Methylmercury Activates Enhancer-of-Split and Bearded Complex Genes Independent of the Notch Receptor
Advance Access publication March
Methylmercury Activates Enhancer-of-Split and Bearded Complex Genes Independent of the Notch Receptor
Matthew D. Rand 1 2
Christin E. Bland 1 2
Jeffrey Bond 0 1
0 Department of Microbiology and Molecular Genetics , Bioinformatics Core , College of Medicine, University of Vermont , Vermont 05405 , USA
1 The Author 2008. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved
2 Department of Anatomy and Neurobiology, College of Medicine, University of Vermont , USA
Methylmercury (MeHg) is a persistent environmental toxin that has targeted effects on fetal neural development. Although a number of cytotoxic mechanisms of MeHg have been characterized in cultured cells, its mode of action in the developing nervous system in vivo is less clear. Studies of MeHg-affected rodent and human brains show disrupted cortical and cerebellar architecture suggestive of mechanisms that augment cell signaling pathways potentially affecting cell migration and proliferation. We previously identified the Notch receptor pathway, a highly conserved signaling mechanism fundamental for neural development, as a target for MeHg-induced signaling in Drosophila neural cell lines. Here we have expanded our use of the Drosophila model to resolve a broader spectrum of transcriptional changes resulting from MeHg exposure in vivo and in vitro. Several Notch target genes within the Enhancer-of-split (E(spl)C) and Bearded (BrdC) complexes are upregulated with MeHg exposure in the embryo and in cultured neural cells. However, the profile of MeHg-induced E(spl)C and BrdC gene expression differs significantly from that seen with activation of the Notch receptor. Targeted knockdown of Notch and of the downstream coactivator Suppressor of Hairless (Su(H)), shows no effect on MeHg-induced transcription, indicating a novel Notch-independent mechanism of action for MeHg. MeHg transcriptional activation is partially mimicked by iodoacetamide but not by N-ethylmaleimide, two thiol-specific electrophiles, revealing a degree of specificity of cellular thiol targets in MeHginduced transcriptional events.
Methylmercury (MeHg) is a persistent environmental toxin
that targets the developing nervous system of the fetus and of
young children. Cytotoxicity of MeHg stems from its high
affinity for cysteine thiol groups on proteins
number of studies in cultured cells and in vivo have
documented effects of MeHg on fundamental cellular processes
including intracellular calcium ion homeostasis
, protein phosphorylation (Parran et al., 2003),
generation of reactive oxygen species (ROS)
cell cycle control
(Ponce et al., 1994)
, induction of apoptosis
(Wilke et al., 2003)
and disruption of protein synthesis and
(Cheung and Verity, 1985; Miura et al.,
. These pleiotropic effects of MeHg reflect its ability to
interact with a number of essential proteins that are pivotal to
the above-mentioned cellular processes.
Less clear are mechanisms of MeHg toxicity that are specific
to developing neural tissues in vivo. Fetal Minimata Disease,
a condition resulting from high-level MeHg exposure akin to
the catastrophic exposures that occurred in Minimata, Japan
and Iraq, is characterized by atrophy and disorganization of
cortical layers, loss of cells in the cerebellum and cerebrum,
and ectopic cells in the white matter of the cortices in human
brains, as well as in primate and rodent brains
(Choi et al.,
1978; Kakita et al., 2002; Peckham and Choi, 1988)
. It is of
note that cell death is not a hallmark of prenatal MeHg
exposure and that aberrant cell migration is thought to
contribute to the abnormal patterning of cellular layers
et al., 1978)
. Thus, the neuropathology is consistent with the
notion that MeHg can redirect the program of normal neural
development, presumably through altering cell–cell signaling
events. In support of this hypothesis, studies in embryonic
carcinoma cells demonstrate MeHg-induced changes in
expression levels of Eph and Ephrin, a receptor–ligand signaling
pair important for axonal guidance
(Wilson et al., 2005)
addition, MeHg alters neurotrophin signaling through the TrkA
receptor in the neuronal PC12 cell line
(Parran et al., 2004)
and in developing rat brains, which may parallel disruption of
(Barone et al., 1998)
. Aside from these
examples, our understanding of the signaling pathways targeted by
MeHg during neural development is incomplete.
In an effort to identify the fundamental developmental
signaling pathways targeted by MeHg, we have begun to
exploit Drosophila as a model. Using Drosophila-derived
neural cell lines, we previously identified the Notch receptor
pathway as a target for MeHg-induced signaling
. Notch signals are a means of cell–cell
communication used to orchestrate cell fate decisions and early
morphogenic activities of neural cells
et al., 1999; Skeath and Doe, 1998; Udolph et al., 2001)
stimulation by its ligand, Delta, the Notch receptor is activated
by sequential cleavages resulting in an intracellular domain
product that signals in the nucleus
(Kopan et al., 1996; Struhl
and Adachi, 1998)
. Nuclear Notch, in complex with the
Suppressor of Hairless (Su(H)) transcription factor, upregulates
transcription of genes in the Enhancer of Split complex
(E(spl)C) and Bearded complex (BrdC)
(Lai et al., 2000;
Wurmbach et al., 1999)
. The E(spl)C contains 13 genes
encoding three types of proteins: m3, m5, m7, m8, mb, md, mc
encode basic helix-loop-helix (bHLH) transcriptional repressor
proteins; ma, m2, m4, and m6 encode Bearded family-related
proteins; and m1 encodes a putative protease inhibitor
(Delidakis and Artavanis-Tsakonas, 1992; Klambt et al.,
1989; Wurmbach et al., 1999)
. The Bearded complex encodes
six proteins: Bearded (Brd), twin of m4 (Tom), Ocho, and
Brother of Bearded (Bob) A, B and C genes
(Lai et al., 2000;
Leviten et al., 1997)
. Brd proteins are small basic amphipathic
helical proteins that are antagonistic toward Notch signaling
(Apidianakis et al., 1999; Bardin and Schweisguth, 2006; Lai
et al., 2000)
. Notch activity induces expression of these target
genes in several developmental contexts to exert control of cell
fate decisions, most notably in the embryonic neurectoderm
(Bardin and Schweisguth, 2006; Zaffran and Frasch, 2000)
a previous study we demonstrated that MeHg promotes
upregulation of E(spl)mc and mb genes in several Drosophila
cell lines, suggesting that MeHg activates Notch signaling
(Bland and Rand, 2006)
. We also identified a potential role for
MeHg in promoting Notch receptor cleavage, consistent with
the current model for activation of the receptor
In this study we expand our use of Drosophila to further
characterize MeHg influence on Notch signaling. Using
microarray analysis of embryonic transcripts and quantitative
real-time PCR (qPCR) analysis of neuronal cell line messenger
RNA (mRNA), we show an increased expression of several
E(spl)C and BrdC target genes in response to MeHg exposure.
Unexpectedly, we find that activation of E(spl)C and BrdC
target genes by MeHg does not require the Notch receptor or
the downstream transcriptional coactivator Su(H).
Furthermore, MeHg activation of E(spl)C and BrdC target genes is
partially mimicked with the thiol-alkylating reagent
iodoacetamide (IA), whereas a second thiol-specific reagent,
N-ethylmaleimide (NEM), fails to induce Notch target genes.
Our results support the hypothesis that MeHg acts through
specific thiol-containing protein targets to regulate
transcriptional activation of E(spl)C and BrdC genes.
Fly stocks and MeHg treatments. All treatments and assays were
performed with the standard laboratory wild type Canton S strain. MeHg was
administered through additions to food preparations. Food consisted of
a cornmeal, molasses, and agar mixture (Jazz Mix AS153, Fisher Scientific),
prepared in batches and distributed in culture vials or in bottles. Methylmercury
chloride (Aldrich 442534) was prepared as a 50mM stock solution in dimethyl
sulfoxide (DMSO) and was added to the warm food mixture before it solidified.
For embryo exposures a mating population of > 300 flies were starved
overnight and then fed on control (DMSO) or 5–20lM MeHg food for 5 days.
Embryos were collected in population cages on grape juice-agar plates with
yeast paste and developmentally staged by aging for appropriate times to allow
nervous system development (e.g., 18 h at 18 C for stage 14) prior to fixation
for immunostaining or lysis for RNA isolation or total mercury analyses.
Total mercury determination and development assays. Larval hatching
rates were determined on embryos from control- and MeHg-treated flies.
Embryos were transferred to a new grape juice plate in batches of 50. Hatching
of first instar larvae was determined manually under a stereo dissecting
microscope 24 h after transfer of embryos to the new plate.
For mercury determinations, pooled embryos (> 50 embryos per sample)
were solubilized with 50 ll of tetramethyl ammonium hydroxide (Alfa Aesar,
Ward Hill, MA) and total mercury in each tube was determined by
a combustion-trapping-atomic absorption technique
by the Trace
Element Research Laboratory (Texas A&M, College Station, TX). Samples
were analyzed with a Milestone DMA 80 Hg analyzer for quantification of Hg
over the range 0.001-0.700 lg. Instrument calibration utilized dry certified
reference materials as standards. Laboratory quality control samples included
a method blank and certified reference materials (National Research Council
Canada DOLT-2 and DORM-2). The method blank was below the detection
limit of 0.00004 lg Hg, and the reference material recoveries were 107 and
99% of the certified values, respectively.
Immunostaining. Embryos were dechorionated with 2% bleach for 3 min,
rinsed thoroughly, and fixed in 4% paraformaldehyde 50% heptane by vigorous
shaking for 18 min. Vitelline membranes were removed by manually shaking in
methanol:heptane 50:50 and washed and stored at 4 C in methanol until
staining. For immunostaining, embryos were permeablized in 0.5% triton
X-100 in phosphate buffered saline (PBST). Subsequent blocking, primary and
secondary antibody incubations were done in PBST with 5% normal donkey
serum. Primary antibody was rabbit anti-horseradish peroxidase (Jackson Labs,
West Grove, PA) and secondary antibody was horseradish peroxidase (HRP)–
conjugated anti-rabbit (ECL, GE Healthcare, Piscataway, NJ) for
diaminobenzidine substrate detection using a standard protocol. Embryos were
visualized by brightfield microscopy methods and recorded with digital
imaging using a Spot One camera and software (MVI, Avon, MA).
RNA isolation and microarray analyses. Total RNA was isolated from
dechorionated embryos pooled and homogenized in TRIzol reagent (Invitrogen,
Carlsbad, CA). RNA was DNase treated (Ambion, Austin, TX) and processed
for probe generation and hybridization to ‘‘Affy’’ GeneChips, Drosophila
Version 2, by the National Institutes of Health Neuroscience Microarray
Consortium (NIH NMC). Data sets are posted at the NIH NMC data repository
Affymetrix data set analyses were performed using tools made available
through the Bioconductor Project
(Balasubramanian et al., 2004)
. Probe set
statistics were calculated from probe intensities (.CEL files) using the
(RMA; Bolstad et al., 2003)
(Workman et al., 2002)
, and median polish summary
statistic RMA; Bolstad et al., 2003).
A linear model
of the RMA-like probe set expression statistic
that included batch and MeHg exposure factors was used to calculate the
response associated with the treatment, M, as well a p value, p, reflecting the
probability of the response under the null hypothesis. It is generally recognized
(Yang and Speed, 2003)
that for small sample sizes, it is not advisable to
partition genes based exclusively on either M (because different probe sets
exhibit different variances) or p (because of the high false discovery rate with
small sample sizes). Instead it has proven useful to use both of these differential
expression statistics simultaneously; for example, via Boolean filters or
Volcano plots. We identified the 5% most differentially expressed genes based
on the joint distribution of M and log(p) under the null hypothesis, obtained by
(1) permuting sample labels, (2) binning the results, and (3) contouring the
resulting density (see Fig. 2D).
Gene Ontology Annotation
(GOA, Camon et al., 2004)
, the DAVID 2.0
functional annotation tool (http://david.abcc.ncifcrf.gov/home.jsp)
et al., 2003)
, and Gentleman’s GOHyperG
(Falcon and Gentleman, 2007)
procedure were used to identify probe sets associated with a biological process,
molecular function, or cellular component.
Variability was observed in the signal intensity for individual probe sets
among replicate microarray data sets. Of the 10 Notch-related genes that were
upregulated with MeHg only four (m4, ma, mc, and m8) gave p values < 0.05
among the three replicate Genechip data sets. For example, where E(spl)mc
gave replicates of 2.05-, 2.03-, and 1.31-fold increase and p ¼ 0.021, BobA
showed high variability with replicates of 6.98-, 1.62-, and 1.23-fold increase
with p ¼ 0.94. Statistical significance was therefore obtained through combined
analyses of M and p values and annotation based analyses using GO terms as
explained and justified above. It is likely the variability in this system comes
from the developmental complexity and dynamics of mRNA expression in the
tissue source (i.e. the whole embryo). In addition, the point at which MeHg
accumulates to a toxic threshold during embryo development is likely to vary
from treatment to treatment and among embryos of a single treatment. This
notion is supported by the observation that a variety of neural phenotypes result
within a single treatment population of MeHg-exposed embryos (see Fig. 1C).
Cell culture and treatments. The Drosophila central nervous system
(CNS) cell line ML-DmBG2-C6 (C6)
(Ui et al., 1994; Ui-Tei et al., 1995)
(available at the Drosophila Genome Resource Center) were routinely
maintained at 23–25 C in Sang’s M3 medium (JRH Biosciences, Lenexa,
KS) containing 12.5% fetal calf serum (Atlanta Biologicals, Lawrenceville,
GA), 10 lg/ml insulin, bactopeptone (2.5 g/l) and yeastolate (1 g/l) (13 BPYE)
supplement (Difco, Sparks, MD), 100 units/ml penicillin, and 100 lg/ml
streptomycin (Invitrogen, Carlsbad, CA). Cells were grown to 60–90%
confluence in M3 medium with serum. Media was removed, and cells were
washed with serum-free M3 medium. Cells were then cultured in serum-free
M3 media with various concentrations of MeHg or DMSO-solvent control, for
0–18 h as indicated. Dose selection was based on our earlier observations of the
effectiveness of 10lM MeHg in the C6 cell line
(Bland and Rand, 2006)
number of in vitro studies using insect or mammalian cells have investigated
MeHg effects at 0.1–50lM concentration, which show a wide variety of effects
with respect to the toxic endpoints in question and durations of exposure
(Braeckman et al., 1997; Morken et al., 2005; Sanfeliu et al., 2001; Shanker et
, Although difficult to extrapolate our dose levels in Drosophila to
a relevant dosage in affected human brains, concentrations of 1–10 ppm (~5–
50lM on a wet weight basis) MeHg have been documented in human brains
showing definitive morphological abnormalities (Choi et al., 1978). Cells were
harvested by direct lysis in TRIzol reagent and subsequently processed for
Notch activation in C6 cells was stimulated directly by exposure to
ethylenediaminetetraacetic acid (EDTA)
(Krejci and Bray, 2007; Rand et al.,
. Cells were treated with a 5-min pulse of 5mM EDTA in PBS buffer followed
1-h incubation in complete M3 medium. Cell were harvested by direct lysis in
TRIzol reagent and subsequently processed for RNA isolation and qPCR analyses.
Stock solutions (100mM) of IA (Biorad #163-2109, Hercules, CA) and
NEM (Calbiochem #34115, Gibbstown, NJ) were freshly prepared in water just
prior to use. Cell treatments were carried out in Robb’s Drosophila PBS
(DPBS) (2mM Na2HPO4, 0.35mM KH2PO4, 50mM NaCl, 40mM KCl, 0.5mM
CaCl2, 1.25mM MgCl2, 1.25mM MgSO4, 50mM sucrose, 5mM glucose), at
pH8.3 for IA and pH6.8 for NEM. Cells were washed twice with Robb’s DPBS
and incubated for 1 h with various concentrations of IA or NEM, followed by
2-h incubation in complete M3 medium. Cell were harvested in TRIzol reagent
and subsequently processed for RNA isolation and qPCR analyses.
RNA interference (RNAi) used to knockdown expression of Notch and
Su(H) was done essentially as described previously
(Bland and Rand, 2006;
Clemens et al., 2000)
. For Su(H) RNAi double stranded RNA encompassing
the 5# coding region (bp 1067–1579) was synthesized from a PCR DNA
FIG. 1. Mercury accumulation, larval hatching and nervous system defects
in maternally dosed embryos. (A) Total mercury in embryos from female flies
fed the indicated concentration of MeHg-containing food. (B) Percent of
embryos hatching to larva subsequent to feeding of female flies on indicated
concentration of MeHg. Total number of embryos counted for each
determination is indicated above each bar. (C) Immunostaining with anti-HRP
reveals all neural cells in the ventral nerve cord (VNC) of embryos. (I–III)
Stage 16 (18 h) embryos from wild type Canton S females fed on
MeHgcontaining food (20lM) for 5 days prior to embryo collection. Arrows in III
denote regions devoid of neural cells. Control embryos (Cont.), show a regular
segmental pattern of condensed neurons in the VNC. The ‘‘ladder-like’’
formation of the longitudinal tracts and anterior and posterior commissural
bundles (‘‘rungs’’) is clearly developed. See text for discussion.
template with terminal T7 sequence amended primers. Primer design was
assisted by reference to the GenomeRNAi web site (http://www.dkfz.de/
signaling2/rnai/) and utilized probe ID HFA03445.
Quantitative PCR. qPCR was done using SYBR-Green JumpStart Taq
ready mix (Sigma, St Louis, MO) on an ABI PRISM 7500 Fast Sequence
Detection System. Primer sequences for E(spl) m1, m2, m3, m7, mb, mc, md,
BobA, Brd, Notch, Su(H), and the RP49 control gene can be found in
supplemental data. Primers for E(spl)m1, m2, m3, m7, and md are derived from
Krejci and Bray (2007)
. The linear response of amplification with each primer
set was validated using template dilutions, and single product amplification was
determined by gel analysis and melt curve determination. Gene expression
levels were determined by the comparative CT method
(Livak and Schmittgen,
and compared between control- and MeHg-treated samples.
Western blotting. Cell lysates were prepared and separated by sodium
dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and
transferred to polyvinylidene flouride PVDF membranes for Western blot analyses
by standard protocols and as previously described
(Bland and Rand, 2006)
Antibodies used included: monoclonal C17-9C6 anti-dNotch intracellular
domain (gift from Spyros Artavanis-Tsakonas, Harvard Medical School,
also available at the Developmental studies hybridoma-bank, http://dsub.
biology.uiowa.edu/, and polyclonal anti-Su(H) (Santa Cruz Biotechnology,
#sc15813). Secondary IRDye700 and 800 secondary antibodies (Rockland,
Gilbertsville, PA) were used for development on a LiCor Odyssey (Li-Cor,
Lincoln, NE) infrared scanner.
Delivery and Developmental Effects of MeHg in the
We previously demonstrated the ability of MeHg to activate
expression of two Notch target genes, E(spl) mc and mb, in
cultured Drosophila cells
(Bland and Rand, 2006)
with the hypothesis that MeHg can stimulate Notch receptors.
We sought to further characterize this activity, first by examining
whether a similar effect of MeHg occurs in vivo. We approached
this by examining changes in transcripts in whole embryos
exposed to MeHg (see below). We first set out to validate
whether MeHg exposure can be achieved in Drosophila
embryos. Dosages were introduced by culturing female flies
on MeHg-containing food for 5 days prior to embryo collection
(see ‘‘Methods’’). Total mercury content in embryos from
MeHgfed female flies was measured to determine the possibility that
maternally ingested MeHg accumulates in the embryo. Total
mercury amounts were determined in lysates of a pool of > 50
embryos using combustion-trapping atomic absorption (see
‘‘Methods’’). We observed a linear increase in total mercury
accumulation in embryos with respect to MeHg concentration in
the food (Fig. 1A). To estimate the concentration of MeHg in
the embryo we assume that the total Hg is essentially all MeHg due
to the stability of MeHg. By approximating the embryo volume
from its average dimensions (500 lm 3 180 lm) and its
nearellipsoid geometry, we calculate an estimated volume of 8.4 nl per
embryo. With 0.12 ng Hg/embryo achieved at the 20lM food
concentration, an approximate 3.5-fold increase (71.4lM) in
MeHg is seen in the embryo relative to the food.
Embryos exposed to a maternal feeding of MeHg showed
developmental defects as seen by a decrease in larval hatching
rates and malformation of the nervous system. From
controltreated females, 98.5% of embryos successfully hatched to
larvae (Fig. 1B). A significant decrease in hatching was
observed with embryos from MeHg-fed females, with 82.5%
hatching (16% decrease) on 10lM food and 52.7% hatching
(42% decrease) on 20lM food (Fig. 1B). As an additional line
of evidence that MeHg was reaching the embryo, the gross
morphology of the embryonic central nervous system was
assayed using the pan-neural specific antibody anti-HRP. This
antibody recognizes a surface antigen expressed on all neurons
(Sun and Salvaterra, 1995)
. Compared with
control embryos, treatment of embryos with MeHg results in
distinct developmental defects as seen by abnormal morphology
of the nervous system (Fig. 1C). MeHg-exposed embryos show
a range of phenotypes including expanded regions of
HRPpositive cells (Fig. 1CI) or alternatively, regions where neurons
fail to form (Fig. 1CIII, arrows). These two phenotypes suggest
that neurogenic pathways (such as Notch in the latter case) are
targets for MeHg. As well, deficits in formation of axon tracts
(not shown) and an overall ‘‘twisted’’ formation of the CNS
(Fig. 1CII) was observed indicative of additional developmental
signaling pathways being disrupted. The percent of embryos
demonstrating an abnormal neural phenotype was found to
correlate with the concentration of MeHg in the food: 5lM
MeHg food produces 3.4% (39/1099) of embryos with a neural
phenotype, whereas 20lM MeHg food gives a neural phenotype
in 45% (370/798) of embryos. Overall, these data confirm that
embryonic MeHg exposure, which results in developmental
abnormalities, can be achieved through maternal dosage of the
fly. The diversity of neural phenotypes we have observed so far
fall into discrete classes that implicate several distinct pathways,
including Notch, as targets for MeHg and will be the subject of
a separate investigation.
Transcript Profiling Identifies the Notch Pathway as a MeHg
Target in the Embryo
A whole-genome transcriptional profiling approach was
taken to determine genes that are up- and downregulated in
expression subsequent to MeHg exposure in the embryo. RNA
from embryos of female flies fed either control or 20lM MeHg
food was analyzed. This concentration of MeHg was chosen
because it gives a significant fraction of neural phenotypes by
immunostaining (see above). Three replicates of control and
MeHg treatments were analyzed for a total of six Affymetrix
GeneChip data sets.
The results show that the experimental system of transcript
profiling provides power for detecting differentially expressed
genes. The median standard deviation of the RMA gene
expression statistic is 0.13 (Fig. 2A). A threshold of |DRMA| ¼ 1
(customarily taken to represent twofold change in expression)
is therefore ~8 times the median standard deviation. For the
data set the median standard deviation corresponds to
a coefficient of variation of < 9% in the expression signal
intensity. A histogram shows that for most genes, the p values
are consistent with the null hypothesis that MeHg treatment
does not ubiquitously alter transcription (Fig. 2B, 97% of probe
sets are not differentially expressed). We identified 287 ‘‘hits’’
that change at least 1.5-fold up or down (more precisely,
|DRMA| > 0.59). Of these, the signal for 74 probe sets is
upregulated and 213 probe sets are downregulated (see Fig. 3
and Supplementary Data).
We find a nonrandom association of Notch signaling
pathway genes with the pool of differentially expressed genes
with MeHg exposure. Of the 286 differentially expressed hits,
146 are associated with a biological process. GOA
(Camon et al.,
associates 42 genes with the Notch Signaling Pathway GO
term represented among the > 18,000 probe sets on the
Affymetrix GeneChips. Ten of these 42 Notch genes are
differentially regulated with MeHg exposure (Fig. 2C), with
a significant (p < 4 3 10 9) association between differential
expression and the Notch Signaling Pathway as determined using
Gentleman’s Bioconductor GOstats package
et al., 2004)
. A Volcano plot for the entire data set is illustrated in
Figure 2D. The contour illustrated demarks the 5% quantile under
the null hypothesis. Superimposed on the plot are the points for
the 42 Notch pathway genes identified by GO. The skewed
distribution of the Notch-related genes toward the ‘‘right’’ of the
plot reflects the significant increase of several Notch pathway
probe sets with MeHg treatment. The 10 Notch family genes
showing upregulation and lying outside the 5% quantile contour
are illustrated (Fig. 2D).
The Notch-related genes upregulated with MeHg are
E(spl)ma, mc, md, m4, m5, m8, Bearded (Brd), Brother of
Bearded (BobA), Ocho, and Bigbrain (Bib) (Fig. 3A).
Remarkably, all but one of these Notch pathway genes share
the common property of being a downstream target for Notch
receptor activation and comprise nearly half of the 19 Notch
target genes contained in the E(spl)C and BrdC complexes. It is
of note that the E(spl)C and BrdC loci are separated by a large
distance, being located on different arms of chromosome 3
(Fig. 3B). The Bib gene, which encodes an aquaporin-related
protein, has genetic interactions with the Notch pathway,
however, its regulation by Notch signaling is not well
(Doherty et al., 1997)
. Identification of these
targets is consistent with our previous studies in cell-based
assays, which similarly shows upregulated transcription of the
E(spl)mc target gene with MeHg exposure in Drosophila
(Bland and Rand, 2006)
Additional Responses in Transcript Profiles with MeHg
Aside from the Notch pathway, annotation clusters with
enrichments greater than 1.5 as identified by the DAVID
functional annotation tool include: genes with lipid transporter
activity (p < 5.5 3 10 6), thiol-specific antioxidants (p < 8.8 3
10 4), and defense response (p < 1.2 3 10 3). The lipid
transporter-related genes identified correspond with transcripts
for CG6300, CG4830, and CG11407, that are identified by
sequence similarity and have not been functionally
characterized. The antioxidant gene with the greatest differential
expression is the peroxiredoxin (Prx) 2540 (CG11765), for
which an increase was seen in two independent probe sets. The
increase in Prx probe sets also contributes to the annotation
cluster of defense response. Other defense response probe sets
showing upregulation are the heat shock proteins Hsp70A and
B (CG18743 and CG6489, respectively). Notable annotation
clusters identified by probe sets with downregulated intensity
include insect cuticle genes (p < 1.0 3 10 11), proteases of the
C1 (cysteine)-type (p < 2.1 3 10 4), and EF-hand containing
calcium binding proteins (p < 1.9 3 10 3).
Concentration and Time-Dependent Activation of E(spl)mc
Results from microarray analyses of MeHg-treated embryos
highlight the potential for a broader range of Notch targets
(BrdC genes in addition to E(spl)C genes) to respond to MeHg
than previously observed. We therefore sought to determine
more precisely the spectrum of Notch targets that respond to
MeHg stimulation. Although effective in producing a global
perspective on the transcriptional response in vivo, the embryo
presents a complexity of cell types and developmental
dynamics that could confound interpretation of transcriptional
sensitivity in specific tissues, e.g. developing neural tissues.
We therefore turned to the C6 neural-derived cell system with
which we have considerable experience in the more
quantitative method of qPCR
(Bland and Rand, 2006)
. We refined our
experimental conditions by first analyzing the time- and
concentration-dependent MeHg response of a representative
target, the E(spl)mc mRNA (mc). Treatments of 10lM MeHg
induce a time-dependent increase in mc with a 14-fold rise in
expression seen at 18 h (Fig. 4A). Closer examination of early
time points shows a significant increase of mc at 3-h treatment.
We next examined various concentrations of MeHg at a 3-h
exposure time and found a concentration-dependent increase
with a nearly 15-fold increase in expression of mc with 50lM
MeHg (Fig. 4B). To account for potential changes in receptor
expression we examined the level of Notch mRNA. We find
that Notch transcript levels remain essentially unchanged over
the time and concentration ranges examined (Figs. 4A and 4B),
indicating changes in E(spl)mc mRNA were not merely
resulting from upregulation of Notch expression.
It remained possible that the increase in E(spl)mc mRNA
with MeHg was occurring by one of two mechanisms: an
increase in mRNA synthesis or, alternatively, inhibition of
constitutive degradation of the mRNA transcript. To
distinguish between these two possibilities we examined the mc
transcriptional response to MeHg in the presence versus
absence of actinomycin D (AMD), an inhibitor of RNA
. We see that AMD
completely abolishes the MeHg-induced increase of mc
transcript in C6 cells (Fig. 5). These data confirm that MeHg
functions by inducing transcription of the mc gene and is
therefore likely to exert a similar effect upon the other E(spl)C
and BrdC genes. In addition, these data define an appropriate
treatment regimen to analyze increases in other E(spl)C and
BrdC transcript levels in response to MeHg in these cells.
MeHg Exposure and Notch Activation Differentially Activate
Transcription of E(spl)C and BrdC Genes
Our previous model proposed that MeHg activates Notch by
stimulating cleavage of the receptor
(Bland and Rand, 2006)
This model predicts that stimulation of the Notch receptor and
MeHg exposure would result in an identical pattern of Notch
target gene activation. We therefore examined the profile of
expression of several representatives of the E(spl) and Brd
family genes in C6 cells where endogenous Notch is stimulated
directly as compared with MeHg exposure. Notch receptor
activation is achieved with a brief exposure to the
calciumchelating agent EDTA, which causes dissociation of the Notch
heterodimer and acute activation of the receptor (Rand et al.,
2000). This approach has proven a useful tool in analysis of
Notch activation of E(spl) genes in cultured Drosophila and
(Krejci and Bray, 2007; Rand et al., 2000)
A comprehensive analysis of E(spl) and Brd genes using qPCR
was restricted by the ability to generate valid primer sets for
this quantitative method, preventing us from analyzing all
E(spl) and Brd genes (e.g., m4, Ocho were excluded).
Nonetheless, we found a robust upregulation of several E(spl)C
genes in response to EDTA activation, notably a 95-fold
increase in the m3 gene (Fig. 6A). As well, the m7 and mc
transcripts show a marked increase of 55-fold and 38-fold
respectively (Fig. 6A). In contrast, the md gene is minimally
responsive (twofold increase, Fig. 6A). Also, there are
insignificant changes in the BrdC genes, BobA and Brd, with
Notch activation (1.3- and 1.2-fold, respectively, Fig. 6A). To
confirm that the EDTA response relies exclusively on the
Notch receptor, we examined cells where Notch expression is
knocked down with RNAi. We see that EDTA response of m3
and mc expression is effectively abolished with Notch RNAi
(inset Fig. 6A), confirming that the profile of EDTA
stimulation of E(spl)C and BrdC expression directly reflects
the activity of the Notch receptor.
In contrast to Notch activation, MeHg shows a markedly
different profile of E(spl) and Brd gene activation with
a relatively small induction of m3 (sixfold) and the greatest
stimulation of md (35-fold) (Fig. 6B). In addition, a significant
increase in the BrdC genes (BobA and Brd) is seen with MeHg
(six to eightfold increase, Fig. 6A), again differing from Notch
activation where BobA and Brd fail to be activated. Altogether,
these data demonstrate MeHg-induced transcription of
conventional Notch target genes is distinct from activation of the
Notch receptor itself.
MeHg Activation of Notch Target Genes does not Require
The contrasting profile of E(spl)C and BrdC gene expression
with MeHg and Notch activation prompted us to reinvestigate
the role of the Notch receptor, as well as core components that
propagate Notch signaling in the MeHg response. We
examined the requirement for the Notch receptor directly by
knocking down Notch expression with RNAi. First, we observe
in control cells that 3 h of 50lM MeHg treatment shows no
change in Notch mRNA transcript levels (Figs. 4A and 4B,
7B). With RNAi treatment, Notch transcript levels can
effectively be reduced to < 30% of control and even greater
reduction at the protein level (Figs. 7A and 7B). We find that
knockdown of Notch shows no effect on MeHg stimulation of
mc expression (Fig. 7C). This is in contrast to the robust effect
Notch RNAi has on suppressing EDTA activation of E(spl)
targets (see Fig. 6A inset).
To further examine a possible role for the Notch pathway,
we performed targeted RNAi knockdown of Su(H), the
essential transcriptional coactivator that is required for
Notchdependent E(spl) expression. First, we observed that 3 h of
50lM MeHg alone caused a significant reduction in the Su(H)
transcript level (Fig. 8A). RNAi treatment similarly achieved
~50% reduction in Su(H) mRNA, which was further knocked
down (~90% reduction) in combination with MeHg (Fig. 8A).
Reduction in transcript level correlated with significant
reduction in Su(H) protein expression (Fig. 8B). Despite
substantial knockdown of Su(H), an insignificant reduction of
MeHg-induced mc expression was observed (Fig. 8C).
Altogether, these data demonstrate that activation of mc by
MeHg does not require the Notch receptor or Su(H), two
essential core components of the Notch signaling pathway.
MeHg Activation of Notch Target Genes is Likely to Operate via a Thiol-Selective Mechanism
The observation that MeHg acts independently of the Notch pathway to induce Notch target genes raised questions as to the
FIG. 9. Effects of IA and NEM on E(spl) and Brd gene activation. C6 cells
were treated with indicated concentration of IA or NEM as described in the
methods and E(spl) or Brd gene expression was determined by qPCR (mean
values, n ¼ 2). Expression levels of E(spl) mc (A) and E(spl) m3 (B) in
response to various concentrations of IA or NEM. (C) Expression of E(spl) or
Brd genes in response to 80lM IA relative to no IA control (see ‘‘Methods’’).
specificity with which MeHg acts in this potentially novel
mechanism. Because MeHg has a high affinity for protein thiols
, we asked whether a similar activation of
E(spl)C and BrdC genes could be achieved with two common
thiol-specific alkylating reagents, IA and NEM. Although these
compounds are both considered thiol-specific, they react via
distinct mechanisms and have previously been shown to
selectively modify different thiol-containing proteins
et al., 2006)
. Exposure of C6 cells to IA results in a
concentrationdependent increase in both mc and m3 expression (Figs. 9A and
9B). In contrast, NEM at the same or higher concentrations fails
to significantly induce mc or m3 expression (Fig. 9 and data not
shown). Further analysis shows upregulated expression of all the
E(spl) and Brd genes in a pattern that shows both similarity and
differences with that of MeHg induction (Fig. 9C). As with
MeHg, IA upregulated mc and m3 to similar levels relative to
each other; however, unlike MeHg, these two targets were the
most highly induced (Fig. 9C). Again, similar to MeHg but
distinct from Notch activation, the Brd and BobA genes were
upregulated with IA (Fig. 9C). As well, downregulation of Notch
with RNAi showed no effect on subsequent induction of E(spl)
and BrdC gene expression with IA exposure (data not shown),
indicating that IA similarly works through a Notch-independent
mechanism. In contrast to MeHg, md is not as highly responsive
to IA exposure. Overall, IA demonstrates a specific activity in the
induction of E(spl) and Brd genes confirming the sensitivity of
expression of these Notch targets to cellular thiol modification.
MeHg Induces E(spl)C and BrdC Genes Independent of Notch
We show that MeHg is a direct activator of several E(spl)C
and BrdC genes, the well documented targets of the Notch
signaling pathway. In the developing Drosophila embryo
MeHg exposure upregulates expression of six transcripts of the
E(spl)C (md, m4, ma, m5, mc, and m8) and three transcripts of
the BrdC complex (BobA, Brd and Ocho). Similarly, we find
MeHg induces expression of seven E(spl) genes (m1, m2, m3,
m7, mb, mc, and md) and two Brd genes (BobA and Brd) in
cultured Drosophila neural cells. All of these genes are
characterized as having Su(H) binding sites in their upstream
regulatory regions, a hallmark of Notch responsive genes
(Nellesen et al., 1999)
. Yet we demonstrate that MeHg induces
expression of E(spl)C and BrdC genes independent of the
Notch receptor and the Su(H) transcription factor, pointing to
a novel mechanism for transcriptional activation by MeHg. The
fact that distinct profiles of activation of E(spl)C and BrdC
gene are seen with MeHg versus Notch stimulation further
supports the hypothesis that MeHg induces transcription
through a novel mechanism. It is of note that the lack of
effect of Notch RNAi seen here differs significantly from our
previously reported effect on Notch RNAi on MeHg activation
(Bland and Rand, 2006)
. However, our former study
examined effects of a longer exposure regimen (16-h treatment
with 10lM MeHg), which may engage Notch in secondary
regulatory responses controlling mc expression by mechanisms
distinct from the acute effects seen in this study.
Of the E(spl)C and BrdC genes, the activity of the bHLH
encoding genes is best understood. In Drosophila, loss of
function of the bHLH E(spl) genes, most commonly achieved
through disabling Notch signaling, disrupts neural
development resulting in hypertrophy of the embryonic CNS
et al., 1994; Lieber et al., 1993)
. This activity of E(spl)s
predicts that MeHg could elicit direct effects on neural
development via altering E(spl) levels. Although several
vertebrate E(spl) bHLH homologs
(the HES and HERP genes
reviewed in Iso et al., 2003)
have been identified, a direct
functional correlation between individual Drosophila and
vertebrate E(spl) homologs has not been well characterized.
However, the HES genes display similar control over neuronal
differentiation in mice
(Ohtsuka et al., 1999)
. A similar
transcriptional activity of MeHg on HES genes in a mammalian
system is currently under investigation.
The ability of MeHg to act as a transcriptional activator
remains an important point of interest. Two studies suggest
MeHg can support transcription by acting at the level of
RNAPolII function or infuencing nucleosome stability. MeHg
can stimulate RNAPolII activity in isolated nuclei of HeLa
cells in a selective manner that is not achieved with other
organic and inorganic mercury compounds
(Chao and Frenkel,
. Furthermore, NEM is only weakly effective at inducing
(Chao and Frenkel, 1983)
, consistent with
our results. Alternatively, MeHg may serve to stabilize
transcriptionally active conformations of nucleosomes, as
demonstrated by the use of orgonomercurial supports to isolate
transcriptionally active fragments of chromosomes
ChenCleland et al., 1993
). Both of these mechanisms predict MeHg
would globally upregulate transcriptional activity in the cell.
However, two observations support the notion that MeHg is
acting with a higher degree of specificity toward individual
regulatory regions of genes: first, our whole-genome transcript
analyses show that a very limited number of transcripts change
levels in response to MeHg in embryos (Fig. 2B); next, we see
that some E(spl) genes respond to MeHg to a greater extent
than others (e.g., md vs. mb in C6 cells, Fig. 6B), suggesting
that upstream regulatory regions of certain genes harbor
selective ‘‘MeHg-responsive’’ elements.
This latter observation focuses attention on the upstream
regions of individual E(spl) and Brd genes, which have been
the topic of a number of studies. A recent analysis of
conservation of E(spl) regulatory regions among nine species
(Maeder et al., 2007)
binding sites for as many as 27 distinct transcription factors
distributed across the various E(spl) genes, with notable
heterogeneity between individual E(spl) genes. Our data
suggests that MeHg induces the association of one or more
transcription factors with their cognate binding sites in the
E(spl) and Brd genes, thereby stimulating transcription.
Alternatively, MeHg may cause dissociation of repressors in
these regions, similarly upregulating expression.
Distinguishing the factors involved in mediating MeHg-induced
transcriptional activation warrants a more in-depth analysis of the E(spl)
and Brd regulatory region, which will be addressed in future
experiments. Overall, these data implicate a potentially direct
mechanism for MeHg to influence transcription at specific loci
that are critical in neurogenesis.
Thiol-Selective Mechanisms of Transcriptional Activation
Additional MeHg-Sensitive Pathways in the Embryo
By comparing MeHg, IA, and NEM exposure to C6 cells, Maternal dosage of flies results in a 3.5-fold increase in
a profile of thiol-specific activation of E(spl)C and BrdC genes MeHg in embryos (20lM MeHg food concentrations give
is revealed. Similar to MeHg, IA induces mc expression in ~70lM MeHg in the embryo). At these levels of MeHg we
a concentration-dependent manner that is not seen with NEM. observe a significant drop in embryonic hatching rate, which
We infer from these results that MeHg and IA have over- correlates with the appearance of a number of neural
lapping protein targets. Previous studies show that IA and phenotypes. These concentrations are within the same range
NEM can be differentiated by their thiol-selective reactivity where pronounced effects of MeHg on transcription in C6
toward cellular protein substrates
(Dennehy et al., 2006)
. This neuronal cells are observed. It is reasonable to predict that
difference likely reflects the fact that IA is a smaller aliphatic a similar transcriptional response should occur in the embryo
electrophile that reacts via a different chemistry than the with this exposure level. Yet the level of MeHg-induced
bulkier heterocyclic dicarbonyl NEM
(Dennehy et al., 2006)
. It E(spl)C and BrdC gene expression in embryos, as determined
follows that the relatively small size of IA and MeHg may by probe hybridization methods in microarrays, differs
contribute to an ability to modify the same proteins. However, significantly from that seen in C6 cells using the qPCR
a broader examination of the E(spl)C and BrdC genes induced methodology. Some of this discrepancy is likely due to the
by IA demonstrates similarities and differences in transcrip- fundamentally different sensitivities of the respective
technitional activation compared with MeHg. IA stimulates Brd and ques. However, differences in expression between embryos and
BobA expression in a similar manner as MeHg. In contrast to cells are more likely to stem from the diversity of cell types in
MeHg, IA only modestly stimulates md expression. These the embryo in contrast to the relatively homogeneous C6 cell
results indicate that there is a level of specificity in MeHg line. This observation reinforces the importance of
investigatreactivity that is not shared with IA. This difference may reflect ing MeHg effects in the context of the developing organism.
a selective reactivity toward cysteine thiols based on the Nonetheless, the fact that E(spl) and Brd genes are targeted in
composition of the neighboring amino acids in the protein, both contexts strongly supports the overall hypothesis that
a factor previously reported to influence the specificity of IA a fundamental underlying MeHg-sensitive mechanism
reguversus NEM reactivity
(Dennehy et al., 2006)
. lates their expression.
With respect to cellular thiol modification, a particularly We observe changes in several transcripts in MeHg-exposed
relevant thiol-responsive element is the Nrf2/Keap1system, embryos indicating a number of other pathways are potentially
which regulates transcription in response to electrophiles and affected, consistent with the observation that several
(Copple et al., 2008)
. MeHg has been shown to types arise in MeHg-exposed embryonic nervous system. For
activate this pathway in mammalian cells (Toyama et al., example, elevation of heat shock proteins (HSP70) in embryos
2007). Interestingly, a biotinylated form of iodacetamide shows with MeHg indicates a response shared with other stress
a preferential activation of this pathway as compared with paradigms examined in the Drosophila model, including
a bitotinylated maleimide
(Hong et al., 2005)
, in agreement temperature and starvation.
(Landis et al., 2004; Sorensen
with the differential effects of IA and NEM we observe here. It et al., 2005)
. As well, upregulation of HSPs is seen with metal
is of great interest that a homologous Nrf2/Keap1 signaling ion exposure in Drosophila cells (Bournias-Vardiabasis et al.,
pathway has recently been described in Drosophila (Sykiotis 1990). Heat shock proteins act as chaperones, aiding in protein
and Bohmann, 2008), presenting the opportunity to investigate folding in under conditions of stress. Interestingly HSP70 has
MeHg effects on this important antioxidant pathway in a fly been shown to suppress neurodegeneration in polyglutamine
model. disease and preserve locomotor function under stress
An important implication stemming from our observations ditions
(Feder et al., 1996; Klose et al., 2005)
. Thus, one
of thiol-selective transcriptional activity is the potential for possibility is that HSPs counter the effects of MeHg in
other environmental oxidants or electrophiles to stimulate disrupting thiol disulfide bond formation in newly synthesized
E(spl)C, BrdC, or other genes via modification of the same proteins. Although induction of HSPs is a potential marker for
thiol targets. IA shares a thiol-reactive chemistry with aliphatic MeHg toxicity (Sacco et al., 1997), a functional role for HSPs
epoxides and alkyl halides
(Dennehy et al., 2006)
, suggesting in attenuating MeHg toxicity has yet to be resolved, but
of that the latter may function similarly to invoke transcription. remains a testable hypothesis in the Drosophila system.
Our current observations and our system provide the rationale An additional family of genes upregulated in MeHg-exposed
and an experimental model by which to compare, for example, embryos are the perioxiredoxins (Prx), a highly conserved
inorganic mercury (Hg2þ) and other organomercurials such as family of proteins that exhibit thiol-dependent peroxidase
ethylmercury (Thimerosal) with MeHg in influencing tran- activity and that are critical to maintaining intracellular redox
scription. Overall, these data support the hypothesis that status
(Rhee et al., 2005)
. Drosophila carry five paralogs of Prx
a discrete set of thiol-presenting proteins are involved in
(Radyuk et al., 2001)
. Two of these members are of the
1a MeHg-sensitive mechanism that regulates gene expression. Cysteine class (Prx2540, Prx6005) and three of these members
(Prx 4156, 4783, 5037) are of the 2-Cysteine class
et al., 2001)
. Curiously, we see that with MeHg exposure, only
the 1-Cys class is affected with upregulation of Prx2540 (Fig.
3A) and downregulation of Prx6005 (see Supplementary Data).
Prx genes are characterized as playing a role in defense
response to ROS
(Rhee et al., 2005)
. It is less clear why
Prx6005 is downregulated; however, this result highlights
a potential mechanism of differential expression in this gene
family in response to MeHg. A role for Prx in attenuating
MeHg toxicity has not been demonstrated directly.
A number of microarray probe sets show downregulation of
transcripts with MeHg and represent genes annotated as insect
cuticle proteins by sequence similarity (Supplemental Data).
This observation is consistent with the notion that cuticle
formation is compromised in late embryos and leads to our
observed decrease in larval hatching of MeHg-exposed
embryos. Another possibility is that decreased cuticle gene
expression is a downstream effect of altered Notch signaling
because epidermal cell fate is compromised with altered Notch
(Hoppe and Greenspan, 1990)
The observed decrease in cysteine protease gene expression
with MeHg exposure is difficult to interpret. The fact that
mercury compounds can bind and inactivate cysteine proteases
(Muller and Saenger, 1993)
would predict a compensatory
increase in expression of these gene transcripts, contracy to
what we observe. In contrast, the observed downregulation of
calcium binding proteins, particularly EF-hand proteins, may
stem from the ability of mercury to substitute for Caþþ ions in
activation of EF-hand proteins
(e.g., calmodulin; Chao et al.,
. Overall, our microarray data are indicative of the
complexity of events downstream of MeHg exposure in the
developing embryo and highlight the necessity to evaluate
transcriptional changes with MeHg in additional contexts and
at various developmental stages to deduce the most universally
To conclude, our results demonstrate that MeHg can regulate
transcriptional activation of E(spl)C and BrdC genes
independent of the Notch pathway in Drosophila. Our study also
presents evidence of a thiol-selective mechanism in these events.
Finally, our results demonstrate the utility of the Drosophila
model in elucidating potentially fundamental mechanisms of
Supplementary data are available online at http://toxsci. oxfordjournals.org/.
National Institute of Health, National Institute of Environmental Health Science (R21ES013754 and R01ES015550) awarded to M.D.R.
We thank the NIH Neuroscience Microarray consortium for
assistance in sample handling, processing and probe set data
analysis. This work was carried out with the excellent technical
assistance of Julie Dao. We are grateful to Alena Krejci and
Sarah Bray (University of Cambridge, UK) for sharing primer
sequences. We thank Lucy Cherbas for cell culture consults
and materials received through the Drosophila Genomics
Resource Center. We also thank Felix Eckenstein (University
of Vermont) for critical review of the manuscript. We
acknowledge the help and services of the University of
Vermont COBRE molecular core facility for use of qPCR
instrumentation and assistance with data analysis. We
acknowledge the Vermont Cancer Center and Vermont Genetics
Network for resources made available to J.B. and the
Bioinformatics Facility. We are grateful for the services of
Dr Robert Taylor at the Texas A&M Trace Element Research
Laboratory for total mercury analyses.
Apidianakis , Y. , Nagel , A. C. , Chalkiadaki , A. , Preiss , A. , and Delidakis , C. ( 1999 ). Overexpression of the m4 and malpha genes of the E(spl)-complex antagonizes Notch mediated lateral inhibition . Mech. Dev . 86 , 39 - 50 .
Artavanis-Tsakonas , S. , Rand , M. D. , and Lake , R. J. ( 1999 ). Notch signaling: Cell fate control and signal integration in development . Science 284 , 770 - 776 .
Balasubramanian , R. , LaFramboise, T., Scholtens , D. , and Gentleman , R. ( 2004 ). A graph-theoretic approach to testing associations between disparate sources of functional genomics data . Bioinformatics 20 , 3353 - 3362 .
Bardin , A. J. , and Schweisguth , F. ( 2006 ). Bearded family members inhibit neuralized-mediated endocytosis and signaling activity of Delta in Drosophila . Dev. Cell 10 , 245 - 255 .
Barone , S. , Jr., Haykal-Coates , N. , Parran , D. K. , and Tilson , H. A. ( 1998 ). Gestational exposure to methylmercury alters the developmental pattern of trk-like immunoreactivity in the rat brain and results in cortical dysmorphology . Brain Res. Dev. Brain Res . 109 , 13 - 31 .
Bland , C. E. , and Rand , M. R. ( 2006 ). Methylmercury induces activation of Notch signaling . Neurotoxicology 27 , 982 - 991 .
Bolstad , B. M. , Irizarry , R. A. , Astrand , M. , and Speed , T. P. ( 2003 ). A comparison of normalization methods for high density oligonucleotide array data based on variance and bias . Bioinformatics 19 , 185 - 193 .
Bournias-Vardiabasis , N. , Buzin , C. , and Flores , J. ( 1990 ). Differential expression of heat shock proteins in Drosophila embryonic cells following metal ion exposure . Exp. Cell Res . 189 , 177 - 182 .
Braeckman , B. , Raes , H. , and Van Hoye , D. ( 1997 ). Heavy-metal toxicity in an insect cell line. Effects of cadmium chloride, mercuric chloride and methylmercuric chloride on cell viability and proliferation in Aedes albopictus cells . Cell Biol . Toxicol . 13 , 389 - 397 .
Camon , E. , Magrane , M. , Barrell , D. , Lee , V. , Dimmer , E. , Maslen , J. , Binns , D. , Harte , N. , Lopez , R. , and Apweiler , R. ( 2004 ). The Gene Ontology Annotation (GOA) Database: Sharing knowledge in Uniprot with Gene Ontology . Nucleic Acids Res . 32 , D262 - D266 .
Chao , E. S. , and Frenkel , G. D. ( 1983 ). Studies on the mechanism of the stimulation of polymerase II-catalyzed RNA synthesis by mercury compounds . J. Biol. Chem . 258 , 9861 - 9866 .
Chao , S. H. , Suzuki , Y. , Zysk , J. R. , and Cheung , W. Y. ( 1984 ). Activation of calmodulin by various metal cations as a function of ionic radius . Mol. Pharmacol . 26 , 75 - 82 .
Chen-Cleland , T. A. , Boffa , L. C. , Carpaneto , E. M. , Mariani , M. R. , Valentin , E. , Mendez , E. , and Allfrey , V. G. ( 1993 ). Recovery of transcriptionally active chromatin restriction fragments by binding to organomercurial-agarose magnetic beads. A rapid and sensitive method for monitoring changes in higher order chromatin structure during gene activation and repression . J. Biol. Chem . 268 , 23409 - 23416 .
Cheung , M. K. , and Verity , M. A. ( 1985 ). Experimental methyl mercury neurotoxicity: Locus of mercurial inhibition of brain protein synthesis in vivo and in vitro . J. Neurochem . 44 , 1799 - 1808 .
Choi , B. H. , Lapham , L. W. , Amin-Zaki , L. , and Saleem , T. ( 1978 ). Abnormal neuronal migration, deranged cerebral cortical organization, and diffuse white matter astrocytosis of human fetal brain: A major effect of methylmercury poisoning in utero . J. Neuropathol. Exp. Neurol . 37 , 719 - 733 .
Clemens , J. C. , Worby , C. A. , Simonson-Leff , N. , Muda , M. , Maehama , T. , Hemmings , B. A. , and Dixon , J. E. ( 2000 ). Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways . Proc. Natl. Acad. Sci . U. S. A. 97 , 6499 - 6503 .
Copple , I. M. , Goldring , C. E. , Kitteringham , N. R. , and Park , B. K. ( 2008 ). The Nrf2-Keap1 defence pathway: Role in protection against drug-induced toxicity . Toxicology 246 , 24 - 33 .
Delidakis , C. , and Artavanis-Tsakonas , S. ( 1992 ). The Enhancer of split [E(spl)] locus of Drosophila encodes seven independent helix-loop-helix proteins . Proc. Natl. Acad. Sci . U. S. A. 89 , 8731 - 8735 .
Dennehy , M. K. , Richards , K. A. , Wernke , G. R. , Shyr , Y. , and Liebler , D. C. ( 2006 ). Cytosolic and nuclear protein targets of thiol-reactive electrophiles . Chem. Res. Toxicol . 19 , 20 - 29 .
Dennis , G. , Jr.., Sherman , B. T. , Hosack , D. A. , Yang , J. , Gao , W. , Lane , H. C. , and Lempicki , R. A. ( 2003 ). DAVID: Database for annotation, visualization, and integrated discovery . Genome Biol . 4 , P3 .
Doherty , D. , Jan , L. Y., and Jan , Y. N. ( 1997 ). The Drosophila neurogenic gene big brain, which encodes a membrane-associated protein, acts cell autonomously and can act synergistically with Notch and Delta . Development 124 , 3881 - 3893 .
EPA. ( 1998 ). SW- 846 Method 7473. Mercury in solids and solutions by thermal decomposition, amalgamation, and atomic absorption spectrophotometry .
Falcon , S. , and Gentleman , R. ( 2007 ). Using GOstats to test gene lists for GO term association . Bioinformatics 23 , 257 - 258 .
Feder , M. , Cartano , N. , Milos , L. , Krebs , R. A. , and Lindquist , S. L. ( 1996 ). Effect of engineering Hsp70 copy number on Hsp70 expression and tolerance of ecologically relevant heat shock in larvae and pupae of Drosophila melanogaster . J. Exp. Biol . 199 , 1837 - 1844 .
Hong , F. , Sekhar , K. R. , Freeman , M. L. , and Liebler , D. C. ( 2005 ). Specific patterns of electrophile adduction trigger Keap1 ubiquitination and Nrf2 activation . J. Biol. Chem . 280 , 31768 - 31775 .
Hoppe , P. E. , and Greenspan , R. J. ( 1990 ). The Notch locus of Drosophila is required in epidermal cells for epidermal development . Development 109 , 875 - 885 .
Hughes , W. L. ( 1957 ). A physicochemical rationale for the biological activity of mercury and its compounds . Ann. N. Y. Acad. Sci . 65 , 454 - 460 .
Iso , T. , Kedes , L. , and Hamamori , Y. ( 2003 ). HES and HERP families: Multiple effectors of the Notch signaling pathway . J. Cell. Physiol . 194 , 237 - 255 .
Jennings , B. , Preiss , A. , Delidakis , C. , and Bray , S. ( 1994 ). The Notch signalling pathway is required for Enhancer of split bHLH protein expression during neurogenesis in the Drosophila embryo . Development 120 , 3537 - 3548 .
Kakita , A. , Inenaga , C. , Sakamoto , M. , and Takahashi , H. ( 2002 ). Neuronal migration disturbance and consequent cytoarchitecture in the cerebral cortex following transplacental administration of methylmercury . Acta Neuropathol . 104 , 409 - 417 .
Klambt , C. , Knust , E. , Tietze , K. , and Campos-Ortega , J. A. ( 1989 ). Closely related transcripts encoded by the neurogenic gene complex enhancer of split of Drosophila melanogaster . EMBO J . 8 , 203 - 210 .
Klose , M. K. , Chu , D. , Xiao , C. , Seroude , L. , and Robertson , R. M. ( 2005 ). Heat shock-mediated thermoprotection of larval locomotion compromised by ubiquitous overexpression of Hsp70 in Drosophila melanogaster . J. Neurophysiol . 94 , 3563 - 3572 .
Kopan , R. , Schroeter , E. H. , Weintraub , H. , and Nye , J. S. ( 1996 ). Signal transduction by activated mNotch: Importance of proteolytic processing and its regulation by the extracellular domain . Proc. Natl. Acad. Sci . U. S. A. 93 , 1683 - 1688 .
Krejci , A. , and Bray , S. ( 2007 ). Notch activation stimulates transient and selective binding of Su(H)/CSL to target enhancers . Genes Dev . 21 , 1322 - 1327 .
Lai , E. C. , Bodner , R. , and Posakony , J. W. ( 2000 ). The Enhancer of split Complex of Drosophila includes four Notch- regulated members of the Bearded gene family . Development 127 , 3441 - 3455 .
Landis , G. N. , Abdueva , D. , Skvortsov , D. , Yang , J. , Rabin , B. E. , Carrick , J. , Tavare , S. , and Tower , J. ( 2004 ). Similar gene expression patterns characterize aging and oxidative stress in Drosophila melanogaster . Proc. Natl. Acad. Sci . U. S. A. 101 , 7663 - 7668 .
Leviten , M. W. , Lai , E. C. , and Posakony , J. W. ( 1997 ). The Drosophila gene Bearded encodes a novel small protein and shares 3# UTR sequence motifs with multiple enhancer of split complex genes . Development 124 , 4039 - 4051 .
Lieber , T. , Kidd , S. , Alcamo , E. , Corbin , V. , and Young , M. W. ( 1993 ). Antineurogenic phenotypes induced by truncated Notch proteins indicate a role in signal transduction and may point to a novel function for Notch in nuclei . Genes Dev. 7 , 1949 - 1965 .
Livak , K. J. , and Schmittgen , T. D. ( 2001 ). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method . Methods 25 , 402 - 408 .
Maeder , M. L. , Polansky , B. J. , Robson , B. E. , and Eastman , D. A. ( 2007 ). Phylogenetic footprinting analysis in the upstream regulatory regions of the Drosophila enhancer of split genes . Genetics 177 , 1377 - 1394 .
Miura , K. , Himeno , S. , Koide , N. , and Imura , N. ( 2000 ). Effects of methylmercury and inorganic mercury on the growth of nerve fibers in cultured chick dorsal root ganglia . Tohoku J. Exp. Med . 192 , 195 - 210 .
Morken , T. S. , Sonnewald , U. , Aschner , M. , and Syversen , T. ( 2005 ). Effects of methylmercury on primary brain cells in mono- and co-culture . Toxicol. Sci . 87 , 169 - 175 .
Muller , A. , and Saenger , W. ( 1993 ). Studies on the inhibitory action of mercury upon proteinase K. J. Biol. Chem . 268 , 26150 - 26154 .
Nellesen , D. T. , Lai , E. C. , and Posakony , J. W. ( 1999 ). Discrete enhancer elements mediate selective responsiveness of enhancer of split complex genes to common transcriptional activators . Dev. Biol . 213 , 33 - 53 .
Ohtsuka , T. , Ishibashi , M. , Gradwohl , G. , Nakanishi , S. , Guillemot , F. , and Kageyama , R. ( 1999 ). Hes1 and Hes5 as Notch effectors in mammalian neuronal differentiation . EMBO J . 18 , 2196 - 2207 .
Parran , D. K. , Barone , S. , Jr. , and Mundy , W. R. ( 2003 ). Methylmercury decreases NGF-induced TrkA autophosphorylation and neurite outgrowth in PC12 cells . Brain Res. Dev. Brain Res . 141 , 71 - 81 .
Parran , D. K. , Barone , S. , Jr. , and Mundy , W. R. ( 2004 ). Methylmercury inhibits TrkA signaling through the ERK1/2 cascade after NGF stimulation of PC12 cells . Brain Res. Dev. Brain Res . 149 , 53 - 61 .
Peckham , N. H. , and Choi , B. H. ( 1988 ). Abnormal neuronal distribution within the cerebral cortex after prenatal methylmercury intoxication . Acta Neuropathol . 76 , 222 - 226 .
Ponce , R. A. , Kavanagh , T. J. , Mottet , N. K. , Whittaker , S. G. , and Faustman , E. M. ( 1994 ). Effects of methyl mercury on the cell cycle of primary rat CNS cells in vitro . Toxicol. Appl. Pharmacol . 127 , 83 - 90 .
Radyuk , S. N. , Klichko , V. I. , Spinola , B. , Sohal , R. S. , and Orr , W. C. ( 2001 ). The peroxiredoxin gene family in Drosophila melanogaster . Free Radic. Biol. Med . 31 , 1090 - 1100 .
Rand , M. D. , Grimm , L. M. , Artavanis-Tsakonas , S. , Patriub , V. , Blacklow , S. C. , Sklar , J. , and Aster , J. C. ( 2000 ). Calcium depletion dissociates and activates heterodimeric Notch receptors . Mol. Cell Biol . 20 , 1825 - 1835 .
Rhee , S. G. , Chae , H. Z. , and Kim , K. ( 2005 ). Peroxiredoxins: A historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling . Free Radic. Biol. Med . 38 , 1543 - 1552 .
Sacco , M. G. , Zecca , L. , Bagnasco , L. , Chiesa , G. , Parolini , C. , Bromley , P. , Cato , E. M. , Roncucci , R. , Clerici , L. A. , and Vezzoni , P. ( 1997 ). A transgenic mouse model for the detection of cellular stress induced by toxic inorganic compounds . Nat. Biotechnol . 15 , 1392 - 1397 .
Sanfeliu , C. , Sebastia , J. , and Ki , S. U. ( 2001 ). Methylmercury neurotoxicity in cultures of human neurons, astrocytes, neuroblastoma cells . Neurotoxicology 22 , 317 - 327 .
Sarafian , T. A. ( 1999 ). Methylmercury-induced generation of free radicals: Biological implications . Met. Ions Biol. Syst . 36 , 415 - 444 .
Seber , G. ( 1977 ). Linear Regression Analysis . J. Wiley and Sons, New York.
Shanker , G. , Hampson , R. E. , and Aschner , M. ( 2004 ). Methylmercury stimulates arachidonic acid release and cytosolic phospholipase A2 expression in primary neuronal cultures . Neurotoxicology 25 , 399 - 406 .
Sirois , J. E. , and Atchison , W. D. ( 2000 ). Methylmercury affects multiple subtypes of calcium channels in rat cerebellar granule cells . Toxicol. Appl. Pharmacol . 167 , 1 - 11 .
Skeath , J. B. , and Doe , C. Q. ( 1998 ). Sanpodo and Notch act in opposition to Numb to distinguish sibling neuron fates in the Drosophila CNS . Development 125 , 1857 - 1865 .
Sobell , H. M. ( 1985 ). Actinomycin and DNA transcription . Proc. Natl. Acad. Sci . U. S. A. 82 , 5328 - 5331 .
Sorensen , J. G. , Nielsen , M. M. , Kruhoffer , M. , Justesen , J. , and Loeschcke , V. ( 2005 ). Full genome gene expression analysis of the heat stress response in Drosophila melanogaster . Cell Stress Chaperones 10 , 312 - 328 .
Struhl , G. , and Adachi , A. ( 1998 ). Nuclear access and action of notch in vivo . Cell 93 , 649 - 660 .
Sun , B. , and Salvaterra , P. M. ( 1995 ). Characterization of nervana, a Drosophila melanogaster neuron-specific glycoprotein antigen recognized by antihorseradish peroxidase antibodies . J. Neurochem . 65 , 434 - 443 .
Sykiotis , G. P. , and Bohmann , D. ( 2008 ). Keap1/Nrf2 signaling regulates oxidative stress tolerance and lifespan in Drosophila . Dev. Cell 14 , 76 - 85 .
Toyama , T. , Sumi , D. , Shinkai , Y. , Yasutake , A. , Taguchi , K. , Tong , K. I. , Yamamoto , M. , and Kumagai , Y. ( 2007 ). Cytoprotective role of Nrf2/Keap1 system in methylmercury toxicity . Biochem. Biophys. Res. Commun . 363 , 645 - 650 .
Udolph , G. , Rath , P. , and Chia , W. ( 2001 ). A requirement for Notch in the genesis of a subset of glial cells in the Drosophila embryonic central nervous system which arise through asymmetric divisions . Development 128 , 1457 - 1466 .
Ui , K. , Nishihara , S. , Sakuma , M. , Togashi , S. , Ueda , R. , Miyata , Y. , and Miyake , T. ( 1994 ). Newly established cell lines from Drosophila larval CNS express neural specific characteristics . In Vitro Cell Dev. Biol. Anim. 30A , 209 - 216 .
Ui-Tei , K. , Sakuma , M. , Watanabe , Y. , Miyake , T. , and Miyata , Y. ( 1995 ). Chemical analysis of neurotransmitter candidates in clonal cell lines from Drosophila central nervous system, II: Neuropeptides and amino acids . Neurosci. Lett . 195 , 187 - 190 .
Wilke , R. A. , Kolbert , C. P. , Rahimi , R. A. , and Windebank , A. J. ( 2003 ). Methylmercury induces apoptosis in cultured rat dorsal root ganglion neurons . Neurotoxicology 24 , 369 - 378 .
Wilson , D. T. , Polunas , M. A. , Zhou , R. , Halladay , A. K. , Lowndes , H. E. , and Reuhl , K. R. ( 2005 ). Methylmercury alters Eph and ephrin expression during neuronal differentiation of P19 embryonal carcinoma cells . Neurotoxicology 26 , 661 - 674 .
Workman , C. , Jensen , L. J. , Jarmer , H. , Berka , R. , Gautier , L. , Nielser , H. B. , Saxild , H. H. , Nielsen , C. , Brunak , S. , and Knudsen , S. ( 2002 ). A new nonlinear normalization method for reducing variability in DNA microarray experiments . Genome Biol . 3 , research0048.
Wurmbach , E. , Wech , I. , and Preiss , A. ( 1999 ). The enhancer of split complex of Drosophila melanogaster harbors three classes of notch responsive genes . Mech. Dev . 80 , 171 - 180 .
Yang , Y. H. , and Speed , T. P. ( 2003 ). Design and analysis of comparative microarray experiments . In Statistical Analysis of Gene Expression Microarray Data (E. T. P. Speed , Ed.), pp. 35 - 91 . Chapman & Hall/CRC, New York.
Zaffran , S. , and Frasch , M. ( 2000 ). Barbu: An E(spl) m4/ma-related gene that antagonizes Notch signaling and is required for the establishment of ommatidial polarity . Development 127 , 1115 - 1130 .