A single nucleotide polymorphism in the acetylcholinesterase gene of 1 the predatory mite Kampimodromus aberrans (Acari Phytoseiidae) is 2 associated with chlorpyrifos resistance

17 18 The predatory mite Kampimodromus aberrans (Oudemans) (Acari Phytoseiidae) is one of 19 the most important biocontrol agents for herbivorous mites in European perennial crops. 20 The use of pesticides, such as organophosphate insecticides (OP), is a major threat to the 21 success of biocontrol strategies based on predatory mites in these cropping systems. 22 However, resistance to OP in K. aberrans has recently been reported. The present study 23 investigated the target site resistance mechanisms that are potentially involved in OP 24 insensitivity. In the herbivorous mite Tetranychus urticae, resistance to OP is due to a 25 modified and insensitive acetylcholinesterase (AChE) that bears amino acid substitution 26 F331W (AChE Torpedo numbering). To determine whether the predators and prey had 27 developed analogous molecular mechanisms to withstand the same selective pressure,

field release or in marker-assisted selection of improved populations of K. aberrans to achieve multiple resistance phenotypes through gene pyramiding.The latent complexity of the target site resistance in K. aberrans vs. that of T. urticae is also discussed by exploiting data from the genome project of the predatory mite Metaseiulus occidentalis (Nesbitt).

Introduction
Kampimodromus aberrans (Oudemans) (Acari: Phytoseiidae) is a predatory mite that occurs in various European cropping systems, such as grapevines, apples and hazelnuts (Ivancich Gambaro, 1973;El Borolossy and Fischer-Colbrie, 1989;Tsolakis et al., 2000;Ozman-Sullivan, 2006).This predatory mite is also common on the uncultivated plants that surround crops and represent a potential reservoir for biocontrol agents (Tixier et al., 2002(Tixier et al., , 2006)).K. aberrans is considered to be generalist predator (McMurtry and Croft, 1997;Kreiter et al., 2002;Broufas et al., 2007;Lorenzon et al., 2012) and an effective biocontrol agent of tetranychid and eriophyoid mites in European vineyards (Duso 1989;Girolami et al., 1992;Duso and Pasqualetto, 1993;Duso et al., 2012).In addition to several ecological factors, insecticide and fungicide applications strongly affect naturally occurring and artificially introduced K. aberrans populations (Ivancich Gambaro 1973;Girolami, 1987;Pozzebon et al., 2002;Peverieri et al., 2009).However, a K. aberrans strain has been detected in North Italian vineyards under conditions of integrated pest management strategies (IPM) that rely on common ethylene-bis-dithiocarbamate (EBCD) fungicides and organophosphate (OP) insecticides (Posenato 1994).This strain was also successfully released in other vineyards and apple orchards following organic or IPM strategies (Duso et al., 2007;Duso et al., 2009;Ahmad et al., 2013).Recently, laboratory studies have confirmed significant levels of chlorpyrifos resistance in this same strain (Tirello et al., 2012).The biochemical basis of OP resistance in phytoseiid mites depends on the active ingredients involved in the selective pressure and on species/strain-specific genetic backgrounds.The resistance phenotype might rely on high detoxifying enzyme activities and/or on a modified and insensitive target AChE.For example, laboratory selection with methidathion in Amblyseius womersleyi (Schicha) leads to increases in monooxygenase activity and CYP4-d isoform overexpression (Sato et al., 2001(Sato et al., , 2006(Sato et al., , 2007)), while in Phytoseiulus persimilis (Athias-Henriot), this selection results in an enhancement of glutathione transferase (Fournier et al., 1987).High rate in-vitro degradation of azinphosmethyl has been observed under both polygenic and monogenic control in OP-resistant strains of Amblyseius fallacis (Garman) (Motoyama et al., 1971;Croft et al., 1976).Target site resistance to OP has also been detected biochemically either in isolation or combination with enhanced OP detoxifying pathways.Resistance to certain OP and carbamate compounds, such as parathion and propoxur, in a Dutch strain of Typhlodromus pyri Scheuten has been found to be under monogenic control and to be associated with an insensitive target AChE (Overmeer and van Zon, 1983).In a paraoxonresistant strain of Amblyseius andersoni, (Chant) the resistant phenotype has been revealed to be due to an insensitive AChE coupled with modified carboxylesterases (Anber et al., 1988(Anber et al., , 1989)).
Although reductions in chlorpyrifos susceptibility have been reported in other predatory mites, e.g., T. pyri (Fitzgerald and Solomon 1999;Cross and Berrie 1994;Bonafos et al., 2008), little is known about the underlying molecular mechanisms.Among the Acari, high levels of chlorpyrifos resistance in Tetranychus urticae Koch have been found to be due to a F331W amino acid substitution in the target enzyme acetylcholinesterase (AChE) (Khajehali et al., 2010).Knowledge of a genetic marker associated with chlorpyrifos insensitivity in K. aberrans could be useful for understanding the amplitude of this phenomenon and managing predatory mite populations with IPM strategies.Therefore, we report the cloning and sequencing of a T. urticae-like acetylcholinesterase cDNA in K.
aberrans and its genotyping in chlorpyrifos-susceptible and resistant strains.The potential complexity of the target site resistance that occurs in predatory mites was also inferred by inspecting the annotated genome of M. occidentalis.

Kampimodromus aberrans populations
This study was performed on seven K. aberrans strains collected in North-eastern Italy (Veneto Region).Four strains were collected from commercial vineyards, and three strains were collected from untreated European nettle trees (Celtis australis L.) (Table 1).
All strains were reared without insecticide exposure in separate rearing units at the Department of Agronomy, Food, Natural Resources, Animals and the Environment of the University of Padova, Italy.Grapevine leaves on pads of wet cotton were used as a substrate for the predatory mites, and small pieces of PVC were placed for shelter and oviposition.Typha latifolia L. pollen was provided as food (Lorenzon et al., 2012).
Information about the effects of OP was available for only two strains; specifically, the PO strain is resistant to chlorpyrifos, and the LE strain is highly susceptible to this insecticide (Tirello et al., 2012).

Insecticide bioassays
Laboratory bioassays were conducted for the preliminary screening for resistant and susceptible phenotypes.The bioassays were performed with a commercial formulate (Dursban ® 75WG, 75% a.i., Dow AgroSciences).The discriminant concentration for the resistant and susceptible phenotypes was set at 70 g/hl of formulate, which is the recommended field dose for use in vineyards against grape berry moths and leafhoppers.
The pesticide formulate was diluted in distilled water before the toxicological test procedures (Tirello et al., 2013).The latter procedures were performed using rectangular leaf sections (approximately 6 cm 2 ).The sections were immersed in the insecticide solution for 30 s, and distilled water was used in the control treatments.When the pesticide residues completely dried out, the leaf sections were placed on wet cotton pads, and cotton barriers were created along their perimeters to prevent predatory mite escape.
Two 12-d-old K. aberrans females were gently transferred to each leaf section, and fresh pollen was provided as food.The experimental units were maintained in a climate chamber at 25 ± 2° C and 70 ± 10% relative humidity with a 16L:8D photoperiod.Female mortality was assessed 72 h after the treatments.The females that drowned or escaped were removed from the initial test number.In total, we assessed 40-45 females per strain.

Primer design for cloning AChE cDNA in K. aberrans
The annotated version of the genome assembly (release Mocc_1.0,March 2012) of the predatory mite Metaseiulus occidentalis (Nesbitt) (WOPM genome project) was used to search for putative AChE-like proteins with the tBlastn algorithm using the AChE sequence that was amplified from the susceptible strain of T. urticae (GenBank accession n.ADK12697.1)as the query sequence.
Transcripts predicted to code for putative AChE-like proteins were extracted from the scaffolds, and their open reading frames (ORFs) were compared to the T. urticae AChE protein using Lasergene sequence analysis tools EditSeq and MegAlign 5.0 (DNASTAR, Inc., Madison, WI, USA).Degenerate primers were designed by manual inspection of the conserved domains after the alignments of T. urticae AChE and putative orthologous AChE-like proteins in M. occidentalis.The resulting primers were used to amplify the cDNA core fragments of the orthologous AChE in K. aberrans.To complete the cloning, walking steps and 3'-5' RACEs, were performed using no degenerate primers and outlined with PrimerSelect 5.0 (DNASTAR, Inc., Madison, WI, USA).

mRNA extraction and AChE cDNA cloning
Total RNA was extracted by homogenising 200 adults in 500 µl Tri-Reagent (Sigma), according to the manufacturer's instructions.The sample integrities were examined by electrophoresis in 1.2% agarose and 2.2 M formamide/formaldehyde denaturing gel.
Quality and quantity assessments of the extracted RNA were performed in a Nanodrop ND-1000 Spectrophotometer (NanoDrop, Fisher Thermo, Wilmington, DE, USA).Firststrand cDNA was synthesised according to the protocol recommended by the supplier using Improm-II reverse transcriptase (Promega) and random primers.Amplification of a cDNA fragment for a putative AChE in K. aberrans was achieved through two consecutive rounds of reverse-transcription PCR (RT-PCR) with degenerate primers.The PCR mixtures (25 µl) contained GoTaq Flexi 1x buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 30 pmol forward and reverse degenerate primers, 1 U GoTaq and 2.5 µl of cDNA.The degenerate primers were based on partially conserved functional domains of homologous AChEs proteins in T. urticae and M. occidentalis.For the first RT-PCR round, the forward and reverse primers were KaAChEF1d (GIPYAKP domain) and KaAChER1d or KaAChER2d (WVYGGSF motif) (Table 2).The PCR product was then diluted 10-fold and used as the template for a second RT-PCR in which the KaAChEF1 primer was replaced with the more internal primer KaAChEF2d, which was designed based on the PYAKPP domain (Table 2).
The two PCR rounds shared the following profile: an initial denaturation step of 3 min at 94 °C; 5 cycles at 94 °C for 30 s, 45 °C for 30 s and 72°C for 60 s; 5 cycles at 94 °C for 30 s, 45 °C plus +1 °C/cycle and 72°C for 60 s; 25 cycles at 94 °C for 30 s, 50 °C for 30 s and 72°C for 60 s; and a final extension step at 72 °C for 10 min.PCR products of the expected size (approximately 300 bp) were purified from 1% (w/v) agarose/TBE 1x gel using a EuroGOLD Gel Extraction Kit (Euroclone) and cloned using a pGEM-T easy vector (Promega).The plasmids were purified with a EuroGOLD Plasmid Miniprep Kit (Euroclone) and sent for sequencing at BMR genomics (Padua, Italy).The sequences were assembled and analysed using SeqMan 5.0 (DNASTAR, Inc., Madison, WI, USA).
Identification of the AChE-like sequences was performed via a BLASTX search in GenBank (http://www.ncbi.nkm.nih.gov) using the ORFs deduced from the cloned cDNA fragments.The cDNA clones were further extended in the 3' direction by performing an RT-PCR that used a forward primer that was designed based on the first cloned cDNA fragment in K. aberrans (KaAChEF3) and a reverse primer (KaAChER3) that was designed based on the sequence coding for the conserved domain CAFWKNFL in both of the AChE transcripts found in M. occidentalis without any primer degeneration (Table 2).
The RT-PCR mixture had the same composition described above except that the primer concentration was reduced to 15 pmol.The PCR was performed as follows: 1 cycle of 94 °C for 2 min; 5 cycles that included the three steps of 94°C for 30 s, 50 °C for 30 s and 72 °C for 60 s; 5 cycles of 94 °C for 30 s, 50 °C for 30 s (+ 1 °C/cycle) and 72 °C for 60 s; 20 cycles of 94°C for 30 s, 55 °C for 30 s and 72 °C for 60 s; and a final extension step at 72 °C for 10 min.The PCR product was purified, sequenced and analysed as described above.Three prime and 5' rapid amplification of cDNA ends reactions (RACEs) were performed to complete the AChE cDNA sequences.In the RACE reactions, the first strand cDNAs were synthesised using total RNA and polyT-adaptor primer for 3' RACE or KaAChE-R4 for 5' RACE (Table 2) according to the manufacturer's protocol (5´ RACE System for Rapid Amplification of cDNA Ends, Invitrogen).The 3' RACE product spanning across the unknown 3'-end of the AChE cDNA was amplified in two consecutive PCR rounds with the KaAChEF4-Adaptor1 and KaAChEF5-Adaptor2 primer pairs.To obtain the 5' end of the AChE transcript, the cDNA was subjected to polyC-tailing of its 3'-end with terminal deoxynucleotidyl-transferase (TdT) following the protocol of the kit (5´ RACE System for Rapid Amplification of cDNA Ends, Invitrogen).The upstream cDNA sequence encompassing the 5' untranslated region was amplified with two PCR rounds using the coupled primers KaAChER5-TS-primer and KaAChER6-TS-PCR (Table 2).The first 5' RACE round was performed as follows: 94 °C for 2 min (1 cycle); 5 cycles at 94 °C, 56 °C for 30 s and 72 °C for 60 s; 5 cycles at 94 °C, 57 °C for 30 s and 72 °C for 60 s; and 20 cycles at 94 °C for 30 s, 58 °C for 20 s and 72 °C for 60 s.The second 5' RACE round consisted of the following: 1 cycle at 94 °C for 2 min; and 30 cycles at 94 °C for 30 s, 55 °C for 30 s and 72 °C for 60 s.The 5' RACE fragment was purified from the agarose gel and sequenced as previously described.

Full length AChE cDNA sequencing
Total RNA was extracted from adults of both sensitive (LE) and resistant (PO) strains with TRI-Reagent described for the cDNA cloning.First-strand cDNA was synthesised from total RNA with Improm-II reverse transcriptase (Promega) and random primers as indicated by manufacturer's protocol.To sequence the ORF of the cloned cDNA, three RT-PCR fragments that partially overlapped were generated using the following primer couples: KaAChEF6-R7, KaAChEF7-R8, and KaAChEF8-R9 (Table 2).The PCR reaction (25 µl) included 2 µl of cDNA, a final concentration of GoTaq Flexi 1x buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 0.6 µM of each primer and 0.625 U/µl GoTaq (Promega).The thermal profile adopted was as follows: 94°C for 2 min (1 cycle); 30 cycles of 94 °C for 30 s, 56 °C for 30 s and 72 °C 60 s; and a final extension step at 72 °C for 10 min.The PCR products were checked by electrophoresis on 1% agarose in TBE 0.5x buffer, purified with the EuroGOLD Cycle-Pure Kit (Euroclone) and sent to BMR genomics (Padua, Italy) for sequencing.To this aim, the same primers used for the RT-PCR amplifications and new internal primers (KaAChEF9, R10, and F10) were used (Table 2).Chromatograms were assembled with SeqMan tools (DNAstar, Lasergene), and the alignments of the cDNA consensus sequences from sensitive and resistant strains were manually inspected for non-synonymous SNPs with the MegAlign program (DNAstar, Lasergene).

DNA extraction and exon-intron junction amplification
DNA extraction was performed according to the methods described by Tixier et al. (2008) while scaling up the reagents.Two hundred frozen adults of each strain were homogenised in 150 µl of extraction buffer (2% CTAB, 1.4 M NaCl, 0.2% 2-mercaptoethaol, 100 mM EDTA, and 100 mM Tris-HCl at pH 8.0) using a micro tissue grinder (Wheaton, Millville, NJ).The homogenate was transferred to a 0.5-ml test tube and incubated for one hour at 65°C with periodic hand mixing.One hundred and fifty microliters of a chloroform:isoamyl alcohol mixture (24:1) was added, the solution was mixed by inversion, and tubes were centrifuged at 6°C for 5 min at 1000 g.The aqueous solution was collected in a new test tube, and 80 µl of isopropanol was added to the decanted aqueous phase, which was then chilled at -20 °C for 20 min for DNA precipitation.After centrifugation (15 min, 6 °C, 1000 g), the pellet was suspended in 100 µl of 96% alcohol at 4 °C.After a final centrifugation of 10 min (6 °C, 1000 g), the dried pellet was suspended in 30 µl of deionised water.The quality and quantity of the extracted DNA were assayed by spectrophotometric analyses with a Nanodrop ND-1000, and the integrities were verified through electrophoresis on 1% agarose/TBE 0.5x gel.Exon-intron boundary predictions were made by aligning the AChE cDNA sequence cloned in K. aberrans with the scaffold form the M. occidentalis genome project from which the transcript XR_145413 had been predicted (Genbank accession n.AFFJ01003151.1).Relying on hypothetical gene structure conservation, the primers were designed on the exon sequences to generate partially overlapping PCR fragments that encompassed the putative introns in the K.
aberrans AChE gene.The PCR products were purified and sequenced as described for the cDNA sequencing using the same primers that were employed for the DNA amplifications.

Insecticide bioassays
Laboratory trials confirmed the findings reported by Tirello et al. (2012).At the discriminant dose, 100% corrected mortalities were observed for the LE, PA and PD strains, which originated from untreated nettles, and high survival rates were observed for the PO, ME, SF and BX strains (4.08%, 9.57%, 7.69% and 4.26% corrected mortalities, respectively), which were collected from commercial vineyards.

T. urticae AChE-like gene in the M. occidentalis genome
The tBlastn search on the annotated genome of the predator mite M. occidentalis using the AChE cloned from T. urticae as the query sequence resulted in two predicted mRNAs that codes for putative AChEs and had sequences that were significant similar to that of the query (XR_145413 and XR_145279; identity 54%, positive 69%, e-value 0.0).These mRNAs originated from genes in partially overlapping contigs (AFFJ01003151 and AFFJ01002402).The corresponding open reading frames differed primarily in the amino terminal due to a diverse prediction of the first splicing site, while they shared the remaining five, which resulted in only 8 mismatches out of the 593 conserved amino acid residues.These mismatches arose from indels in the coding regions of the two genes, which did not differ in the intronic sequences with the exception of the first intron that originated from alternative splicing paths.Because the algorithms used for automatic splicing site predictions often fail to identify splicing sites at the 5' end of putative transcripts and because of the low level of sequence divergence between the two genes, it was unclear whether there were two copies per genome or if they were derived from in silico mis-assembling of the high-throughput sequencing reads.In any case, when the M. occidentalis transcriptome shotgun assembly was interrogated with Blastn with the two putative transcripts, a pair of cDNA fragments were retrieved that covered both mRNAs (JL046593.1 and JL050556.1;identities 99% and 98%), which confirmed that they were actually transcribed.Altogether, these findings suggested that the two very similar predicted mRNAs could be informative for cloning T. urticae-like AChE cDNA in K. aberrans.
3.3.AChE cDNA in the susceptible strains of K. aberrans cDNA of 2329 was isolated from the susceptible LE strain (Genbank accession number: HF934042).The deduced precursor was composed of 655 amino acids (Fig. 1) with a signal peptide that was predicted to encompass the first 32 amino acids from the amino terminal (Shen et al. 2007).The cloned KaAChE displayed most of the amino acids responsible for the functional integrity of the enzyme that are typically well conserved both in insect and mite AChEs; i.e., the KaAChE residues involved in the intramolecular disulphide bonds (C139, C166, C325, C336, C471, and C593), the catalytic triad (S271, E395, and H509), the anionic subsite (W156), the oxianion-hole (G189, G190, and A274), and the acyl pocket (W304, F360, and F399) (Fig. 1).The highest identity (> 93%) was observed for the AChE that was annotated in M. occidentalis from the transcript XR_145413 because the first splicing path was consistent with that predicted in this putative mRNA.No alternative cDNA sequence similar to the M. occidentalis transcript XR_145279 was detected in K. aberrans.As expected, the greatest divergences in the amino acid sequences between the KaAChE and XR_145413 predicted AChEs were restricted to the amino and carboxy terminals of the protein outside of the functional domains.The amino acid identities with the other cloned and predicted AChEs in the Acari genomes that carry multiple AChE loci ranged from 61% (Ixodes scapularis putative AChE, XP_002413212) to 33% (Rhipicephalus microplus, AChE3, AAP92139).The amino acid identity was 52% between the AChEs coded by single copy genes in the T. urticae and T.
evansi that carry mutations associated with reduced chlorpyrifos sensitivity (GQ461344, ADK12694, and AFS60097) This divergence was compatible with that observed in the AChEs from different species of Acari and even between AChEs from multiple loci in the I. scapularis or R. microplus genomes.AChEs of insects are divided in two groups, i.e., those orthologous and those paralogous to the D. melanogaster AChE (Kim et al., 2012), and KaAChE exhibited a high level of similarity to the paralogous AChEs found in Nephotettix cincticeps (Hemiptera: Deltocephalidae) and Blattella germanica (Blattodea: Blattellidae) with an amino acid identity of approximately 57% (ADZ15146; ABB89946).

Organisation of the clone AChE locus in K. aberrans
The intron-spanning amplifications of the K. aberrans AChE locus confirmed the exonintron junctions that were predicted in silico in the M. occidentalis genome scaffold AFFJ01003151, which lead to XR_145413 transcript annotation and coding for a putative M. occidentalis AChE (MoAChE).However, the first 106 nucleotides of the 5' UTR region of the KaAChE cDNA did not match with any portion of the scaffold sequence AFFJ01003151.In contrast, the unmatched 5'UTR portion of the KaAChE cDNA exhibited an 81% identity with segments of two partially overlapping scaffolds in the M. occidentalis genome (Genbank accession n.AFFJ01002403 and AFFJ01002403).The GT-AG consensus rule for donor and acceptor splice sites was also respected using the KaAChE cDNA sequence to guide the joining of the putative and still unannotated 5'UTR portion of MoAChE on the scaffolds AFFJ01002403 and AFFJ01002403 to the 5' end of the remaining open reading frame relying on the AFFJ01003151 scaffold.Because the AFFJ01002403 and AFFJ01002403 scaffolds do not overlap with AFFJ01003151, a long intronic sequence has to be envisaged in the MoAChE locus and is likely excluded from the assembly step.Assuming intron size conservation between the two phytoseiids, this hypothesis was supported by the unsuccessful amplification of this intron in the K. aberrans AChE locus.Although we were able to characterise 5 introns experimentally and an additional putative splicing site bioinformatically, we suggest that the KaAChE gene includes seven exons (I-VII) that are separated by 6 introns (Table 3, (Genbank accession n.HG328327).Exon I is non-coding, whereas exon II contains the initiation codon (ATG), which is similar to the observations of the majority of the AChE gene loci that have been annotated in insects and mites.Exons III-VI formed the catalytic domain and exhibited partially amino acid conservation across the AChEs that were cloned from the mites.Exon exon seven contains the stop codon (TAG) and the 3' UTR region.All intron-exon boundaries followed the GT-AG rule (Breathnach et al., 1978); furthermore, these boundaries contained the YTNAN consensus sequence for lariat formation at the branch point close to the 5' end of the acceptor-splicing site.In addition to the positions, the lengths of the amplifiable introns were also conserved in the homologous AChE loci from the two phytoseiidae species with the exception of the third intron, which was slightly longer in the K. aberrans than in the M. occidentalis AChE gene (1162 bp vs. 936 base pairs, respectively).Sequence inspection of the third intron in the K. aberrans AChE locus revealed the presence of short microsatellite repeats and a long inverted repeat (LIR) (Wang et al., 2006).These nucleotide motifs can cause sliding of the intron sequences during DNA replication and might account for the different sizes of the third intron in the KaAChE gene.

Comparison the AChE cDNA sequences across different strains
Full-length sequencing of the KaAChE cDNA of the susceptible (LE) and resistant (PO) strains revealed a non-synonymous G to A mutation at position 687 that led to a G191S substitution in the protein sequence (G119S AChE Torpedo numbering; Fig. 1).This residue is involved in the oxianion hole, which is one of the functional domains of AChE activity (Zhang et al. 2002).The susceptible and resistant strains also differed in another single nucleotide polymorphism (SNP) at position 1499 of the cloned cDNA; this C to T transition did not affect the codon for the D461 residue.The resistant strain was homozygous at this site and carried only the T allele, while the sensitive strain exhibited both SNPs with a preference for C over T as indicated by the electropherograms.The phenylalanine residue (F339) that was replaced by a tryptophan in the chlorpyrifosresistant stains of T. urticae (F439W mutation, or F331W AChE, Torpedo numbering) was still conserved both in the susceptible and resistant strains of K. aberrans.The same was true for the glycine residue (G336) that was found to be replaced with alanine (G328A) in the F331W-bearing strains of T. urticae.The cDNA KaAChE sequences of two additional susceptible (PA and PD) and three resistant (ME, SF and BX) strains of K. aberrans were also examined.The resistant strains were all homozygous for the G191S substitution, while the susceptible strains carried only the G191 allele.The resistant BX strain sequence differed from the other strains in a SNP in the 3' UTR that consisted of a G to A substitution.

Discussion
Resistance to pesticides can be a desirable feature in K. aberrans because this predatory mite is an effective biocontrol agent for spider mites in perennial crops.Indeed, strains of this predatory mite that are apparently resistant to OP have successfully been released in vineyards and apple orchards in which the pest control strategies included chlorpyrifos and many other pesticides (Duso et al. 2009;Duso et al., 2012;Ahmad et al., 2013).The resistance to chlorpyrifos of these strains has been definitively demonstrated (Tirello et al., 2012), but the underlying molecular mechanisms remain poorly understood.An initial clue about this issue came from the chlorpyrifos-resistant strain of T. pyri that exhibited a lower level of AChE activity that the susceptible strain, which suggests that the reduced substrate affinity observed in the biochemical assay might be associated with a modified AChE (Fitzgerald and Solomon, 1999).In Acari, target site resistance due to a modified AChE that confers high levels of insensitivity to OPs, including chlorpyrifos, has been described in T. urticae and Tetranychus kanzawai Kishida (Aiki et al., 2004;Van Leeuwen et al. 2010;Khajehali et al., 2010).A G119S substitution (AChE torpedo numbering) in the single copy AChE gene has been associated with the moderate decreases in chlorpyrifos susceptibility between the resistant compared to the sensitive strains of T. urticae (resistance ratio at LD50, RR50 = 31), and a greater resistance ratio has been detected in cases of F331W replacement (RR50 > 400).In vitro expression of the AChE isoforms of T. urticae that carry F331W and/or G119S substitutions has revealed a reduction in sensitivity to another organophosphate (monocrotophos) and a decrease in the catalytic efficiency of the enzyme (Kwon et al., 2012); however, no data had been reported for chlorpyrifos.Although these features appeared much more evident in the F331W-mutated AChE, the two substitutions acted synergistically when they were associated in vitro and thus were favourably co-selected in the resistant strains in vivo (Kwon et al., 2010b;Ilias et al., 2014).In mosquitoes, two acetylcholinesterase genes are present, both substitutions affect the paralogous AChE in the highly OP-and carbamate-resistant strains of Culex pipiens L. and Anopheles gambiae Giles (G119S; Weill et al., 2004a) and in Culex tritaeniorhynchus Giles (F331W; Alout et al., 2007).The role of these mutations in reducing the AChE sensitivity to OP was confirmed via inhibition analysis of the expression of AChE from mutated mosquitos S2 cells (Weill et al., 2003;Oh et al., 2006).The F331W substitution has also been detected in AChE1 of a chlorpyrifos-resistant strain of the sweet potato whitefly Bemisia tabaci Gennadius (Alon et al., 2008).The K. aberrans strain with the highest level of insensitivity to chlorpyrifos (PO strain) described by Tirello et al. (2012) has a RR50 = 539,602, and this ratio is even higher than that found in the T. urticae and B.
tabaci populations with the F331W AChE genotype.There a target site resistance might be present in that strain.Because no AChE sequences for K. aberrans are stored in databases, the annotated genome project of the predatory mite M. occidentalis was inspected.More than a dozen AChEs-like sequences were predicted by the curators of the M. occidentalis genome project using an automated computational analysis, although some of the transcripts represented uncompleted open reading frames or differed only in their splicing paths.To identify a suitable AChE candidate that is potentially responsible for target site resistance in K. aberrans, the AChE protein sequence from T. urticae was used to probe probing the annotated genome of M. occidentalis.Once a putative homologous AChE in M. occidentalis was found, its sequence was used to speed up the cloning of the corresponding AChE cDNA in K. aberrans.Full sequencing of the cloned AChE cDNA revealed that the resistant strain (PO) differed from the susceptible strain (LE) in terms of non-synonymous G to A mutation that introduced a G191S substitution in the AChE open reading frame.That mutation corresponds to the aforementioned G119S substitution in AChE Torpedo numbering.Strangely, the corresponding amino acid position in the homologous AChE that was found in the annotated genome of M. occidentalis is occupied by a serine.Unfortunately no information is available concerning the chlorpyrifos susceptibilities of the M. occidentalis strains employed for the genome project.Notably, the G119 in the K. aberrans AChE is encoded by a GGC codon, which could easily be converted to the AGC codon for serine.This substitution in the mosquito paralogous AChE seems not to be neutral under the selective pressure produced by organophosphate and carbamate treatments (Weill et al. 2004b).Indeed, when KaAChE cDNA from an additional two chlorpyrifos-susceptible and three chlorpyrifos-resistant unrelated strains of K. aberrans, the G191S substitution was absent only in the resistant strains in the in homozygous condition.The silent nucleotide polymorphisms found in the KaAChE cDNA of the resistant strains might may also suggest that different G119S mutation events occurred independently.Although, in Culex quinquefasciatus Say, a chlorpyrifos inhibition study of a paralogous AChE bearing the G119S substitution revealed a reduced sensitivity to the insecticide (Liu et al., 2005) that likely resulted from the reduced accessibility of the catalytic site (Weill et al., 2004a).In vivo, T. urticae strains with the same mutated AChE genotype display only a moderate resistance to chlorpyrifos.In contrast, all examined resistant strains of K. aberrans are highly resistant to the insecticide (Tirello et al., 2012).
Thus, the role of the G119S remains unclear, although the possibility that the same mutation has different effects on chlorpyrifos AChE sensitivity in the predatory mite cannot be ruled out.Nevertheless, the non-silent nucleotide polymorphism responsible for the G119S substitution represents a reliable molecular marker that is associated with the detected resistant phenotype.Alternative target site resistance mechanisms not explored in this study might also rely on the amplification of modified AChEs as occurs in T. urticae and T. evansi (Kwon et al., 2010a;Carvalho et al., 2012) or on mutations that affect multiple AChE loci with additive effects such as has been observed in R. microplus (Temeyer et al., 2009(Temeyer et al., , 2010(Temeyer et al., , 2012)).In either case, the co-expression of sensitive and insensitive AChEs might contribute to reducing the fitness costs associated with OP resistance (Carvalho et al., 2012, Temeyer et al. 2013a).Alternatively, the overexpression of distinct AChEs from different loci it is thought to result in to bio-scavenging due to the supply of excess targets for xenobiotics, including OP and carbamate insecticides (Lee et al. 2014, in press).While the detection of homozygosity for the G119S substitution argues against the co-existence of duplicated sensitive and insensitive AChEs in chlorpyrifosresistant strains, the role of multiple AChE loci in K. aberrans cannot be not excluded.
Although these putative AChEs have lower amino acid identities (< 34%) to those found in T. urticae, they exhibit conserved functional residues for acetylcholinesterase activity (i.e., the form the catalytic triad and the acetylcholine binding pocket), exhibit conserved amino acid positions, are potentially involved in substitutions that affect AChE sensitivity to organophosphates and carbamates, have and received support from transcriptomics analyses (Hoy et al., 2013).Together, these findings suggest that, in predatory mites, multiple AChEs resemble the composite picture observed in ticks (Temeyer et al. 2013b).
Thus, herbivorous and predatory mites can differ not only in detoxification pathways (Mullin et al., 1982;Grbic et al., 2001;Dermauw et al., 2012) but also in AChE repertoires, which potentially offers alternative solutions for the development of target site resistance.
However, chlorpyrifos inhibition assays of AChE activity should be performed to support this hypothesis in examined OP resistant strains.Although, the contribution of enhanced detoxifying activities to chlorpyrifos resistance in predatory mites has not yet been reported as it has for other organophosphates (Sato et al., 2001, Fournier et al., 1987;Motoyama et al., 1971;Anber et al., 1988), the use of synergists in bioassays should be combined with detoxification enzyme assays to confirm or deny the involvement of metabolic resistance.
In summary, the potential target site resistance to chlorpyrifos in K. aberrans has barely been dissected compared to that in T. urticae.The F331W substitution that is responsible for AChE that is highly insensitive to chlorpyrifos in Tetranychidae was absent in a putative homologous gene that was cloned from the resistant strain K. aberrans.However, a G119S mutation that was detected in the same gene appeared to be associated with the resistant phenotype.Because pesticide treatments strongly affect the success of predatory mites release (Ahmad et al., 2013), this polymorphism might be useful as a molecular marker for tracing the resistant phenotype in ecological studies or in gene pyramiding and marker-assisted selection of desirable traits for multiple insecticide resistance.

Not included
Fig.1Alignment of the AChEs cloned in from the chlorpyrifos -susceptible strain of Kampimodromus aberrans (KaAChE) and predicted from Metaseiulus occidentalis transcript XR_145413 (MoAChE).Identical amino acids are indicated by asterisks, and conservative substitutions are indicated by dots.The cleavage site of signal peptide is indicated by a slash.The mutated residue (G191S) in the chlopryrifos-resistant strain is in reverse in the background, the cysteine residues that form the intramolecular disulphide bonds are numbered and on the light-gray background (C139-C166, C325-C336, C471-C593), the catalytic triad residues are boxed (S271, E395, H509), and the following conserved residues are indicated with plus signs: anionic subsite (W156), oxianion-hole (G189, G190, A274), acyl pocket (W304, F360, F399), and cysteine residue forming

Table 3 .
Genome organisation of the K. aberrans ace locus The numbering of nucleotides is based on the K. aberrans AChE cDNA in which +1 corresponds to start codon.