Meta-Type Analysis of Dopaminergic Effects on Gene Expression in the Neuroendocrine Brain of Female Goldfish
Total Page:16
File Type:pdf, Size:1020Kb
ORIGINAL RESEARCH ARTICLE published: 02 November 2012 doi: 10.3389/fendo.2012.00130 Meta-type analysis of dopaminergic effects on gene expression in the neuroendocrine brain of female goldfish JasonT. Popesku 1*†, Christopher J. Martyniuk 2 and Vance L.Trudeau 1* 1 Centre for Advanced Research in Environmental Genomics, Department of Biology, University of Ottawa, Ottawa, ON, Canada 2 Canadian Rivers Institute and Department of Biology, University of New Brunswick, Saint John, NB, Canada Edited by: Dopamine (DA) is a major neurotransmitter important for neuroendocrine control and recent Wei Ge, The Chinese University of studies have described genomic signaling pathways activated and inhibited by DA agonists Hong Kong, China and antagonists in the goldfish brain. Here we perform a meta-type analysis using microar- Reviewed by: José A. Muñoz-Cueto, University of ray datasets from experiments conducted with female goldfish to characterize the gene Cadiz, Spain expression responses that underlie dopaminergic signaling. Sexually mature, pre-spawning Anderson O. Wong, The University of [gonadosomatic index (GSI) D 4.5 ± 1.3%] or sexually regressing (GSI D 3 ± 0.4%) female Hong Kong, Hong Kong goldfish (15–40 g) injected intraperitoneally with either SKF 38393, LY 171555, SCH 23390, *Correspondence: sulpiride, or a combination of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and a-methyl-p- Jason T. Popesku, Centre for Advanced Research in Environmental tyrosine. Microarray meta-type analysis identified 268 genes in the telencephalon and hypo- Genomics, Department of Biology, thalamus as having reciprocal (i.e., opposite between agonism and antagonism/depletion) University of Ottawa, Ottawa, ON, fold change responses, suggesting that these transcripts are likely targets for DA-mediated Canada K1N 6N5. regulation. Noteworthy genes included ependymin, vimentin, and aromatase, genes that e-mail: [email protected]; Vance L. Trudeau, Department of support the significance of DA in neuronal plasticity and tissue remodeling. Sub-network Biology, University of Ottawa, Room enrichment analysis (SNEA) was used to identify common gene regulators and binding 160, Gendron Hall, 30 Marie Curie, proteins associated with the differentially expressed genes mediated by DA. SNEA analy- Ottawa, ON, Canada K1N 6N5. sis identified gene expression targets that were related to three major categories that e-mail: [email protected] included cell signaling (STAT3, SP1, SMAD, Jun/Fos), immune response (IL-6, IL-1b, TNFs, †Present address: Jason T. Popesku, Department of cytokine, NF-kB), and cell proliferation and growth (IGF1, TGFb1). These gene networks Cellular and Physiological Sciences, are also known to be associated with neurodegenerative disorders such as Parkinsons’ Life Sciences Institute, University of disease, well-known to be associated with loss of dopaminergic neurons. This study iden- British Columbia, Vancouver, BC, tifies genes and networks that underlie DA signaling in the vertebrate CNS and provides Canada V6T 1Z3. targets that may be key neuroendocrine regulators. The results provide a foundation for future work on dopaminergic regulation of gene expression in fish model systems. Keyword: dopamine; sub-network enrichment analysis; neurodegeneration; reproduction; immune response INTRODUCTION expression in the neuroendocrine brain (Popesku et al., 2011a). Dopamine (DA) is a neurotransmitter important in disorders such Additionally, we have previously described the effects of a combi- as schizophrenia (Seeman and Kapur, 2000) and Parkinson’s dis- nation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP; ease (Baik et al., 1995), but is also the major neurotransmitter a selective DA neurotoxin) and a-methyl-p-tyrosine (aMPT; a controlling teleost reproduction (reviewed in Dufour et al., 2005; tyrosine hydroxylase inhibitor) on the goldfish hypothalamic tran- Dufour et al., 2010). In this regard, DA inhibits the release of scriptome (Popesku et al., 2008). Using microarray datasets from luteinizing hormone (LH) in fish through multiple mechanisms: two of these experiments, and an additional novel microarray data (a) DA inhibits gonadotropin-releasing hormone (GnRH) release presented here, we further elucidate the mechanistic effects of DA from GnRH neurons through the D1 receptor (Yuand Peter,1992); on gene expression in the neuroendocrine brain by performing a (b) DA directly inhibits LH release from gonadotrophs in the ante- meta-type analysis of these datasets. rior pituitary through the D2 receptor (Peter et al., 1986; Omelja- In transcriptomics, there are a number of bioinformatics niuk et al., 1987); (c) DA decreases the expression of GnRH recep- approaches to globally assess gene expression data and to organize tor mRNA in the pituitary (Kumakura et al., 2003; Levavi-Sivan expression data into a larger biological context. These methods et al., 2004); and (d) DA inhibits the synthesis of GABA (Hibbert include Gene Ontology (GO) characterization, functional enrich- et al., 2004, 2005), an important stimulator of LH release (Mar- ment, and pathway analysis. Many of these approaches have been tyniuk et al., 2007). Furthermore, it is well understood that DA, successfully performed using genomic data in neuroendocrine acting through the D1, stimulates growth hormone in fish (Wong regions of teleost fishes to better describe cellular events that are et al., 1992). Our recent studies using goldfish have investigated mediated by neurotransmitters,hormones,or exogenous neuroac- the effects of DA agonists on the hypothalamic transcriptome and tive agents (Marlatt et al., 2008; Popesku et al., 2008; Zhang et al., proteome (Popesku et al., 2010) or of DA antagonists on gene 2009a; Martyniuk et al., 2010). New bioinformatics tools are now www.frontiersin.org November 2012 | Volume 3 | Article 130 | 1 Popesku et al. Dopaminergic regulation of gene expression available to construct gene networks using gene expression pro- 1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol] or LY filing and have been used successfully in teleost fish (e.g., reverse 171555 [D2 agonist; LY; (−)-Quinpirole hydrochloride] pur- engineering of adverse pathways for ecotoxicology (Perkins et al., chased from Tocris (Ballwin, MO, USA). The experimental design 2011). Sub-network enrichment analysis (SNEA; Ariadne’s Path- and doses chosen were based on Otto et al. (1999) who showed way Studio v7.0 Sivachenko et al., 2007) offers a unique approach rapid effects on goldfish brain somatostatin mRNAs. LY was dis- to protein interaction networks that are described in the litera- solved in physiological saline (0.6% NaCl) to yield a dose of ture as well as a curated mammalian database. Specifically, SNEA 2 mg/g body weight of fish. SKF was first dissolved in a minimal builds sub-networks by mapping experimental data onto known amount of dimethylsulfoxide (DMSO), and subsequently diluted bio-molecular interactions. The interactions include promoter- to 40 mg/g body weight of fish with physiological saline (0.6% binding, protein modification, and common targets of expression. for fish). The final concentration of DMSO was 0.099%; DMSO This algorithm has been used to identify gene sub-networks in up to 0.1% does not affect basal GH or LH levels (Otto et al., breast cancer cell lines (Chuang et al., 2007) and is a useful tool for 1999). While 0.1% DMSO may (Mortensen and Arukwe, 2006) identifying interaction or signaling networks that involve differ- or may not (Nishimura et al., 2008) affect gene expression, all of entially expressed genes. As such, this method can provide insight our gene expression work is relative to control fish which received in gene regulatory pathways. an equivalent amount of DMSO. The fish received two sequential In this study, we identify genes and sub-networks that are likely i.p. injections at 5 mL/g body weight each according to the sched- regulated by DA based on their reciprocal response to DA ago- ule shown in Table 1. The experiment was conducted this way to nism or antagonism/depletion. These data have implications for ensure that all fish received an equivalent volume of vehicle. our understanding of DA action in fish neuroendocrine systems. DOPAMINE ANTAGONIST EXPERIMENT MATERIALS AND METHODS The DA D1-specific antagonist SCH 23390 and DA D2-specific This is a meta-type analysis of published experiments involving antagonist sulpiride were purchased from Tocris (Ballwin, MO, treatments of goldfish with DA agonists (Popesku et al., 2010), USA). The antagonists were first dissolved in a minimal amount antagonists (Popesku et al., 2011a), and after pharmacological of DMSO, and subsequently diluted with 0.6% saline. The final depletion of DA (Popesku et al., 2008). The abbreviated Mate- concentration of DMSO was 0.099%. Sexually regressing (June; rials and Methods pertaining to the experiments are included here GSI D 3 ± 0.4%; n D 18 each) female goldfish received a single for completeness. It should be noted that, while published, the injection at 5 mL/g body weight of either SCH 23390 or sulpiride previous DA depletion studies offered only a cursory analysis of to give a dose of 40 mg/g or 2 mg/g body weight of fish, respectively, the microarray data in the context of neurotransmitter effects on or saline containing an equivalent amount of DMSO. gene expression and did not specifically address global dopaminer- gic control of transcriptional responses. Furthermore, we present DOPAMINE DEPLETION EXPERIMENT novel transcriptomic data for specific DA antagonism for which 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine and a-methyl-p- the physiological response to these antagonists has been published tyrosine (aMPT) were purchased from Sigma-Aldrich (St. Louis, (Popesku et al., 2011a), but for which microarray analysis was MO, USA). Sexually mature (May; GSI D 4.7 ± 0.6%) female gold- not performed at that time. We used this novel dataset to com- fish (n D 5 each) were injected with MPTP (50 mg/g; day 0) and pare these DA antagonism responses to agonist and DA depletion aMPT (240 mg/g; day 5) or saline (control) in order to severely responses to improve identification of DA-regulated transcripts in deplete catecholamines.