Microarray analysis reveals transcriptional plasticity in the reef building coral Acropora millepora LINE K. BAY,*† KARIN E. ULSTRUP,‡ H. BJØRN NIELSEN,§ HANNE JARMER,§ NICOLAS GOFFARD, – BETTE L. WILLIS,*** DAVID J. MILLER*†† and MADELEINE J. H. VAN OPPEN*† *ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Qld 4811, Australia, †Australian Institute of Marine Science, PMB #3, Townsville MC, Qld 4810, Australia, ‡Marine Biological Laboratory, Department of Biology, University of Copenhagen, Strandpromenaden 5, DK-3000 Helsingør, Denmark, §Center for Biological Sequence Analysis, Danish Technical University, Kemitorvet, Building 208, DK-2800 Lyngby, Denmark, –Institut Louis Malarde ´, PO Box 30, Papeete 98713, Tahiti, French Polynesia, **School of Marine and Tropical Biology, James Cook University, Townsville, Qld 4811, Australia, ††ARC Centre for the Molecular Genetics of Development, Research School of Biological Sciences, Australian National University, ACT 2601, Canberra, Australia Abstract We investigated variation in transcript abundance in the scleractinian coral, Acropora millepora, within and between populations characteristically exposed to different turbidity regimes and hence different levels of light and suspended particulate matter. We examined phenotypic plasticity by comparing levels of gene expression between source populations and following 10 days of acclimatization to a laboratory environment. Analyses of variance revealed that 0.05% of genes were differentially expressed between source populations, 1.32% following translocation into a common laboratory and 0.07% in the interaction (source population-dependent responses to translocation). Functional analyses identified an over-representation of differentially expressed genes associated with metabolism and fluorescence categories (primarily downregulated), and environ- mental information processing (primarily upregulated) following translocation to a lower light and turbidity environment. Such metabolic downregulation may indicate non- oxidative stress, hibernation or caloric restriction associated with the changed environ- mental conditions. Green fluorescent protein-related genes were the most differentially expressed and were exclusively downregulated; however, green fluorescent protein levels remained unchanged following translocation. Photophysiological responses of corals from both locations were characterized by a decline when introduced to the common laboratory environment but remained healthy (F v ⁄ F m > 0.6). Declines in total lipid content following translocation were the greatest for inshore corals, suggesting that turbid water corals have a strong reliance on heterotrophic feeding. Keywords: coral reef, green fluorescent protein, metabolism, phenotypic plasticity, transcripto- mics Received 24 February 2009; revision received 20 April 2009; accepted 26 April 2009 Introduction Tropical coral reefs are highly diverse ecosystems with representatives from most phyla and kingdoms, co-exist- ing through a range of biological and ecological interac- tions (Margulis 1993). Coral reefs are also important socio-economically (Moberg & Folke 1999). They generate >US$30 billion annually and contribute significantly to the economies of many tropical countries by providing food and other natural resources and income from tour- ism and recreation (Moberg & Folke 1999; Access Eco- nomics 2005). However, the biodiversity, ecosystem function and economic resources of coral reefs are under increasing threat from a diverse range of impacts primarily associated with human activities (Hughes et al. 2003). These include overfishing and eutrophication Correspondence: Line K. Bay, Fax: +7 4725 1570; E-mail: line.bay@jcu.edu.au Ó 2009 Blackwell Publishing Ltd Molecular Ecology (2009) 18, 3062–3075 doi: 10.1111/j.1365-294X.2009.04257.x