We investigated the response of a marine bivalve, the Pacific oyster, Crassostrea gigas, using a suppression subtrac-tive hybridization method to identify up- and down-regulated genes aft
Trang 1differentially regulated genes in the Pacific oyster
Crassostrea gigas in response to pesticide exposure
Arnaud Tanguy1, Isabelle Boutet1,2, Jean Laroche1and Dario Moraga1
1 Laboratoire des Sciences de l’Environnement Marin (LEMAR), UMR-CNRS 6539, Institut Universitaire Europe´en de la Mer, Universite´ de Bretagne Occidentale, Plouzane´, France
2 UMR CNRS-IFREMER 5171 ‘Ge´nome, Populations, Interactions, Adaptation’, Station Me´diterrane´enne de l’Environnement Littoral, Se`te, France
For several decades, coastal ecosystems have been
sub-jected to increased pesticide contamination, mainly
from agricultural practices For example, herbicide
compounds such as atrazine, diuron, isoproturon,
sim-azine, alachlor, metolachlor, and, more recently,
glyphosate, are widely applied to cereal crops and
reach coastal waters by runoff Moreover, the exposure
of animals in these ecosystems to pesticides is never
limited to a single pollutant, but to an assortment of
chemicals from a variety of sources whose interactions
could result in additive, synergistic or antagonistic
effects with regard to toxic outcome Pesticides, similar
to other xenobiotics, are metabolized by many
enzymes, including those of the cytochrome
P450-dependent monooxygenase system, flavin-containing
monooxygenase, prostaglandin synthetase, alcohol
dehydrogenase, esterases, and a variety of transferases [1,2] Pesticides are also known to alter the progression
of some human cancers by reducing immune defenses against cancer [3] They have also been reported to affect reproductive and developmental processes in wildlife, possibly by disrupting endocrine pathways [4–6] Some particular herbicides have been well stud-ied in terms of toxicology effects in various organisms Atrazine [2-chlor-4-ethylamino-6-isopropylamino-1,3,5-triazin] has a known toxicity to blood-forming organs and the immune system, and can induce the produc-tion of cytokinines such as interferon c or tumor nec-rosis factor a [7,8] In fish, atrazine affects different tissues, particularly liver tissue which shows a substan-tial increase in the size of lipid inclusions followed
by lipoid degeneration, enlargment of the secondary
Keywords
Crassostrea gigas; environment; gene
expression; pesticides; subtractive libraries
Correspondence
D Moraga, UMR-CNRS 6539 Laboratoire
LEMAR, Institut Universitaire Europe´en de
la Mer, Universite´ de Bretagne Occidentale,
Place Nicolas Copernic, F-29280 Plouzane´,
France
Fax: +33 2 98 49 86 45
Tel: +33 2 98 49 86 42
E-mail : Dario.Moraga@univ-brest.fr
(Received 3 August 2004, revised 5 November
2004, accepted 12 November 2004)
doi:10.1111/j.1742-4658.2004.04479.x
The effects of pesticide contamination on the metabolism of marine mol-luscs are poorly documented We investigated the response of a marine bivalve, the Pacific oyster, Crassostrea gigas, using a suppression subtrac-tive hybridization method to identify up- and down-regulated genes after a 30-day exposure period to herbicides (a cocktail of atrazine, diuron and isoproturon, and to the single herbicide glyphosate) A total of 137 unique differentially expressed gene sequences was identified, as well as their asso-ciated physiological process The expression of 18 of these genes was ana-lyzed by RT-PCR under laboratory experimental conditions The metabolic functions they are associated with include xenobiotic detoxifica-tion, energy producdetoxifica-tion, immune system response and transcription This study provides a preliminary basis for studying the response of marine bivalves to long-term herbicide exposure in terms of regulated gene expres-sion and characterizes new potential genetic markers of herbicide contami-nation
Abbreviations
AChE, acetylcholinesterase; SSH, suppression subtractive hybridization.
Trang 2lysosomes, mitochondrial malformation and
vacuoliza-tion, and a reduction in glycogen content [9]
Iso-proturon[3-(4-ospropylphenyl)-1,1-dimethylurea] is a
nonhalogenated, lipophilic-substituted phenylurea
her-bicide that is used to protect cereal crops It has a
strong affinity for interaction with membrane
phosp-holipids [10] It has been shown recently that high
concentrations of isoproturon could affect
metall-othi-oneins by reducing their metal content in the aquatic
oligochaete Tubifex tubifex [11] and could affect
enzyme activities in amphibians [12] Diuron
[3-(3,4-dichlorphenyl)-1,1-dimethyl-harnstoff] is a phenylurea
herbicide [13] used for pre- and post-emergence of
weeds in agriculture The toxicity of this herbicide has
mainly been studied in water phytoplankton [14] In
aquatic organisms, LC50(48 h) values for diuron range
from 4.3 to 42 mgÆL)1 in fish and range from 1 to
2.5 mgÆL)1 in aquatic invertebrates It was also shown
that its principal biodegradation product,
3,4-dichloro-aniline, exhibits a higher toxicity and is also persistent
in soil, water and groundwater [15] Significant
inhibi-tions (9–12%) of brain acetylcholinesterase (AChE)
activity were also observed in response to diuron in
the juvenile goldfish (Carassius auratus) [16]
Glypho-sate [N-(phosphonomethyl)glycine], also known as
‘Round-Up’, is a herbicide used to control grasses,
her-baceous plants, including deep-rooted perennial weeds,
brush, some broadleaf trees and shrubs, and some
conifers Glyphosate acts by preventing the plant from
producing an essential amino acid and by
inhibit-ing the enzyme enolpyruvylshikimate-phosphate
syn-thase which reduces the production of protein in the
plant, thereby inhibiting plant growth Studies showed
that glyphosate caused the appearance of myelin-like
structures in Cyprinus carpio hepatocytes, swelling of
mitochondria and disappearance of the internal
mito-chondrial membrane in carp at both exposure
concen-trations [17] However, little information about the
effect of glyphosate on marine invertebrate species is
available
Despite the vast commercial use of these herbicides,
there is little published data about their effects on
mar-ine molluscs In invertebrates, resistance to pesticides
by detoxification has previously been directly
correla-ted with some enzyme biomarkers, such as
mixed-func-tion oxidase activity, glutathione S-transferase activity,
AChE or other esterase activity [18,19] Aquatic
macro-invertebrates such as molluscs are widely used
in biological monitoring programs, but their use in
biomarker studies remains limited To identify
bio-markers of exposure, it is first necessary to identify
and describe the mechanisms of ecotoxicological
dam-age and those involved in the response of the
organisms to the pollutant In previous studies, we demonstrated that atrazine and isoproturon could select particular alleles at some enzymatic loci, such as adenylate kinase, phosphoglucomutase and phospho-glucoisomerase [20]
In this paper, we report regulated genes involved in the molecular response induced by herbicides in
C gigas As a first step, we identified the down and up-regulated genes after one month of exposure to a cocktail of globally employed herbicides using a sup-pression subtractive hybridization (SSH) method We then analyzed the regulation of specific herbicide-regulated gene expression
Results
Identification of herbicide-regulated genes Two forward and reverse SSH libraries were made from pooled digestive glands and gills of C gigas after
30 days of exposure to a herbicide cocktail of atrazine, diuron and isoproturon Two other forward and reverse SSH libraries were made from the same tissue types from oysters exposed to glyphosate for 30 days The search for homology using the blastx program revealed a total of 137 different sequences, including
56 sequences corresponding to known genes and 81 sequences corresponding to new expressed sequence tags The sequences obtained from the various SSH libraries are listed in four tables: 30-days, up- and down-regulated after exposure to an atrazine⁄ diu-ron⁄ isoproturon (ADI) cocktail (Tables 1 and 2), and 30-days, up-and down-regulated after exposure to glyphosate (Tables 3 and 4) These genes regulated by herbicide exposure can be assigned to six major cellu-lar physiological functions (a) Xenobiotic detoxifica-tion; (b) nucleic acid and protein regulation (including transcription, cell cycle regulation and metabolism of nucleic acid components); (c) respiration; (d) cell com-munication (including immune system and membrane receptors); (e) cytoskeleton production and mainten-ance; and (f) energy metabolism Several ribosomal proteins were also found in both the forward and reverse libraries from the two experiments There were more regulated genes from the libraries associated with exposure to the ADI cocktail than in the libraries asso-ciated with exposure to glyphosate, particularly among the down-regulated genes
Expression of herbicide-regulated genes The time-dependent expression of 13 genes that were up-regulated by exposure to herbicide and five that
Trang 3were down-regulated was analyzed by RT-PCR using
RNA from both the gills and digestive glands of
oys-ters after 0, 7, 14, 21 and 30 days of pesticide
expo-sure First, an RNA pool of the eight oysters collected
at each exposure time was used to identify the tissue in
which the differential expression when compared to
the control was significant RT-PCR was then carried
out on each oyster sample for the target tissues, to
estimate the variation in gene expression between
sam-ples A summary of these results is presented in
Table 5 and Fig 1 Among the 18 genes analyzed, only
two showed no clear regulation – fucolectin and ribo-somal L18A All of the other genes showed some regu-lation with exposure to herbicide The reguregu-lation of expression was strongly tissue-dependent for 10 genes, and most of these are chiefly regulated in the digestive gland (Table 5) Student’s t-test showed that despite clear patterns of increased or decreased expression at the different times of exposure, the differences in gene expression are most significant when we compare sam-ples exposed for 7 and 14 days with those exposed for 21–30 days A summary of the significant value of sta-tistical test is presented in Tables 6 and 7 Less signifi-cant differences can be explained by the variation seen
in expression among the samples In the ADI (Fig 1C) and glyphosate (Fig 1B) experiments, identified genes from the SSH libraries were expressed differentially compared to the control In the control oysters, no sig-nificant variation in gene expression was observed between samples from the different sampling periods, though high inter-individual variations were detected (Fig 1A)
Discussion
Despite the intensity of pesticide use along the marine coast, few studies have investigated the response to
Table 1 Up-regulated genes identified in the SSH libraries of the
atrazine ⁄ diuron ⁄ isoproturon experiment (after 30 days of exposure)
with significant database matches.
Homolog (protein)
BLASTX
value GenBank accession no Xenobiotics detoxification
Unknown function
Clone 44 unnamed human protein product 6e-15 CF369128
C380A1.2.2 Homo sapiens novel protein 0.048 CF369136
Cellular cycle, protein regulation and transcription
Hypertension-related
calcium-regulated gene (HSCR)
1e-18 CF369127
Respiratory chain
NADH dehydrogenase subunit 4 8e-94 AF177226
Metabolism
Cellular communication, membrane
receptors and immune system
b-1,3-Glucan binding protein 3e-6 CF369126
Putative senescence-associated protein 5e-46 CF369134
Cytoskeleton
Ribosomal proteins
CF369174
a Sequences with nonsignificant e-value (< 0.01) or with unknown
proteins.
Table 2 Down-regulated genes identified in the SSH libraries of the atrazine ⁄ diuron ⁄ isoproturon experiment (after 30 days of expo-sure) with significant database matches.
Homolog (protein)
BLASTX
value
GenBank accession no Unknown function
EbiP2667 Anopheles gambiae 5e-70 CF369178 CG1524-PC Drosophila melanogaster 8e-33 CF369179 Unnamed protein Mus musculus 6e-18 CF369181
EbiP2667 Anopheles gambiae 3e-10 CF369182 Cellular communication, membrane
receptors and immune system Tripartite motif protein TRIM2 2e-4 CF369175 Apoliphorin precursor protein 3e-15 CF369184 Transport protein Sec61 alpha subunit 7e-10 CF369183
Ribosomal proteins
CF369220
a Sequences with nonsignificant e-value (< 0.01) or with unknown proteins.
Trang 4exposure in marine organisms at the level of gene
tran-scription In this study, we characterized the response
of the Pacific oyster, C gigas, to herbicide exposure
under experimental conditions Using a suppression subtractive hybridization method, we obtained 137 unique partial sequences of cDNA (56 corresponding
to known genes) encoding proteins being transcribed
in oysters after 30 days exposure to herbicides Use of this method in conjunction with differential display PCR has previously identified 242 differentially expressed genes in the zebra mussel, Dreissena poly-morpha, treated with various contaminants such as Aroclor 1254, 3-methylcholanthrene, chrysene and atr-azine [21] Our results are difficult to compare with those obtained in D polymorpha because of the higher concentrations of pesticides (between 1 gÆL)1 and
2 mgÆL)1, according to the herbicide) and shorter time
of exposure (between 8 and 24 h) used by the authors
In another study, 258 differentially expressed genes were identified in C gigas in response to exposure to hydrocarbons for 7 and 21 days [22], and some of these genes were also present in the herbicide SSH lib-raries (including glutamine synthetase and cathepsin)
In our experiments, we tested a cocktail of the three more common herbicides detected in the French Atlan-tic estuaries and used concentrations that can be found
in seawater The experiment using only glyphosate was conducted to test the effect of this newly introduced herbicide in French ecosystems Glyphosate is known
to be not bioaccumulated, biomagnified or persisting
in a biologically available form in the environment Its mechanism of action is specific to plants and it is relat-ively nontoxic to animals [23] But several studies have demonstrated that glyphosate and more especially the surfactants used to increase its efficacy showed that glyphosate could be toxic for many organisms [24–26] Isoproturon appears not to bioaccumulate in molluscs [27], and the LC50 at 9 days in C gigas larvae has been estimated at 0.37 mgÆL)1 In acute tests, diuron was had limited toxicity to fish and invertebrates [28] Few results are available from chronic tests, especially for aquatic invertebrates The LC50(at 48 or 96 h) for diuron varied from 1 to 30 mgÆL)1 in fish and inver-tebrate species [29] Fish species data are primarily from acute exposures, and the lethal effects ranged from 2.8 to 31 mgÆL)1in 1–4 day exposures [28,30,31] Glyphosate is considered relatively effective with little
to no hazard to animals [32] However, at sublethal concentrations, glyphosate affected the reproduction and development of Pseudosuccinea columella snails [32] Glyphosate was also shown to be relatively non-toxic in certain animal species and presents virtually
no effects in some aquatic organisms (the 96 h LC50in rainbow trout, Oncorhynchus mykiss, and other fish ranges from 86 to 168 mgÆL)1, and the 48 h LC50was
780 mgÆL)1 in Daphnia) [33] The LC50 for oyster
Table 3 Up-regulated genes identified in the SSH libraries of the
glyphosate experiment (after 30 days of exposure) with significant
database matches.
Homolog (protein)
BLASTX
value
GenBank accession no Xenobiotics detoxification
Respiratory chain
NADH dehydrogenase subunit 4 8e-94 AF177226
NADH dehydrogenase subunit 5 6e-16 AF177226
Metabolism
Cellular communication,membrane receptors
and immune system
b-1,3-Glucan binding protein 3e-6 CF369126
Meningioma expressed antigen 5 7e-19 CF369226
Lipoprotein receptor related protein 5 2e-24 CF369227
Cytoskeleton
Ribosomal proteins
CF369244
a
Sequences with nonsignificant e-value (< 0.01) or with unknown
proteins.
Table 4 Down-regulated genes identified in the SSH libraries of
the glyphosate experiment (after 30 days of exposure) with
signifi-cant database matches.
Homolog (protein)
BLASTX
value
GenBank accession no Cellular cycle, protein regulation and transcription
Ribosomal proteins
Unknown genes (nine sequences) a CF369253 to
CF369261
a
Sequences with nonsignificant e-value (< 0.01) or with unknown
proteins.
Trang 5larvae has been estimated at more than 10 mgÆL)1[34].
Glyphosate closely resembles naturally occurring
sub-stances and does not possess chemical groups that
would confer great reactivity or biological persistence,
and its chemical properties indicate that it is not
bio-accumulate [35] Although primarily aimed at
reversi-bly inhibiting photosynthesis in plants [36], atrazine
has also been found to affect a variety of physiological
processes in aquatic animals Atrazine accumulation has
been seen in a number of tissues [37,38] In freshwater
invertebrates, atrazine affected hydromineral balance or
gill function in crabs [39,40], as well as hemocyanin
function [41] In fish it affects hematology [42,43] and
metabolism [41,44,45] More recently, low levels of
atra-zine have been shown to impair sexual development in
male frogs [6] In the mollusc C virginica, the 96 h
LC50 has been estimated at 1 mgÆL)1 The supposed
low level of toxicity of glyphosate to oysters, and the
weak concentration used in our experiment could
partly explain the difference observed in the number of
genes identified from the libraries (98 in the ADI SSH libraries vs 48 in the glyphosate SSH libraries)
In previous experiments studying the effect of atra-zine on fish, the authors showed that the concentra-tions quantified in water were about 70% less than the concentrations introduced in the tanks, probably due
to adsorption of atrazine to the surfaces of the tanks and possibly due to removal by the fish themselves [46] Moreover, the authors suggested that some of the atrazine added may be liable to biotic (e.g bacteria) and abiotic (e.g light) degradation and therefore that the physiological responses of the fish to atrazine may
be occurring at lower water concentrations than indi-cated Until a better understanding of atrazine dynam-ics can be applied to our experimental set-up, we prefer to discuss the physiological changes in C gigas
in relation to the concentrations of pesticides intro-duced in our tanks More, the fact that sea-water was changed every day in our tanks and the corresponding herbicide concentration was added at each water
Table 5 Summary of the results of expression studies in the two tissues used in SSH experiments.
Fig 1 Analysis of differential expression of up- and down-regulated genes in C gigas exposed to pesticides Expression of the gene studied
is presented as the calculated ratios ODCgGSII ⁄ OD28S after RT-PCR For each gene and each sampling time, the bar represents the aver-age value of gene expression (ratios ODCgGSII ⁄ OD28S) for the eight samples and the error bars correspond to the standard deviation for the eight samples at the sampling time considered (A) Expression of the 17 studied genes in the control samples 1, senescence associated protein; 2, Trim2; 3, lysosomal associated protein; 4, elongation factor 2; 5, hypertension-related calcium-regulated gene (HSCR); 6, apolipho-rin precursor protein; 7, glutamine synthetase; 8, ATP syntase; 9, coelomic factor; 10, RNA helicase; 11, guanyl cyclase receptor; 12, ribo-somal protein L13; 13, lipoprotein receptor related protein; 14, meningioma expressed antigen 5; 15, ADP-ribosylation factor 2; 16, procathepsin L (B) Expression of the nine studied genes in the glyphosate experiment 1, lipoprotein receptor related protein; 2, HSCR; 3, meningioma expressed antigen 5; 4, glutamine synthetase; 5, ATP synthase; 6, coelomic factor; 7, ADP-ribosylation factor-2; 8, procathepsin L; 9, ribosomal protein L13 (C) Expression of the 12 studied genes in the ADI experiment 1, senescence associated protein; 2, Trim2; 3, lysosomal associated protein; 4, elongation factor-2; 5, HSCR; 6, apoliphorin precursor protein; 7, glutamine synthetase; 8, ATP syntase; 9, coelomic factor; 10, RNA helicase; 11, guanyl cyclase receptor; 12, ribosomal protein L13.
Trang 62
4
6
8
10
12
T0 T7 T14 T21 T30
0
2
4
6
8
10
12
14
C
B
0
2
4
6
8
10
12
14
A
Trang 7change, allowed to maintain the oysters in a more or
less homogenous concentration of contaminants
Most of the genes we identified function in the
res-piratory chain, cell communication, the immune
sys-tem, or the regulation of protein or the cytoskeleton
Only a few are specific to xenobiotic detoxification
One, the glutamine synthetase gene was previously
shown to be associated with the response of various
organisms to pesticide exposure The response seen in
our libraries was that more genes are up-regulated by
herbicides than are down-regulated This pattern was
stronger in the ADI libraries compared to the
glypho-sate libraries Glyphoglypho-sate seemed not to have a strong
effect on C gigas metabolism However, some of the
genes involved in respiration and energy production
were highly expressed in response to herbicide
expo-sure no matter which pesticide was used Similar
results were observed in previous reports studying the
effect of other stress such as hydrocarbons [22] or
parasite infection [47], showing that any source of
stress generates an increase in energy production A
comparison of the two experiments shows that the
same cellular functions were affected by herbicides
with only a few genes in common to both the ADI
and glyphosate response SSH libraries These were
glu-tamine synthase, ATP synthetase, b-1,3-glucan binding
protein and some housekeeping genes such as tubulin
and actin
Among the 18 genes studied, 16 produced patterns
of differential time- or tissue-dependent expression
Most of the significant differences observed in gene
expression appeared after 21–30 days, suggesting that
at the concentrations used herbicides only affect oyster
metabolism after long periods of exposure Among all
the genes regulated was b-1,3-glucan binding protein,
also named coelomic factor, that has been widely
stud-ied in shrimp [48,49] and other crustaceans [50]
Lipo-polysaccharide and b-1,3-glucan binding protein
combine to form LGBP, a pattern recognition protein,
that plays an important role in the innate immune
response of crustaceans and insects This gene was
pre-viously identified in up-regulated SSH libraries from
C gigas exposed to hydrocarbons [22] and parasites
[47] The binding of lipopolysaccharide and
b-1,3-glu-can binding protein to form LGBP has been shown to
activate the prophenoloxidase cascade The
apolipo-phorin precursor protein we identified is also involved
in the activation of phenoloxidase in invertebrates [51]
The activation of the phenoloxidase cascade appears
to be a general response of C gigas to stress exposure
whatever its nature, abiotic or biotic
We also studied the expression of an RNA helicase
that is known to be a multifunctional protein involved
in various nuclear processes such as transcription, ribo-somal RNA biogenesis and RNA export Several tis-sue-specific RNA helicases have been described in the literature, such as the myocyte enhancer factor-2 pro-tein that acts as an inhibitor of cell proliferation and cardiomyocyte hypertrophy and may also be involved
in cell cycle progression [52] In previous studies, we also showed that this gene was up-regulated by expo-sure to hydrocarbons, and that its over-expression was greatest in the first week after PAH exposure [22]
We saw strong inhibition of guanyl cyclase receptor, particularly in the gills The inhibition of guanyl cyclase has been shown to prevent increased enzyme activity associated with the nitric oxide-mediated dam-age recovery process [53] Nitric oxide (NO) is a potent, bioactive molecule produced in the presence of
NO synthase, which constitutively mediates numerous physiological functions Over-production of NO (and NO-reaction products) can be induced, and typically leads to cell cycle arrest and apoptosis In vertebrate blood cells, an increase of extracellular levels of NO was detected after exposure to the herbicide paraquat [54] Other studies have likewise shown that pesticides could generate increasing NO levels which led to unre-paired damage caused by free radicals [55] Our experi-ments were not designed to quantify the amount of
NO reaction products; however, the strong down-regu-lation of guanyl cyclase receptor that was measured suggests an activation of physiological process involved
in free radical scavenging, especially NO, that could have been generated during the herbicide exposure in oysters
ATPase is a large family of genes coding for differ-ent enzymes that all participate in the formation of ATP The enzyme identified from our libraries could not be specifically identified from the partial sequence Nevertheless, over-expression of this gene was observed in both experiments and in both tissue types analysed after 15 days of pesticide exposure In another study, increased Mg2+-ATPase and Na+⁄ K+ -ATPase activity was detected in the liver and erythro-cytes of rats exposed to the insecticides malathion and anilofos [56] Similar results were obtained in snails exposed to fungicide and herbicides [57] Because ATPase is constitutively expressed and probably regu-lated by multiple environmental and⁄ or physiological factors, a more complete study of the regulation of its expression has to be done before its potential use as a biomarker for pesticide-exposure monitoring can be determined
Other interesting genes that were identified from our libraries are the lysosomal associated protein and pro-cathepsin L, a protein protease Both are involved in
Trang 8lysosomal functions Earlier studies have shown that
pesticides can affect lysosome membrane stability [58]
In insecticide-resistant Musca domestica strains, the
mechanism by which proteases may confer advantages
to insecticide resistant insects could involve providing
an increased supply of precursor amino acids from
proteolytic degradation products to the intracellular pool, prior to the de novo synthesis of detoxifying enzymes [59] We previously showed that the lysosomal associated protein was up-regulated in response to para-site exposure [47] suggesting that membrane stability seems to be a target for stress factors The lysosomal membrane destabilization is a common parameter used
to to investigate the impact of environmental pollution
in disturbed ecosystems and is considered as a general biomarker of stress [60,61]
We also identified the glutamine synthetase (GS) gene
as being up-regulated with pesticide exposure GS per-forms the fundamental functions of ammonia fixation and glutamine biosynthesis, but also plays a role in the central nervous system where it clears the excitatory neurotransmitters of glutamic acid [62] In Arabidopsis thaliana, GS was shown to be induced under amino acid starvation conditions as well as by exposure to the herbi-cide acifluorfen [63] In the fish, Cyprinus carpio, an increase in GS was observed after several days of expo-sure to cypermethrin At the same time, there was a decrease in the amount of free amino acids coincident with changes that occurred in the transamination pro-cess, again, related to the formation of nitrogenous end products [64] For the other differentially transcribed genes identified in our libraries, we found no informa-tion in the literature concerning their regulainforma-tion by pes-ticides Some of these genes, such as meningioma expressed antigen, hypertension-related calcium-regula-ted gene, and tripartite motif protein TRIM2, have been described as being regulated in cell proliferation [65] Their presence in our SSH libraries could suggest an effect of pesticides on these physiological processes Interestingly, no sequence corresponding to AChE was identified in our libraries The monitoring of AChE activity, particularly its inhibition, is commonly used as a biomarker of pesticide exposure in experi-ments on marine species such as the mussel, Mytilus
sp [66,67], the shore crab, Carcinus maenas [68] and other invertebrates [69] This result can be explained in several ways First, AChE has different distributions and physiological roles in different species [70,71], which results in a highly variable degree of inhibition associated with toxic effects [68] Also, AChE activity
is modulated by seasonal and nutritional variables [66] Finally, in bivalves, only a few studies have been pub-lished on the use of AChE activity as a biomarker of pesticide exposure, reflecting both low endogenous activity and a relative insensitivity to inhibition by these pollutants compared with other species [72–75] The results given here provide a preliminary basis for further study of detoxification processes and other physiological responses to herbicides in the marine
Table 6 Summary of the statistical test (Student’s t-test)
per-formed on gene expression in the glyphosate experiment The
gene expressions are compared by pair of sampling time *,
Signifi-cant value at 0.05%.
Lipoprotein receptor
T7
HSCR
T7
Meningioma antigen
T7
Glutamine synthetase
T7
T14
ATP synthase
T7
Coelomic factor
T7
T14
T21
ADP-ribosylation factor-2
T7
T14
T21
Procathepsin L
T7
Ribosomal protein L13
T7
T14
Trang 9bivalve, C gigas This is the first published investiga-tion into the mechanisms of these responses at the molecular level in this species We will focus our next efforts on a more complete study of the regulation of these genes with respect to pollutant concentration, their use in surveys of wild populations of oysters, and also in the search for functional polymorphism
Experimental procedures
Oyster conditioning and treatment
Adult Crassostrea gigas were collected from La Pointe du Chaˆteau (Brittany, France) After an acclimatization per-iod of 7 days in aerated 0.22 lm filtered seawater at
respectively), oysters were challenged as follows Two groups of 40 oysters were exposed to two experimental conditions One group was exposed to a cocktail of
(1 lgÆL)1), and isoproturon (0.5 lgÆL)1) that represent the three most used and toxic herbicides detected in all French Bay waters The other group is exposed to
2 lgÆL)1 of glyphosate, a herbicide that has been recently introduced in marine ecosystems, and for which no infor-mation about its toxicity at low concentration is known in marine molluscs Another group of 40 oysters was main-tained in aerated 0.22 lm filtered seawater without con-taminant as a control The experiment lasted for 4 weeks and no mortality was observed Herbicide concentrations were chosen based on data reported by the Bay of Brest monitoring program [76] and by the National Observation Network of IFREMER-France The concentrations used
in our experiment correspond to the highest concentra-tions observed in the various French Bay waters and cor-respond to about 1⁄ 100 of the LC50 values observed for the herbicides tested [77–79]
Suppression subtractive hybridization
After 4 weeks, total RNA was extracted from the digestive gland and gills of a pool of 10 control and 10 exposed
Table 7 Summary of the statistical test (Student test) performed
on gene expression in the atrazine ⁄ diuron ⁄ isoproturon experiment.
The gene expressions are compared by pair of sampling time.
*, Significant value at 0.05%.
Senescence associated protein
Trim2
T7
T14
Lysosomal associated protein
T7
Elongation factor-2
HSCR
T7
Apoliphorin precursor
T7
Glutamine synthetase
ATP syntase
T7
Coelomic factor
T7
T14
RNA helicase
T7
Guanyl cyclase receptor
Table 7 (Continued).
Ribosomal protein L13 T7
T14 T21
Trang 10oysters using RNAble (Eurobio, les Ulis, France)
accord-ing to the manufacturer’s instructions Poly(A+) mRNA
was isolated from total RNA using the PolyATtract
mRNA Isolation System (Promega, Madison, WI, USA)
according to the manufacturer’s instructions Both forward
and reverse subtracted libraries were made on 2 lg of
mRNA extracted from the 10 oysters collected after
30 days of exposure and pooled for RNA extraction
mRNA from the gill and the digestive gland (1 lg from
each) were used for the construction of SSH libraries
First and second strand cDNA synthesis, RsaI
endonuc-lease enzyme digestion, adapter ligation, hybridization and
PCR amplification were performed as described in the
PCR-select cDNA subtraction manual (Clontech, Palo
Alto, CA, USA) The differentially expressed PCR
prod-ucts were cloned into pGEM-T vector (Promega,
Madi-son, WI, USA) Two hundred white colonies per library
ampicillin) from which the vector was extracted using an
alkaline lysis plasmid minipreparation, and screened by
size after digestion A total of 250 clones from forward and
(Sciencetech, Lincoln, NE, USA) and Thermo Sequenase
Pri-mer Cycle Sequencing Kit (APri-mersham Bioscience, Uppsala,
Sweden) and a AB3100 sequencer (PerkinElmer, Boston,
MA, USA) and Big Dye Terminator V3.1 Kit
(PerkinEl-mer) All sequences were subjected to a homology search
through the blastx program (http://www.ncbi.nlm.nih.gov/
BLAST/)
Pesticide detoxification gene expression analysis
by semiquantitative RT-PCR
Total RNA was extracted from the gill and digestive gland of eight control and eight exposed oysters at days 0, 7, 15, 21 and 30 days using a method based on extraction in guani-dium isothiocyanate [80] For each oyster, 50 lg of total RNA were reverse transcribed using the oligo(dT) anchor
and M-MLV reverse transcriptase (Promega, Madison, WI, USA) Amplification of 18 regulated genes was carried out using cDNA from both the control and exposed samples in
primer sequences for the genes studied are listed in Table 8 For an internal PCR control, 28S ribosomal DNA was amplified under the same conditions with sense (5¢-AAGG GCAGGAAAAGAAACTAAC-3¢) and antisense (5¢-GT TTCCCTCTAAGTGGTTTCAC-3¢) primers The resulting PCR products were electrophoresed in a 0.5· TBE ⁄ 1.5% ag-arose gel, and visualized with UV light after staining with ethidium bromide To minimize differences in RT efficiency, template for all the PCRs were done on 2 lg of the same reverse transcribed product The number of PCR cycles nee-ded to show differential expression between the control and exposed samples was the same (40 cycles) for each gene except for the 28S, where 25 cycles were used to avoid band intensity saturation for optical determination Band intensi-ties were quantified using the gene profiler software (ver-sion 4.03, Scanalytics, Inc., Lincoln, NE, USA)
Table 8 Sequences of the primers used in the expression study.
Glutamine synthetase ADI and G (up) 5¢-GTGCATCAAAGAATTTTGGATAC-3¢ 5¢-TGCAATAATTTTTGAAGCCCCGG-3¢ ATP syntase beta ADI and G (up) 5¢-AGAGAAGTGGCAGCTTTCGCTCAGTTTGG-3¢ 5¢-TTAGCATCTGTGGCCTCTGTGATTTGTCC-3¢ Coelomic factor ADI and G (up) 5¢-CTCGGCAAAGAAACCGCTGGTTCCTCCCA-3¢ 5¢-GCCCCTACCATAACATAGAGGACCCCTGG-3¢ RNA helicase 2 ADI (up) 5¢-GAGACGTCCAGGAAATCTTCCGCAACACC-3¢ 5¢-CAAATCTACCTGCACGAGCCACTCTGTGC-3¢ Procathepsin L G (up) 5¢-CAGAGTGTGCACTAGCATGCGGTCCCGT-3¢ 5¢-CACAACCTGGTCGCCGACCGCGGGGACT-3¢ Meningioma
associated protein
G (up) 5¢-AGGTGTCCTAGATACCGGCCATGTACCA-3¢ 5¢-TGGACACCTTAGAGACGGTGGCCAGAC-3¢ Lipoprotein receptor G (up) 5¢-AGCCTTGATGAGCCAAGGGCAGTGACCT-3¢ 5¢-GGCCCGACGGGTGTCTCTCCAGACCCGT-3¢ Fucolectin ADI and G (up) 5¢-CATGGCTTCGAATTGATCTTGGAGCTGT-3¢ 5¢-TAACCTCCAAAAACTTGCACGTCGGCAA-3¢ RiboL18A ADI and G (up) 5¢-ATACCGTGACCTGACATCCGCTGGTGCT-3¢ 5¢-GCACAGTTCCCCTACAGTCCCGCTTTAG-3¢
Senescente
associated prot
ADI (up) 5¢-TTGCAACGACTGCAGTCATCAGTAGGGT-3¢ 5¢-GAGCTCAGCGAGGACGGAAACCTCGCGT-3¢ Lysosomalassociated
protein
ADI (up) 5¢-CCAATCAGGTAGGCCTTCATGGAGAGGA-3¢ 5¢-CCCAGAGATCCTCCAAGAGACAGCCAGT-3¢ Elongation factor2 ADI (up) 5¢-ATCTGGAGAGCACATCATTGCTGGTGCA-3¢ 5¢-CTTTCTGGCCTCTCCAACATCCATGCCA-3¢ Apolipophorin ADI (down) 5¢-ACATCGAGGAAGAGTTTTCTATCCTGGA-3¢ 5¢-ATGCCAAGGTAGTTTATGATGATCGAGA-3¢ Tripartite motif
protein TRIM2
ADI (down) 5¢-ACATCGCTGAGAATGTCAACGGGGATAT-3¢ 5¢-TCTCCTACGATCATCTCACCGTCACCGA-3¢ Guanyl cyclase ADI (down) 5¢-GGTTGTCAATTTATTGAATGATCTCTACA-3 5¢-CCTCGGATCTTGAGACCCTCGACTGGA-3¢