Changes in Brain MicroRNAs Contribute to Cholinergic Stress Reactions The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. Citation Meerson, Ari, Luisa Cacheaux, Ki Ann Goosens, Robert M. Sapolsky, Hermona Soreq, and Daniela Kaufer. “Changes in Brain MicroRNAs Contribute to Cholinergic Stress Reactions.” Journal of Molecular Neuroscience 40, no. 1 2 (January 27, 2010): 47-55. As Published http://dx.doi.org/10.1007/s12031-009-9252-1 Publisher Springer-Verlag Version Final published version Citable link http://hdl.handle.net/1721.1/82659 Detailed Terms http://creativecommons.org/licenses/by/3.0/ J Mol Neurosci (2010) 40:47–55 DOI 10.1007/s12031-009-9252-1 Changes in Brain MicroRNAs Contribute to Cholinergic Stress Reactions Ari Meerson & Luisa Cacheaux & Ki Ann Goosens & Robert M. Sapolsky & Hermona Soreq & Daniela Kaufer Received: 6 July 2009 /Accepted: 20 July 2009 /Published online: 27 August 2009 # The Author(s) 2009. This article is published with open access at Springerlink.com Abstract Mental stress modifies both cholinergic neuro- SC35. Stress was previously shown to upregulate SC35, transmission and alternative splicing in the brain, via which promotes the alternative splicing of acetylcholines- incompletely understood mechanisms. Here, we report that terase (AChE) from the synapse-associated isoform stress changes brain microRNA (miR) expression and that AChE-S to the, normally rare, soluble AChE-R protein. some of these stress-regulated miRs regulate alternative Knockdown of miR-183 expression increased SC35 splicing. Acute and chronic immobilization stress differen- protein levels in vitro, whereas overexpression of miR- tially altered the expression of numerous miRs in two 183 reduced SC35 protein levels, suggesting a physiolog- stress-responsive regions of the rat brain, the hippocampal ical role for miR-183 regulation under stress. We show CA1 region and the central nucleus of the amygdala. miR- stress-induced changes in miR-183 and miR-134 and 134 and miR-183 levels both increased in the amygdala suggest that, by regulating splicing factors and their following acute stress, compared to unstressed controls. targets, these changes modify both alternative splicing Chronic stress decreased miR-134 levels, whereas miR-183 and cholinergic neurotransmission in the stressed brain. remained unchanged in both the amygdala and CA1. Importantly, miR-134 and miR-183 share a common Keywords Stress . microRNA . miR-183 . miR-134 . SC35 . predicted mRNA target, encoding the splicing factor Cholinergic Proceedings of the XIII International Symposium on Cholinergic Mechanisms A. Meerson H. Soreq (*) Department of Biological Chemistry, Department of Biological Chemistry The Hebrew University of Jerusalem, and Interdisciplinary Center of Neural Computation, 91904 Jerusalem, Israel The Hebrew University of Jerusalem, 91904 Jerusalem, Israel L. Cacheaux e-mail: [email protected] Helen Wills Neuroscience Institute, UC Berkeley, 3140 VLSB, Berkeley, CA 94720-3140, USA D. Kaufer (*) Department of Integrative Biology, K. A. Goosens University of California, Berkeley, McGovern Institute for Brain Research, Berkeley, CA 94720-3140, USA Department of Brain and Cognitive Sciences, e-mail: [email protected] MIT Building, 46-2171B, Cambridge, MA, USA D. Kaufer R. M. Sapolsky Helen Wills Neuroscience Institute Department of Biological Sciences, Stanford School of Medicine, and the Department of Integrative Biology, Stanford University, UC Berkeley, 3140 VLSB, Stanford, CA, USA Berkeley, CA 94720-3140, USA 48 J Mol Neurosci (2010) 40:47–55 Introduction miR-18 and miR-124 were recently reported to regulate glucocorticoid receptors, suggesting involvement in a variety Mammalian psychological stress is known to induce of systemic stress responses (Vreugdenhil et al. 2009). prominent changes in neuronal activity and gene regulation However, the mechanistic involvement of miRs at large in across multiple brain regions (McEwen 2007). Acute and psychological stress remains unknown. We predicted that chronic stress both lead to the remodeling of dendrites in miR-mediated regulation contributes to the yet incompletely the hippocampus (McEwen 1999), which controls learning understood link(s) between the molecular and the physio- functions via establishing spatial, episodic, and contextual logical reactions of particular brain regions to psychological memory formation (Kenney and Gould 2008; Otto and stress and more specifically to the regulation of alternative Eichenbaum 1992). Specifically, in the CA1 region of the splicing. To test for miR involvement in governing the hippocampus, both neurons and glia are affected by mental region-specific stress-induced changes in alternative splicing, stress (Espinosa-Oliva et al. 2009; Hirata et al. 2009). In the we studied the expression profiles of miRs in the hippocam- amygdala, stress reactions impact emotion, addiction, fear, pal CA1 and the central amygdala brain regions in stressed and anxiety (Vyas et al. 2002; Walker and Davis 2008). and control rats and challenged the relevance of observed Chronic stress also increases aggression, likely to reflect changes by manipulating miR-183 levels in cultured cells. hyperactivity of the amygdala (Wood et al. 2003). At the physiological level, cholinergic neurotransmission is altered under stress (Kaufer et al. 1998; Soreq and Seidman 2001). Materials and Methods These stress-induced changes are mostly attributed to the combinatorial regulation of many genes’ altered transcrip- Immobilization Stress Adult male rats (Charles River; tion levels (Anguelova et al. 2000). However, the contri- 200–225 g) were subjected to a single 4-h session of bution of posttranscriptional regulation mechanisms to immobilization stress (acute stress group), 4 h of complete these stress-associated responses is increasingly acknowl- immobilization stress per day for 14 days (chronic stress edged (Battaglia and Ogliari 2005; Gattoni et al. 1996; group), or brief daily handling (no-stress group). Twenty- Meshorer et al. 2005). One such mechanism is the production four hours after the last stress or handling session, the of multiple proteins with divergent or even opposite animals were sacrificed, and tissues were carefully functions from a single transcript via alternative splicing, dissected, flash-frozen, and stored at −80°C. RNA was which accounts for much of the proteome's flexibility then purified using the mirVana kit (Ambion), which (Stamm et al. 2005). A case in point is the stress-induced preserves short RNAs, from the dissected central amyg- alternative splicing of the primary ACh-hydrolyzing enzyme dala and hippocampus CA1 region. For each brain region acetylcholinesterase (AChE), which modifies cholinergic in each of the three groups, pooled samples were neurotransmission (Meshorer and Soreq 2006) and affects generated from the RNA of three to four rats. neuronal processes in a pathway involving the splicing factors SC35 and ASF/SF2 (Meshorer et al. 2005). However, Spotted Array Methods Spotted array methods were adap- the molecular mechanism(s) underlying the apparent rela- ted from Ben-Ari et al. (2006). The mirVana oligo set tionship between stress, alternative splicing, and cholinergic (Ambion, Austin, TX, USA; Cat. Num. 1564V1) was used neurotransmission remain incompletely understood. to construct an in-house array with >200 spotted probes An important posttranscriptional mechanism for gene complementary to known human and mouse miRs. Dye- regulation involves microRNAs (miRs). miRs are 20–28- swapping tests served to exclude dye-specific labeling nucleotide noncoding RNAs encoded in the genomes of differences (Dombkowski et al. 2004). Labeling used the plants and animals, exerting translational repression and/or CyDye reactive dye pack (Amersham, NSW, Australia), as degradation of target mRNAs via complementary binding instructed 2 dye-swapped arrays were used for every to the 3′ untranslated region (UTR; Maniataki and comparison, each with at least 6 replicate spots per miRNA Mourelatos 2005). miR levels vary across cell type, tissue, probe. Hybridization was performed in chambers (Corning, and developmental stages (Baek et al. 2008; Liang et al. NY, USA) for 15 h at 64°C. Scanning used an Affymetrix 2007; Plasterk 2006; Sood et al. 2006). Each miR targets 428 Array Scanner at 532 and 650 nm, controlled by the multiple mRNAs (Krek et al. 2005; Lewis et al. 2005), “Jaguar” software (Affymetrix, CA, USA), and results were which in some cases code for proteins participating in the exported to the “Imagene” program (BioDiscovery Inc., same signaling pathway (Li et al. 2007). Thus, both the CA, USA) for quantification. Data normalization, exclusion overall profile of miR expression and the expression levels of unreliable spots, and combination of the information of particular key miRs impact diverse biological processes from all slides were performed using the Normalize Suite (Chen et al. 2004; He et al. 2005; Kluiver et al. 2007; (Beheshti et al. 2003). Significantly altered transcripts (with Laneve et al. 2007; Li et al. 2007; Thai et al. 2007). a p value of the sign-test smaller than 0.05), which were not J Mol Neurosci (2010) 40:47–55 49 disqualified due to any quality parameter, were identified ton, DE, USA). Reverse transcription (RT) of miRs was by setting an arbitrary threshold. performed using SuperScript III First-Strand Synthesis Systems kit reagents for RT-polymerase chain reaction Cell