Environ. Sci. Technol. 2008, 42, 2181–2188
Linking Molecular and Population Stress Responses in Daphnia magna exposed to cadmium R I C H A R D C O N N O N , †,[ HELEN L. HOOPER,† RICHARD M. SIBLY,† F E I - L I N G L I M , ‡,@ L A R S - H E N R I K H E C K M A N N , †,¶ DAVID J. MOORE,‡ HAJIME WATANABE,§ ANNELEEN SOETAERT,4 KATIE COOK,⊥ STEVE J. MAUND,# T H O M A S H . H U T C H I N S O N , ∇,+ JONATHAN MOGGS,‡ WIM DE COEN,4 TAISEN IGUCHI,O AND A M A N D A C A L L A G H A N * ,† School of Biological Sciences, University of Reading, PO Box 68, Reading RG6 6BX, U.K., Syngenta Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire, SK10 4TJ, U.K., Center for Integrative Bioscience, Okazaki National Research Institutes 5-1, Higashiyama, Okaza, Myodaiji, JP 444-8585, Department of Zoology, University of Antwerp, Groenenborgerlaan 171, 2020 Antwerp, Belgium, Syngenta, Jealott’s Hill International Research Centre, Bracknell, Berks, RG42 6EY, U.K., Syngenta Crop Protection AG, 4002 Basel, Switzerland, AstraZeneca R&D, Global Safety Assessment, SE-151 85 Södertälje, Sweden, and National Institutes of Natural Sciences, Okazaki 444-8787, Japan
Received October 1, 2007. Revised manuscript received November 19, 2007. Accepted November 21, 2007.
DNA microarrays can be used to measure environmental stress responses. If they are to be predictive of environmental impact, we need to determine if altered gene expression translates into negative impacts on individuals and populations. A large cDNA microarray (14000 spots) was created to measure molecular stress responses to cadmium in Daphnia magna, the most widely used aquatic indicator species, and relate responses to population growth rate (pgr). We used the array to detect differences in the transcription of genes in juvenile D. magna (24 h old) after 24 h exposure to a control and three cadmium concentrations (6, 20, and 37 µg Cd2+ L-1). Stress responses at the population level were estimated * Corresponding author phone: 0118 378 4428; fax 0118 931 0180; e-mail:
[email protected]. † University of Reading. ‡ Syngenta Central Toxicology Laboratory. § Okazaki National Research Institutes 5-1. 4 University of Antwerp. ⊥ Syngenta, Jealott’s Hill International Research Centre. # Syngenta Crop Protection AG. ∇ Global Safety Assessment. O National Institutes of Natural Sciences. [ Current address: University of California, Davis, One Shields Avenue, School of Veterinary Medicine: Anatomy, Physiology and Cell Biology, Davis, California, 95616. ¶ Current address: University of Aarhus, National Environmental Research Institute, Department of Terrestrial Ecology, Vejlsøvej 25, P.O. Box 314, DK-8600, Silkeborg, Denmark. + Current address: Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth PL1 3DH, UK. @ Current address: Unilever, Colworth Science Park, Sharnbrook, Beds, MK44 1LQ, UK. 10.1021/es702469b CCC: $40.75
Published on Web 02/14/2008
2008 American Chemical Society
following a further 8 days exposure. Pgr was approximately linear negative with increasing cadmium concentration over this range. The microarray profile of gene expression in response to acute cadmium exposure begins to provide an overview of the molecular responses of D. magna, especially in relation to growth and development. Of the responding genes, 29% were involved with metabolism including carbohydrate, fat and peptide metabolism, and energy production, 31% were involved with transcription/translation, while 40% of responding genes were associated with cellular processes like growth and moulting, ion transport, and general stress responses (which included oxidative stress). Our production and application of a large Daphnia magna microarray has shown that measured gene responses can be logically linked to the impact of a toxicant such as cadmium on somatic growth and development, and consequently pgr.
Introduction In the field of toxicology, microarray-based assays have already provided insight into the mode of toxicity of chemicals (1) and offer hope that a similar insight can be obtained in the field of ecotoxicology. Ecotoxicogenomics is a rapidly growing field, but if it is to realize its potential, conclusive linkages must be made between gene expression profiles and adverse outcomes at the individual and population level (2–4). Linking the molecular, cellular, whole organism and population responses to stressors remains one of the great challenges in ecology and ecotoxicology (5). Linking molecular changes with relevant ecological responses will greatly improve the predictive powers of tests based on molecular responses. Genes up- or down-regulated in response to acute stress may predict chronic effects on individuals and populations before any such effect is apparent. Thus components and sometimes pathways that underlie physiological processes can be identified and investigated (6) and aid further understanding of the mode of action of stressors (4). Stressor-specific signatures in gene expression profiles could offer a diagnostic approach to identify the cause of an aquatic pollution event (3), but this approach assumes that exposure to toxicants will transiently alter gene expression at a detectable level. It is also dependent on a library of responses that have been previously defined and an understanding of stress pathways in the test organism. Despite the lack of extensive genomic data, invertebrates that have traditionally been used in ecotoxicology are being utilized in ecotoxicogenomic research. For this study, we chose the widely used indicator species Daphnia magna Straus as our model organism. Recent molecular studies on D. magna include studies on heavy metal toxicity (7, 8), an assessment of the impact of an azole fungicide on embryonic development (9), expression of target and reference genes in D. magna exposed to ibuprofen (10), and the cloning and analysis of D. magna expressed sequence tags (11). Studies on the closely related D. pulex include the expression of CYP450 genes (CYP4) in response to dietary polyphenols (12). There has, however, been little attempt to link these molecular stress responses to population responses, which are best measured by per capita population growth rate (hereafter pgr), which depends on the survivorship, somatic growth, time to maturity, and reproduction rate of individuals (13). We report here the development of a large-scale D. magna microarray to measure widespread changes in gene expression following acute exposure to chemical stress at nonlethal concentrations. We also assess the predictive value of a suite VOL. 42, NO. 6, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2181
TABLE 1. Stressor Conditions Used to Obtain DNA Fragments by Suppressive Subtractive Hybridization (SSH) for Microarray Construction number of daphnids stressor
exposure condition (24 h)
exposed
control
number of clones
kerosene cadmium 1 cadmium 2 lufenuron ibuprofen pH calcium (limitation)
25% of saturated fraction in media 100 µg L-1 400 µg L-1 0.867 µg L-1 63, 72 and 90 mg L-1 (pooled) 5 (adjusted with HCl) not incorporated in culture medium
100 100 100 100 900 100 100
100 100 100 100 300 100 100
14 280 714 192 384 181 264
of genes responding after a short (24 h) exposure upon population dynamics, thus juvenile populations, exposed to the same concentrations, were monitored for 9 days to estimate the effects of the stressor on pgr. We used cadmium as our model stressor because biochemical responses and adverse effects on the whole organism are relatively well understood. A highly toxic heavy metal with no reported biological function, cadmium has been shown to impact on daphnid growth and cellular energy allocation, and recent work has suggested that it affects molecular pathways such as digestion, oxygen transport, cuticular metabolism, and embryo development (8). Many studies have demonstrated that cadmium is highly reactive and forms complexes with nucleophilic ligands of target molecules. Cadmium is known to induce various stress-related genes including heat shock proteins (HSP) and metallothioneins (MT), with one of the main cellular impacts being oxidative stress. However, the induction of stress related proteins other than MT and glutathione (GSH) is generally only found with levels of cadmium approaching cytotoxic levels (14). Oxidative stress has a profound impact on most organisms, including arthropods. Oxidation of thiol groups leads to the depletion of reductants such as glutathione, which results in an overall oxidizing environment in the cell (14). This can trigger a series of biochemical changes that may include a switching of metabolic processes at the expense of glycolysis and the Krebs cycle. Using our microarray approach, we would therefore expect to be able to measure significant changes in genes involved in these metabolic processes, as well as those involved in growth and development. To link stress responses at the molecular level to those observed on individuals and populations, we classified responding genes into three major functional groups, which were further subdivided to aid a comparison of the measured effects. Clearly, the lack of complete genome coverage and the single time point exposure will limit the measurable molecular responses, but this data is rapidly being generated and the information gathered to date allows us to begin to understand stress pathways in Daphnia. As the body of literature in this area develops, using different microarray platforms and experimental design, we have growing confidence in the reproducibility of meaningful results (7, 8).
obtained from a cDNA library from unexposed mixed aged organisms obtained from a 3–4 week old culture (11) (10272 clones). (iii) cDNAs isolated following SSH, as above, between 15 adults carrying eggs and 75 juveniles (9) (1143 clones). Spots were pin-printed on Corning CMT-UltraGAPS glass slides (Fisher, UK), in a 17 × 18 block format with 48 blocks per microarray (grid ) 14688), using an Omnigrid 100 (Genomics Solutions, U.S.). Dimethyl sulfoxide (DMSO) was added, to a final concentration of 50%, to the purified PCR products prior to printing. After printing, microarray slides were cross-linked by UV (150 mJ/cm2) followed by baking for 2 hrs at 80 °C, and stored at room temperature under vacuum in total darkness until required. A number of positive and negative hybridization control spots were printed in each of the 48 blocks on the arrays. The positive hybridization controls comprised of D. magna genomic DNA and four Spot Report System PCR products from Arabidopsis thaliana; CAB, RCA, RBCL, and LPT4 (Stratagene, U.S.). Negative hybridization controls consisted of salmon sperm DNA, mouse Cot 1, human Cot 1, yeast transfer RNA (tRNA), and poly A RNA. Blank spots consisting of 50% DMSO were interspaced with the controls and jointly used to qualitatively assess the microarray hybridization efficiency and provide orientation for the grid overlap during assessment. A common reference pool, against which all samples were hybridized, was prepared using total RNA from over 30000 unexposed 24–48 h old D. magna collected from cultures over a period of 6 months. Experimental Design. Experimental Animals and Test Conditions. D. magna were obtained from the Water Research Centre (WRc), UK and were cultured at the University of Reading for a year prior to this experiment. Full details of culturing methods are given in Hooper, et al 15. Tests were carried out at 20 ( 1 °C with a 16 h: 8 h photoperiod in 5 L glass aquaria, which for the first 24 h contained an inner exposure vessel (a 600 mL glass cylinder, length 13 cm,
Materials and Methods Microarray Construction. The D. magna microarray was constructed using DNA fragments from three sources: (i) Stress-specific cDNA (cDNA) was generated using suppressive subtractive hybridization (SSH) on organisms exposed to the stressors shown in Table 1. SSH was performed with forward and reverse subtraction using a PCR-Select cDNA Subtraction Kit (Clontech Laboratories Inc., USA). cDNA was prepared from mRNA extracted from batches of 100 D. magna, 48 h old, using a Straight A’s mRNA isolation kit system (Novagen Inc., UK) (2029 clones). (ii) Expressed sequence tags (ESTs) 2182
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 6, 2008
FIGURE 1. Mean (n ) 4) population growth rate (pgr) per day of Daphnia magna after 9 days exposure to CdCl2 in 5 L of reconstituted water. Bars indicate standard errors, * notation indicates significant differences between treatments and controls (Fisher’s pairwise comparisons: *** p < 0.001, ** p < 0.01, * p < 0.05).
FIGURE 2. Microarray gene tree (heat map) showing clusters of genes with similar expression profiles (Pearsons’ correlation). Red indicates genes that are up-regulated and green those that are down-regulated. diameter 9 cm), with a 0.25 mm nylon mesh bottom to allow free movement of the test water between the two vessels. Replicate populations (n ) 4) were initiated with 250 third brood neonates (
