The Pharmacogenomics Journal (2012) 12, 328–341 & 2012 Macmillan Publishers Limited. All rights reserved 1470-269X/12 www.nature.com/tpj ORIGINAL ARTICLE Interaction networks of lithium and valproate molecular targets reveal a striking enrichment of apoptosis functional clusters and neurotrophin signaling A Gupta1, TG Schulze2, The overall neurobiological mechanisms by which lithium and valproate 3 1 stabilize mood in bipolar disorder patients have yet to be fully defined. The V Nagarajan , N Akula , therapeutic efficacy and dissimilar chemical structures of these medications 1 1 W Corona , X-y Jiang , suggest that they perturb both shared and disparate cellular processes. To N Hunter1, FJ McMahon1 and investigate key pathways and functional clusters involved in the global action SD Detera-Wadleigh1 of lithium and valproate, we generated interaction networks formed by well- supported drug targets. Striking functional similarities emerged. Intersecting 1Human Genetics Branch, National Institute of nodes in lithium and valproate networks highlighted a strong enrichment Mental Health, National Institutes of Health, of apoptosis clusters and neurotrophin signaling. Other enriched pathways 2 Bethesda, MD, USA; Department of Psychiatry included MAPK, ErbB, insulin, VEGF, Wnt and long-term potentiation and Psychotherapy, University Medical Center, Georg-August-Universita¨t, Go¨ttingen, Germany indicating a widespread effect of both drugs on diverse signaling systems. and 3Bioinformatics and Computational MAPK1/3 and AKT1/2 were the most preponderant nodes across pathways Biosciences Branch, Office of Cyber Infrastructure suggesting a central role in mediating pathway interactions. The conver- and Computational Biology, National Institute of gence of biological responses unveils a functional signature for lithium and Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA valproate that could be key modulators of their therapeutic efficacy. The Pharmacogenomics Journal (2012) 12, 328–341; doi:10.1038/tpj.2011.9; Correspondence: published online 8 March 2011 Dr SD Detera-Wadleigh, Human Genetics Branch, National Institute of Mental Health, Keywords: expression; pathways; MiMI; DAVID; GLay National Institutes of Health, 35/1A205, 35 Convent Drive, Bethesda, MD 20892 USA. E-mail: [email protected] Introduction Lithium and valproate salts are popularly prescribed mood stabilizers for patients with recurrent episodes of mania and depression. The therapeutic effect of lithium was serendipitously discovered more than half a century ago,1 in later years valproate was also shown to be effective in treating mania.2 Such a similar efficacy by agents of dissimilar chemical structures—lithium, an alkali and valproate, a 5-carbon carboxylate—raises the possibility of both shared and distinct mechanisms of action that remain to be fully defined. Two major mechanisms have been invoked to explain the therapeutic action of lithium. The inositol depletion hypothesis that assigns a major role to the phosphatidylinositol cycle is based on the demonstration that lithium disrupts the activity of enzymes involved in inositol turnover.3 Later studies showed that lithium also inhibited GSK3b, thereby implicating the Wnt signaling pathway.4 Received 27 November 2010; revised 23 January 2011; accepted 30 January 2011; However, these may be just two of the many lithium-responsive biological published online 8 March 2011 processes. Lithium and valproate highlight apoptosis and neurotrophin signaling A Gupta et al 329 Studies have documented diverse cellular effects of Assembly of lithium-responsive molecules lithium in organisms spanning the evolutionary After identifying a series of publications that found lithium- scale. Lithium has been shown to extend the life of the induced expression changes in various species, we utilized a nematode, Caenorhabditis elegans,5 shift the development multi-pass system to extract as much relevant information of the sea urchin, Strongylocentrotus purpuratus, toward as possible from each study’s data set. The end goal was to vegetalization,6 inhibit fermentation in the yeast convert each study’s data into mouse gene homologs for Saccharomyces cerevisiae grown in galactose,7 and disrupt cell direct comparison. Accession numbers cited in various fate determination in the social amoeba, Dictyostelium studies and relevant databases including eukaryotic orthology discoideum.4 In various vertebrate model systems lithium (YOGY) (http://www.sanger.ac.uk/PostGenomics/S_pombe/YOGY has been shown to influence multiple processes including /index.shtml) and the Sea Urchin Genome Project (http:// cell survival, neuroprotection, neurogenesis and inhibition annotation.hgsc.bcm.tmc.edu/Urchin/cgi-bin/pubLogin.cgi) were of apoptosis.8,9 used. A Uniprot ID was established for each differentially Valproic acid, an anticonvulsant, found its use in treating regulated transcript and BLAST was performed against the bipolar disorder in the 1980s.2 The therapeutic efficacy mousegenome(http://www.ncbi.nlm.nih.gov/blast/Blast.cgi). and side effects of valproate and lithium have been Only e-values o1.00E-5 were considered. investigated. Valproate has been shown to inhibit histone Once a gene name was identified for a transcript, NCBI’s deacetylases (HDACs),10,11 which may lead to activation Homologene was queried (http://www.ncbi.nlm.nih.gov/ of repressed genes. Studies have shown that valproate homologene). Using a universal point of reference helped promotes neurogenesis,12,13 and in cortical neurons, to minimize confusion regarding alternate gene names. If this effect is mediated through a MAPK-dependent path- there was no match in Homologene or by protein BLAST, the way.12 data point was not considered. A list of mouse homologs was A unitary picture has yet to emerge from published studies then generated, and compared against all other studies for on the global effects of lithium and valproate. We explored replication. the molecular repertoire that underpins the overall cellular responses to these medications, by selecting well-documen- ted targets that showed consistent differential expression or Classic targets of lithium activity in at least two studies, and using these to build To populate the network seed, priority was assigned to well- interaction networks. To identify significantly enriched documented targets of lithium: GSK3B, IMPA1, AKT1, BCL2 functional clusters and signaling pathways recruited by and BDNF (Table 1). Inhibition of GSK3B that leads to the these networks, functional annotation of nodes was per- activation of Wnt signaling is one of the best-supported formed. This approach generated drug-responsive inter- cellular responses to lithium.4,32 A recent report has shown a action networks that integrated diverse pathway interac- dose-dependent decrease in Gsk3b mRNA in rat hippocam- tions and highlighted apoptosis-related functional clusters pal cultures.33 and multiple pathways, foremost of which was neurotro- IMPase catalyzes the release of free inositol and is phin signaling. inhibited by lithium.3,43 Lithium-induced overexpression of Impa1 has been demonstrated in yeast,7 in rat brain34 and mouse hippocampus35 (Table 1). Materials and methods Lithium-induced activation of Akt1 accompanied by neuroprotection has been demonstrated in cerebellar gran- Drug target selection: basic principles ule cells14 (Table 1). Akt1 activation has been shown also in We attempted to capture all molecules that showed rat brain15 and renal epithelial cells deprived of growth consistent lithium- and valproate-induced changes in factor.16 Lithium elicited opposite effects on AKT1 protein expression and/or enzyme activity in various studies, concentration in Huh7 and Hep40 cells.17 Chronic lithium including those directed at the transcriptome or proteome, treatment of bovine adrenal chromaffin cells inhibited and those that focused on specific molecules in various cells Gsk3b activity and reduced Akt1 expression.18 and organisms. We found that the majority of genes across Potent upregulation of the anti-apoptotic gene, Bcl2,by these studies showed inconsistent expression patterns, lithium has been well documented,44 as illustrated by particularly in microarray-based genome-wide profiling. studies in rat cerebellar cells,23 rat frontal cortex,24 mouse Experimental differences across studies that could influence retinal ganglion cells25and chick cochlear magnocellularis the level and pattern of gene expression were apparent. neurons26 (Table 1). These include the use of various organisms and cells, Lithium stimulated the expression of Bdnf, a neurogenesis variability in methods of drug administration, doses and and neuroprotective factor, in rat frontal cortex,27,28 lengths of drug exposure, ages of animals, sacrifice techni- hippocampus29 and in cortical neurons via activation of ques, tissues and brain regions isolated for analysis, micro- the Bdnf promoter IV,30 but failed to elicit neuroprotection array chip batches and other divergences in experimental in Bdnf deficient mice challenged with glutamate.45 Incon- design. It was therefore important to establish a stringent sistent direction of Bdnf brain expression has been reported inclusion threshold for drug-regulated molecules in order to in lithium-treated FSL and FRL rats.31 The variable effects of create a robust seed for interaction networks. lithium on AKT1, BDNF and IMPA1, and possibly other The Pharmacogenomics Journal Lithium and valproate highlight apoptosis