The DHR96 nuclear receptor regulates xenobiotic responses in Drosophila
Exposure to xenobiotics such as plant toxins, pollutants, or prescription drugs triggers a defense response, inducing genes that encode key detoxification enzymes. Although xenobiotic responses have been studied in vertebrates, little effort has been made to exploit a simple genetic system for characterizing the molecular basis of this coordinated transcriptional response. We show here that approximately 1000 transcripts are significantly affected by phenobarbital treatment in Drosophila. We also demonstrate that the Drosophila ortholog of the human SXR and CAR xenobiotic receptors, DHR96, plays a role in this response. A DHR96 null mutant displays increased sensitivity to the sedative effects of phenobarbital and the pesticide DDT as well as defects in the expression of many phenobarbital-regulated genes. Metabolic and stress-response genes are also controlled by DHR96, implicating its role in coordinating multiple response pathways. This work establishes a new model system for defining the genetic control of xenobiotic stress responses.
The DHR96 nuclear receptor regulates xenobiotic responses in
Michael A. Horner,
and Carl S. Thummel
Howard Hughes Medical Institute, Department of Human Genetics, University of Utah School of Medicine, 15 N 2030 E 5100,
Salt Lake City, Utah 84112
These authors contributed equally to this work.
Present address: University of Alberta, CW-405 Biological Sciences Building, Edmonton, Alberta TGG 2E9, Canada.
Exposure to xenobiotics such as plant toxins, pollutants, or prescription drugs triggers a defense response, inducing genes
that encode key detoxiﬁcation enzymes. Although xenobiotic responses have been studied in vertebrates, little effort has
been made to exploit a simple genetic system for characterizing the molecular basis of this coordinated transcriptional
response. We show here that w1000 transcripts are signiﬁcantly affected by phenobarbital treatment in Drosophila.We
also demonstrate that the Drosophila ortholog of the human SXR and CAR xenobiotic receptors, DHR96, plays a role in
this response. A DHR96 null mutant displays increased sensitivity to the sedative effects of phenobarbital and the pesticide
DDT as well as defects in the expression of many phenobarbital-regulated genes. Metabolic and stress-response genes are
also controlled by DHR96, implicating its role in coordinating multiple response pathways. This work establishes a new
model system for deﬁning the genetic control of xenobiotic stress responses.
Higher organisms are constantly challenged by a wide range of
toxins in their environment. These compounds, referred to as
xenobiotics, enter the body by physical contact, inhalation, or
ingestion and can originate from many sources including phar-
maceuticals, pesticides, plant toxins, and pollutants. In order
to deal with the deleterious effects of xenobiotics, higher organ-
isms induce enzymes that metabolize these compounds into
less harmful substances, aiding in their inactivation and excre-
tion. This detoxiﬁcation machinery includes four classes of en-
zymes that are conserved from insects to humans: cytochrome
P450 monooxygenases (P450s), glutathione S-transferases
(GSTs), carboxylesterases, and UDP-glucuronosyl transferases
(UGTs). The most abundant class of xenobiotic metabolizing
enzymes is the P450s, represented by 57 genes in humans
and 90 genes in the fruit ﬂy Drosophila melanogaster (Maurel,
1996; Ranson et al., 2002). They are referred to as phase I en-
zymes because they catalyze the ﬁrst step in the detoxiﬁcation
process, decreasing the biological activity of a broad range of
substrates. The other classes of detoxifying enzymes, including
GSTs, carboxylesterases, and UGTs, are classiﬁed as phase II
enzymes. Carboxylesterases catalyze the hydrolysis of ester-
containing xenobiotics leading to their detoxiﬁcation, while
GSTs and UGTs add bulky side groups onto toxic compounds
to increase their hydrophilicity, facilitating their excretion from
Detailed studies in vertebrates have deﬁned a central role for
two nuclear receptors in sensing xenobiotic compounds and
regulating detoxiﬁcation gene expression: the human Steroid
and Xenobiotic Receptor (SXR, PXR in mice, NR1I2), and
Constitutive Androstane Receptor (CAR, NR1I3) (Chawla et al.,
2001; Francis et al., 2003; Willson and Kliewer, 2002). SXR/
PXR and CAR are most abundantly expressed in the liver and
small intestine, organs that provide a ﬁrst line of defense against
xenobiotics. SXR/PXR directly bind a wide range of lipophilic
xenochemicals, while CAR interacts with a more restricted set
of compounds and can be activated indirectly (Goodwin and
Moore, 2004; Kliewer et al., 2002). These receptors appear to
function in a partially redundant manner, with a range of overlap-
ping target genes, including genes that encode phase I and
phase II detoxiﬁcation enzymes, multidrug-resistance enzymes,
sulfotransferases, and proteins involved in the transport and
conjugation of small lipophilic compounds (Maglich et al.,
2002; Rosenfeld et al., 2003; Ueda et al., 2002).
In contrast to these studies in humans and mice, relatively lit-
tle effort has been made to characterize the regulation of insect
xenobiotic responses. Rather, studies in insects have been
largely restricted to insecticide-resistant strains that are the re-
sult of extreme selective pressures, analyzing populations that
have adapted to the presence of speciﬁc compounds in their
environment (Ffrench-Constant et al., 2004; Wilson, 2001). For
example, overexpression of a single P450 gene, Cyp6g1, is suf-
ﬁcient to confer DDT resistance in Drosophila (Daborn et al.,
2002), and resistance to organophosphates, malathion, and car-
bamates have been linked to overproduction of carboxyles-
terases (Hemingway et al., 2004; Zhu et al., 2004). There are
also examples of acquired resistance to naturally occurring
toxins. For example, some Drosophila species that live in the
Sonoran Desert use hazardous cactus species as a food source
due to their expression of appropriate detoxiﬁcation enzymes
(Danielson et al., 1997, 1998; Fogleman, 2000). This ability of
insects to adapt to speciﬁc xenobiotic compounds remains
the greatest impediment to the development of effective insec-
ticides, with serious consequences for human health and wel-
fare. Better forms of insect population control are critical for in-
creasing agricultural crop yields as well as combating lethal
insect-borne human diseases such as malaria.
CELL METABOLISM 4, 37–48, JULY 2006 ª2006 ELSEVIER INC. DOI 10.1016/j.cmet.2006.06.006 37
SXR/PXR and CAR are represented by a single ortholog
in Drosophila, DHR96 (NR1J1), and by three family members
in C. elegans, DAF-12, NHR-8, and NHR-48 (Escriva et al.,
2004). Genetic studies in C. elegans have shown that nhr-8 mu-
tants display reduced resistance to colchicine and chloroquine,
suggesting a role in xenobiotic responses (Lindblom et al.,
2001). Overall, however, little is known about the roles for nu-
clear receptors in invertebrate xenobiotic responses. We show
here that treatment with phenobarbital results in a dramatic re-
programming of Drosophila gene expression, affecting genes
that encode both phase I and phase II enzymes as well as novel
xenobiotic response genes that are likely to facilitate drug de-
toxiﬁcation and excretion. DHR96 is selectively expressed in
the primary digestive, metabolic, and excretory organs of the
animal, providing the capacity to respond to toxic compounds.
Consistent with this expression pattern, a DHR96 null mutant
displays increased sensitivity to the sedative effects of pheno-
barbital and defects in the expression of many phenobarbital-
regulated genes, deﬁning a role for this gene in responses to
a xenobiotic compound. These studies establish Drosophila as
a genetic model system for characterizing the regulation of xe-
nobiotic responses and provide new possible directions for
the rational design of more effective pesticides.
Phenobarbital treatment results in global changes
in detoxiﬁcation gene expression
The GABA agonist phenobarbital (PB) is one of the most widely
studied xenobiotic drugs and provides a highly effective inducer
of detoxiﬁcation genes in organisms ranging from bacteria to
humans (Waxman, 1999; Zelko and Negishi, 2000). PB appears
to have a similar effect in Drosophila, leading to signiﬁcant tran-
scriptional induction of several cytochrome P450 genes (Brun
et al., 1996; Danielson et al., 1997; Dombrowski et al., 1998;
Dunkov et al., 1997; Maitra et al., 1996; Waxman, 1999). We
thus selected this drug as a means of determining whether a
xenobiotic compound is sufﬁcient to direct a global reprogram-
ming in insect detoxiﬁcation gene expression. Staged wild-type
Canton S (CanS) ﬂies were treated for 10 hr with either sucrose
alone or sucrose supplemented with 0.3% PB, a concentration
known to induce Drosophila P450 gene transcription (Brun et al.,
1996; Dunkov et al., 1997). RNA was extracted from these
animals, labeled, and hybridized to Affymetrix Drosophila 2.0 mi-
croarrays. Raw data was analyzed using gcRMA (Wu et al.,
2004), and signiﬁcant gene expression changes were deter-
mined by SAM 2.0 (Tusher et al., 2001).
A total of 503 genes are upregulated and 484 genes are down-
regulated upon PB treatment, most of which encode enzymes
(Table 1, lines 1 and 2). The overall effects on transcript levels
are remarkable, with over 250 genes displaying from 3-fold to
700-fold changes in expression level. Members of the four clas-
sic detoxiﬁcation gene families are signiﬁcantly overrepresented
in the upregulated gene set. These include 29 P450 genes (Table
1, line 1), with eight such genes among the 30 top PB-inducible
genes, including Cyp6a8 (Figure 1A), Cyp6a21, Cyp12d1-p, and
Cyp6a2. Lower levels of PB-induced expression are seen for
Cyp6g1 and Cyp12a4. In addition, 35 other oxidoreductase
genes are induced by PB, with a total of 64 upregulated genes
in this class (Table 2, line 1). Sixteen GST genes (e.g., GstD2,
Figure 1A), seven UGT genes, and seven carboxylesterase
genes are also upregulated by PB.
In addition to the classic detoxiﬁcation genes, several other
gene sets are over-represented among the PB-regulated genes
as revealed by protein families with common InterPro domains
(Table 1) or searches for enriched gene ontology terms (Table 2).
These include 13 genes characterized by a putative choline
kinase domain known as DUF227, which is implicated in insec-
ticide resistance in Drosophila (Aminetzach et al., 2005)(Table 1,
rows 1 and 2). Members of this family are induced as well as re-
pressed; for example, CG6908 is induced w26-fold while
CG10514 is downregulated w31-fold by PB (Figure 1B). The
Turandot stress-response genes and two acyl-CoA synthetase
genes, CG4500 (Figure 1A) and CG6300/CG11659, are also
highly induced by PB. Acyl-CoA synthetases have been im-
plicated in xenobiotic metabolism in mammals (Knights and
Drogemuller, 2000). The only large group based on InterPro
domains that is signiﬁcantly downregulated includes 25
members of the serine peptidase S1 gene family (Table 1, line
2), although many other peptidase genes are also downregu-
lated (Table 2, line 2).
Table 1. Occurrence of gene families in microarray results based on InterPro domains
array results (n)
GST (45) Carboxylesterase
n p value n p value n p value n p value n p value n p value n p value n p value n p value n p value
1 CS6PB [ (503) 4 2.2E-02 7 1.8E-09 16 1.5E-42 7 3.4E-09 29 4.3E-66 13 1.0E-30 10 3.5E-02 2 >0.05 0 >0.05 41.1E-16
2 CS6PB Y (484) 3 >0.05 2 >0.05 1 >0.05 2 >0.05 3 >0.05 4 3.7E-03 25 1.6E-19 2 >0.05 1 >0.05 0 >0.05
+PB [ (326)
5 1.0E-05 2 >0.05 1 >0.05 0 >0.05 5 6.5E-03 0 >0.05 6 >0.05 4 1.9E-07 1 2.0E-02 0 >0.05
+PB Y (199)
1 >0.05 0 >0.05 1 >0.05 2 1.3E-02 0 >0.05 0 >0.05 5 4.1E-02 1 >0.05 2 3.8E-11 4 1.2E-41
[ (183) 0 >0.05 0 >0.05 3 1.0E-04 0 >0.05 4 9.7E-04 2 1.1E-02 4 >0.05 2 6.6E-04 1 9.1E-04 0>0.05
Y (386) 3 >0.05 1 >0.05 1 >0.05 43.0E-04 2 >0.05 2 >0.05 12 5.5E-05 2 5.0E-02 3 1.6E-12 0>0.05
7 w:hsDHR96 [ (174) 0 >0.05 1 >0.05 2 1.4E-02 0 >0.05 1 >0.05 0 >0.05 3 >0.05 0 >0.05 0 >0.05 0 >0.05
8 w:hsDHR96 Y (500) 6 5.3E-05 3 4.5E-02 2 >0.05 7 3.0E-09 9 2.1E-05 7 9.7E-09 12 2.5E-03 4 8.7E-05 3 9.3E-10 0>0.05
Pairwise comparisons between microarray gene lists (in rows) and gene lists that represent different protein families as determined by InterPro domains (in columns). The
number of overlapping genes is shown in each cell (n) along with the signiﬁcance of this overlap as represented by a p value determined by a c
test (bold indicates p value <
). The p values show the signiﬁcance of the difference between the observed number of genes in the overlap and the number of genes that would be expected on
average when two equally sized lists of randomly picked Drosophila genes are compared. Arrows indicate up- or downregulated gene sets. The number of genes in
each data set is shown in parentheses. CS, Canton S control stock; PB, Phenobarbital; w, w
control stock; transp., transporter; UDP Gluc., UDP-glucuronosyl trans-
ferase; GST, glutathione S-transferase; DUF227, Domain of Unknown Function 227; JHBP, Juvenile Hormone Binding Protein; ML, MD-2-related Lipid-recognition.
38 CELL METABOLISM : JULY 2006
Figure 1. Expression proﬁles and microarray validation of genes that are signiﬁcantly affected by PB treatment and/or the DHR96
A–D) Data from three microarray replicates were averaged for each data point, with error bars representing the standard deviation. The highest expression level was set to
100% for each group. CanS controls, untreated (dark blue) or PB-treated (light blue), and DHR96
mutants, untreated (dark yellow) or PB-treated (light yellow) , are depicted.
(A) Xenobiotic and stress-response genes. Cyp6a8, GstD2,andCG4500 are the three most highly PB-induced genes in wild-type animals. The TotM, TotA, and TotC Tur-
andot genes are stress-response genes that depend on DHR96 for their maximal response to PB. (B) Genes that encode either DUF227 domain proteins or members of the
Juvenile Hormone Binding Protein (JHBP) family. (C) Three genes encoding Jonah family proteases, the transcription factor gene sugarbabe, and the Est-9 carboxylester-
ase gene. (D) Four of the eight ML domain (MD-2-related lipid recognition) encoding genes found in the ﬂy genome are regulated by PB and DHR96.
E) Northern blot analysis of selected genes shown in panels (A)–(D). Blots were hybridized with rp49 as a control for loading and transfer. Of 27 PB-regulated genes tested in
this manner, all but two were validated on Northern blots.
Regulation of detoxiﬁcation in Drosophila
CELL METABOLISM : JULY 2006 39
Searches for enriched gene ontology terms revealed overrep-
resentation of pathways associated with carbohydrate, amino
acid, purine, and cholesterol metabolism among the upregu-
lated genes and lipid metabolic pathways among both the up-
and downregulated genes, indicating that toxin exposure alters
the metabolic state of the animal (Table 2, lines 1 and 2). For ex-
ample, 26 genes involved in fat breakdown, including genes
encoding triacylglycerol lipases, hydrolases, and dehydroge-
nases, are coordinately downregulated (Table 2, line 2). Interest-
ingly, the only group that is statistically underrepresented in the
PB-regulated gene set are those genes encoding DNA and RNA
binding proteins (represented by the term ‘‘nucleus’’ in Table 2,
lines 1 and 2), suggesting that the genomic response to PB relies
upon existent transcription factors rather than de novo syn-
thesis of these regulatory proteins. Taken together, this study
demonstrates that insects can mount a coordinated tran-
scriptional response to a xenobiotic compound and identiﬁes
many new genes that are likely to play a role in detoxiﬁcation
DHR96 is selectively expressed in tissues that
monitor and metabolize xenobiotics
As a ﬁrst step toward determining whether DHR96 might act in
Drosophila xenobiotic responses, we determined its spatial pat-
tern of expression. Organs were dissected from wandering third
instar larvae and stained with afﬁnity-puriﬁed antibodies. DHR96
protein is restricted to the nucleus of all expressing cells, consis-
tent with its proposed function as a transcription factor (Figure 2).
It is most abundantly expressed in four tissues: the fat body
(Figure 2A), salivary glands (Figure 2C), gastric caeca of the mid-
gut (Figure 2D), and Malpighian tubules (Figure 2E). Upon pro-
longed staining, DHR96 can also be detected in the proventricu-
lus and anterior region of the midgut. No speciﬁc expression was
detected in the imaginal discs, muscle, epidermis, brain, or tra-
chea (data not shown). The gastric caeca represent the primary
site for secretion of digestive enzymes and absorption of dietary
components into the circulatory system. Nutrients are metabo-
lized and stored in the fat body, the insect equivalent of the mam-
malian liver. Waste products, along with metabolized toxic com-
pounds, are then transferred back into the circulatory system
and absorbed by the Malpighian tubules, the principal osmoreg-
ulatory and excretory organ of the insect. The spatial pattern of
DHR96 expression thus tracks the course of toxic compounds
as they are absorbed through the gut epithelium (gastric caeca),
metabolized (fat body), and eliminated from the body (Malpigh-
ian tubules). The function of DHR96 in the larval salivary glands
is less clear. This organ opens onto the esophagus and thus
could be exposed to toxic compounds that enter through the
diet or it could provide enzymes for digestion.
Generation of DHR96 mutations
‘‘Ends-in’’ gene targeting was used to introduce speciﬁc muta-
tions into the DHR96 locus in an effort to determine if this gene
Figure 2. DHR96 protein is expressed in a restricted subset of tissues
Tissues were dissected from wild-type late third instar larvae and stained with af-
ﬁnity-puriﬁed antibodies directed against DHR96. The protein is restricted to the
nucleus and detected in the fat body (A), salivary glands (C), gastric caeca of
the midgut (D), and the Malpighian tubules (E). No protein is detected in the
mutant, as depicted for the fat body (B).
Table 2. Occurrence of gene ontology terms in microarray results
array results (n)
n p value n p value n p value n p value n p value n p value n p value n p value n p value
1 CS6PB [ (503) 24 7.7E-12 36 1.0E-47 12 4.1E-09 11 5.5E-14 6 4.1E-09 5 -7.0E-08 25 >0.05 64 3.9E-93 3 >0.05
2 CS6PB Y (484) 26 3.3E-15 10 1.1E-02 2 >0.05 0 >0.05 2 >0.05 22 -2.8E-02 57 6.2E-20 15 5.7E-03 7 3.5E-03
+PB [ (326)
7 >0.05 5 >0.05 4 >0.05 59.0E-050 >0.05 47 1.8E-07 17 >0.05 14 5.9E-05 2 >0.05
+PB Y (199)
7 6.6E-03 3 >0.05 1 >0.05 0 >0.05 2 2.7E-03 19 >0.05 16 2.2E-03 4 >0.05 3 4.8E-02
[ (183) 3 >0.05 3 >0.05 0 >0.05 1 >0.05 0 >0.05 22 8.7E-03 14 7.7E-03 6 >0.05 0 >0.05
Y (386) 7 >0.05 4 >0.05 1 >0.05 1 >0.05 0 >0.05 48 3.3E-05 23 3.1E-02 7 >0.05 3 >0.05
7 w:hsDHR96 [ (174) 3 >0.05 1 >0.05 3 2.7E-02 1 >0.05 0 >0.05 15 >0.05 11 >0.05 5 >0.05 0 >0.05
8 w:hsDHR96 Y (500) 19 9.6E-07 16 1.6E-07 7 5.8E-03 4 5.4E-02 3 1.3E-02 28 >0.05 35 2.3E-04 20 8.2E-06 8 5.4E-04
Pairwise comparisons between microarray gene lists (in rows) and gene lists that repres ent different protein families as determined by gene ontology terms (in columns).
The number of overlapping genes is shown in each cell (n) along with the signiﬁcance of this overlap as represented by a p value determined by a c
test (bold indicates
p value <10
). The p values show the signiﬁcance of the difference between the observed number of genes in the overlap and the number of genes that would be expected
on average when two equally sized lists of randomly picked Drosophila genes are compared. Arrows indicate up- or downregulated gene sets. The number of genes in each
data set is shown in parentheses. CS, Canton S control stock; PB, Phenobarbital; w, w
control stock; metab., metabolism.
40 CELL METABOLISM : JULY 2006
might function in xenobiotic response pathways (Rong and
Golic, 2000). Two deletions were introduced into the donor
DHR96 sequences, one of which removes the translational start
codon and the second of which removes exon four, the down-
stream intron, and the splice acceptor site for exon 5, disrupting
the ligand binding domain-coding region (Figure S2A in the Sup-
plemental Data available with this article online.).
A screen for speciﬁc targeted mutations resulted in the isola-
tion of DHR96
, which carries a GFP reporter gene between two
nonfunctional copies of DHR96 (Figure S2A). Southern blot anal-
ysis of DNA isolated from homozygous DHR96
vealed a restriction pattern that is consistent with a duplication
event (Figures S2A and S2B). In addition, DNA sequencing of
PCR-ampliﬁed fragments revealed that both mutations are
present in the inserted sequences in DHR96
, with the second
deletion also present in the 3
duplicated copy of the gene
(Figure S2A). Similar unpredicted sequence changes have
been reported for Drosophila ends-in gene targeting (Rong
et al., 2002). Western blot analysis using afﬁnity-puriﬁed anti-
bodies directed against DHR96 demonstrated that the protein
is undetectable in mutant animals (Figure S2C). Similar results
were seen in antibody stains of tissues dissected from
mutant larvae (Figure 2B). Taken together, the molecu-
lar defects in the mutant and the antibody studies demonstrate
is a strong loss-of-function allele and most likely
a null mutation.
To minimize the effect of genetic background on xenobiotic
sensitivity (Miyo et al., 2001, 2003; Pyke et al., 2004), we out-
crossed the DHR96
mutation to wild-type CanS ﬂies through
nine generations of free recombination. CanS is widely used
as a control for pesticide resistance studies (Bogwitz et al.,
2005; Brandt et al., 2002; Daborn et al., 2001; Pedra et al.,
2004). All subsequent studies were conducted using the out-
mutant in combination with a CanS control.
DHR96 mutants are sensitive to phenobarbital and DDT
mutant is viable and fertile when raised under stan-
dard conditions, similar to PXR, CAR, and nhr-8 mutants (Lind-
blom et al., 2001; Wei et al., 2000; Xie et al., 2000). DHR96
tants, however, display increased sensitivity to PB. Staged
CanS and DHR96
mutant ﬂies were fed either sucrose alone
or sucrose and 0.1%–10% PB for 6 to 24 hr. No signiﬁcant effect
was seen in mutant or wild-type animals treated with 0.1% PB,
but both mutant and wild-type displayed reduced activity and
uncoordination at higher PB concentrations. DHR96
however, were markedly less active than their wild-type coun-
terparts, as quantiﬁed by a negative geotaxis assay (Figure 3A).
DHR96 is thus required for proper resistance to the sedative ef-
fects of PB.
We also tested sensitivity to the pesticide DDT as a model in-
sect xenobiotic. Although acute exposure to DDT did not reveal
a signiﬁcant difference in the sensitivity of control and DHR96
mutant animals, chronic exposure to lower concentrations of
pesticide resulted in a reproducibly lower survival rate for mu-
tant animals relative to the controls. This is depicted for a range
of DDT concentrations in Figure 3B, where only a few DHR96
mutant ﬂies survive chronic exposure to a higher concentration
of DDT relative to the survival of about half of the CanS controls.
Xenobiotic gene transcription is affected
in DHR96 mutants
Studies of PXR and CAR mutant mice focus primarily on the
transcription of speciﬁc P450 target genes, with whole-animal
experiments largely involving hepatic damage and sensitivity
to paralytic sedatives (Wei et al., 2000; Xie et al., 2000; Zhang
et al., 2004). We thus chose to restrict our initial studies to char-
acterizing possible roles for DHR96 in the regulation of genes af-
fected by PB, attempting to link the PB sensitivity of DHR96
mutants with changes in the expression of speciﬁc detoxiﬁca-
tion genes. For this purpose, microarray analysis was used to
compare the transcriptional responses of PB-treated CanS
wild-type ﬂies with PB-treated DHR96
mutants. A total of 525
genes displayed signiﬁcantly different levels of expression be-
tween these data sets (Tables 1 and 2, lines 3 and 4). Of these
differentially expressed genes, 102 genes are present in the
list of 987 genes regulated by PB in wild-type animals (p value =
2.9 3 10
, Figure 4A). Among these 102 genes are 21 genes
that are PB-induced but which require DHR96 for their maximal
response to the drug. These include four members of the Turan-
dot (Tot) gene family (Table 1,line4).TotM and TotX are w5-fold
lower in DHR96 mutants relative to controls, whereas TotC
and TotA reach w50% of wild-type induction levels (Figure 1A).
This list also includes the CG4500 acyl-CoA synthetase gene
(Figure 1A), which is the third most highly PB-induced gene and
Figure 3. DHR96 mutants are sensitive to pheno-
barbital and DDT
A) Staged young adult ﬂies were fed either sucrose
alone or sucrose and 1% phenobarbital (PB) for 18
hr. A negative geotaxis assay was used to quantitate
activity, showing the percentage of the tota l popula-
tion that remained active after treatment. Each bar
represents the average from three repeats on each
of three vials (a total of nine samples), with error
bars depicting the standard deviation.
B) Young adult ﬂies were maintained on medium
containing either 10, 50, or 250 mg DDT per vial.
The percentage of surviving ﬂies was determined af-
ter three weeks. Each bar depicts the average from
four vials with 30 ﬂies each, with the error bars show-
ing standard deviation.
Regulation of detoxiﬁcation in Drosophila
CELL METABOLISM : JULY 2006 41
which is likely to contribute to xenobiotic responses (Knights and
A group of 20 genes, including the Jonah protease genes and
sugarbabe (Figure 1C), are repressed by PB but are further
downregulated in DHR96 mutants, suggesting that this class
of drug-responsive genes requires DHR96 for its normal expres-
sion. The remaining genes either fail to be properly repressed
(26 genes) or are superinduced (35 genes) in PB-treated mu-
tants. Interestingly, the latter group harbors numerous potential
xenobiotic genes, including the Cyp309a2, Cyp4s3, Cyp6a14
P450 genes, two additional oxidoreductase genes (CG9509
and CG3597), Ugt86Dd, GstD7, and the CG6300/CG11659
acyl-CoA synthetase genes. Taken together, these ﬁndings sup-
port the proposal that DHR96 contributes to xenobiotic tran-
We also identiﬁed genes that are normally unresponsive to PB
treatment but can be induced by PB in DHR96
include four genes that encode so-called Juvenile Hormone
Binding Proteins (JHBPs): CG7916, CG7968, CG8997, and
CG7953 (Table 1, lines 3 and 4) (Figure 1B). This ﬁnding indi-
cates that DHR96 is required for the repression of some genes
under normal conditions and may also inhibit their PB-respon-
Untreated DHR96 mutants reveal links
to xenobiotic responses
We next asked whether differential gene expression patterns
between untreated CanS controls and DHR96
vealed regulatory pathways that require DHR96 under normal
conditions. Interestingly, this comparison identiﬁed a number
of genes encoding likely xenobiotic enzymes (Table 1, lines 5
and 6). These include moderate upregulation of three P450
genes, Cyp18a1, Cyp9h1, and Cyp4s3, and reduced expression
of Cyp4ac1 and Cyp4e3, the latter being strongly induced by PB
in wild-type animals. The array data also indicated reduced
expression of four carboxylesterase genes, Nrt, Ace, Est-8,
mutants (Est-9 depicted in Figure 1C),
which are not strongly affected by PB but may play a role in
the detoxiﬁcation of other compounds (Menozzi et al., 2004).
Half of the eight genes in the ﬂy genome that encode ML do-
main proteins are also signiﬁcantly affected in both untreated
and PB-treated DHR96 mutants (Table 1, lines 3–6). The ML do-
main (MD-2-related lipid-recognition, also known as Def2/Der2)
is a widespread protein motif, found in plants, fungi, and ani-
mals, that has been implicated in binding to speciﬁc lipids (Ino-
hara and Nunez, 2002). The most strikingly affected ML domain
genes in DHR96 mutants are CG12813 and CG31410. CG12813
mRNA is virtually absent in DHR96
ﬂies, resulting in expression
that is >100-fold below wild-type (Figure 1D). In contrast,
CG31410 expression levels are w20-fold higher in mutant ani-
mals than in controls, representing the most strongly dere-
pressed gene in DHR96
ﬂies (Figure 1D). A second gene pair
encoding ML domain proteins, CG11314 and CG11315, display
almost identical patterns in mutant animals, as both are sub-
maximally expressed and display a moderate downregulation
when PB is added to the diet (Figure 1D).
We also identiﬁed a highly signiﬁcant overlap of 83 genes that
are downregulated in both untreated DHR96
mutants and PB-
treated wild-type ﬂies (Figure 4B). Interestingly, these genes dis-
play similar fold changes in expression levels, suggesting that
a loss of DHR96 function in untreated ﬂies is functionally equiv-
alent to PB-dependent downregulation (Table S1). For example,
the DUF227 and ML domain-encoding genes CG10514 and
CG12813 are signiﬁcantly downregulated by PB in wild-type an-
imals but are expressed at an equally low level in untreated
mutants (Figures 1B and 1D). In addition, we ﬁnd that
this set harbors 11 peptidase-encoding genes, including three
Jonah protease genes (Jon99Fi, Jon25Bii, and Jon65Aii), that
are signiﬁcantly reduced in DHR96
mutants (Figure 1C and Ta-
ble 1, line 1). This set of 83 genes also contains a signiﬁcant clus-
ter of 13 DNA/RNA binding proteins, demonstrating that more
than half of the 22 PB-repressed genes associated with the
Figure 4. Many PB-regulated genes depend on
DHR96 for their proper expression
A) Comparison of genes that change their expression
in wild-type ﬂies treated with PB (CanS:CanS 6PB)
with genes that are differentially expressed between
PB-treated wild-type controls and DHR96
+PB), or genes that are differentially
expressed between w
controls and heat-induced
hsDHR96 transformants (w
indicate up- and downregulation.
B) Comparison of genes that are downregulated by
PB in wild-type ﬂies (CanS:CanS 6PB) and genes
that are downregulated in untreated DHR96
42 CELL METABOLISM : JULY 2006
ontology term ‘‘nucleus’’ (Table 2, line 2) are identical to the 48
‘‘nuclear’’ genes that are submaximally expressed in DHR96
mutants (Table 2, line 6). This observation suggests that the
detoxiﬁcation pathway and the DHR96 regulatory network con-
verge on a common set of regulatory factors.
Ectopic DHR96 expression represses genes associated
with xenobiotic and metabolic pathways
As a ﬁnal step in our analysis of DHR96 function, we determined
the effects of ectopic DHR96 expression on the patterns of ge-
nome-wide transcription and compared these effects with those
seen in DHR96
loss-of-function mutants as well as our list of
PB-regulated genes. Staged transgenic animals that carry
a heat-inducible promoter fused to the DHR96 gene (hsDHR96)
and control wild-type animals of the same genetic background
) were subjected to a brief heat treatment and allowed to
recover for 4 hr. RNA was extracted from these animals and an-
alyzed on Affymetrix microarrays, essentially as described
above. Comparison of expression levels in these data sets re-
vealed 500 genes that are downregulated upon ectopic
DHR96 expression and 174 genes that are upregulated (Tables
1 and 2, lines 7 and 8). Among the 500 repressed genes are six
genes encoding ABC transporters, including the Mdr49 multi-
drug resistance family member, three UGTs, two GSTs, seven
carboxylesterases, nine P450s, seven DUF227 domain proteins,
12 S1 peptidases, four JHBPs, and three ML domain proteins
(Table 1, line 8). The observation that these same gene families
are regulated by PB (Table 1, lines 1 and 2) supports the proposal
that ectopic DHR96 expression affects at least part of the xeno-
biotic response pathway. Additional evidence for this arises from
a comparison of the genes that change their expression upon ec-
topic DHR96 expression with the genes regulated by PB in wild-
type animals. This study reveals a signiﬁcant overlap of 102
genes (p value = 1.3 3 10
, Figure 4A). These genes include
four P450 genes, all of which are induced by PB treatment and
downregulated upon ectopic DHR96 expression, along with
two carboxylesterase genes, one GST gene, two genes harbor-
ing a DUF227 domain, one ML domain gene, and four genes in-
volved in lipid metabolism. Taken together with the transcrip-
tional proﬁles of the DHR96 loss-of-function mutant, this study
indicates that this receptor participates in the control of PB-reg-
ulated genes. In addition, analysis of the hsDHR96 gain-of-func-
tion dataset using gene ontology terms revealed that the down-
regulated genes display a signiﬁcant enrichment for terms that
include lipid and carbohydrate metabolism, proteolysis, oxido-
reductases, and ion transporters (Table 2, line 8), suggesting
that DHR96 contributes to metabolic gene regulation.
Overlaps between xenobiotic and
The coordinate regulation of the Turandot stress-response
genes under most conditions tested prompted us to examine
the possibility that PB-treatment, and/or DHR96, contribute to
stress-response pathways. To do this, we compared our PB-
regulated microarray data sets with the stress-response micro-
array study of Girardot et al. (2004), identifying 69 genes that
are upregulated and 78 genes that are downregulated under
both conditions (Table 3, lines 1 and 2). Of the 25 top PB-induced
genes, only two are identiﬁed as general stress-response genes,
GstD2 and Cyp309a1, while the next 25 PB-induced genes
harbor 11 such stress-inducible genes, including DptB, GstD5,
and GstE6. If one includes the four Turandot genes, roughly
one third of the top 50 PB-induced genes represent a general
When the DHR96
mutant data is compared to the stress ar-
ray data, we ﬁnd that DHR96 mutants, whether treated or not
treated with PB, display signiﬁcant misregulation of genes that
are normally downregulated in response to stress, while a mod-
erate but signiﬁcant number of stress upregulated genes have
elevated expression levels in DHR96 mutants (Table 3, lines 3–
6). Comparison of DHR96 gain-of-function gene sets with the
stress microarray data further indicates that these animals exe-
cute a partial stress response (Table 3, lines 7 and 8).
The studies described here represent a ﬁrst step toward using
Drosophila to deﬁne the coordinate transcriptional regulation
of defensive responses to toxic compounds. We show that in-
sect xenobiotic responses are not limited to the adaptive effects
of insecticide resistance but also include a highly inducible tran-
scriptional response of key detoxiﬁcation enzymes, analogous
to that found in mammalian systems. We also provide evidence
that the Drosophila ortholog of the mammalian SXR/PXR and
CAR nuclear receptors, DHR96, plays a role in insect xenobiotic
responses, providing resistance to the sedative PB and contrib-
uting to the proper regulation of detoxiﬁcation genes. Below, we
expand upon our studies of PB-regulated transcription and
DHR96 functions, comparing and contrasting our work with
studies in vertebrates and in other insects.
Drosophila execute a massive transcriptional
response to a xenobiotic compound
Although two studies have used genome-wide microarrays to
deﬁne the transcriptional proﬁle of insecticide-resistant strains
of Drosophila (Pedra et al., 2004) and the mosquito Anopheles
gambiae (Vontas et al., 2005), none have characterized the induc-
ible transcriptional response to a toxic compound in wild-type in-
sects. The data presented here, examining the effects of PB on
Table 3. DHR96 and Phenobarbital microarray data compared to stress
array results (n)
stress [ (222) stress Y (211)
n p value n p value
1 CS6PB [ (503) 69 1.5E-140 6 >0.05
2 CS6PB Y (484) 2 >0.05 78 1.9E-187
+PB[ (326) 11 5.8E-04 16 2.6E-08
+PBY (199) 3 >0.05 9 6.4E-05
[ (183) 8 3.7E-04 17 6.1E-19
Y (386) 5 >0.05 18 2.3E-09
7 w:hsDHR96 [ (174) 18 6.2E-32 4 >0.05
8 w:hsDHR96 Y (500) 5 >0.05 21 3.5E-08
The stress microarray data from Girardot et al. (2004) was condensed to sets of
222 upregulated genes and 211 downregulated genes (see Experimental Proce-
dures) and compared to our microarray data sets. The number of overlapping
genes is shown in each cell (n) along with the signiﬁcance of this overlap as repre-
sented by a p value determined by a c
test (bold indicates p value < 10
p values show the signiﬁcance of the difference between the observed number of
genes in the overlap and the number of genes that would be expected on average
when two equally sized lists of randomly picked Drosoph ila genes are compared.
Arrows indicate up- or downregulated gene sets. The number of genes in each
data set is shown in parentheses. CS, Canton S control stock; PB, Phenobarbital;
Regulation of detoxiﬁcation in Drosophila
CELL METABOLISM : JULY 2006 43
transcription in wild-type ﬂies, provide insights into how insects
mount a defense response to toxic compounds. Approximately
1000 transcripts are affected by PB treatment, with many show-
ing high-fold changes in expression. The vast majority of these
genes encode enzymes, and many of these correspond to
known detoxiﬁcation pathways, including multiple P450s,
GSTs, carboxylesterases, and UGTs (Table 1, lines 1 and 2). Al-
though relatively few genes have been linked to speciﬁc xenobi-
otic functions in insects, the second most highly PB-induced
gene (677-fold), Cyp6a8, can metabolize organophosphates,
cyclodiene insecticides (Dunkov et al., 1997), and promutagens
(Saner et al., 1996). Similarly, the PB-inducible Cyp6g1 and
Cyp12a4 genes are sufﬁcient to confer resistance to the pesti-
cide Lufenuron (Bogwitz et al., 2005; Daborn et al., 2002). The
overall types of enzymes regulated by PB in Drosophila resemble
those seen in insecticide-resistant strains of Anopheles and Dro-
sophila (Pedra et al., 2004; Vontas et al., 2005), as well as the en-
zymes affected by xenobiotics in mammalian studies ( Kume
et al., 2005; Maglich et al., 2002; Ueda et al., 2002), providing fur-
ther evidence that the core detoxiﬁcation machinery has been
conserved through evolution, from insects to humans.
In addition to the main classes of detoxiﬁcation genes, several
gene families emerge from searches for overrepresented Inter-
Pro or gene ontology terms. These include 13 PB-inducible
genes that encode proteins with a DUF227 domain (Table 1,
line 1). This apparently insect-speciﬁc domain is proposed to
act as a choline/ethanolamine kinase (InterPro IPR004119). A
recent study reported that a speciﬁc mutation in a DUF227
domain-encoding gene (CG10618) confers resistance to organ-
ophosphates in Drosophila (Aminetzach et al., 2005). Two acyl-
CoA synthetase genes are also among the ten most abundantly
induced PB-responsive genes (CG4500 and CG6300/CG11659).
Acyl-CoA synthetases have been implicated in xenobiotic me-
tabolism in mammals (Knights and Drogemuller, 2000), and an
acyl-CoA synthetase is upregulated in an insecticide-resistant
strain of Anopheles (Vontas et al., 2005).
Remarkably, many PB-regulated genes not associated with
overrepresented gene families may also contribute to general
detoxiﬁcation responses. For example, Jheh1, which is induced
16-fold by PB, encodes an epoxide hydrolase. These enzymes
can detoxify epoxides by increasing their solubility and aiding
in their excretion (Lu and Miwa, 1980). Similarly, the glycine N-
methyltransferase encoded by CG6188, which is induced 77-
fold by PB, can bind polyaromatic hydrocarbons and contribute
to mammalian Cyp1A1 induction (Bhat and Bresnick, 1997).
Thus, the genome-wide transcriptional response to PB has re-
vealed not only new members of the known classes of detoxiﬁ-
cation enzymes but also a number of other pathways that are
likely to contribute to xenobiotic responses, providing a new ba-
sis for understanding how insects defend themselves against
environmental toxins. Our recent studies have also shown that
many of these genes are regulated in an identical manner by
a different drug, the dopamine antagonist chlorpromazine, sug-
gesting that this transcriptional pattern reﬂects a general de-
fense response to xenobiotics (M.H., unpublished data).
DHR96 is selectively expressed in tissues that
function in xenobiotic detoxiﬁcation
DHR96 protein is ideally positioned to monitor the entry and
exit of dietary nutrients and foreign compounds, coordinating
xenobiotic stress and metabolic responses within the animal.
DHR96 displays a highly restricted pattern of expression, limited
primarily to organs that are involved in nutrient and xenobiotic
absorption (gastric caeca), metabolism (fat body), and waste
elimination (Malpighian tubules) (Figure 2).This pattern is similar
to that of PXR and CAR, which are most highly expressed in the
liver and intestine. This expression pattern also reﬂects that of
the two PB-inducible P450 transcripts that have been spatially
localized. Cyp6a2 is expressed in the midgut, fat bodies, and
Malpighian tubules of adult ﬂies while Cyp12a4 is expressed pri-
marily in the midgut and Malpighian tubules of third instar larvae
(Bogwitz et al., 2005; Brun et al., 1996). Unlike its mammalian or-
thologs, however, which reside in the cytoplasm and translocate
to the nucleus upon xenobiotic challenge (Kobayashi et al.,
2003; Squires et al., 2004), DHR96 protein appears to be re-
stricted to the nucleus.
DHR96 is required for proper responses
to a xenobiotic compound
DHR96 mutants display a signiﬁcant increase in their sensitivity
to the sedative effects of PB (Figure 3A). This observation is sim-
ilar to the prolonged sleep phenotype of PB-treated CAR mutant
mice (Swales and Negishi, 2004) and the effects of PB treatment
on resistance to the paralytic effects of the muscle relaxant zox-
azolamine in CAR mutants (Wei et al., 2000). Of 144 genes that
change their expression upon intraperitoneal injection of PB in
mice, about half are dependent upon CAR for their proper re-
sponse to the drug (Ueda et al., 2002). This is similar to the effect
we see in Drosophila, where 102 PB-regulated genes are af-
fected by either the DHR96
loss-of-function mutation or ec-
topic DHR96 gain-of-function (Figure 4A). In addition, many of
these genes encode members of the four classic detoxiﬁcation
enzyme families, demonstrating that the insect xenobiotic re-
sponse depends upon DHR96 for its proper implementation.
Similar effects on xenobiotic gene regulation were seen in
PXR-VP16 gain-of-function studies and with PXR mutant mice
(Maglich et al., 2002; Rosenfeld et al., 2003).
Interestingly, some genes are only PB-inducible in a DHR96
mutant background. The most strongly affected genes in this
group are members of the Juvenile Hormone Binding Protein
family (JHBPs, Figure 1B). JHBPs are hemolymph carrier pro-
teins that are capable of binding lipophilic compounds such as
juvenile hormone, facilitating their transport within the animal
and protecting them against nonspeciﬁc esterases (Touhara
and Prestwich, 1993). Whereas JHBP genes are not responsive
to PB in wild-type animals, they are induced by PB in DHR96
mutants, indicating that DHR96 normally acts to block their ex-
pression. This could reﬂect a protective function for DHR96 in
that JHBPs may normally protect lipophilic compounds from
degradation, thus interfering with drug clearance. Several other
genes are also induced by PB only in a DHR96 mutant back-
ground, including Glutathione Synthetase and the putative tran-
scription factor CG14965. It remains unclear, however, how
these genes might contribute to the detoxiﬁcation response. Re-
markably, a similar effect has been reported for CAR function in
mice, where a subset of genes is only responsive to PB in a CAR
mutant background (Ueda et al., 2002). These genes include two
P450 genes and genes encoding a calcium binding protein and
glucosamine phosphate N-acetyl transferase. This so-called
CAR-dependent blocking of gene expression may reﬂect an
evolutionarily conserved aspect of the xenobiotic response
44 CELL METABOLISM : JULY 2006
Untreated DHR96 mutants display signiﬁcant changes
in detoxiﬁcation gene expression
Further evidence of a role for DHR96 in xenobiotic response
pathways arises from the ﬁnding that many potential detoxifying
enzymes change their expression in untreated DHR96 mutants
(Table 1, lines 5 and 6). Among this set, we found 83 genes
that are downregulated in both untreated DHR96 mutants and
PB-treated wild-type ﬂies, including genes that encode tran-
scription factors and numerous proteases and peptidases
(Table S1). Curiously, the relative fold changes of these genes
are almost identical, in spite of their responding to two very dif-
ferent conditions, suggesting that untreated DHR96 mutants
display some aspects of a toxin response (data not shown).
There are two possible models to explain this observation. First,
DHR96 may regulate a speciﬁc branch of the xenobiotic network
and the loss of DHR96 may result in the accumulation of toxic
metabolites (endobiotics) that activate DHR96-independent
xenobiotic pathways. Although at ﬁrst glance, this ‘‘endobiotic
model’’ is attractive, we note that relatively few PB-inducible
genes are upregulated in untreated DHR96 mutants (18 genes,
Table S1). Thus, the mutants do not appear to display a xenobi-
otic response, but rather misregulate a set of genes that are nor-
mally repressed by PB treatment. As a result, we favor an alter-
native model which proposes that DHR96 is required for normal
expression of a subset of genes involved in the xenobiotic re-
sponse pathway (exempliﬁed by the 83 genes described here).
PB then acts to inhibit this function of DHR96, resulting in the re-
duced expression of these DHR96 target genes. In addition, as
discussed below, other xenobiotic receptors are likely to pro-
vide input to this pathway.
PB and DHR96 regulate stress-response
and metabolic pathways
Given that treatment with a toxic compound is likely to impose
stress on the animal, it is not surprising that we observe a strong
correlation between genes up- and downregulated by PB and
genes regulated in the same manner by stress (Table 3). Our
studies with loss-of-function and gain-of-function DHR96 muta-
tions also indicate a role for this receptor in regulating stress re-
sponses (Tables 1 and 2). Stress-response genes are upregu-
lated in insecticide-resistant strains of Anopheles and some
insecticide-resistant strains of insects are more resistant to ox-
idative stress, providing a functional link between these two
pathways (Abdollahi et al., 2004; Vontas et al., 2005; Vontas
et al., 2001). Our results suggest that this link is conferred at
the level of speciﬁc stress-response gene regulation. In addition,
PXR has been recently implicated in mediating oxidative stress
responses, suggesting that at least some aspects of this regula-
tion have been conserved through evolution (Gong et al., 2006).
PB treatment is also associated with a signiﬁcant upregulation
of genes involved in energy and sugar metabolism. The ﬁrst two
steps in gluconeogenesis are catalyzed by pyruvate carbox-
ylase (CG1516) and phosphoenolpyruvate carboxykinase
(PEPCK), enzymes encoded by genes that are upregulated by
PB. Other glucose-generating processes show a similar re-
sponse to PB, including six a-amylase genes that are induced
more than 3-fold by the drug. Amylases are secreted by the sal-
ivary glands and midgut epithelia to break down dietary starch
and glycogen into dextrins. Two genes encoding glucosidases,
which can further degrade dextrins into monosaccharides, are
also upregulated upon PB treatment. This effect is the opposite
of that seen in mammals, where gluconeogenesis is downregu-
lated by PB (Kodama et al., 2004). It makes sense, however, that
this pathway would be upregulated as part of the detoxiﬁcation
response. As noted by Reichert and Menzel (2005), toxin metab-
olism is energetically costly. P450s consume NADPH or NADH
for their oxidation of xenobiotics, while UGTs consume glucose
and GSTs consume glutathione. Thus, part of the metabolic
response to xenobiotics appears to provide the appropriate
energetic requirements for detoxiﬁcation.
Although our loss-of-function and gain-of-function genetic data
indicate a role for DHR96 in mediating xenobiotic responses, it is
interesting to note that the majority of PB-regulated genes are
unaffected by DHR96 mutations. This is similar to studies in
mice, where at least half of the detoxiﬁcation gene network is
unaffected by PXR or CAR null mutations (Maglich et al., 2002;
Ueda et al., 2002). Given the massive coordinate regulation of
the PB response, it seems likely that one or more additional tran-
scriptional regulators feed into this pathway. These could in-
clude PAS-bHLH family members, analogous to the role of the
mammalian aryl hydrocarbon receptor in regulating xenobiotic
responses (Rowlands and Gustafsson, 1997). Functional stud-
ies of PB-regulated promoters should provide insights into
how their activity is controlled by the drug and identify additional
players in their regulation. Similarly, the identiﬁcation and char-
acterization of direct targets for DHR96 transcriptional control
will allow us to deﬁne how this receptor exerts its regulatory
functions. Finally, DHR96 provides a potential target for the ra-
tional design of novel pesticides. By developing compounds
that alter DHR96 activity it may be possible to increase the effec-
tiveness of pesticide treatment for insect population control.
Taken together, the studies described here provide a basis for
using Drosophila as a genetic model for dissecting the reg-
ulation of xenobiotic responses, with implications for better
understanding how these pathways are controlled in all higher
Histidine-tagged DHR96 protein was puriﬁed from pET24c-DHR96 bacteria,
resolved by preparative SDS-PAGE, excised from the gel, injected into three
rabbits (Covance), and antisera were screened by Western blotting. Antise-
rum was afﬁnity-puriﬁed as described (Carroll and Laughon, 1987), using
protein from pMAL-DHR96 bacteria. Wandering third instar larval tissues
were dissected and ﬁxed as described (Boyd et al., 1991). DHR96 protein
was detected using afﬁnity-puriﬁed DHR96 antibodies diluted 1:100 and in-
cubated overnight at 4ºC. Donkey anti-rabbit CY3 secondary antibodies
(Jackson) were used at a 1:200 dilution. The stains were visualized on a Bio-
rad confocal laser scanning microscope. For Western blots, protein from
adult ﬂies was extracted in SDS sample buffer. The equivalent of approxi-
mately one adult ﬂy was loaded in each lane of an 8% polyacrylamide gel,
separated by electrophoresis, and transferred to a PVDF membrane.
hsDHR96 transformants were treated at 37.5ºC for 30 min followed by a 3
hr recovery at room temperature. DHR96 protein was detected by incubating
the membrane with a 1:500 dilution of afﬁnity-puriﬁed anti-DHR96 anti-
bodies, followed by a 1:1000 dilution of goat anti-rabbit HRP secondary an-
tibody (Pierce). A supersignal chemiluminescence kit was used to detect the
Negative geotaxis assays and DDT sensitivity
Five-day-old adult ﬂies were raised on standard cornmeal/agar food and
starved overnight under humid conditions at 25ºC before treatment with
Regulation of detoxiﬁcation in Drosophila
CELL METABOLISM : JULY 2006 45
phenobarbital (PB). Treatment was conducted in plastic vials that contained
a strip of Whatman ﬁlter paper soaked with 500 ml of either 5% sucrose or 5%
sucrose and 1% PB. Eight adult ﬂies were placed in each vial. Activity was
scored 18 hr later at room temperature. To test for activity, the ﬂies were
banged to the bottom of the vial and, after 30 s, the number of ﬂies that
climbed 6.5 cm from the bottom was determined. Each experiment con-
sisted of three trials on each vial, with three vials tested for each condition,
for a total of nine data points. DDT resistance was determined by transferring
newly eclosed ﬂies to medium th at was supplemented with either 10, 50, or
250 mg DDT (Sigma) per vial. For each vial, 3 g of instant medium (Carolina 4-
24) was mixed with 10 ml of deionized water that contained the appropriate
DDT amounts. Flies were transferred to fresh medium once a week, and
scored after three weeks.
Total RNA was isolated from staged animals using phenol/chloroform extrac-
tion or Trizol (Gibco) and puriﬁed on RNAeasy columns (Qiagen). One- to two-
day-old adult ﬂies were raised on standard cornmeal/agar food and starved
overnight under humid conditions at 25ºC before treatment with either 5%
sucrose or 5% sucrose and 0.3% PB for 10 hr. This is a nonlethal concentra-
tion of PB that is sufﬁcient to direct signiﬁcant upregulation of Cyp6a2 and
Cyp6a8 transcription. For DHR96 overexpression, third instar larvae staged
at 4 hr after the second-to-third instar larval molt were treated at 37ºC for 1
hr and allowed to recover at 25ºC for 4 hr. All samples were prepared in trip-
licate to allow subsequent statistical analysis. Probe labeling, hybridization to
Affymetrix GeneChip Drosophila 2.0 Genome Arrays, and scanning were
performed by the University of Maryland Biotechnology Institute Microarray
Core Facility. Background/signal correction, normalization, and calculation
of probe set expression values was determined by gcRMA (Wu et al.,
2004). Calculation of fold changes and t-statistics was performed by SAM
2.0 (Tusher et al., 2001). Data sets were ranked by signiﬁcance and the top
3% upregulated and downregulated genes (563 genes each) that also dis-
played fold changes R1.3 were used for further analysi s. Microarray data-
sets were compared using Microsoft Access. Gene ontology terms are
based on Affymetrix annotation ﬁles (June 2005). To compare our data to
the stress microarray study of Girardot et al. (2004), we created a condensed
stress data set that included all stress-responsive genes independent of
treatment (the authors used ﬁve different stress conditions). For this, we cal-
culated the geometric mean of the fold changes and ﬁltered for gene expres-
sion changes that were on average >1.5, obtaining 222 upregulated and 211
Microarray data from this study can be accessed at NCBI GEO, with the
following accession numbers: GSE5096 for the PB-treated studies and
GSE5097 for the hsDHR96 gain-of-function studies. Excel spreadsheets
for the array data described in this paper can be accessed at the Thummel
lab website: http://thummel.g enetics.utah.edu/.
Supplemental data include Supplemental Experimental Procedures, Supple-
mental References, two ﬁgures, and two tables and can be found with this arti-
cle online at http://www.cellmetabolism.org/cgi/content/full/4/1/37/DC1/.
We thank E. Wimmer for the p5LP3-EGFP plasmid, B. Hall for the pCaSpeR-
FRT vector, Y. Rong and K. Golic for advice and stocks for gene targeting,
P. Batterham for suggesting the use of Canton S as a control for testing xeno-
biotic sensitivity, C. Keller for help with obtaining phenobarbital, A. Godinez
for assistance with microarray hybridizations and data collection, R. Weiss
for helpful sugges tions for microarray analysis, and K. Baker and R. Beck-
stead for helpful comments on the manuscript. This research was supported
by the Howard Hughes Medical Institute.
Received: November 22, 2005
Revised: April 27, 2006
Accepted: June 8, 2006
Published: July 4, 2006
Abdollahi, M., Ranjbar, A., Shadnia, S., Nikfar, S., and Rezaiee, A. (2004).
Pesticides and oxidative stress: a review. Med. Sci. Monit. 10, RA141–
Aminetzach, Y.T., Macpherson, J.M., and Petrov, D.A. (2005). Pesticide re-
sistance via transposition-mediated adaptive gene truncation in Drosophila.
Science 309, 764–767.
Bhat, R., and Bresnick, E. (1997). Glycine N-methyltransferase is an example
of functional diversity. Role as a polycyclic aromatic hydrocarbon-binding re-
ceptor. J. Biol. Chem. 272, 21221–21226.
Bogwitz, M.R., Chung, H., Magoc, L., Rigby, S., Wong, W., O’Keefe, M.,
McKenzie, J.A., Batterham, P., and Daborn, P.J. (2005). Cyp12a4 confers lu-
fenuron resistance in a natural population of Drosophila melanogaster. Proc.
Natl. Acad. Sci. USA 102, 12807–12812.
Boyd, L., O’Toole, E., and Thummel, C.S. (1991). Patterns of E74A RNA and
protein expression at the onset of metamorphosis in Drosophila. Develop-
ment 112, 981–995.
Brandt, A., Scharf, M., Pedra, J.H., Holmes, G., Dean, A., Kreitman, M., and
Pittendrigh, B.R. (2002). Differential expression and induction of two Dro-
sophila cytochrome P450 genes near the Rst(2)DDT locus. Insect Mol.
Biol. 11, 337–341.
Brun, A., Cuany, A., Le Mouel, T., Berge, J., and Amichot, M. (1996). Induc-
ibility of the Drosophila melanogaster cytochrome P450 gene, CYP6A2, by
phenobarbital in insecticide susceptible or resistant strains. Insect Biochem.
Mol. Biol. 26, 697–703.
Carroll, S.B., and Laughon, A. (1987). Production and puriﬁcation of poly-
clonal antibodies to the foreign segment of b-galactosidase fusion proteins.
In DNA cloning, D.M. Glover, ed. (Oxford: IRL Press), pp. 89–111.
Chawla, A., Repa, J.J., Evans, R.M., and Mangelsdorf, D.J. (2001). Nuclear
receptors and lipid physiology: opening the X-ﬁles. Science 294, 1866–1870.
Daborn, P., Boundy, S., Yen, J., Pittendrigh, B., and Constant, R. (2001). DDT
resistance in Drosophila correlates with Cyp6g1 over-expression and con-
fers cross-resistance to the neonicotinoid imidacloprid. Mol. Genet. Geno-
mics 266, 556–563.
Daborn, P.J., Yen, J.L., Bogwitz, M.R., Le Goff, G., Feil, E., Jeffers, S., Tijet,
N., Perry, T., Heckel, D., Batterham, P., et al. (2002). A single p450 allele as-
sociated with insecticide resistance in Drosophila. Science 297, 2253–2256.
Danielson, P.B., Foster, J.L., McMahill, M.M., Smith, M.K., and Fogleman,
J.C. (1998). Induction by alkaloids and phenobarbital of Family 4 Cytochrome
P450s in Drosophila: evidence for involvement in host plant utilization. Mol.
Gen. Genet. 259, 54–59.
Danielson, P.B., MacIntyre, R.J., and Fogleman, J.C. (1997). Molecular clon-
ing of a family of xenobiotic-inducible drosophilid cytochrome p450s: evi-
dence for involvement in host-plant allelochemical resistance. Proc. Natl.
Acad. Sci. USA 94, 10797–10802.
Dombrowski, S.M., Krishnan, R., Witte, M., Maitra, S., Diesing, C., Waters,
L.C., and Ganguly, R. (1998). Constitutive and barbital-induced expression
of the Cyp6a2 allele of a high producer strain of CYP6A2 in the genetic back-
ground of a low producer strain. Gene 221, 69–77.
Dunkov, B.C., Guzov, V.M., Mocelin, G., Shotkoski, F., Brun, A., Amichot, M.,
Ffrench-Constant, R.H., and Feyereisen, R. (1997). The Drosophila cyto-
chrome P450 gene Cyp6a2: structure, localization, heterologous expression,
and induction by phenobarbital. DNA Cell Biol. 16, 1345–1356.
Escriva, H., Bertrand, S., and Laudet, V. (2004). The evolution of the nuclear
receptor superfamily. Essays Biochem. 40, 11–26.
Ffrench-Constant, R.H., Daborn, P.J., and Le Goff, G. (2004). The genetics
and genomics of insecticide resistance. Trends Genet. 20, 163–170.
Fogleman, J.C. (2000). Response of Drosophila melanogaster to selection for
P450-mediated resistance to isoquinoline alkaloids. Chem. Biol. Interact.
46 CELL METABOLISM : JULY 2006
Francis, G.A., Fayard, E., Picard, F., and Auwerx, J. (2003). Nuclear receptors
and the control of metabolism. Annu. Rev. Physiol. 65, 261–311.
Girardot, F., Monnier, V., and Tricoire, H. (2004). Genome wide analysis of
common and speciﬁc stress responses in adult drosophila melanogaster.
BMC Genomics 5, 74.
Gong, H., Singh, S.V., Singh, S.P., Mu, Y., Lee, J.H., Saini, S.P., Toma, D.,
Ren, S., Kagan, V.E., Day, B.W., et al. (2006). Orphan nuclear receptor preg-
nane X receptor sensitizes oxidative stress responses in transgenic mice and
cancerous cells. Mol. Endocrinol. 20, 279–290.
Goodwin, B., and Moore, J.T. (2004). CAR: detailing new models. Trends
Pharmacol. Sci. 25, 437–441.
Hemingway, J., Hawkes, N.J., McCarroll, L., and Ranson, H. (2004). The mo-
lecular basis of insecticide resistance in mosquitoes. Insect Biochem. Mol.
Biol. 34, 653–665.
Inohara, N., and Nunez, G. (2002). ML–a conserved domain involved in innate
immunity and lipid metabolism. Trends Biochem. Sci. 27, 219–221.
Kliewer, S.A., Goodwin, B., and Willson, T.M. (2002). The nuclear pregnane X
receptor: a key regulator of xenobiotic metabolism. Endocr. Rev. 23, 687–702.
Knights, K.M., and Drogemuller, C.J. (2000). Xenobiotic-CoA ligases: kinetic
and molecular characterization. Curr. Drug Metab. 1, 49–66.
Kobayashi, K., Sueyoshi, T., Inoue, K., Moore, R., and Negishi, M. (2003). Cy-
toplasmic accu mulation of the nuclear receptor CAR by a tetratricopeptide
repeat protein in HepG2 cells. Mol. Pharmacol. 64, 1069–1075.
Kodama, S., Koike, C., Negishi, M., and Yamamoto, Y. (2004). Nuclear recep-
tors CAR and PXR cross talk with FOXO1 to regulate genes that encode
drug-metabolizing and gluconeogenic enzymes. Mol. Cell. Biol. 24,
Kume, E., Aruga, C., Ishizuka, Y., Takahashi, K., Miwa, S., Itoh, M., Fujimura,
H., Toriumi, W., Kitamura, K., and Doi, K. (2005). Gene expression proﬁling in
streptozotocin treated mouse liver using DNA microarray. Exp. Toxicol.
Pathol. 56, 235–244.
Lindblom, T.H., Pierce, G.J., and Sluder, A.E. (2001). A C. elegans orphan nu-
clear receptor contributes to xenobiotic resistance. Curr. Biol. 11, 864–868.
Lu, A.Y., and Miwa, G.T. (1980). Molecular properties and biological func-
tions of microsomal epoxide hydrase. Annu. Rev. Pharmacol. Toxicol. 20,
Maglich, J.M., Stoltz, C.M., Goodwin, B., Hawkins-Brown, D., Moore, J.T.,
and Kliewer, S.A. (2002). Nuclear pregnane x receptor and constitutive an-
drostane receptor regulate overlapping but distinct sets of genes involved
in xenobiotic detoxiﬁcation. Mol. Pharmacol. 62, 638–646.
Maitra, S., Dombrowski, S.M., Waters, L.C., and Ganguly, R. (1996). Three
second chromosome-linked clustered Cyp6 genes show differential consti-
tutive and barbital-induced expression in DDT-resistant and susceptible
strains of Drosophila melanogaster. Gene 180, 165–171.
Maurel, P. (1996). Cytochrome P450s: Metabolic and Toxicological Aspects.
In Cytochrome P450s: Metabolic and Toxicological Aspects, C. Ioannnides,
ed. (Boca Raton: CRC Press), pp. 241–270.
Menozzi, P., Shi, M.A., Lougarre, A., Tang, Z.H., and Fournier, D. (2004). Mu-
tations of acetylcholinesterase which confer insecticide resistance in Dro-
sophila melanogaster populations. BMC Evol. Biol. 4,4.
Miyo, T., Oguma, Y., and Charlesworth, B. (2003). The comparison of intrinsic
rates of increase among chromosome-substituted lines resistant and sus-
ceptible to organophosphate insecticides in Drosophila melanogaster.
Genes Genet. Syst. 78, 373–382.
Miyo, T., Takamori, H., Kono, Y., and Oguma, Y. (2001). Genetic variation and
correlations among responses to ﬁve insecticides within natural populations of
Drosophila melanogaster (Diptera: Drosophilidae). J. Econ. Entomol. 94, 223–232.
Pedra, J.H., McIntyre, L.M., Scharf, M.E., and Pittendrigh, B.R. (2004). Ge-
nome-wide transcription proﬁle of ﬁeld- and laboratory-selected dichlorodi-
phenyltrichloroethane (DDT)-resistant Drosophila. Proc. Natl. Acad. Sci. USA
Pyke, F.M., Bogwitz, M.R., Perry, T., Monk, A., Batterham, P., and McKenzie,
J.A. (2004). The genetic basis of resistance to diazinon in natural populations
of Drosophila melanogaster. Genetica 121
Ranson, H., Claudianos, C., Ortelli, F., Abgrall, C., Hemingway, J., Shara-
khova, M.V., Unger, M.F., Collins, F.H., and Feyereisen, R. (2002). Evolution
of supergene families associated with insecticide resistance. Science 298,
Reichert, K., and Menzel, R. (2005). Expression proﬁling of ﬁve different
xenobiotics using a Caenorhabditis elegans whole genome microarray.
Chemosphere 61, 229–237.
Rong, Y.S., and Golic, K.G. (2000). Gene targeting by homologous recombi-
nation in Drosophila. Science 288, 2013–2018.
Rong, Y.S., Titen, S.W., Xie, H.B., Golic, M.M., Bastiani, M., Bandyopadhyay,
P., Olivera, B.M., Brodsky, M., Rubin, G.M., and Golic, K.G. (2002). Targeted
mutagenesis by homologous recombination in D. melanogaster. Genes Dev.
Rosenfeld, J.M., Vargas, R., Jr., Xie, W., and Evans, R.M. (2003). Genetic pro-
ﬁling deﬁnes the xenobiotic gene network controlled by the nuclear receptor
pregnane X receptor. Mol. Endocrinol. 17, 1268–1282.
Rowlands, J.C., and Gustafsson, J.A. (1997). Aryl hydrocarbon receptor-me-
diated signal transduction. Crit. Rev. Toxicol. 27, 109–134.
Saner, C., Weibel, B., Wurgler, F.E., and Sengstag, C. (1996). Metabolism of
promutagens catalyzed by Drosophila melanogaster CYP6A2 enzyme in
Saccharomyces cerevisiae. Environ. Mol. Mutagen. 27, 46–58.
Squires, E.J., Sueyoshi, T., and Negishi, M. (2004). Cytoplasmic localization
of pregnane X receptor and ligand-dependent nuclear translocation in mouse
liver. J. Biol. Chem. 279, 49307–49314.
Swales, K., and Negishi, M. (2004). CAR, driving into the future. Mol. Endocri-
nol. 18, 1589–1598.
Touhara, K., and Prestwich, G.D. (1993). Juvenile hormone epoxid e hydro-
lase. Photoafﬁnity labeling, puriﬁcation, and characterization from tobacco
hornworm eggs. J. Biol. Chem. 268, 19604–19609.
Tusher, V.G., Tibshirani, R., and Chu, G. (2001). Signiﬁcance analysis of mi-
croarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci.
USA 98, 5116–5121.
Ueda, A., Hamadeh, H.K., Webb, H.K., Yamamoto, Y., Sueyoshi, T., Afshari,
C.A., Lehmann, J.M., and Negishi, M. (2002). Diverse roles of the nuclear or-
phan receptor CAR in regulating hepatic genes in response to phenobarbital.
Mol. Pharmacol. 61, 1–6.
Vontas, J., Blass, C., Koutsos, A.C., David, J.P., Kafatos, F.C., Louis, C.,
Hemingway, J., Christophides, G.K., and Ranson, H. (2005). Gene expres-
sion in insecticide resistant and susceptible Anopheles gambiae strains con-
stitutively or after insecticide exposure. Insect Mol. Biol. 14, 509–521.
Vontas, J.G., Small, G.J., and Hemingway, J. (2001). Glutathione S-transfer-
ases as antioxidant defence agents confer pyrethroid resistance in Nilapar-
vata lugens. Biochem. J. 357, 65–72.
Waxman, D.J. (1999). P450 gene induction by structurally diverse xeno-
chemicals: central role of nuclear receptors CAR, PXR, and PPAR. Arch. Bio-
chem. Biophys. 369, 11–23.
Wei, P., Zhang, J., Egan-Haﬂey, M., Liang, S., and Moore, D.D. (2000). The
nuclear receptor CAR mediates speciﬁc xenobiotic induction of drug metab-
olism. Nature 407, 920–923.
Willson, T.M., and Kliewer, S.A. (2002). PXR, CAR and drug metabolism. Nat.
Rev. Drug Discov. 1, 259–266.
Wilson, T.G. (2001). Resistance of Drosophila to toxins. Annu. Rev. Entomol.
Wu, Z., Irizarry, R., Gentlemen, R., Martinez-Murillo, F., and Spencer, F.
(2004). A model-based background adjustment for oligonucleotide expres-
sion arrays. J. Am. Stat. Assoc. 99, 909–917.
Xie, W., Barwick, J.L., Downes, M., Blumberg, B., Simon, C.M., Nelson, M.C.,
Neuschwander-Tetri, B.A., Brunt, E.M., Guzelian, P.S., and Evans, R.M.
Regulation of detoxiﬁcation in Drosophila
CELL METABOLISM : JULY 2006 47
(2000). Humanized xenobiotic response in mice expressing nuclear receptor
SXR. Nature 406, 435–439.
Zelko, I., and Negishi, M. (2000). Phenobarbital-elicited activation of nuclear
receptor CAR in induction of cytochrome P450 genes. Biochem. Biophys.
Res. Commun. 277, 1–6.
Zhang, J., Huang, W., Qatanani, M., Evans, R.M., and Moore, D.D. (2004).
The Constitutive Androstane Receptor and Pregnane X Receptor Function
Coordinately to Prevent Bile Acid-induced Hepatotoxicity. J. Biol. Chem.
Zhu, Y.C., Snodgrass, G.L., and Chen, M.S. (2004). Enhanced esterase gene
expression and activity in a malathion-resistant strain of the tarnished plant
bug, Lygus lineolaris. Insect Biochem. Mol. Biol. 34, 1175–1186.
Microarray data from this study can be accessed at NCBI GEO, with the fol-
lowing accession numbers: GSE5096 for the PB-treated studies and
GSE5097 for the hsDHR96 gain-of-function studies.
48 CELL METABOLISM : JULY 2006