© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society

125

Insect Molecular Biology (2008)

17

(2), 125–135 doi: 10.1111/j.1365-2583.2007.00788.x

Blackwell Publishing Ltd

Characterization of inhibitors and substrates of

Anopheles gambiae

CYP6Z2

L. A. Mclaughlin*, U. Niazi†, J. Bibby†, J.-P. David‡, J. Vontas§, J. Hemingway¶, H. Ranson¶, M. J. Sutcliffe† and M. J. I. Paine¶

*

Biomedical Research Centre, University of Dundee, Ninewells Hospital and Medical School, Dundee, United Kingdom;

†

Manchester Interdisciplinary Biocentre & School of Chemical Engineering and Analytical Science, University of Manchester, Manchester, United Kingdom;

‡

Laboratoire d’Ecologie Alpine, UMR 5553 Université Joseph Fourier – CNRS, Grenoble, France;

§

Laboratory of Pesticide Science, Agricultural University of Athens, Greece; and

¶

Liverpool School of Tropical Medicine, Liverpool, United Kingdom

Abstract

Three

CYP6Z

genes are linked to a major pyrethroidresistance locus in the mosquito

Anopheles gambiae

.We have expressed CYP6Z2 in

Escherichia coli

andproduced a structural model in order to examine itsrole in detoxification.

E. coli

membranes co-express-ing CYP6Z2 and

An. gambiae

P450 reductase (AgCPR)catalysed the dealkylation of benzyloxyresorufin withkinetic parameters

K

m

====

0.13 μμμμ

M;

K

cat

====

1.5 min

−−−−

1

. TheIC

50

values of a wide range of compounds were meas-ured. Pyrethroids cypermethrin and permethrin pro-duced low IC

50

values, but were not metabolized. Plantflavanoids were the most potent inhibitors. Severalcompounds were shown to be substrates, suggestingthat CYP6Z2 has broad substrate specificity and playsan important chemo-protective role during the her-bivorous phase of the life-cycle.

Keywords: P450, insecticide metabolism, mosquito,malaria, P450 reductase.

Introduction

The mosquito

Anopheles gambiae

is the major vector for

Plasmodium falciparum

, the parasite responsible for malariawhich affects several million people worldwide. The use ofinsecticides, in particular those belonging to the pyrethroidclass, is an important component of malaria control pro-grammes. However, resistance to pyrethroids has startedto emerge, which poses a major threat to the control ofmosquito-borne disease (N’guessan

et al

., 2007). Thecytochrome P450 gene family, which catalyses the chemicaltransformation of both xenobiotic and endogenous com-pounds, is associated with insecticide resistance throughincreased metabolism of insecticides (Feyereisen, 1999;Ranson

et al

., 2002). Although the

An. gambiae

genomecontains over one hundred P450 genes, only a handful ofthese have thus far been linked to insecticide resistance.

CYP6Z1

and

CYP325A3

are overexpressed in an estab-lished pyrethroid-resistant laboratory strain of

An. gambiae

(David

et al

., 2005), whereas

CYP6Z2

and

CYP6M2

areoverexpressed in a pyrethroid-resistant field strain fromOdumasy, Ghana (Muller

et al

., 2007).

CYP6Z

genes havealso been associated with constitutive pyrethroid resistanceby genetic mapping and quantitative PCR (Nikou

et al

., 2003).As yet none of these enzymes have been functionallycharacterized.

We have focused here on the

CYP6Z

gene family, whichis comprised of four genes, three of which are tightly clus-tered on chromosome 3R (Nikou

et al

., 2003). Interestingly,all three have distinctive differences in life stage expression;

CYP6Z1

is predominately expressed in adults, whereas

CYP6Z2

is expressed at high levels in the larvae and to alesser extent in adults, and finally

CYP6Z3

is expressedpredominantly in larvae and pupae (Nikou

et al

., 2003). Themolecular basis for this variation in developmental expres-sion remains unclear but presumably reflects differencesin genetic control mechanisms and endogenous function(Nikou

et al

., 2003). Here we have sought to define thefunction of these P450s by expressing CYP6Z2 in

Escherichiacoli

with its cognate redox partner

An.

gambiae

NADPHP450 reductase (AgCPR) and examining its inhibition andsubstrate preferences. Alongside, we have used homologymodelling and docking techniques to gain insight at the

Received 10 September 2007; accepted after revision 6 December 2007;first published online 15 February 2008. Correspondence: Mark J. I. Paine,Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L35QA, United Kingdom. Tel.:

+

44 (0)151 705 3310; fax:

+

44 (0)151 705 3370;e-mail: m.j.paine@liverpool.ac.uk

126

L. A. Mclaughlin

et al.

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society,

17

, 125–135

atomic level into the mode of binding of ligands (includingpyrethroids) in CYP6Z2.

Results

Functional expression of CYP6Z2 in

E. coli

To express functional CYP6Z2 in

E. coli omp

A and

pel

Bsignal sequences were used to direct P450 and AgCPR tothe inner bacterial membrane to form a functional monoox-ygenase complex – a strategy used previously for humanP450s (Pritchard

et al

., 1997, 1998). For expression,

E. coli

cells were co-transformed with the expression plasmids

ompA

-CYP6Z2 and pACYC-AgCPR. High yields of between600–700 nmol P450/litre of bacterial culture were generatedas measured by Fe

2

+

-CO vs. Fe

2

+

difference spectra inbacterial whole cells and membranes (Fig. 1). CYP6Z2

was expressed predominantly as P450 with only a low levelof P420, indicative of stable good quality functional enzyme.Consistent with previous observations, P450 yields were~25% higher in the absence of AgCPR. AgCPR expressionwas measured from bacterial membranes by reduction ofcytochrome

c

. Membrane preparations contained 0.9–1.5 nmoles P450/mg, whereas the P450 reductase activitywas in the range of 100–120 nmol/min/mg.

Identification of a fluorescent substrate probe for CYP6Z2

We examined the metabolism of ethoxy-, methoxy-, benzy-loxy-, and pentoxyresorufin (ER, MR, BR and PR) to identifypotential fluorogenic substrate probes for CYP6Z2. As shownin Table 1, BR produced the highest specific activity andwas further examined with respect to the kinetic parametersof benzyloxyresorufin O-debenzylation (BROD) (Fig. 2).

Table 1. Specific activities of CYP6Z2 towards a series of resorufin analogues

Figure 1. Fe2+-CO vs. Fe2+ difference spectrum determined from Escherichia coli membranes expressing CYP6Z2. The spectrum was determined from 0.25 mg protein in 100 mM Tris-Cl, pH 7.4; containing 20% (v/v) glycerol, 10 mM CHAPS and 1 mM EDTA. P450 content was 1.5 nmol P450/mg protein.

Figure 2. Michaelis–Menten plot of CYP6Z2-benzyloxyresorufin kinetics. Km is expressed as μM, and Kcat as pmol resorufin produced/min/pmol CYP6Z2.

Mosquito CYP6Z2 characterization

127

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society,

17

, 125–135

The reaction followed simple Michaelis–Menten kineticswith a

K

m

of 0.13

μ

M and

K

cat

of 1.5 min

−

1

, and this compoundwas used as a fluorogenic probe for inhibition assays.

Identifying inhibitors and substrates of CYP6Z2

Forty-two compounds of diverse size and compositionwere screened for their ability to inhibit CYP6Z2-mediatedBROD activity through measurement of IC

50

; i.e. the valueat which the compound inhibited P450 activity by 50%. Thesewere divided into six functional categories that includeddietary phytoestrogens and phytoallexins, insecticides,insect hormones/precursors, steroids and drug molecules.The compounds and resulting IC

50

values are listed inTables 2–6.

In general, the phytoestrogens/phytoallexins showedthe lowest IC

50

values (Table 2). Strongest inhibition wasdemonstrated by

α

-naphthoflavone (ANF) with an IC

50

of7 nM. However, the majority of the other compounds hadIC

50

values in the low micromolar/nanomolar range demon-strating a clear preference for this group of plant-derivedcompounds.

Three of the insecticides tested showed significant inhi-bition including the pyrethroids cypermethrin (IC

50

=

0.3

μ

M)and permethrin (IC

50

=

3.0

μ

M) and the organochlorinemethoxychlor (IC

50

=

3.2

μ

M) (Table 3). The two per-

methrin metabolites n-phenoxy benzyl alcohol and 3-phenoxybenzoic acid showed increased IC

50

s of 47.2 and911

μ

M respectively.The only endogenous insect hormone to show significant

inhibition was juvenile hormone III (JH-III) with an IC

50

of6.6

μ

M (Table 4), whereas the ecdysteroids were essen-tially non-inhibitory. By contrast, the mammalian/syntheticsteroid compounds inhibited with IC

50

values in the range2.6

μ

M (megasterol acetate) to 25.6

μ

M (corticosterone)(Table 5). Finally, the drug compounds exhibited a broadspectrum of IC

50

values reflecting their structural diversity(Table 6). The two azole antifungals, clotrimazole and keto-conazole, produced low IC

50

values of 3.2 and 7.4

μ

M,respectively, while ivermectin, a large macrocyclic lactone,also had a low IC

50

of 6.6

μ

M.The ability of CYP6Z2 to metabolize the pyrethroid insec-

ticides permethrin and cypermethrin was examined along-side several compounds selected from the other functionalgroups (Table 7). Neither permethrin nor cypermethrin weremetabolized – consistent with the binding mode observedin our modelling studies (Fig. 3A,B). However, the plantcompounds resveratrol (consistent with our modellingstudies), enterodiol, butein, ANF, piceatannol and 8-MOPwere metabolized with turnover rates between 1.5–0.53 min

−

1

,as were the drug compound diltiazem (0.75 min

−

1

) and the

Table 2. IC50 values for phytoestrogens/phytoalexins

128

L. A. Mclaughlin

et al.

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society,

17

, 125–135

Table 2. Continued

*Mechanistic inhibition was observed for methoxypsoralen (8-MOP) under standard assay conditions. IC50 data was therefore generated using 0.75 pmol P450 and 1 min incubation, during which rates were linear.

Mosquito CYP6Z2 characterization

129

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society,

17

, 125–135

steroid testosterone (0.21 min

−

1

) (Table 7). Overall thesedata suggest that CYP6Z2 binds and metabolizes a broadrange of xenobiotic substrates which is consistent with arole in detoxification.

Discussion

Recent studies have linked CYP6Z2 with pyrethroid resist-ance in both laboratory and field strains of

An. gambiae

Table 3. IC50 values for insecticides

Table 4. IC50 values for insect hormones and precursors

130 L. A. Mclaughlin et al.

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

(Nikou et al., 2003; David et al., 2005; Muller et al., 2007).We have therefore investigated interactions of this enzymewith pyrethroids and other compounds to understand itsrole in detoxification. Adopting systems used for the heter-ologous expression of human drug-metabolizing P450s inE. coli, we have produced high yields of functional CYP6Z2.A wide variety of chemical compounds was screened todetermine ligand specificity and functional links with xeno-biotic and endogenous metabolism. CYP6Z2 demonstrateda preference for the plant compounds, particularly thosewith planar polyphenolic structures; ANF, a prototypeflavanoid, had the lowest IC50 of 7 nM and several other fla-vones bound in the nanomolar range (Table 2). To examinelinks with endogenous metabolism we assessed inhibitionby ecdysteroids, which promote moulting, and JH-III, whichis involved with insect development and metamorphosis.CYP6Z2 is not inhibited by the ecdysteroid compounds(Table 4), while a speculative link with the biosynthesis ofJH-III was indicated by the inhibition of BROD activity withan IC50 of 6.3 μM.

Increased constitutive gene expression of CYP6Z1, Z2and Z3 is elevated in pyrethroid-resistant populations ofAn. gambiae. CYP6Z1 expression is elevated in the RSPstrain, originating from East Africa and used in the presentstudy (Nikou et al., 2003), whereas CYP6Z2 and CYP6Z3are expressed at higher levels in the permethrin-resistantOdumasy strain from Ghana, West Africa, comparatively toa standard insecticide-susceptible strain (Muller et al., 2007).These findings, together with the location of the CYP6Zgene cluster within the boundaries of a major quantitativetrait locus associated with permethrin resistance in An. gambiae(Ranson et al., 2004), strongly support their involvement ininsecticide resistance.

Our data indicate that although CYP6Z2 binds permethrinand cypermethrin, neither of these pyrethroids are metab-olized. This could be a result of the fact that CYP6Z2 is unableto metabolize these pyrethroids because of structural con-straints within the active site, or it may be that cytochromeb5 is required to promote pyrethroid metabolism as is thecase for house-fly CYP6D1 (Zhang & Scott, 1996). The roleof cytochrome b5 is still under investigation. However, anonproductive binding mode is supported by our modellingstudies. These predict the two aromatic residues – Phe 212and Phe 222 – to be a major determinant in pyrethroid bind-ing (Fig. 3A,B), positioning an inert chlorine atom close tothe catalytic haem iron (additionally Lys 48 and/or Asn 369formed hydrogen bonds to the ethoxy benzyl moiety insome of our modelled complexes). Phe 212 and Phe 222are not conserved in CYP6D1 (Fig. 4), where the equivalentresidues are Lys 213 and Asn 222. This leads to a differentbinding mode in CYP6D1 – a mode consistent with theexperimentally observed 4′ hydroxylation (Fig. 3C,D,E)(Scott et al., 1998).

Overall, CYP6Z2 displays broad substrate specificitywhich one might associate with xenobiotic metabolism anddetoxification. Interestingly, the active site appears selectivefor planar polyaromatic hydrocarbon molecules, which werethe strongest inhibitors – consistent with our modelling,which suggests that a key determinant is the interaction ofthe two aromatic rings on the ligand with the aromaticsidechains of Phe 212 and Phe 222. There was also anapparent intolerance of the six-carbon chain at positionC17 in the ecdysteroids against the structurally relatedmammalian steroids, which bound reasonably well, as wellas a preference for an extra phenolic group in BR comparedwith methox-, ethoxy- and pentoxy-resorufin derivatives.

Table 5. IC50 values for steroids

Mosquito CYP6Z2 characterization 131

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

Table 6. IC50 values for drugs

132 L. A. Mclaughlin et al.

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

Interestingly, Table 1 shows that increasing the substituentsize of the alkyl group lowers the specific activity. Our dock-ings suggest that this can be explained in terms of the alkylmoieties being accommodated above the heme – in con-trast to the larger benzyl moiety – and that increased stericscreening of the oxidation site (O-alkylation) occurs withincreasing alkyl chain length, as previously suggested(Liu et al., 2003).

An. gambiae larvae are aquatic herbivorous feeders andmost are exposed to dietary and environmental toxins.Thus, the expression of CYP6Z2 in larvae (Nikou et al., 2003)coupled with the obvious preference of CYP6Z2 for planttoxins indicates that this gene may play an important chemo-protective role during this phase of the life-cycle. This is inagreement with the finding of a closely related CYP6 gene(56% protein similarity) in the yellow fever mosquito Aedes

Figure 3. Modelled binding modes of permethrin and cypermethrin in the active sites of CYP6Z2 and CYP6D1. (A) Permethrin in CYP6Z2; (B) cypermethrin in CYP6Z2; (C) permethrin in CYP6D1; (D) cypermethrin in CYP6D1; and (E) arrowed point of metabolic attack in cypermethrin and permethrin. Predicted protein-ligand hydrogen bonds are depicted by dashed black lines. R = H in permethrin and R = CN in cypermethrin.

Table 7. CYP6Z2 metabolism: substrate turnover

Compound Type Turnover*

Resveratrol Stilbene 1.53 ± 0.03Enterodiol Lignan 1.25 ± 0.16Butein Chalcone 0.93 ± 0.004ANF Flavone 0.79 ± 0.04Diltiazem Drug 0.75 ± 0.08Piceatannol Stilbene 0.55 ± 0.028-MOP Furanocoumarin 0.53 ± 0.004Testosterone Steroid 0.21 ± 0.10Morin Flavanol NDMethoxychlor Organochlorine NDPermethrin Pyrethroid NDCypermethrin Pyrethroid NDQuercetin Flavanol NDIvermectin Macrocyclic lactone ND

*Turnover measured as nmoles substrate depleted/nmole P450/min, ND: not detected.

Mosquito CYP6Z2 characterization 133

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

aegypti (CYP6AL1) showing an induction in fourth stagelarvae fed on decomposed tree leaf litter.

Experimental procedures

Chemicals

All chemicals unless otherwise stated, were purchased fromSigma-Aldrich, Gillingham, U.K.

Cloning and co-expression of and AgCPR

CYP6Z2 and AgCPR cDNAs were isolated from the pyrethroid-resistant strain An. gambiae RSP as described in Nikou et al.(2003). There were 19 single nucleotide differences relative to theGENBANK sequence (AF487780) (Nikou et al., 2003), all of whichwere all silent. However, a six-nucleotide insertion sequence wasfound that encoded two extra amino acids, Leu-Arg, at positions441 and 442, respectively. An identical Leu-Arg sequence is present

Figure 4. Sequence alignment of CYP6Z2, CYP6D1 and CYP3A4. Amino acid residues that are identical in the structural template (CYP3A4) and CYP6Z2 and/or CYP6D1 are shown in white on a black background. Residue numbering shown above the alignment is that for CYP6Z2, secondary structure shown above the alignment is that for the X-ray structure of CYP3A4 (PDB accession code 1TQN, Yano et al., 2004), and the positions of the substrate recognition sites (SRS1-6; Gotoh, 1992) are indicated below the alignment. Swissprot accession codes for CYP6Z2 and CYP6D1 are Q8T625 and AAC99340, respectively.

134 L. A. Mclaughlin et al.

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

in both CYP6Z1 and CYP6Z3, indicating this is the bona fidesequence. To make the construct ompA 6Z2, the CYP6Z2 cDNAwas modified at the 5′ end by mutating position +2 to alanine andmaking the next seven amino acids AT rich (silent mutations),which improves P450 yields. This modified cDNA was fused to abacterial ompA leader sequence by fusion PCR as previouslydescribed (Pritchard et al., 1997, 1998). The following PCR prim-ers were used; ompA forward primer: 5′-G GAA TTC CAT ATGAAA AAG ACA GCT AT-3′ (NdeI site underlined, initiation codon inbold); ompA/CYP6Z2 fusion primer (reverse orientation): 5′-GAGCAC GAG AAA GAT CAC GGC CGC AAC AAG AGC AAG AGTATA AAC AGC CAT CGG AGC GGC CTG CTG CGC TAC GGT AGCGAA-3′; CYP6Z2 reverse primer: 5′-CGG GAA TTC TCA CTT TCTATG GTC TAT CCT CAT-3′ (EcoRI site underlined, stop codon in bold).The PCR fragment was digested with NdeI/EcoRI and ligated intopB13 (modified pCW vector Pritchard et al., 1998). For co-expression,AgCPR cDNA was fused by PCR to a pelB leader sequence (forwardprimer: 5′-C GGG ATC CAT ATG AAA TAC CTG CTG CCG ACCGCT GCT GCT GCT CTG CTG CTC CTC GCT GCC CAG CCGGCG ATG GCC ATG GAC GCC CAG ACA GAA ACG GAA GTG-3′(BamHI site underlined and start codon in bold) and reverse primer:5′-CCG GAA TTC TTA GCT CCA CAC GTC CGC CGA GTA TCGTTT-3′ (EcoRI site underlined and stop codon in bold). The PCRproduct was digested with BamHI/EcoRI and cloned into pB13. ThecDNA sequence was confirmed by sequencing. A cassette containingthe Ptac-Ptac promotor from pB13 and pelB AgCPR was excised fromthis construct by EcoRV/BglII digest and cloned into EcoRV/BamHIdigested pACYC 184 (New England Biolabs, Hitchin, UK), which dis-rupts the tetracycline resistance gene, to produce pACYC-AgCPR.

For functional expression, competent E. coli JM109 cells wereco-transformed with ompA 6Z2 and pACYC-AgCPR, and mem-branes co-expressing CYP6Z2 and AgCPR prepared as previouslydescribed (Pritchard et al., 1998; Paine et al., 2003). Membraneswere characterized based on recombinant P450 content as esti-mated by Fe2+-CO vs. Fe2+ difference spectra, and AgCPR content,as estimated by cytochrome c reductase activity (Pritchard et al.,1998; Paine et al., 2003).

Enzyme activity measurements

O-dealkylation (OD) reactions were carried out in white 96-wellplates (Thermo Labsystems, Basingstoke, U.K.) in 50 mM potas-sium phosphate pH 7.4, 5 μM substrate BR, ER, MR or PR, bac-terial membranes containing 5 pmol of CYP6Z2 and 0.125 mMNADPH, in a total volume of 200 μl. For determination of BROD Km

reactions contained 0–1.2 μM BR, 1 pmol of CYP6Z2. Plates werepre-warmed for 5 min at 30 °C before reactions were initiated byaddition of NADPH. Production of the resorufin metabolite wasmeasured for 2 min at λex530; λex585 using a Fluoroskan AscentFL microtitre plate reader (λex = 530, λem = 585 nm) (ThermoLabsystems). Turnover rates were calculated with reference to anauthentic resorufin standard curve.

Ligand IC50 determinations

Assays were performed in duplicate using 96-well microtitre platesin a final volume of 200 μl consisting of 50 mM potassium phosphatepH 7.4; 7.9 pmol/ml CYP6Z2, 0.12 μM BR (the Km for CYP6Z2), andvariable concentrations of test ligand. Ligand stocks were dissolvedin methanol or DMSO and a solvent control was included to correctfor any solvent effects across the dilution range. IC50 values were cal-culated using GRAFIT 5.0.4 (Erithacus Software Ltd., Surrey, U.K.).

Metabolism studies

Substrates were incubated with 50 pmol of CYP6Z2 at 30 °C for 0,8, 16 and 32 min in the presence or absence of an NADPH gener-ating system (5 mM glucose 6-phosphate, 1 U glucose 6-phosphatedehydrogenase, 1 mM NADP) in a final volume of 200 μl. Reactionswere terminated by the addition of 100 μl of ice-cold methanol.After 10 min ice-incubation, samples were centrifuged at 16 000 gbefore transferring supernatant to fresh high-performance liquidchromatography (HPLC) vials. Substrate concentrations used were25 μM for cypermethrin, permethrin, methoxychlor, invermectin,enterodiol, testosterone and diltiazem, and 5 μM for 8-MOP, ANF,piceatannol, resveratrol, butein, quercetin and morin. Most substratepeaks were separated using a 250 × 4.6 mm (5 μm) SupelcosilLC-18-DB column (Supelco, Sigma-Aldrich, Gillingham, U.K.) at25 °C, except for diltiazem: [250 × 4.6 mm (5 μm) Hypersil C18-BDS column (Agilent Technologies, Stockport, U.K.)] and testo-sterone: [250 × 4.6 mm (5 μm) Hypersil ODS column]. Elutionconditions and monitoring wavelengths are detailed in Table 8.

Modelling CYP6Z2-ligand complexes

Proteins of known three-dimensional structure homologous toCYP6Z2 (ie suitable structural templates) were identified by scan-ning the amino acid sequence of CYP6Z2 against the sequencesof those structures in the Protein Data Bank (PDB; Berman et al.,2000) using PSI-BLAST (Altschul et al., 1997). This resulted in CYP3A4(Yano et al., 2004; PDB accession code 1TQN) being identified asa suitable template. The amino acid sequence of CYP6Z2 wasaligned against the sequence of CYP3A4 (Fig. 4) using 3DCoffee(O’Sullivan et al., 2004). The resultant alignment was used to gen-erate a set of 30 models using MODELLER (Sali & Blundell, 1993).A final model was selected on the basis of the best MODELLER scoreand the best Errat profile (Colovos & Yeates, 1993). Docking stud-ies were carried out using the program GOLD v3.1 (Jones et al.,1997) with the ChemScore scoring function (Eldridge et al., 1997;

Table 8. HPLC conditions used for xenobiotic metabolism assays

Compound Mobile phase* (v/v) UV wavelength

Permethrin 80% A: 20% B 250 nmCypermethrin 80% A: 20% B 250 nmMethoxychlor 80% A: 20% B 280 nmIvermectin 90% A: 10% B 240 nm8-MOP 60% A: 40% B 300 nmANF 80% A: 20% B 280 nmPiceatannol Gradient I 320 nmResveratrol 30% A: 70% B 300 nmButein Gradient I 380 nmQuercetin Gradient I 380 nmMorin Gradient I 380 nmGenestein 30% A: 70% B 280 nmEnterodiol 35% A: 65% B 280 nmTestosterone Gradient II 240 nmDiltiazem 35% A: 65% 0.1 M

Ammonium acetate pH 5.8238 nm

*A, acetonitrile; B, dH2O; C, methanol; Gradient I, 0 min = 10% A: 90% B; 4 min = 50% A: 50% B; 5 min = 5% A: 95% B; 9 min = 5% A: 95% B; 10 min = 10% A: 90% B; 0 min = 73% A: 27% B (All v/v); Gradient II, 0 min = 2.5% A: 45% B: 52.5% C; 3 min = 2.5% A: 45% B: 52.5% C; 8 min = 2.5% A: 25% B:72.5% C; 11 min = 2.5% A: 25% B: 72.5% C; 11.5 min = 0 min = 2.5% A: 45% B: 52.5% C (All v/v)

Mosquito CYP6Z2 characterization 135

© 2008 The AuthorsJournal compilation © 2008 The Royal Entomological Society, 17, 125–135

Verdonk et al., 2003) and an active site radius of 20 Å; ligandstructures were obtained from ChemIDPlus (http://chem.sis.nlm.nih.gov/chemidplus/). Fifty binding modes were obtained for eachligand (the pyrethroids cypermethrin and permethrin, the plantcompounds resveratol and morin and the resorufin analoguebenzyloxyresorufin).

Modelling CYP6D1-ligand complexes

The same procedure was followed for CYP6D as is described abovefor CYP6Z2. Scanning the amino acid sequence of CYP6D1(Swissprot accession code AAC99340) against the sequences ofthose structures in the PDB identified CYP3A4 (Yano et al., 2004;1TQN) as a suitable template. Using GOLD v3.1, 50 binding modeswere obtained for each ligand (the pyrethroids cypermethrin andpermethrin).

Acknowledgements

This work was supported by grants from the Wellcome Trust(to H.R. and J.H.), the Innovative Vector Control Consortium(IVCC) (J.H. and M.P.) and studentships from the NationalUniversity of Sciences and Technology, Pakistan and theEPSRC for U.N. and J.B., respectively. We are grateful to Prof.C. Roland Wolf for the supply of reagents and plasmids.

References

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z.,Miller, W. et al. (1997) Gapped BLAST and PSI-BLAST: a newgeneration of protein database search programs. NucleicAcids Res 25: 3389–3402.

Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N.,Weissig, H. et al. (2000) The Protein Data Bank. Nucleic AcidsRes 28: 235–242.

Colovos, C. and Yeates, T.O. (1993) Verification of protein struc-tures: patterns of nonbonded atomic interactions. Protein Sci 2:1511–1519.

David, J.P., Strode, C., Vontas, J., Nikou, D., Vaughan, A., Pignatelli,P.M. et al. (2005) The Anopheles gambiae detoxification chip:a highly specific microarray to study metabolic-based insecti-cide resistance in malaria vectors. Proc Natl Acad Sci USA102: 4080–4084.

Eldridge, M.D., Murray, C.W., Auton, T.R., Paolini, G.V. and Mee, R.P.(1997) Empirical scoring functions: I. The development of a fastempirical scoring function to estimate the binding affinity of ligandsin receptor complexes. J Comput Aided Mol Des 11: 425–445.

Feyereisen, R. (1999) Insect P450 enzymes. Annu Rev Entomol44: 507–533.

Gotoh, O. (1992) Substrate recognition sites in cytochrome P450family 2 (CYP2) proteins inferred from comparative analyses ofamino acid and coding nucleotide sequences. J Biol Chem267: 83–90.

Jones, G., Willett, P., Glen, R.C., Leach, A.R. and Taylor, R. (1997)Development and validation of a genetic algorithm for flexibledocking. J Mol Biol 267: 727–748.

Liu, J., Ericksen, S.S., Besspiata, D., Fisher, C.W. and Szklarz, G.D.(2003) Characterization of substrate binding to cytochromeP450 1A1 using molecular modeling and kinetic analyses: caseof residue 382. Drug Metab Dispos 31: 412–420.

Muller, P., Donnelly, M.J. and Ranson, H. (2007) Transcription pro-filing of a recently colonised pyrethroid resistant Anophelesgambiae strain from Ghana. BMC Genomics 8: 36.

N’guessan, R., Corbel, V., Akogbeto, M. and Rowland, M. (2007)Reduced efficacy of insecticide-treated nets and indoor residualspraying for malaria control in pyrethroid resistance area, Benin.Emerg Infect Dis 13: 199–206.

Nikou, D., Ranson, H. and Hemingway, J. (2003) An adult-specificCYP6 P450 gene is overexpressed in a pyrethroid-resistantstrain of the malaria vector, Anopheles gambiae. Gene 318:91–102.

O’Sullivan, O., Suhre, K., Abergel, C., Higgins, D.G. and Notre-dame, C. (2004) 3DCoffee: combining protein sequences andstructures within multiple sequence alignments. J Mol Biol 340:385–395.

Paine, M.J.I., Mclaughlin, L.A., Flanagan, J.U., Kemp, C.A., Sutcliffe,M.J., Roberts, G.C. et al. (2003) Residues glutamate 216 andaspartate 301 are key determinants of substrate specificity andproduct regioselectivity in cytochrome P450 2D6. J Biol Chem278: 4021–4027.

Pritchard, M.P., Ossetian, R., Li, D.N., Henderson, C.J., Burchell, B.,Wolf, C.R. et al. (1997) A general strategy for the expression ofrecombinant human cytochrome P450s in Escherichia coli usingbacterial signal peptides: expression of CYP3A4, CYP2A6, andCYP2E1. Arch Biochem Biophys 345: 342–354.

Pritchard, M.P., Glancey, M.J., Blake, J.A., Gilham, D.E., Burchell, B.,Wolf, C.R. et al. (1998) Functional co-expression of CYP2D6and human NADPH-cytochrome P450 reductase in Escherichiacoli. Pharmacogenetics 8: 33–42.

Ranson, H., Claudianos, C., Ortelli, F., Abgrall, C., Hemingway, J.,Sharakhova, M.V. et al. (2002) Evolution of supergenefamilies associated with insecticide resistance. Science 298:179–181.

Ranson, H., Paton, M.B., Jensen, B., Mccarroll, L., Vaughan, A.,Hogan, J.R. et al. (2004) Genetic mapping of genes conferringpermethrin resistance in the malaria vector, Anopheles gam-biae. Insect Mol Biol 13: 379–386.

Sali, A. and Blundell, T.L. (1993) Comparative protein model-ling by satisfaction of spatial restraints. J Mol Biol 234: 779–815.

Scott, J.G., Liu, N. and Wen, Z. (1998) Insect cytochromes P450:diversity, insecticide resistance and tolerance to plant toxins.Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 121:147–55.

Verdonk, M.L., Cole, J.C., Hartshorn, M.J., Murray, C.W. andTaylor, R.D. (2003) Improved protein-ligand docking usingGOLD. Proteins 52: 609–623.

Yano, J.K., Wester, M.R., Schoch, G.A., Griffin, K.J., Stout, C.D.and Johnson, E.F. (2004) The structure of human microsomalcytochrome P450 3A4 determined by X-ray crystallography to2.05-A resolution. J Biol Chem 279: 38091–3804.

Zhang, M. and Scott, J.G. (1996) Cytochrome b5 is essential forcytochrome P450 6D1-mediated cypermethrin resistance inLPR house flies. Pestic Biochem Physiol 55: 150–156.