Phenotypic Plasticity of the Drosophila Transcriptome Shanshan Zhou1,2, Terry G. Campbell2,3, Eric A. Stone2,3, Trudy F. C. Mackay2,3, Robert R. H. Anholt1,2,3* 1 Department of Biology, North Carolina State University, Raleigh, North Carolina, United States of America, 2 W. M. Keck Center for Behavioral Biology, North Carolina State University, Raleigh, North Carolina, United States of America, 3 Department of Genetics, North Carolina State University, Raleigh, North Carolina, United States of America Abstract Phenotypic plasticity is the ability of a single genotype to produce different phenotypes in response to changing environments. We assessed variation in genome-wide gene expression and four fitness-related phenotypes of an outbred Drosophila melanogaster population under 20 different physiological, social, nutritional, chemical, and physical environments and we compared the phenotypically plastic transcripts to genetically variable transcripts in a single environment. The environmentally sensitive transcriptome consists of two transcript categories, which comprise ,15% of expressed transcripts. Class I transcripts are genetically variable and associated with detoxification, metabolism, proteolysis, heat shock proteins, and transcriptional regulation. Class II transcripts have low genetic variance and show sexually dimorphic expression enriched for reproductive functions. Clustering analysis of Class I transcripts reveals a fragmented modular organization and distinct environmentally responsive transcriptional signatures for the four fitness-related traits. Our analysis suggests that a restricted environmentally responsive segment of the transcriptome preserves the balance between phenotypic plasticity and environmental canalization. Citation: Zhou S, Campbell TG, Stone EA, Mackay TFC, Anholt RRH (2012) Phenotypic Plasticity of the Drosophila Transcriptome. PLoS Genet 8(3): e1002593. doi:10.1371/journal.pgen.1002593 Editor: Gregory S. Barsh, Stanford University School of Medicine, United States of America Received October 14, 2011 Accepted January 28, 2012 Published March 29, 2012 Copyright: �� 2012 Zhou et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by National Institutes of Health grants GM59469 and GM45146. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: anholt@ncsu.edu Introduction Phenotypic plasticity is the ability of a single genotype to give rise to different phenotypes in different environments [1]. Phenotypic plasticity is the counterpoint to environmental canalization [2���3], whereby genotypes produce the same pheno- type in different environments. Phenotypic plasticity allows organisms to respond rapidly to changing environmental condi- tions without the time lag required for response to natural selection on segregating allelic variants and without the cost of selection, while environmental canalization buffers phenotypes against environmental perturbations. The balance between plasticity and robustness is thus crucial for optimal fitness [3���4] in variable environments, but the genetic basis for phenotypic plasticity has remained poorly defined. Elucidating the genetic underpinnings of phenotypic plasticity (and its converse, environmental canalization) requires that we determine what fraction of the genome is environmentally sensitive, which genes respond to the same or different environmental perturbations and how expression of environmen- tally sensitive genes is correlated with plasticity of organismal phenotypes. It is also necessary to determine what the relationship is between genetic variance and phenotypic plasticity, whether the same genes affecting phenotypic plasticity for a trait also affect genetic variation for that trait, and whether environmentally plastic and environmentally robust genes evolve at different rates. Although previous studies have analyzed changes in gene expression under one or few different environmental or physio- logical conditions [5���11], the study presented here is the first comprehensive study that analyzes co-variation among environ- mentally responsive genes across a wide range of environments in a defined outbred population reconstructed from inbred lines with documented expression profiles, enabling us to compare genotypic and environmental variation. We examined phenotypic plasticity in genome-wide gene expression and four organismal phenotypes related to reproductive fitness in a population generated by crossing 40 wild-derived inbred D. melanogaster lines [12]. The majority of the transcriptome shows robust expression across a range of environmental challenges, including different nutritional rearing conditions, physical stress conditions, chemical exposures, social crowding during larval or adult stages, and aging. Approximately 15% of transcripts are phenotypically plastic. By comparing genotypic variation among the original 40 wild-derived inbred lines under standard growth conditions, documented earlier [12], with environmental variation of transcript abundance levels in the reconstituted outbred population, we were able to discriminate two distinct classes of environmentally responsive transcripts, which we have designated Class I and Class II transcripts. Results Phenotypic Plasticity of the Transcriptome To identify phenotypically plastic and environmentally cana- lized transcripts, we assessed genome-wide gene expression of flies exposed to 20 treatments, including a control treatment of mated flies reared under standard conditions, and different nutrient or drug supplements, exposure to different physical and social environments, and maintenance at different reproductive states. Of the 18,800 transcripts represented on the microarray, 14,400 PLoS Genetics | www.plosgenetics.org 1 March 2012 | Volume 8 | Issue 3 | e1002593

(76.6%) generated signal intensities above background under at least one treatment, similar to the proportion of the transcriptome detected in a previous study, in which transcript profiles were obtained separately for the 40 individual genotypes that gave rise to our outbred population [12]. Analysis of variance of microarray intensity signals across all 20 rearing conditions revealed 1,249 transcripts that showed a significant treatment effect (8.7%), 6,745 transcripts that showed a sex effect (46.8%), and 200 transcripts with a significant treatment by sex interaction term (1.4%) at a false discovery rate of 0.05. Thus, the majority of the transcriptome is remarkably robust and buffered against diverse environmental challenges. We refer to the 1,249 transcripts exhibiting phenotypic plasticity as quantified by the significant treatment term in the ANOVAs as Class I transcripts. To simplify statistical analyses and maintain optimum power we excluded 166 Class I transcripts that also had significant treatment by sex interaction terms, giving 1,133 phenotypically plastic transcripts for further analyses (Table S1). We grouped these transcripts according to their relative expression levels across the 20 conditions (Figure S1). The highest relative gene expression levels were observed in flies exposed to low temperature, dopamine, nicotine or high sugar, and the lowest relative levels in flies exposed to heat shock or grown on high yeast medium. Surprisingly, overall relative gene expression levels are either uniformly higher or lower (more than 70% show higher abundance levels than the median) across the 20 conditions however, starvation stress resistance, aging, larval crowding and exposure to high temperature, result in substantial variation among relative expression levels (Figure S1). To further examine the relationship between gene expression and environmental exposure, we compared transcript abundance levels of the Class I transcripts under the different treatments to the standard rearing condition with post hoc least square difference (LSD) tests (p,0.05 Table S2). Heat shock has the greatest impact on gene expression (589 transcripts), whereas fluoxetine changes expression of only four transcripts (Figure 1A). Many transcripts show altered expression under multiple treatments for example, 167 transcripts show altered expression both after heat shock and exposure to starvation (Figure 1A). The majority of transcripts do not undergo significant changes compared to the standard condition (Figure 1B). Among the 1,133 Class I transcripts, 691 are computationally predicted with unknown function, 14 probe sets correspond to intergenic regions, non-coding RNAs and transposons, and 428 are annotated. The transcripts that show altered expression after heat shock include 13 Heat shock proteins (Hsps), 60 proteases, 17 members of the cytochrome P450 family (Cyps), two glutathione-S- transferases (Gsts), six UDP-glucose glycoprotein glycosyltransfer- ases (Ugts), and six immune-induced molecules (IM) (Figure 2, Table S2 Figure S2). A variety of additional transcripts in diverse gene ontology (GO) categories also respond to environmental challenges (Table S3). Nine of the 13 Hsps upregulated after heat shock are also upregulated after induction of chill coma (Figure S2). The abundance of heat shock proteins and proteases encoded by environmentally sensitive genes reflect mechanisms for environmental adaptation of the proteome. Heat shock proteins may offer protection for nascent polypeptides under adverse temperature conditions, while one function of environmentally sensitive proteases may be to facilitate de novo protein synthesis by providing a pool of amino acids through degradation of dispensable proteins. Modules of Phenotypically Plastic Transcripts We asked to what extent expression patterns of phenotypically plastic transcripts are co-regulated across environmental treat- ments. A previous study on the 40 inbred lines from which our outbred population is derived demonstrated that the genetically variable transcriptome (10,096 transcripts) is highly inter-correlat- ed and can be subdivided into 241 co-regulated modules [12], identified by modulated modularity clustering (MMC) [12���13]. Here, we used MMC to identify transcripts that covary across different treatments (Figure 3A, Table S1). This analysis partitioned the 1,133 Class I transcripts into 103 small, but highly correlated transcriptional modules (the average absolute correla- tion coefficient, |r|, within modules is at least 0.56). Extensive cross-module correlations are also evident. Negative correlations are rare, in agreement with the overall uniform up- or down- regulation of transcripts (Figure S1). All seven IM transcripts group together in module 71. A putative IM, CG15065, which is genetically correlated with IM1 and IM3 [12], is also contained in this module (Figure S2). These results show that changes in environmental conditions can cause fragmentation of the highly intercorrelated structure of the transcriptome observed under the standard growth condition [12]. Phenotypically Plastic Transcription Factors What are the cellular mechanisms that regulate transcriptional responses to environmental changes? As a first step to investigating how environmental stimuli may influence transcriptional regula- tion, we asked which transcription factors show altered expression under the different environmental conditions. Among the Class I transcripts, we identified 26 transcripts that encode transcriptional regulators, of which 25 were differentially expressed relative to the standard growth condition (Figure 3B, Figure 4A). Many of these transcription factors occur together in the same transcriptional modules (Figure 3B). The complexity of the interrelationships between transcript abundance levels of different transcription Author Summary Unlike Mendelian traits, where the genotype allows a direct prediction of the phenotype, predicting phenotypic values is not straightforward for complex traits, which arise from multiple segregating genes and their interactions with the environment. Here, a single genotype can often express different phenotypes in different environments. Such phenotypic plasticity is the counterpoint to ������envi- ronmental canalization,������ whereby genotypes produce the same phenotype in different environments. Whereas phenotypic plasticity allows organisms to respond rapidly to changing environments, environmental canalization buffers phenotypes against environmental perturbations. The balance between plasticity and robustness is crucial for optimal fitness, but the genetic basis for phenotypic plasticity is poorly defined. Here, we present the most comprehensive analysis to date of variation in genome- wide gene expression of an outbred Drosophila melano- gaster population under 20 different environments. We find that a restricted environmentally responsive segment of the transcriptome (,15%) preserves the balance between phenotypic plasticity and environmental canali- zation. Environmentally plastic transcripts can be grouped into two categories. Class I transcripts are genetically variable and associated with detoxification, metabolism, proteolysis, heat shock proteins, and transcriptional regulation. Class II transcripts have low genetic variance and show sexually dimorphic expression enriched for reproductive functions. Despite low genetic variance these transcripts evolve rapidly. Plasticity of the Drosophila Transcriptome PLoS Genetics | www.plosgenetics.org 2 March 2012 | Volume 8 | Issue 3 | e1002593