http:\/\/www.nwf.org\/News-and-Magazines\/Media-Center\/News-by-Topic\/Wildlife\/2011\/09-26-11-New-Study-Documents-BP-Oils-Impact-on-Gulf-Ecosystem.aspx<\/p>\n
Alarming New Study Documents BP Oil’s Impact on Gulf Ecosystem<\/strong><\/p>\n Washington, DC (September 26, 2011) – A study published today in the Proceedings of the National Academy of Sciences documents the effect of BP oil on the Gulf killifish. The minnow-like wetlands resident, also known as bull minnow or cacahoe, is a critical part of the Gulf’s food chain and was chosen for study by a team of researchers because of its abundance and sensitivity to any effects of toxic pollution. The study finds that oil exposure has altered the killifish’s cellular function in ways that are known to be predictive of developmental abnormalities, decreased hatching success, and decreased embryo and larval survival.<\/p>\n Doug Inkley, senior scientist with the National Wildlife Federation, said today:<\/p>\n “This study is alarming because similar health effects seen in fish, sea otters, and harlequin ducks following the Exxon Valdez spill in Alaska were predictive of population impacts, from decline to outright collapse. While up to 210 million gallons of oil were involved in the Gulf oil disaster, the study is a reminder that even small amounts of oil can have a large and lasting impact on individual fish and wildlife. Wherever oil continues to be found in the Gulf, it should be removed if doing so won’t cause more environmental harm than good.<\/p>\n “The Gulf killifish provides us with a reminder that oil’s impacts on wildlife can’t be separated from its impacts on people. Not only are Gulf killifish a food source for sport fish like redfish and speckled trout, but killifish eat mosquitoes, helping to keep the pest population in check.<\/p>\n “The study is also a reminder that Congress has yet to act to protect the Gulf’s people and wildlife by passing comprehensive response legislation. Action is urgently needed, both to improve oil and gas drilling safety regulations so this doesn’t happen again, and to dedicate fines and penalties to Gulf Coast restoration.”<\/p>\n Learn more about the National Wildlife Federation’s response to the Gulf oil disaster at NWF.org\/OilSpill and visit the National Wildlife Federation Media Center at NWF.org\/News.<\/p>\n Celebrating 75 years of inspiring Americans to protect wildlife for our children’s future. Miles Grant Genomic and physiological footprint of the Deepwater Horizon oil spill on resident marsh fishes<\/strong><\/p>\n by Andrew Whiteheada,1, Benjamin Dubanskya, Charlotte Bodiniera, Tzintzuni I. Garciab, Scott Milesc, Chet Pilleyd, Vandana Raghunathane, Jennifer L. Roacha, Nan Walkere, Ronald B. Walterb, Charles D. Ricef, and Fernando Galveza Departments of a Biological Sciences, c Environmental Sciences, and e Oceanography and Coastal Sciences, and d Coastal Studies Institute, Louisiana State University, Baton Rouge, LA 70803; b Department of Chemistry and Biochemistry, Texas State University, San Marcos, TX 78666; and f Department of Biological Sciences, Clemson University, Clemson, SC 29634<\/p>\n Edited by Paul G. Falkowski, Rutgers, The State University of New Jersey, New Brunswick, NJ, and approved September 1, 2011 (received for review June 13, 2011)<\/p>\n The biological consequences of the Deepwater Horizon oil spill are unknown, especially for resident organisms. Here, we report results from a field study tracking the effects of contaminating oil across space and time in resident killifish during the first 4 mo of the spill event. Remote sensing and analytical chemistry identified exposures, which were linked to effects in fish characterized by genome expression and associated gill immunohistochemistry, despite very low concentrations of hydrocarbons Oil-contaminated waters are also associated with aberrant protein expression in gill tissues of larval and adult fish. These data suggest that heavily weathered crude oil from the spill imparts significant biological impacts in sensitive Louisiana marshes, some of which remain for over 2 mo following initial exposures.<\/strong><\/p>\n ecological genomics | ecotoxicology | microarray | RNA-seq |vtoxicogenomics<\/p>\n Following the Deepwater Horizon (DWH) drilling disaster on http:\/\/www.nwf.org\/News-and-Magazines\/Media-Center\/News-by-Topic\/Wildlife\/2011\/09-26-11-New-Study-Documents-BP-Oils-Impact-on-Gulf-Ecosystem.aspx Alarming New Study Documents BP Oil’s Impact on Gulf Ecosystem Washington, DC (September 26, 2011) – A study published today in the Proceedings of the National Academy of Sciences documents the effect of BP oil on the Gulf killifish. The minnow-like wetlands resident, also known as bull minnow or cacahoe, is a critical part … Continue reading National Wildlife Federation: Alarming New Study Documents BP Oil’s Impact on Gulf Ecosystem “Genomic and physiological footprint of the Deepwater Horizon oil spill on resident marsh fishes”<\/span>
\n——————–<\/p>\n
\nOnline Communications Manager
\nNational Wildlife Federation
\nGrantM@NWF.org
\n202-797-6855 (office) – 703-864-9599 (cell)<\/p>\n
\nremaining in water and tissues. Divergence in genome expression coincides with contaminating oil and is consistent with genome responses that are predictive of exposure to hydrocarbon-like chemicals and indicative of physiological and reproductive impairment.<\/strong><\/p>\n
\nApril 20, 2011, in the Gulf of Mexico, acute oiling and the
\nresulting mortality of marine wildlife were evident. In contrast,
\nthe sublethal effects, critically important for predicting longterm
\npopulation-level impacts of oil pollution (1), have not
\nbeen well described following the DWH disaster. Here, we report
\nthe results of a 4-mo field study monitoring the biological
\neffects of oil exposure on fish resident in Gulf of Mexico coastal
\nmarsh habitats.
\nGulf killifish (Fundulus grandis) were used as our model species
\nbecause they are among the most abundant vertebrate animals
\nin Gulf of Mexico-exposed marshes (2\u20134). Furthermore, the
\nAtlantic-distributed sister species to F. grandis (Fundulus heteroclitus)
\nhas a narrow home range and high site fidelity, especially
\nduring the summer (5, 6), and, among fishes, it is relatively
\nsensitive to the toxic effects of organic pollutants (7). Although
\nhome range and toxicology studies are lacking for F. grandis, we
\ninfer that F. grandis is also relatively sensitive to pollutants and
\nexhibits high site fidelity, such that the biology of this species is
\nlikely affected primarily by the local environment, given the recent
\nshared ancestry of F. grandis with F. heteroclitus (8) and
\nsimilar physiology, life history, and habitat (9\u201313). We sampled
\nfrom populations resident in Gulf of Mexico-exposed marshes
\nbefore oil landfall (May 1\u20139, 2010), during the peak of oil
\nlandfall (June 28\u201330, 2010), and after much of the surface oil was
\nno longer apparent 2 mo later (August 30\u2013September 1, 2010) at
\nsix field sites from Barataria Bay, Louisiana, east to Mobile Bay,
\nAlabama (Fig. 1 and Dataset S1).
\nResults and Discussion
\nRemote sensing and analytical chemistry were used to characterize
\nexposure to DWH oil, where remote sensing data are
\nspatially and temporally comprehensive but of low resolution
\nand chemistry data are of high resolution but patchy in space and
\ntime. Ocean surface oil was remotely detected through the
\nanalysis of images from synthetic aperture radar (SAR) (14).
\nProximity of the nearest oil slick to each field site (e.g., Fig. S1)
\nwas measured for each day that SAR data were available, from
\nMay 11 through August 13, 2010, to approximate the location,
\ntiming, and duration of coastal oiling (Fig. 1C). Although surface
\noil came close to many of our field sites in mid-June, only the
\nGrande Terre (GT) site was directly oiled (Fig. 1 B and C).
\nAlthough the GT site had been clearly contaminated with crude
\noil for several weeks before our sampling (Fig. 1C and Fig. S2)
\nand retained much oil in sediments (Dataset S2), only trace
\nconcentrations of oil components were detected in subsurface
\nwater samples collected from the GT site on June 28, 2010, and
\ntissues did not carry abnormally high burdens of oil constituents
\nat any site or time point (Dataset S2). Despite a low chemical
\nsignal for oil in the water column and tissues at the time of
\nsampling, we detected significant biological effects associated
\nwith the GT site postoil.
\nWe sampled multiple tissues from adult Gulf killifish (average
\nweight of 3.5 g) from each of six field sites for each of three time
\npoints [only the first two time points for the Mobile Bay (MB)
\nsite] spanning the first 4 mo of the spill event (Fig. 1C). We
\ncompared biological responses across time (before, at the peak,
\nand after oiling) and across space (oiled sites and sites not oiled)
\nand integrated responses at the molecular level using genome
\nexpression profiling with complimentary protein expression and
\ntissue morphology. Genome expression profiles, using microarrays
\nand RNAseq, were characterized for livers because the
\norgan is internal and integrates xenobiotic effects from multiple
\nroutes of entry (gill, intestine, and skin), and because liver is the
\nprimary tissue for metabolism of toxic oil constituents. Tissue
\nmorphology and expression of CYP1A protein, a common biomarker
\nfor exposure to select polycyclic aromatic hydrocarbons
\n(PAHs), was characterized for gills, the organ that provides the
\ngreatest surface area in direct contact with the surrounding
\naquatic environment. In addition, we exposed developing
\nembryos to field-collected water samples to document bioavailability
\nand bioactivity of oil contaminants for this sensitive
\nearly life stage.
\nAuthor contributions: A.W. and F.G. designed research; A.W., B.D., C.B., T.I.G., S.M., C.P.,
\nV.R., J.L.R., N.W., R.B.W. and F.G. performed research; C.D.R. contributed new reagents\/
\nanalytic tools; A.W., B.D., C.B., T.I.G., S.M., C.P., V.R., N.W., R.B.W. and F.G. analyzed data;
\nand A.W., B.D., C.B., and F.G. wrote the paper.
\nThe authors declare no conflict of interest.
\nThis article is a PNAS Direct Submission.
\nData deposition: Microarray data have been deposited to ArrayExpress (accession no.
\nE-MTAB-663).
\n1To whom correspondence should be addressed. E-mail: andreww@lsu.edu.
\nThis article contains supporting information online at www.pnas.org\/lookup\/suppl\/doi:10.
\n1073\/pnas.1109545108\/-\/DCSupplemental.
\nwww.pnas.org\/cgi\/doi\/10.1073\/pnas.1109545108 PNAS Early Edition | 1 of 5
\nENVIRONMENTAL
\nSCIENCES
\nSCIENCE APPLICATIONS IN SPECIAL FEATURE
\nTHE DEEPWATER HORIZON
\nOIL SPILL SPECIAL FEATURE
\nThe oiling of the GT site at the end of June 2010 is associated
\nwith a clear functional genomic footprint. Of the 3,296 genes
\nincluded in our analysis, expression of 1,600 and 1,257 genes
\nvaried among field sites and throughout the time course, respectively
\n(P < 0.01) (Dataset S3). For the 646 genes that varied\nin expression only among sites (no significant time effect or siteby-\ntime interaction), site variation followed a pattern of population\nisolation by distance, which is consistent with neutral\nevolutionary divergence (Fig. 2A) and population genetic expectations\n(15). Most importantly, 1,500 genes indicated a pattern of\nsite-dependent time course expression (significant interaction,\nfalse discovery rate <0.01), where the trajectory of genome expression\nthrough time was divergent at the GT site compared\nwith all other sites (Fig. 2 B and C), particularly at the second\ntime point, which coincides with oil contamination (Fig. 1C).\nPrevious studies have identified genes that are transcriptionally\nresponsive to planar polychlorinated biphenyl (PCB) exposures\nin killifish (16). Planar PCBs, dioxins, and PAHs (the\nprimary toxic constituents in crude oil) are all mechanistically\nrelated insofar as they exert biological effects, in whole or in part,\nthrough aryl-hydrocarbon receptor (AHR) signaling pathways;\nindeed, morpholino knockdown of the AHR is protective of the\ntoxic effects of PAHs and PCBs in killifish (17), and exposures to\nPCBs and PAHs induce common genome expression responses\nin flounder (18). Of the genes that were transcriptionally responsive\nto PCB exposures (16), 380 were included in the current\nanalysis. Expression of this subset of genes is predictive of\ntranscriptional divergence in fish from the GT site coincident\nwith oil contamination compared with other field sites (Fig. S3),\nespecially for the top 10% of PCB-responsive genes (Fig. 2D).\nTranscriptional activation of these planar PCB-responsive genes\nin developing killifish embryos is predictive of induction of developmental\nabnormalities, decreased hatching success, and decreased\nembryonic and larval survival (16, 19). This set of genes\nincludes members of the canonical battery of genes that are\ntranscriptionally induced by ligand-activated AHR signaling,\nsuch as cytochrome P450s, cytochrome B5, and UDP-glucuronosyltransferase\n(Fig. 2F, set 1), for which increased transcription\nis particularly diagnostic of exposure to select hydrocarbons\n(20). Indeed, many genes that are transcriptionally induced or\nrepressed by AHR activators (dioxins, PCBs, and PAHs) show\ninduction or repression at the GT site coincident with crude oil\ncontamination (Fig. 2F, set 1). An independent measure of genome\nexpression, RNAseq, also indicates AHR activation in GT\nfish from June 28, 2010, compared with reference RNA (e.g., upregulation\nof cytochrome P450s, UDP-glucuronosyltransferase\n(UGT), and AHR itself; Fig. 2E). In parallel, up-regulation of\nCYP1A protein was detected in gills from GT fish sampled\npostoil and in early life-stage fish following controlled exposures\nto GT waters (Figs. 3 and 4). These data appear to be diagnostic\nof exposure to the toxic constituents in contaminating oil (PAHs)\nat a sufficient concentration and duration to induce biological\nresponses in resident fish. Sustained activation of the CYP1A\ngene (Figs. 2F and 3) was predictive of persistent exposure to\nsublethal concentrations of crude oil components and negative\npopulation-level impacts in fish, sea otters, and harlequin ducks\nfollowing the Exxon Valdez oil spill (reviewed in 1), although\nPAH toxicity may be mediated through AHR-independent\npathways as well (21).\nTranscriptional responses in other sets of coexpressed genes\noffer insights into the potential biological consequences of contaminating\noil exposure at the GT site. Several gene ontology\n(GO) categories were enriched in the subset of genes that\nshowed GT-specific expression divergence coincident with siteand\ntime-specific oil contamination (Dataset S4). GO enrichment\nindicates activation of the ubiquitin-proteasome system\n(Fig. 2F, set 2), which, among diverse functions, is important for\ncellular responses to stress, cell cycle regulation, regulation of\nDNA repair, apoptosis, and immune responses (22). The AHR\nprotein itself plays a role as a unique ligand-dependent E3\nubiquitin ligase that targets sex steroid (estrogen and androgen)\nreceptor proteins for proteasomal destruction, thereby impairing\nFig. 1. Location of field study sites and incidence of oil contamination. (A) Location of field sampling sites, which include Grand Terre (GT), Bay St. Louis (BSL),\nBelle Fontaine Point (BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA). Color coding is consistent with other figures. The red star indicates\nthe DWH spill site. (B) Photograph (by A.W.) of the GT field site on June 28, 2010, showing contaminating oil and minnow traps in the marsh. (C) Proximity of\nnearest surface oil to each field site was determined by SAR, where rows are field sites and columns are days. Light gray represents no data, and black\nrepresents the nearest surface oil at a distance of >4 km; the increasing intensity of red indicates closer proximity of oil. Three field sampling trips are
\nhighlighted (blue boxes). BSL; BFP; FMA.
\n2 of 5 | www.pnas.org\/cgi\/doi\/10.1073\/pnas.1109545108 Whitehead et al.
\nnormal cellular responses to sex hormones in reproductive tissues,
\nand this response can be activated by planar PAHs (23).
\nSignificant down-regulation of transcripts for egg envelope proteins
\nzona pellucida (ZP3 and ZP4) and choriogenin (ChgHm
\nand ChgH) that we detect at the GT site coincident with oil
\nexposure (Fig. 2F, set 1) may be linked to this AHR-dependent
\nproteolytic pathway because their transcription is estrogen-dependent
\n(24, 25) and is down-regulated by exposure to PAHs in
\nfish (25\u201327). In corroboration, RNAseq detects dramatically
\ndown-regulated ZP, ChgH, and vitellogenin transcripts in GT
\nfish (Fig. 2E). Although the transcriptional response that we
\ndetect is in male fish, these proteins are synthesized in male livers
\n(reviewed in 25, 27) and down-regulation is consistent with
\nantiestrogenic effects from exposure to PAHs (28). Possible
\nimpacts on reproduction merit attention because water only
\nfrom the GT site induced CYP1A protein in the gills of developing
\nkillifish (Fig. 3) at low concentrations of total aromatics
\nand alkanes (Dataset S2) and more than 2 mo after initial oiling,
\nindicating persistent bioavailability of PAHs. Marsh contamination
\nwith DWH oil coincided with the spawning season for many
\nmarsh animals, including killifish (29), and reproductive effects are
\npredictive of long-term population-level impacts from oil spills (1).
\nControlled exposures of developing killifish to water collected
\nfrom GT on June 28 and August 30, 2010, induced CYP1A protein
\nexpression in larval gills relative to fish exposed to GT water
\npreoil and exposed to Bayou La Batre (BLB) site water that was
\nnot oiled (Fig. 3). This response is consistent with the location and
\ntiming of oil contamination, and it indicates that the remaining oil
\nconstituents dissolved at very low concentrations at GT after
\nlandfall (Dataset S2) were bioavailable and bioactive to developing
\nfish. Although exposures to PAHs stereotypically induce
\ncardiovascular system abnormalities in developing fish at relatively
\nhigh concentrations (e.g., 21), none were observed in these
\nanimals. However, even very low-concentration exposures during
\ndevelopment, insufficient to induce cardiovascular abnormalities
\nin embryos, can impair cardiac performance in adulthood (30).
\nThe adult fish sampled in situ from the oil-contaminated GT site
\nshowed divergent regulation of several genes involved in blood
\nvessel morphogenesis and heme metabolism coincident with oil
\ncontamination (Fig. 2F, set 3). Multigeneration field studies are
\nnecessary to confirm cardiovascular effects from DWH oil contamination
\nof marshes that coincided with spawning.
\nFig. 3. CYP1A protein expression (dark red staining) in larval killifish gills
\n(24 d postfertilization) exposed to waters collected from GT (oiled) and BLB
\n(not oiled) during development. (Magnification 40\u00d7, scale bars = 10 \u03bcm.)
\nCYP1A expression is elevated in the lamellae of larvae exposed during development
\nto waters collected from GT postoil (trips 2 and 3) compared with
\nbackground levels of CYP1A expression in larvae exposed to GT water preoil
\n(trip 1), compared with CYP1A in fish exposed to waters collected from BLB
\n(which was not directly oiled), and compared with CYP1A in fish reared in
\nlaboratory control water. Nuclei were stained using hematoxylin (blue).
\nAnalytical chemistry of exposure waters is reported in Dataset S2.
\nFig. 2. Genome expression between field sites and across time. Field sites include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), Bayou La
\nBatre (BLB), Mobile Bay (MB), and Fort Morgan (FMA). GT was the only site to be directly oiled, which occurred between the first and second sampling times
\n(Fig. 1 and Dataset S2). (A) For genes that vary only among sites (no expression change with time or interaction), pairwise site-specific transcriptome divergence
\nalong principal component (PC) 1, as a function of pairwise geographical distance, shows a pattern consistent with isolation by distance. (B) Trajectory
\nof genome expression responses through time for each of six field sites from the preoil sample time (dot at base of arrow) through the peak-oil sample
\ntime (middle dot), to the latest postevent sample time (dot at head of arrow) following PC analysis of genes showing statistically significant main effects (site
\nand time) and interaction terms. (C) Divergence along PC1 is isolated, where bars for each site from left to right represent sampling times from the earliest to
\nthe latest. (D) Expression divergence along PC1 for the subset of genes that is dose-responsive to PCB exposure (top 10% of PCB-responsive genes). (E) RNAseq
\ndata showing genes up- and down-regulated (x axis positive and negative, respectively) in fish from GT sample time 2 (coincident with oil) compared with
\nreference RNA, where select genes are identified. (Inset) Genes are dramatically down-regulated at GT (detailed RNAseq data are presented in Dataset S5). (F)
\nExpression levels for specific genes (rows) and treatments (columns), where cell color indicates up-regulation (yellow) or down-regulation (blue) scaled
\naccording to site-specific expression level at the preoil sample time, for genes with divergent expression at the GT site. Genes are grouped into functional
\ncategories, and scale bars indicate N-fold up- or down-regulation.
\nWhitehead et al. PNAS Early Edition | 3 of 5
\nENVIRONMENTAL
\nSCIENCES
\nSCIENCE APPLICATIONS IN SPECIAL FEATURE
\nTHE DEEPWATER HORIZON
\nOIL SPILL SPECIAL FEATURE
\nCoastal salt marsh habitats are dynamic and stressful, where
\nchanges in environmental parameters, such as temperature,
\nhypoxia, and salinity, can continuously challenge resident wildlife.
\nRegulation of ion transport in fish is particularly important
\nfor facilitating homeostasis in response to the salinity fluctuations
\nthat are common in estuaries. We found altered regulation
\nof multiple ion transport genes in fish from the GT site coincident
\nwith oil contamination (Fig. 2F, set 4). For example, Vtype
\nproton ATPases are up-regulated and Na+,K+-ATPase
\nsubunits and tight-junction proteins are down-regulated, coincident
\nwith oiling at the GT site, in the absence of substantial
\nchanges in environmental salinity (Dataset S2). Other genes
\nimportant for osmotic regulation in killifish (31) are also divergently
\ndown-regulated at the GT site, including type II
\niodothyronine deiodinase (DIO2), transcription factor jun-B
\n(JUNB), and arginase 2 (ARG2). In corroboration, RNAseq
\ndata show down-regulation of DIO2, JUNB, and ARG2 in GT
\nfish compared with reference fish (Fig. 2E). Although the physiological
\nconsequences of oil exposures are typically studied in
\nisolation, it is reasonable to predict that exposure to oil may
\ncompromise the ability of resident organisms to adjust physiologically
\nto natural stressors.
\nInduction of CYP1A protein expression is a hallmark of AHR
\nsignaling pathway activation, making it a sensitive biomarker of
\nexposure to select planar PAHs and other hydrocarbons (20).
\nAlthough the liver is the key organ for CYP1A-mediated metabolism
\nof these substrates, gill tissues represent the most
\nproximate site of exposure to PAHs. As a result of direct contact
\nwith the environment and the nature of the gill as a transport
\nepithelium, the gill may be a more sensitive indicator of exposure
\nto contaminants than the liver (32). CYP1A protein was markedly
\nelevated in GT fish postoil compared with GT fish preoil
\nand compared with fish from other field sites that were not directly
\noiled (Fig. 4). CYP1A induction was localized predominantly
\nto pillar cells of the gill lamellae and within
\nundifferentiated cells underlying the interlamellar region, which
\nmay have contributed to the filamental and lamellar hyperplasia
\nobserved during trips 2 and 3, as well as the gross proliferation of
\nthe interlamellar region observed during trip 2 in GT fish (Fig.
\n4). These effects imply a decrease in the effective surface area of
\nthe gill, a tissue that supports critical physiological functions,
\nsuch as ion homeostasis, respiratory gas exchange, systemic acidbase
\nregulation, and nitrogenous waste excretion (33). Currently,
\nthe degree to which oil-induced effects may interact with commonly
\nencountered challenges, such as fluctuations in hypoxia
\nand salinity, to compromise physiological resilience is unclear.
\nBy integrating remote sensing and in situ chemical measures of
\nexposure, and linking these with integrated measures of biological
\neffect (genome expression and tissue morphology), we
\nprovide evidence that links biological impacts with exposure to
\ncontaminating oil from the DWH spill within coastal marsh
\nhabitats. Although body burdens of toxins are not high, consistent
\nwith reports indicating that seafood from the Gulf of Mexico
\nis safe for consumption (34), this does not mean that negative
\nbiological impacts are absent. Our data reveal biologically relevant
\nsublethal exposures causing alterations in genome expression and
\ntissue morphology suggestive of physiological impairment persisting
\nfor over 2 mo after initial exposures. Sublethal effects were
\npredictive of deleterious population-level impacts that persisted
\nover long periods of time in aquatic species following the Exxon
\nValdez spill (1) and must be a focus of long-term research in
\nthe Gulf of Mexico, especially because high concentrations of
\nhydrocarbons in sediments (Dataset S2) may provide a persistent
\nsource of exposures to organisms resident in Louisiana marshes.
\nMethods
\nThe locations (latitude and longitude) of our field sampling sites and dates for
\nsampling at each site are summarized in Dataset S1. Gulf killifish (F. grandis)
\nwere caught by minnow trap, and tissues were excised immediately. Liver
\nwas preserved in RNAlater (Ambion, Inc.) for genome expression (microarray
\nand RNAseq) analysis. Gill tissues were fixed in situ in buffered zinc-based
\nformalin Z-Fix (Anatech LTD). Succinct methods follow, and more detailed
\nmethods are available online.
\nSatellite imagery (SAR) was analyzed to provide estimation of the timing,
\nlocation, and duration of coastal oil contamination. The calculated distance
\nfrom each field sampling site to the nearest oil slick was calculated from the
\n\u201cstraight-line\u201d distance from the global positioning system position of the
\nstation (Dataset S1) to that of the observed oil across any and all intervening
\ngeographical barriers (e.g., Fig. S1).
\nFig. 4. CYP1A protein expression in adult killifish gills (dark red staining)
\nsampled in situ from all sampling times (columns) and locations (rows).
\nLocations include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point
\n(BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA).
\n(Magnification 40\u00d7, scale bars = 10 \u03bcm.) The MB site was only sampled on
\ntrips 1 and 2, and gills from trip 1 at the BLB site were not available for
\nprocessing. Fish gills from the GT site during trips 2 and 3 showed high CYP1A
\nexpression and an elevated incidence of hyperplasia of the lamellae and
\ninterlamellar space on the gill filaments coincident with oil contamination.
\nCYP1A protein was elevated at the GT site postoil (trips 2 and 3) compared
\nwith GT preoil (trip 1) as well as with other field sites, none of which were
\ndirectly oiled. Nuclei were stained using hematoxylin (blue). Exact site locations
\nand sampling dates are reported in Dataset S1.
\n4 of 5 | www.pnas.org\/cgi\/doi\/10.1073\/pnas.1109545108 Whitehead et al.
\nAnalytical chemistry of water, tissue, and sediment samples was performed
\nto offer detailed characterization of exposure to contaminating oil (data
\nreported in Dataset S2). Sample dates and locations are summarized in
\nDataset S1. All sample extracts were analyzed using GC interfaced to an MS
\ndetector system. Spectral data were processed by Chemstation Software
\n(Agilent Technologies), and analyte concentrations were calculated based
\non the internal standard method.
\nGenome expression across sites and time was characterized using custom
\noligonucleotide microarrays. Genome expression was measured in liver tissues
\nfrom five replicate individual male fish per site-time treatment (5 biological
\nreplicates) hybridized in a loop design, including a dye swap. Data
\nwere lowess-normalized and then mixed model-normalized using linear
\nmixed models to account for fixed (dye) effects and random (array) effects.
\nNormalized data were then analyzed using mixed model ANOVA, with
\n\u201csite\u201d [Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP),
\nBayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA)] and
\n\u201csampling time\u201d (sampling trips 1, 2, and 3) (Dataset S1) as main effects,
\nincluding an interaction (site-by-time) term. The false discovery rate was
\nestimated using Q-value (35). Principal components analysis was performed
\nusing MeV (36). GO enrichment was tested using DAVID (37).
\nFor RNAseq, transcript abundance was compared between liver mRNA
\nfrom three replicate fish (RNA was not pooled) from the GT site from June 28,
\n2010, and mRNA from two control samples. All RNA samples were sequenced
\non the Illumina Gene Analyzer platform (Expression Analysis, Inc.). Following
\nquality control filtering, quantitative transcript abundance analysis was
\nperformed by mapping sequence reads to target sequences (6,810 unique F.
\nheteroclitus target EST sequences, Dataset S5) using the Bowtie short read
\nalignment software (38). A custom Perl script determined the number of
\nfragments mapped to each target sequence. The Bioconductor package
\nDESeq (version 2.8) (39) was used to determine the statistical significance of
\neach differentially expressed target using a negative binomial method with
\nP values adjusted by the Benjamini\u2013Hochberg procedure.
\nGill tissues were sampled from all field sites for morphological analysis and
\nimmunohistochemical analysis of CYP1A protein expression. Gill tissues from
\nthe first and second gill arches were sectioned along the longitudinal axis at
\na thickness of 4 \u03bcm and probed with mAb C10-7 against fish CYP1A (40).
\nSections were counterprobed using the Vectastain ABC immunoperoxidase
\nsystem (Vector Laboratories), utilizing the ImmPACT Nova RED peroxidase
\nsubstrate kit (Vector Laboratories) to visualize the CYP1A protein in red.
\nTissue sections were counterstained with Vector Hematoxylin QS (Vector
\nLaboratories).
\nF. grandis embryos obtained from parents not exposed to oil (collected
\nfrom Cocodrie, LA) were exposed to water samples from the GT and BLB
\nsites collected subsurface on the dates indicated in Dataset S1. Following
\nfertilization, 20 embryos were randomly transferred in triplicate to one of
\nthe six field-collected waters (2 field sites \u00d7 3 time points) at 3 h postfertilization.
\nEmbryos were also exposed to a laboratory control consisting
\nof artificial 17 parts per thousand (ppt) water. Larvae were sampled at 24
\nd postfertilization and fixed in Z-Fix solution. Sectioning and staining were
\nas described in the previous section.
\nACKNOWLEDGMENTS. K. Carman helped facilitate early field studies. The
\nauthors thank R. Brennan, D. Roberts, E. McCulloch, Y. Meng, A. Rivera,
\nC. Elkins, H. Graber, R. Turner, D. Crawford, and M. Oleksiak, for technical
\nassistance. Funding was from the National Science Foundation (Grants DEB-
\n1048206 and DEB-1120512 to A.W., Grant EF-0723771 to A.W. and F.G., and
\nGrant DEB-1048241 to R.B.W.), the National Institutes of Health (R15-
\nES016905-01 to C.D.R.), and the Gulf of Mexico Research Initiative (A.W.,
\nF.G., and N.W.).
\n1. Peterson CH, et al. (2003) Long-term ecosystem response to the Exxon Valdez oil spill.
\nScience 302:2082\u20132086.
\n2. Rozas LP, Reed DJ (1993) Nekton use of marsh-durface habitats in Louisiana (USA)
\ndeltaic salt marshes undergoing submergence. Mar Ecol Prog Ser 96:147\u2013157.
\n3. Rozas LP, Zimmerman RJ (2000) Small-scale patterns of nekton use among marsh and
\nadjacent shallow nonvegetated areas of the Galveston Bay Estuary, Texas (USA). Mar
\nEcol Prog Ser 193:217\u2013239.
\n4. Subrahmanyam CB, Coultas CL (1980) Studies on the animal communities in 2 North
\nFlorida salt marshes. 3. Seasonal fluctuations of fish and macroinvertebrates. Bull Mar
\nSci 30:790\u2013818.
\n5. Lotrich VA (1975) Summer home range and movements of Fundulus heteroclitus
\n(Pisces: Cyprinodontidae) in a tidal creek. Ecology 56:191\u2013198.
\n6. Teo SLH, Able KW (2003) Habitat use and movement of the mummichog (Fundulus
\nheteroclitus) in a restored salt marsh. Estuaries 26:720\u2013730.
\n7. Van Veld PA, Nacci DE (2008) Toxicity resistance. The Toxicology of Fishes, eds Di
\nGiulio RT, Hinton DE (Taylor and Francis, Boca Raton, FL), pp 597\u2013641.
\n8. Whitehead A (2010) The evolutionary radiation of diverse osmotolerant physiologies
\nin killifish (Fundulus sp.). Evolution 64:2070\u20132085.
\n9. Able KW, Hata D (1984) Reproductive behavior in the Fundulus heteroclitus-F. grandis
\ncomplex. Copeia (4):820\u2013825.
\n10. Kneib RT (1997) The role of tidal marshes in the ecology of estuarine nekton.
\nOceanography and Marine Biology: An Annual Review 35:163\u2013220.
\n11. Nordlie FG (2006) Physicochemical environments and tolerances of cyprinodontoid
\nfishes found in estuaries and salt marshes of eastern North America. Reviews in Fish
\nBiology and Fisheries 16:51\u2013106.
\n12. Rozas LP, Lasalle MW(1990) A comparison of the diets of Gulf killifish, Fundulus grandis
\nBaird and Girard, entering and leaving a Mississippi brackishmarsh. Estuaries 13:332\u2013336.
\n13. Weisberg SB, Lotrich VA (1982) The importance of an infrequently flooded intertidal
\nmarsh surface as an energy source for the mummichog Fundulus heteroclitus: An
\nexperimental approach. Mar Biol 66:307\u2013310.
\n14. Brekke C, Solberg AHS (2005) Oil spill detection by satellite remote sensing. Remote
\nSensing of Environment 95:1\u201313.
\n15. Williams DA, Brown SD, Crawford DL (2008) Contemporary and historical influences
\non the genetic structure of the estuarine-dependent Gulf killifish Fundulus grandis.
\nMar Ecol Prog Ser 373:111\u2013121.
\n16. Whitehead A, Pilcher W, Champlin D, Nacci D (2011) Common mechanism underlies
\nrepeated evolution of extreme pollution tolerance. Proc R Soc B, 10.1098\/rspb.2011.0847.
\n17. Clark BW, Matson CW, Jung D, Di Giulio RT (2010) AHR2 mediates cardiac teratogenesis
\nof polycyclic aromatic hydrocarbons and PCB-126 in Atlantic killifish (Fundulus
\nheteroclitus). Aquat Toxicol 99:232\u2013240.
\n18. Williams TD, et al. (2008) Transcriptomic responses of European flounder (Platichthys
\nflesus) to model toxicants. Aquat Toxicol 90:83\u201391.
\n19. Whitehead A, Triant DA, Champlin D, Nacci D (2010) Comparative transcriptomics
\nimplicates mechanisms of evolved pollution tolerance in a killifish population. Mol
\nEcol 19:5186\u20135203.
\n20. Varanasi U (1989) Metabolism of Polycyclic Aromatic Hydrocarbons in the Aquatic
\nEnvironment (CRC, Boca Raton, FL), p 341.
\n21. Incardona JP, et al. (2005) Aryl hydrocarbon receptor-independent toxicity of weathered
\ncrude oil during fish development. Environ Health Perspect 113:1755\u20131762.
\n22. Glickman MH, Ciechanover A (2002) The ubiquitin-proteasome proteolytic pathway:
\nDestruction for the sake of construction. Physiol Rev 82:373\u2013428.
\n23. Ohtake F, et al. (2007) Dioxin receptor is a ligand-dependent E3 ubiquitin ligase.
\nNature 446:562\u2013566.
\n24. Modig C, et al. (2006) Molecular characterization and expression pattern of zona
\npellucida proteins in gilthead seabream (Sparus aurata). Biol Reprod 75:717\u2013725.
\n25. Yu RMK, Wong MML, Kong RYC, Wu RSS, Cheng SH (2006) Induction of hepatic
\nchoriogenin mRNA expression in male marine medaka: A highly sensitive biomarker
\nfor environmental estrogens. Aquat Toxicol 77:348\u2013358.
\n26. Holth TF, et al. (2008) Differential gene expression and biomarkers in zebrafish (Danio
\nrerio) following exposure to produced water components. Aquat Toxicol 90:277\u2013291.
\n27. Sanchez BC, Carter B, Hammers HR, Sep\u00falveda MS (2011) Transcriptional response of
\nhepatic largemouth bass (Micropterus salmoides) mRNA upon exposure to environmental
\ncontaminants. J Appl Toxicol 31:108\u2013116.
\n28. Thomas P (1990) Teleost model for studying the effects of chemicals on female reproductive
\nendocrine function. J Exp Zool Suppl 4(Suppl 4):126\u2013128.
\n29. Greeley MS, Macgregor R (1983) Annual and semilunar reproductive-cycles of the
\nGulf killifish, Fundulus grandis, on the Alabama Gulf Coast. Copeia (3):711\u2013718.
\n30. Hicken CE, et al. (2011) Sublethal exposure to crude oil during embryonic development
\nalters cardiac morphology and reduces aerobic capacity in adult fish. Proc
\nNatl Acad Sci USA 108:7086\u20137090.
\n31. Whitehead A, Roach JL, Zhang S, Galvez F (2011) Genomic mechanisms of evolved
\nphysiological plasticity in killifish distributed along an environmental salinity gradient.
\nProc Natl Acad Sci USA 108:6193\u20136198.
\n32. Levine SL, Oris JT (1999) CYP1A expression in liver and gill of rainbow trout following waterborne
\nexposure: Implications for biomarker determination. Aquat Toxicol 46:279\u2013287.
\n33. Evans DH, Piermarini PM, Choe KP (2005) The multifunctional fish gill: Dominant site
\nof gas exchange, osmoregulation, acid-base regulation, and excretion of nitrogenous
\nwaste. Physiol Rev 85:97\u2013177.
\n34. State of Louisiana Department of Health and Hospitals (2011) Louisiana Seafood
\nSafety Surveillance Report (Louisiana Department of Health and Hospitals, Baton
\nRouge, LA).
\n35. Storey JD, Tibshirani R (2003) Statistical significance for genomewide studies. Proc
\nNatl Acad Sci USA 100:9440\u20139445.
\n36. Saeed AI, et al. (2006) TM4 microarray software suite. Methods Enzymol 411:134\u2013193.
\n37. Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of
\nlarge gene lists using DAVID bioinformatics resources. Nat Protoc 4:44\u201357.
\n38. Langmead B, Trapnell C, Pop M, Salzberg SL (2009) Ultrafast and memory-efficient
\nalignment of short DNA sequences to the human genome. Genome Biol 10:R25.
\n39. Anders S, Huber W (2010) Differential expression analysis for sequence count data.
\nGenome Biol 11:R106.
\n40. Rice CD, Schlenk D, Ainsworth J, Goksoyr A (1998) Cross-reactivity of monoclonal
\nantibodies against peptide 277-294 of rainbow trout CYP1A1 with hepatic CYP1A
\namong fish. Mar Environ Res 46:87\u201391.
\nWhitehead et al. PNAS Early Edition | 5 of 5
\nENVIRONMENTAL
\nSCIENCES
\nSCIENCE APPLICATIONS IN SPECIAL FEATURE
\nTHE DEEPWATER HORIZON
\nOIL SPILL SPECIAL FEATURE
\nSupporting Information
\nWhitehead et al. 10.1073\/pnas.1109545108
\nSI Methods
\nThe locations (latitude and longitude) of our field sampling sites
\nand dates for sampling at each site are summarized in Dataset S1.
\nGulf killifish (Fundulus grandis) were caught by minnow trap, and
\ntissues were excised immediately. Liver was preserved in RNAlater
\n(Ambion, Inc.) for genome expression (microarray and
\nRNAseq) analysis. Gill tissues were fixed in situ in buffered zincbased
\nformalin Z-Fix (Anatech LTD).
\nSatellite Imagery. Satellite imagery was analyzed to provide
\na coarse but spatially and temporally comprehensive estimation of
\nthe timing, location, and duration of coastal oil contamination.
\nSurface oil from the DWH oil spill was detected through the
\nanalysis of SAR images, which offer the most effective means of
\ndetecting oil remotely. This active radar system operates over
\nlarge spatial scales in all weather and at all times of day (1, 2). Oil
\ndampens the ocean\u2019s smallest capillary waves (3\u20135 cm in length),
\nyielding black regions in the image attributable to the total lack
\nof microwave backscatter from the sea surface to the sensor,
\ncompared with higher backscatter from surrounding regions with
\nwaves (Fig. S1). False-positive results are possible from areas
\nwith low wind (<3 m\/s) and from algal blooms; thus, the use of\nanother satellite sensor or \u201csea truth\u201d (e.g., wind measurements)\nis advisable for confirmation of the SAR signal. Only SAR images\nwith distinct signatures, unrelated to these potential artifacts,\nwere used in this study, although even thin oil sheens\nwould potentially yield a dark return because SAR data yield no\ninformation about oil thickness. We used SAR measurements\nfrom multiple satellites (TerraSARX; ERS-2; CosmoSkymed-1,\n-2, and -3; Radarsat-1 and -2; Palsar; and Envisat-2). Data were\nreceived and processed in real time at the University of Miami\nCenter for Southeastern Tropical Advanced Remote Sensing\n(CSTARS) laboratory and were further processed at the Louisiana\nState University Earth Scan Laboratory. The calculated\ndistance from each field sampling site to the nearest oil slick was\ncalculated from the \u201cstraight-line\u201d distance from the global positioning\nsystem position of the station (Dataset S1) to that of the\nobserved oil across any and all intervening geographical barriers\n(e.g., Fig. S1). Therefore, calculated distances do not necessarily\nrepresent the overall distance oil would have traveled to reach\na sample station, although as the calculated distance approaches\nzero, these two distances (straight line vs. travel distance) become\nextensionally equivalent.\nAnalytical Chemistry. Analytical chemistry of water, tissue, and\nsediment samples was performed to offer detailed characterization\nof exposure to contaminating oil (data reported in Dataset S2).\nSample dates and locations are summarized in Dataset S1. Two\nliters of water was collected subsurface in 1-L amber-glass jars from\neach sample site and date, and it was kept at 4 \u00b0C until extraction,\nwhich was performed within 1 wk of collection. Tissues (whole fish)\nwere collected from each of the field sites from the second (June\n2010) and third (August 2010) sampling time points and frozen at\n\u221220 \u00b0C until extraction. Sediment was collected from each of the\nfield sites after the final sampling time point (September 2010) in\n8-oz glass jars and frozen at \u221220 \u00b0C until extraction.\nThe sediment extraction procedure is as follows. Approximately\n30 g of sediment\/soil was accurately weighed (to the\nnearest 0.01 g) into a precleaned 500-mL beaker. The material\nwas homogenized with anhydrous sodium sulfate sample until\na \u201cdry\u201d sand-like matrix was created. One milliliter of surrogate\nstandard was spiked into the sample, followed by the addition of\n100 mL of pesticide-grade dichloromethane (DCM). The sample\nmixture was sonicated (60% intensity) for \u223c10 min and allowed\nto settle for 15 min. The solvent was poured over\na sodium sulfate funnel to remove any water and drained into\n500-mL flat-bottomed flasks. The extraction process was repeated\ntwo more times, followed by rinsing the funnel with 25\nmL of DCM. The flask was placed on a Buchi evaporative system\nand reduced to a final volume of 5\u201310 mL of DCM. The\nDCM concentrate was pipetted from the flask, placed into a 10-\nmL microextraction thimble, and reduced to a final volume of 1\nmL using a nitrogen blow-down system. The 1-mL extract was\ntransferred to a 2-mL autosampler vial and spiked with 10 \u03bcL of\ninternal standard solution. Autosampler vials were stored at 4 \u00b0C\nuntil ready for analysis.\nThe water extraction procedure is as follows. Approximately\n1,000 mL of water was accurately weighed (to the nearest 1.0 mL)\ninto a precleaned 20,000-mL separatory funnel. One milliliter of\nsurrogate standard was spiked into the sample, followed by the\naddition of 100 mL of pesticide-grade DCM. The sample mixture\nwas hand-shaken for \u223c10 min and allowed to settle for 15 min.\nThe solvent in the bottom of the funnel was drained through\na sodium sulfate funnel to remove any water and drained into\na 500-mL flat-bottomed flask. The extraction process was repeated\ntwo more times, followed by rinsing the funnel with 25\nmL of DCM. The flask was placed on a Buchi evaporative system\nand reduced to a final volume of 5\u201310 mL of DCM. The DCM\nconcentrate was pipetted from the flask, placed into a 10-mL\nmicroextraction thimble, and reduced to a final volume of 1 mL\nusing a nitrogen blow-down system. The 1-mL extract was\ntransferred to a 2-mL autosampler vial and spiked with 10 \u03bcL of\ninternal standard solution. Autosampler vials were stored at 4 \u00b0C\nuntil ready for analysis.\nThe tissue extraction procedure is as follows. Approximately 5\u2013\n10 g of tissue was accurately weighed to the nearest 0.01 g into\na precleaned 500-mL beaker. The material was homogenized\nwith anhydrous sodium sulfate sample until a dry sand-like matrix\nwas created. One milliliter of surrogate standard was spiked\ninto the sample, followed by the addition of 50 mL of pesticidegrade\nDCM. The sample mixture was sonicated (60% intensity)\nfor \u223c10 min and allowed to settle for 15 min. The solvent was\npoured over a sodium sulfate funnel to remove any water and\ndrained into a 250-mL flat-bottomed flask. The extraction process\nwas repeated two more times, followed by rinsing the funnel\nwith 25 mL of DCM. The flask was placed on a Buchi evaporative\nsystem and reduced to a final volume of 3\u20135 mL of DCM.\nThe DCM extract was exchanged to hexane with \u223c25 mL of\npesticide-grade hexane. The flask was returned to the evaporation\nsystem and evaporated down to a final volume of 2\u20135 mL of\nhexane. The sample was fractionated on an alumina\/silica gel\ncolumn by placing the 2- to 5-mL hexane aliquot on the aluminum\/\nsilica gel column, which was then rinsed with high-purity\nhexane. The flow of hexane was stopped before exposing the\nsilica gel to air. This fraction, which contained alkanes, was\ncollected in a graduated thimble. The alumina\/silica gel column\nwas then rinsed with 50% DCM and 50% hexane. The solvents\nwere allowed to elute completely in a separate extraction thimble.\nThis fraction contained the PAHs. The alkane and PAH\nfractions were combined and concentrated to 1.0 mL under\na gentle stream of nitrogen and stored in a 2-mL autosampler\nvial (4 \u00b0C) until GC\/MS analysis.\nAll sample extracts were analyzed using an Agilent 7890A Gas\nChromatography system (Agilent Technologies, Inc.) config-\nWhitehead et al. www.pnas.org\/cgi\/content\/short\/1109545108 1 of 4\nured with a 5% diphenyl\/95% (vol\/vol) dimethyl polysiloxane\nhigh-resolution capillary column (30 m, 0.25-mm inner diameter,\n0.25-\u03bcm film) directly interfaced to an Agilent 5975\ninert XL MS detector system (Agilent Technologies, Inc.). An\nAgilent 7638B series Auto Injector (Agilent Technologies, Inc.)\nwas used for sample introduction into the GC\/MS system. The\nGC flow rates were optimized to provide a required degree of\nseparation, which includes near-baseline resolution of n-C17\nand pristine, and baseline resolution of n-C18 and phytane. The\ninjection temperature was set at 250 \u00b0C, and only high-temperature\nand low-thermal bleed septa were used in the GC\ninlet. GC was performed in the temperature program mode\nwith an initial column temperature of 55 \u00b0C for 3 min, which\nwas then increased to 280 \u00b0C at a rate of 5 \u00b0C\/min and held for 3\nmin. The oven was then heated from 280 \u00b0C to 300 \u00b0C at a rate\nof 1.5 \u00b0C\/min and held at 300 \u00b0C for 2 min. Total run time was\n66.33 min per sample. The interface to the MS was maintained\nat 280 \u00b0C. Ultra-high-purity helium was the carry gas for the\nGC\/MS system.\nSpectral data were processed by Chemstation Software (Agilent\nTechnologies, Inc.). Analyte concentrations were calculated\nbased on the internal standard method. Therefore, an internal\nstandard mixture composed of naphthalene-d8, acenaphthened10,\nchrysene-dl2, and perylene-dl2 (usually at a concentration\nof 10 ng\/\u03bcL) was spiked into the sample extracts just before\nanalysis. The concentration of specific target oil analytes was\ndetermined by a five-point calibration and internal standard\nmethod. Standards containing parent (nonalkylated) hydrocarbons\nwere used in the calibration curve. Alkylated homologs\nwere quantified using the response factor of the parent, and\nwere therefore semiquantitative. This was the standard procedure,\nbecause alkylated standards were not available.\nGenome Expression: Microarrays. Genome expression across sites\nand time was characterized using custom oligonucleotide\nmicroarrays. Genome expression was measured in liver tissues\nfrom five replicate individual male fish per site-time treatment (5\nbiological replicates). Male fish were chosen for genome expression\nanalysis because sampling was conducted during\nspawning season, when female reproductive condition (and associated\nliver genome expression) can be highly variable.\nMicroarray probes (60-mer) were designed from contigs constructed\nfrom F. heteroclitus-expressed sequence tags. F. heteroclitus\nis the Atlantic coast-distributed sister species of Gulf coastdistributed\nF. grandis (3). Microarrays included probes for 6,800\nunique EST sequences, each printed in duplicate on 15,000 element\ncustom Agilent microarrays (design ID no. 027999) (Agilent\nTechnologies, Inc.). Total RNA was extracted using TRIzol\nreagent, antisense RNA (aRNA) prepared using the amino allyl\naRNA amplification kit (Ambion, Inc.), and purified aRNA\ncoupled to Alexa Fluor dyes (Alexa Fluor 555 and 647; Molecular\nProbes, Inc.), and it was hybridized to custom microarrays\nfor 18 h at 60 \u00b0C in a balanced loop design. Microarray images\nwere captured using a Packard Bioscience ScanArray Express\n(PerkinElmer, Inc.) microarray scanner, and images were processed\nusing Imagene (Biodiscovery, Inc.). Spots that were too\nbright (saturated) or too faint (below 2 SDs above background\nintensity) were excluded from normalization, resulting in a final\nset of 3,296 probes included for normalization and statistical\nanalysis (Dataset S3). Data were lowess-normalized and then\nmixed model-normalized using linear mixed models to account\nfor fixed (dye) effects and random (array) effects. Normalized\ndata were then analyzed using mixed model ANOVA, with \u201csite\u201d\n[Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point\n(BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort\nMorgan (FMA)] and \u201csampling time\u201d (sampling trips 1, 2, and 3)\n(Dataset S1) as main effects, including an interaction (site-bytime)\nterm. \u201cDye\u201d was considered a fixed effect, and \u201carray\u201d and\n\u201creplicate individual within site-time treatment\u201d (n = 5) were\ntreated as random effects. The false discovery rate was estimated\nusing Q-value (4). Principal components analysis was performed\nusing MeV (5). GO enrichment was tested using DAVID (6).\nGenome Expression: RNAseq. Transcript abundance was compared\nbetween liver mRNA from three replicate fish (RNA was not\npooled) from the GT site from June 28, 2010, and mRNA from\ntwo control samples. The two control samples are composed of\npooled liver mRNA from six and eight individuals, respectively,\ncollected in April 2008. The individuals for one control sample\nwere collected (2 each) from three sites west of the Mississippi\nriver, including Port Aransas, Texas; Cocodrie, Louisiana; and\nLeeVille, Louisiana. The individuals for the second control\nsample were collected (2 each) from four sites west of the Mississippi\nRiver, including Dauphin Island, Alabama; Weeks Bay,\nAlabama; Santa Rosa Island, Florida; and St. Teresa, Florida. All\nRNA samples were sequenced on the Illumina Gene Analyzer\nplatform (Expression Analysis, Inc.), and the resulting short-read\ndata were summarized in fastq format. Short reads with more than\ntwo uncalled bases were removed. Each read was cut whenever\na position fell below a minimum quality score of 10 or if the\naverage of the qualities of a position and its two neighbors fell\nbelow 20, and the largest remaining fragment was used.\nQuantitative transcript abundance analysis was initiated by\nmapping filtered short reads to target sequences (6,810 unique F.\nheteroclitus target EST sequences, Dataset S5) using the Bowtie\nshort read alignment software (7). A custom Perl script determined\nthe number of fragments mapped to each target sequence.\nThe Bioconductor package DESeq (version 2.8) (8) was\nthen used to determine statistical significance of each differentially\nexpressed target using a negative binomial method with\nP values adjusted by the Benjamini\u2013Hochberg procedure. The\nthree GT site samples were identified as a single \u201cExposed\u201d class\nto DESeq, and the two pooled samples were identified as a single\n\u201cControl\u201d class.\nGill Morphology and Protein Expression: Field Study. Male and female\nfish were sampled from all field sites for analysis of CYP1A\nprotein expression in the gills. Tissues were fixed immediately in ZFix,\nstored on ice, and held at room temperature before further\nprocessing. Gill tissues from at least three fish per site per sampling\ntime were dehydrated in ascending grades of histology-grade\nethanol. Tissues were then transferred to a t-butanol bath before\nclearing in Histochoice Clearing Agent (Amersco) and embedding\nin Paraplast (Sigma). Tissues were cut along the longitudinal axis at\na thickness of 4 \u03bcmusing an American Optical 820 microtome and\ntransferred onto poly-L-lysine\u2013coated microscope slides. After\nrehydration, tissues were processed for antigen retrieval by microwave\nin Tris-buffered saline (pH 9.0) and blocked. Tissues were\nthen probed with mAb C10-7 against fish CYP1A (9). Sections\nwere counterprobed using the Vectastain ABC immunoperoxidase\nsystem (Vector Laboratories), utilizing the ImmPACT Nova\nRED peroxidase substrate kit (Vector Laboratories) to visualize\nthe CYP1A protein in red. Tissue sections were counterstained\nwith Vector Hematoxylin QS (Vector Laboratories). Slides were\nthen observed with a Leica DM RXA2 microscope (Leica Microsystems),\nand images were captured with a Spot Insight 4\nmegapixel camera (Diagnostic Instruments). Representative images\nwere captured at a magnification of 40\u00d7.\nEarly Life-Stage Experiments. Approximately 20 L of water was\ncollected (in coordination with collection of water for analytical\nchemistry; Dataset S2) subsurface from field sites on the dates\nindicated in Dataset S1. Water was stored in airtight stainlesssteel\nsoda kegs and kept at 4 \u00b0C until experiments were conducted.\nWater samples from GT and BLB were utilized in laboratory\nexposures of F. grandis embryos obtained by in vitro\nWhitehead et al. www.pnas.org\/cgi\/content\/short\/1109545108 2 of 4\nfertilization using ova and spermatozoa collected from a brood\nstock of unexposed adult F. grandis derived from Cocodrie,\nLousiana before oiling. Cocodrie parental stock fish were\nmaintained at Louisiana State University, where they were held\nin the aquatics facility at the Department of Biological Sciences\nin 400-L tanks maintained at 17 parts per thousand (ppt) water\n(Instant Ocean) under recirculating conditions.\nFollowing fertilization, 20 embryos were randomly transferred\nin triplicate to one of the six field-collected waters (2 field sites \u00d7 3\ntime points) at 3 h postfertilization. Embryos were also exposed\nto a laboratory control consisting of artificial 17 ppt water.\nLarvae at 24 d postfertilization were sampled and fixed in Z-Fix\nsolution. After fixation, tissues were prepared, sectioned, and\nstained with the mAb C10-7, as described in the previous section.\n1. Brekke C, Solberg AHS (2005) Oil spill detection by satellite remote sensing. Remote\nSensing of Environment 95:1e13.\n2. Fingas MF, Brown CE (1997) Review of oil spill remote sensing. Spill Science and\nTechnology Bulletin 4:199e208.\n3. Whitehead A (2010) The evolutionary radiation of diverse osmotolerant physiologies\nin killifish (Fundulus sp.). Evolution 64:2070e2085.\n4. Storey JD, Tibshirani R (2003) Statistical significance for genomewide studies. Proc Natl\nAcad Sci USA 100:9440e9445.\n5. Saeed AI, et al. (2006) TM4 microarray software suite. Methods Enzymol 411:134e193.\n6. Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large\ngene lists using DAVID bioinformatics resources. Nat Protoc 4:44e57.\n7. Langmead B, Trapnell C, Pop M, Salzberg SL (2009) Ultrafast and memory-efficient\nalignment of short DNA sequences to the human genome. Genome Biol 10:R25.\n8. Anders S, Huber W (2010) Differential expression analysis for sequence count data.\nGenome Biol 11:R106.\n9. Rice CD, Schlenk D, Ainsworth J, Goksoyr A (1998) Cross-reactivity of monoclonal\nantibodies against peptide 277-294 of rainbow trout CYP1A1 with hepatic CYP1A\namong fish. Mar Environ Res 46:87e91.\nFig. S1. Representative measurements of the distance from field sites to ocean surface oil according to the CosmoSkymed2 SAR image captured May 13, 2010,\nat 11:56 UTC (Coordinated Universal Time). Field sites include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), Bayou La Batre (BLB), and Fort\nMorgan (FMA).\nFig. S2. Oil contaminating the marsh at the GT field site on June 16, 2010 (photograph by B.D.).\nWhitehead et al. www.pnas.org\/cgi\/content\/short\/1109545108 3 of 4\nDataset S1. Sites, precise locations, and sampling dates for three field sampling trips\nDataset S1\nDataset S2. Analytical chemistry of subsurface water samples, tissue samples (whole fish), and sediment samples\nDataset S2\nFig. S3. Expression divergence along principal component 1 (PC1) across consecutive sampling times for the subset of 380 genes that was dose-responsive to\nPCB exposure in a study by Whitehead et al. (1). Field sites include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), Bayou La Batre (BLB), and\nFort Morgan (FMA).\nDataset S3. Genome expression microarray data: All probes included in the analysis, including the target EST sequence, probe sequence, annotation, average\nexpression within each treatment (average of n = 5 replicate samples within each site-by-time treatment), and results from statistical analyses\nDataset S3\nDataset S4. Results of GO enrichment analysis using DAVID for the subset of genes that were divergently expressed at the GT site coincident with oil\ncontamination\nDataset S4\nDataset S5. Genome expression RNAseq data: All gene targets included in the analysis, including the target EST sequence, annotation, fold difference in\ntranscript abundance between the average of three replicate fish from GT sample time 2 (June 28, 2010) and two replicate reference RNA pools, and adjusted\nP values\nDataset S5\n1. Whitehead A, Pilcher W, Champlin D, Nacci D (2011) Common mechanism underlies repeated evolution of extreme pollution tolerance. Proc R Soc B, 10.1098\/rspb.2011.0847.\nWhitehead et al. www.pnas.org\/cgi\/content\/short\/1109545108 4 of 4\n\n\n<\/p>\n","protected":false},"excerpt":{"rendered":"