm based on simple statistical analyses to reduce data dimensionality and identify subsets of discriminant genes. Predictive neural networks were built using a multiplayer perceptron to carry out a proper classification task in microarray data in the untreated versus JP-8-treated samples. The pattern of expressions in response to JP-8 provides evidences that detoxificant-related and cell growth regulator genes with the most variability in the level of expression may be useful genetic markers in adverse health effects of personnel exposed to JP-8. The approaches in our analysis provide a simple, safe, novel, and effective method that is reliable in identifying and analyzing gene expression in samples treated with JP-8 or over potential toxic agents. Gene expression data from these studies can be used to build accurate predictive models that separate different molecular profiles. The data establish the use and effectiveness of these approaches for future prospective studies.

Tumor-derived p53 mutants induce oncogenesis by transactivating growth-promoting genes

We have studied the mechanism of mutant p53-mediated oncogenesis using several tumor-derived mutants. Using a colony formation assay, we found that the majority of the mutants increased the number of colonies formed compared to the vector. Expression of tumor-derived p53 mutants increases the rate of cell growth, suggesting that the p53 mutants have 'gain of function' properties. We have studied the gene expression profile of cells expressing tumor-derived p53-D281G to identify genes transactivated by mutant p53. We report the transactivation of two genes, asparagine synthetase and human telomerase reverse transcriptase. Quantitative real-time PCR confirms this upregulation. Transient transfection promoter assays verify that tumor-derived p53 mutants transactivate these promoters significantly. An electrophoretic mobility shift assay shows that tumor-derived p53-mutants cannot bind to the wild-type p53 consensus sequence. The results presented here provide some evidence of a possible mechanism for mutant p53-mediated transactivation.