ORIGINAL PAPER

Glutamate-gated chloride channel subunit cDNA sequencingof Cochliomyia hominivorax (Diptera: Calliphoridae): cDNAvariants and polymorphisms

Alberto Moura Mendes Lopes • Renato Assis de Carvalho •

Ana Maria Lima de Azeredo-Espin

Received: 15 March 2014 / Accepted: 29 May 2014

� Springer-Verlag Berlin Heidelberg 2014

Abstract The New World screwworm (NWS) Cochlio-

myia hominivorax (Coquerel) is one of the major myiasis-

causing flies that injures livestock and leads to losses of

*US$ 2.7 billions/year in the Neotropics. Ivermectin

(IVM), a macrocyclic lactone (ML), is the most used pre-

ventive insecticide for this parasite and targets the gluta-

mate-gated chloride (GLUCLa) channels. Several authors

have associated altered GluCla homologues to MLs resis-

tance in invertebrates, although studies about resistance in

NWS are limited to other genes. Here, we aimed to char-

acterise the NWS GluCla (ChGluCla) cDNA and to search

for alterations associated with IVM resistance in NWS

larvae from a bioassay. The open reading frame of the

ChGluCla comprised 1,359 bp and encoded a sequence of

452 amino acids. The ChGluCla cDNAs of the bioassay

larvae showed different sequences that could be splice

variants, which agree with the occurrence of alternative

splicing in GluCla homologues. In addition, we found

cDNAs with premature stop codons and the K242R SNP,

which occurred more frequently in the surviving larvae and

was located close to mutation (L256F) involved in ML

resistance. Although these alterations were in low fre-

quency, the ChGluCla sequencing will allow further stud-

ies to find alterations in the gene of resistant natural

populations.

Keywords GluCla � Myiasis � Cattle � Macrocyclic

lactone � Resistance

Introduction

The New World screwworm (NWS) fly, Cochliomyia

hominivorax (Coquerel), is one of the worst parasites that

injure livestock population in South American countries,

except Chile. This species has a myiasis-causing larval

stage on its life cycle (Hall and Wall 1995) that leads to

reduced fertility, weight and milk production (Welch and

Hall 2008), and to economic losses of *US$ 2.7 billions/

year in the Neotropics (Vargas-Teran et al. 2005).

Although the releasing of sterile males in infested areas

(sterile insect technique) results in the control of NWS in

North and Central America (Vargas-Teran et al. 2005), in

South America control is accomplished by the use of

chemical insecticides such as the macrocyclic lactones

(MLs) (Welch and Hall 2008), which has led to the

emergence of resistance.

Ivermectin (IVM) is the most profitable macrocyclic

lactone used in the world (Omura and Crump 2004). It

targets the members of the cys-loop ligand-gated ion-

channel (cys-loop LGIC) superfamily, mainly the gluta-

mate-gated chloride channels (GLUCLa) (Wolstenholme

and Rogers 2005), which occur by the co-assemble of five

subunits in the neuromuscular cell membranes of inverte-

brates (Cleland 1996). The IVM binds at the interface of

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10158-014-0172-6) contains supplementarymaterial, which is available to authorized users.

A. M. M. Lopes (&) � R. A. de Carvalho �A. M. L. de Azeredo-Espin

Laboratory of Genetics and Animal Evolution, Department of

Genetics, Evolution and Bioagents, Center of Molecular Biology

and Genetic Engineering (CBMEG), University of Campinas,

6010, Campinas, SP 13083-875, Brazil

e-mail: bmml321@gmail.com

R. A. de Carvalho

e-mail: rassis@gmail.com

A. M. L. de Azeredo-Espin

e-mail: azeredo@unicamp.br

123

Invert Neurosci

DOI 10.1007/s10158-014-0172-6

GLUCLa subunits (Hibbs and Gouaux 2011) and irre-

versibly opens it (Cully et al. 1994), which causes the

parasite death. The members of this ion-channel super-

family extend their diversity by post-transcriptional alter-

ations such as alternative splicing and RNA editing in

many arthropods (Drosophila melanogaster, Apis melli-

fera, Nasonia vitripennis) (Semenov and Pak 1999; Jones

and Sattelle 2006; Jones et al. 2010).

Several authors have associated ML-resistant strains to

mutations in GluCla homologues such as in Cooperia on-

cophora (L256F) (Njue et al. 2004), Tetranychus urticae

(G323D; G326E) (Kwon et al. 2010; Dermauw et al. 2012)

and D. melanogaster (P299S) (Kane et al. 2000). In addi-

tion, artificially induced mutations in Haemonchus con-

tortus GluCla (L256F/W/Y; P316S) (McCavera et al.

2009; Yamaguchi et al. 2012) reduced the level of MLs

binding. Nevertheless, studies about molecular resistance

in NWS are focused on other genes and insecticides. Based

on these mutations in GluCla genes, the high similarity of

amino acid sequence among the cys-loop LGICs (Ortells

and Lunt 1995), and the rise of IVM inefficiency against

NWS, it is possible that the NWS GluCla (ChGluCla)

could be an IVM target-site and that resistance could be

linked to alterations in this gene. Here, we aimed to char-

acterise the ChGluCla cDNA and perform an IVM bioas-

say to search for polymorphisms probably associated with

resistance. This study will help further researchers to

investigate new alterations linked to IVM resistance, which

would improve the NWS control in the field.

Materials and methods

NWS samples

The NWS larvae used were derived from a strain originally

collected in 2006 directly from infested livestock animals

on farms with a history of MLs use, located in the

municipality of Caiaponia (16�57S/51�48 W), State of

Goias, Brazil. They were reared to adults and maintained in

the laboratory conditions for several generations

(*6 years) according to Vargas and Espin (1995), without

selection, prior to RNA extraction. The NWS larvae used

for GluCla characterisation and bioassay evaluation were

of the L3 and L2 life stages, respectively.

Bioassay with IVM

Due to the lack of well-characterised NWS IVM-resistant

and IVM-susceptible reference strains to study molecular

mechanisms of resistance, we carried out an IVM bioassay

and compared the GluCla cDNA of the bioassay larvae to

find alterations probably associated with resistance. The

bioassay was carried out in the laboratory in February

2011. The commercial formulation of IVM 1 % used is a

mixture of C22, C23-dihydroavermectin B1a and B1b, in

the proportion of about 90 and 10 %, respectively (IVO-

MEC�, Merial Saude Animal, Brazil). Four IVM con-

centrations were tested with three replications of 30 larvae

for each concentration replicate (90 larvae for triplicate)

and a control group of 30 larvae, resulting in a total of 390

larvae. The control group had the same characteristics of

the replicates with IVM, but without it. The sample size of

each replicate in the bioassay was chosen based on rec-

ommendations of Robertson et al. (1984). The IVM con-

centrations of 0.32 ppm—3.2 lL of IVOMEC� 1 %

solution in 100 g of culture medium—1.0, 3.2 and 10 ppm,

were prepared mixing the respective IVM dose, separately,

for about 20 s, with a mixture of blood and water (2:1)

(30 mL) into plastic conic tubes (50 mL) capped with a lid.

The respective IVM mixture was subsequently mixed with

ground meat (70 g) in a container (11 cm diameter, 4 cm

high) capped with a screen lid, to allow for air and moisture

exchange. The containers were subsequently put in a dark

room, due to IVM instability under light, at room tem-

perature (*25 �C). The L2 larvae tested were randomly

chosen, removed from the rearing diet medium and sub-

sequently held into the respective containers with (IVM

groups) or without (control group) IVM. We scored mor-

tality 24 h post-treatment, and 15 dead larvae in the lowest

IVM concentration (ChSusc) and 15 surviving larvae in the

highest IVM concentration (ChIVR) were randomly chosen

and removed with tweezers from the containers, cleaned

and dried on paper towels, put into plastic conic tubes and

immediately stored at -70 �C. The larvae that showed

some locomotion were considered alive. The PoloPlus�

software (Jacqueline L. Robertson 2007) was used to

estimate the lethal concentrations (LC50 and LC90), their

confidence intervals (95 % CI) and the significance of the

slope of the regression line by a t test.

RNA extraction, cDNA synthesis, ChGluCla cloning

and sequencing

Total RNA was extracted from NWS L3 larvae (n = 4) for

the GluCla characterisation and from L2 larvae of the

bioassay groups, ChSusc (n = 15) and ChIVR (n = 15),

using Trizol treatment (Life TechnologiesTM) according to

the manufacturer’s instructions. Samples were treated with

TURBOTM DNase (Life TechnologiesTM) to remove the

genomic DNA residues, and 1 lg of total RNA was used

for first-strand synthesis of poly(A) cDNA using the Re-

vertAid First Strand cDNA Synthesis Kit (Fermentas) and

an oligodT18VN primer.

The primers F1 (50 AACCGGCCATTATTTCTGG 30),F2 (50 ATGCAAGAATACGACCATCC 30), F3 (50

Invert Neurosci

123

CCTTTGGATCGTCAAGTCTG 30) and R3 (50 GAT

CAAGCCAGAATGACACC 30), used for PCR amplifica-

tions, were designed based on the previously published

ChGluCla partial sequence (Carvalho et al. 2010). The pri-

mer F1 was located two nucleotides downstream of the ini-

tiation codon (Online Resource 1A), at a corresponding

position on the Lucilia cuprina GluCla. The most down-

stream primer, R4 (50 ACTCATCCTCCTCTTCACGG 30),was designed based on the stop codon region of the L. cup-

rina GluCla considering the sequence alignment of dipteran

GluCla genes (L. cuprina: AF081674.1, Musca domestica:

AB177546.1 and D. melanogaster: FBtr0335417). Two

primer sets were used for the characterisation of the

ChGluCla cDNA: F1/R3 and F3/R4. These primers and

other two combinations, F2/R3 and F1/R4, were used to PCR

amplify the ChGluCla cDNA of the bioassay larvae.

Initially, from the 10 NWS larvae of each bioassay

group that had the ChGluCla cDNA analysed, 16 had it

amplified in a single reaction using the pair of primers F1/

R4—10 from the ChIVR group and 6 from the ChSusc

group—and 4 using the pairs of primers F1/R3 and F3/R4.

Subsequently, we carried out the search for polymorphisms

in a subset of five more larvae from each group using the

pair of primers F2/R3.

The 25 lL PCR mix had 1.5 U of Taq DNA Polymerase

(Life TechnologiesTM) and 1 lL of cDNA. The reaction

concentrations for the pair of primers F1/R3 were 14 lM of

dNTPs, 2.8 mM of MgCl2 and 0.4 lM of primers. For the

primers combination F3/R4, the reaction concentrations

were 10 lM of dNTPs, 2.4 mM of MgCl2 and 0.16 lM of

primers, while the pairs of primers F1/R4 and F2/R3 had

56 lM of dNTPs, 2.8 mM of MgCl2 and 0.4 lM of

primers. The PCR conditions used for the primer sets F1/

R3 and F3/R4 were a denaturation step of 95 �C (3 min)

followed by 35 cycles of 95 �C (30 s), an annealing tem-

perature of 54 and 60 �C (30 s), respectively, and 72 �C

(1 min), finalising with an extension step of 72 �C

(15 min). For the primer sets F1/R4 and F2/R3, the specific

conditions were a denaturation step of 95 �C (5 min) fol-

lowed by 35 cycles of 95 �C (1 min), an annealing tem-

perature of 58 and 61 �C (1 min), respectively, and 72 �C

(1 min 30 s), with a final step of 72 �C (10 min).

The PCR products amplified by the pair of primers F1/

R3 were purified directly from the amplification reaction,

while those amplified by the pairs of primers F3/R4, F1/R4

and F2/R3 had the fragments of the expected size purified

after running in a 1 % agarose gel (w/v) and staining with

0.5 lg/mL ethidium bromide. The purification was carried

out using the Invisorb� Fragment CleanUp kit (Invitek).

After purification, the fragments were cloned into the

pGEMT-Easy vector (Promega) and sequenced bidirec-

tionally with universal vector primers. Three clones per

larva of amplicons amplified by each primer set (F1/R3,

F3/R4, F1/R4 and F2/R3) were sequenced with the Big

DyeTM Terminator Cycle Sequencing Ready Kit v. 3.0

(ABI PrismTM, Perkin-Elmer) on a 96-capillary ABI-

3730xl (Applied Biosystems) sequencer.

Sequence analysis

The chromatograms were analysed using the Sequence

Scanner v.1.0 software (Life TechnologiesTM). The

CAP3 software (Huang and Madan 1999) assembled the

50- and 30-end consensus cDNAs to obtain a baseline

sequence for the characterisation of the open reading

frame (ORF) of ChGluCla. The nucleotide sequences

were translated using the Emboss Transeq software (Rice

et al. 2000), and the BioEdit software (Hall 1999) per-

formed the ClustalW multiple sequence alignment

(Thompson et al. 1997) for the search of polymorphisms,

comparison to other invertebrate GLUCLa subunits, and

analysis of cDNA variants. The analysis of the indels

(insertions and deletions) of cDNA variants were carried

out with the sequences amplified by the pair of primers

F1/R4. The statistical significances were assessed using

the Fisher’s exact test.

The molecular weight and isoelectric point of the

putative protein encoded by ChGluCla were predicted by

Compute pI/Mw tool in Expasy Server (Wilkins et al.

1999). The N-terminal signal peptide and transmembrane

domains were predicted by SignalP v.4.0 (Petersen et al.

2011) and MemPype (Pierleoni et al. 2011), respectively.

The N-terminal loops and the L-glutamate-binding posi-

tions were predicted using homology to corresponding

regions of arthropod cys-loop LGIC receptors (Jones and

Sattelle 2006, 2007; Hibbs and Gouaux 2011). The putative

N-glycosylation and potential phosphorylation sites were

deduced using Prosite (Sigrist et al. 2010), and the database

searches were performed using NCBI BLAST (Altschul

et al. 1990) and FlyBase (Marygold et al. 2013).

Results

ChGluCla cDNA characterisation

We obtained the first ORF of ChGluCla. The consensus

sequences of the 50- and 30-end of the ChGluCla cDNA had

806 bp and 844 bp, respectively, and overlapped by

298 bp. The assembly of these cDNAs resulted in a frag-

ment of 1,352 bp, which comprised the almost full-length

ORF of ChGluCla. We added five nucleotides with the

ATG codon, from the NWS GluCla partial sequence by

Carvalho et al. (2010), and two nucleotides from the stop

codon, in order to obtain the full-length ORF. Thus, the

ORF of ChGluCla comprised 1,359 nucleotides

Invert Neurosci

123

(KF214912) and encoded a deduced protein of 452 amino

acids (Fig. 1).

This sequence showed significant nucleotide identity

with the GluCla of D. melanogaster (82 %, RI variant), M.

domestica (89 %) and L. cuprina (96 %). At the amino acid

level, the ChGluCla showed an even higher identity with

the putative orthologs of D. melanogaster (94 %, PI iso-

form), M. domestica (95 %) and L. cuprina (99 %). The

deduced molecular weight and isoelectric point were

52.29 kDa and 8.76, respectively. We already found four

putative N-glycosylation and 8 potential phosphorylation

sites.

The deduced ChGluCla amino acid sequence has the

characteristic features of members of the cys-loop LGIC

superfamily: four proposed transmembrane domains; a

N-terminal extracellular domain containing distinct loops

(Knipple and Soderlund 2010); the signature pair of

cysteine residues (cys-loop) (Cederholm et al. 2009); a

variable intracellular M3–M4 loop domain; the conserved

residues aspartic acid (D) and arginine (R) (Sine and Engel

2006). Moreover, this sequence has exclusive features of

the ligand-gated chloride channel family such as the PAR,

VT, L and T residues responsible for the chloride selec-

tivity (Jensen et al. 2005) and another pair of cysteine

residues (loop C) (Knipple and Soderlund 2010). We

already found the sites YGWT and TGEY inferred to be

positions of L-glutamate binding (Hibbs and Gouaux 2011)

(Fig. 1).

Bioassay with IVM

The larvae mortality and the insecticide concentration were

directly proportional, and the linear regression showed

significant correlation (p value \0.01, t test). The highest

Fig. 1 Multiple sequence alignment of GluCla protein sequences of

NWS (KF214912), L. cuprina (AF081674.1), M. domestica

(AB177546.1) and D. melanogaster (FBtr0335417). Predicted signal

peptide as well as the four transmembrane domains (M1-4) are

indicated. Highlighted in grey shading: YGWT and TGEY are sites

predicted to bind to L-glutamate; D and R (positions 79 and 243,

respectively) are conserved in the cys-loop LGICs; PAR, VT, L and T

residues responsible for ion charge selectivity. The two cysteines

forming the cys-loop as well as those forming the loop C (LpC) are

highlighted in black shading. N-glycosylation sites are underlined and

potential phosphorylation sites are indicated by black filled rectangles

above the sequences. Positions according to sequence of NWS

Invert Neurosci

123

and the lowest IVM concentrations showed 80 and 39 % of

mortality, respectively, while the control group (0 ppm)

had an insignificant mortality (Table 1).

Analysis of the GluCla cDNA of the bioassay larvae

Sixteen bioassay larvae—10 of the ChIVR group and 6

from the ChSusc group—had the ChGluCla cDNA

amplified by a single reaction that resulted in cDNAs

ranging from 983 to 1,364 bp (Table 2). However, 8 cDNA

variants were present at single copies, and 9 were consis-

tent with two or more copies (Online Resource 1A). The

cDNAs that comprised 1,349 bp and 1,352 bp showed,

each one, two variants: A and B. The cDNA variant A of

1,352 bp was the most represented with 27.1 % of the

sequences of the clones. Other four larvae of the ChSusc

group had the ChGluCla cDNAs amplified by two primer

combinations that resulted in 50- and 30-end fragments

between 735–806 and 691–844 bp (data not shown),

respectively. Therefore, 20 bioassay larvae had the almost

full-length ORF of the ChGluCla amplified.

We compared the predicted amino acid sequence of the

reference ChGluCla cDNA characterised in this study

(1,359 bp) with those of the bioassay larvae that were

amplified by one primer combination and found several

regions probably affected by alternative splicing, as well as

premature stop codons (Fig. 2). Here, we use the nomen-

clature suggested by Jones and Sattelle (2007) in order to

maintain consistency with other cys-loop LGIC subunits.

The ChGluCla possesses exon variants equivalent for

Drosophila GluCla exons 3a, 3b (Modules 1 and 2,

respectively) (Semenov and Pak 1999) and an additional

exon, here denoted as 3c, that generates specific isoforms

of Drosophila (FlyBase: FBtr0335418 and FBpp0307403)

(Marygold et al. 2013) and Tribolium castaneum (Jones

and Sattelle 2007) GluClas (Online Resource 2A, B, C, D).

The putative Cochliomyia GluCla exons 3a and 3b are

identical to the corresponding regions in Drosophila Glu-

Cla; however, they differ by two and one residue, respec-

tively, to those of Tribolium GluCla. The Drosophila exon

3c and the putative corresponding Cochliomyia exon differ

by 7 residues while that of Tribolium GluCla is shorter than

these two. This additional exon generates 4 of the 9

ChGluCla cDNA variants, including those of 1,352 bp,

which could explain its presence in the baseline ChGluClasequence. As with Drosophila GluCla isoform PE (Fly-

Base: FBpp0290592), the cDNA variant comprising

1,281 bp presents the lack of these putative homologous

exons (3a, 3b or 3c), including an inferred different posi-

tion for the starting methionine, which results in the

absence of the leader signal peptide.

The region of the intracellular loop between M3 and M4

in ChGluCla cDNAs clearly undergoes alternative splicing.

This region in Drosophila GluCla isoforms is generated

either by the use of differential splice sites or use of

alternative exons that cause their M3–M4 loop to differ by

Table 1 Results after 24 h of treatment and parameters of the IVM bioassay estimated by PoloPlus�

[IVM] (ppm) 0* 0.32 1 3.2 10 LC50 (CI95 %) LC90 (CI95 %) Slopea ± s.e t test

Dead 2 35 39 51 72 1.74

(0.56-4.34)

104.61

(22.70–13,749)

0.72 0.16 4.62

Alive 28 55 51 39 18

[IVM] ivermectin concentrations, ppm parts per million, LC lethal concentration, s.e. standard error

[IVM] of which the larvae were removed for cDNA analysis are highlighted in bold underlined

* Significantly different mortality from that of the lowest concentration (0.32 ppm), p value \0.001, Fisher’s exact testa Mortality versus [IVM] regression line, p value \0.01, t test

Table 2 cDNAs amplified by the primers combination F1/R4 and

number of sequences analysed (number of clones with the respective

size/total number of clones) in each bioassay group

cDNA(bp)

N� clones/total ChSusc

N� clones/total ChIVR

Prematurestop codona

Predicted lostdomainsb

983 – 1/30 No –

1,117 – 1/30 Yes M3–M4 loop; M4

1,120 3/18 5/30 Yes M3–M4 loop; M4

1,236 1/18 – No N-terminal

1,281 – 2/30 No N-terminal

1,295 – 1/30 No N-terminal

1,304 1/18 2/30 No –

1,307 1/18 1/30 No –

1,316 1/18 2/30 No –

1,319 1/18 – No –

1,340 1/18 – No –

1,349_A 1/18 1/30 No –

1,349_B 1/18 3/30 No –

1,352_A 6/18 7/30 – –

1,352_B – 3/30 Yesc M3–M4 loop; M4

1,361 – 1/30 No –

1,364 1/18 – No –

M4 fourth transmembrane domain, M3–M4 loop loop between M3 and M4a Presence of premature stop codon compared with the reference ChGluClacDNA characterisedb In the deduced amino acid sequence of the cDNAs compared with thereference ChGluCla amino acid sequence characterisedc Premature stop codon due to a SNP

Invert Neurosci

123

Invert Neurosci

123

a stretch of 24 residues and also results in premature stop

codons in isoforms PE and PF (FlyBase: FBpp0290593)

(Online Resource 2E). As with Drosophila GluCla, this

domain in Cochliomyia GluCla cDNAs shows similar

difference—23 residues—among the cDNA variants, as

well as premature stop codons. In each one of the

ChGluCla cDNAs of 1,120 and 1,352 bp (variant B), the

loss of M3–M4 domain eliminates one potential phos-

phorylation site, while the smaller variant additionally lost

one N-glycosylation site. These premature stops were due

to a single nucleotide polymorphism (SNP) or to deletions

resulting in frameshifts (Online Resource 1A). Within the

ChGluCla cDNAs, we still found sequences in the

boundaries of gaps that are similar to consensus splice site

motifs of pre-mRNAs (Online Resource 1B).

The ChGluCla cDNAs of the bioassay larvae also

showed 65 SNPs overall, with 34 being encoding

(Table 3). However, consistent SNPs were detected at two

positions: the K242R SNP (AAG to AGG) occurred in a

lower frequency in the ChIVR group, while the K398*

(AAA to TAA) occurred exclusively in this group. The

other SNPs were present at single copies, probably as

artefacts obtained during amplification. In the ChIVR

group, nine of the ten larvae showed the K242 allele in all

clones and one larva was heterozygote (K242 and R242).

In the ChSusc group, five of the ten larvae showed the

K242 allele in all clones and the other five were hetero-

zygous. Due to proximity to positions involved in ML

resistance and in IVM binding (Fig. 3), we additionally

amplified a region of 672 bp with this SNP in a subset of

five more larvae of each bioassay group. It is worth noting

that the process of pre-mRNA A-to-I editing replaces AAG

by AGG at the position K241 of D. melanogaster GluClaand changes the genomic lysine (K) to arginine

(R) (Semenov and Pak 1999), the same alteration we

observed at the equivalent position K242 in the ChGluClacDNAs.

The analysis showed an insignificant difference

(p [ 0.05) between the frequencies of larvae with the R242

allele on the three clones in both groups. However, a sig-

nificant difference occurred between the frequencies of

clones with this allele (Table 3).

Discussion

The results reveals that the ChGluCla cDNA consensus

sequence have high identity with the GluCla nucleotide

sequence of L. cuprina, D. melanogaster and M. domestica.

In addition, the deduced amino acid sequence of ChGluClahas features that suggest this gene belongs to three groups:

the cys-loop LGICs (Knipple and Soderlund 2010; Sine

and Engel 2006; Cederholm et al. 2009); the ligand-gated

chloride channel family (Jensen et al. 2005); and the glu-

tamate-gated chloride channel subfamily (Hibbs and

b Fig. 2 Multiple sequence alignment of putative C. hominivorax GluClaisoforms. Putative exon 3 splice variants (a, b, c) are highlighted in bold.

The N-terminal signal peptide as well as the four transmembrane domains

are indicated. The two cysteines forming the cys-loop as well as those

forming the loop C (LpC) are highlighted in black shading. Loops that

interfere in ligand binding (LpA-F) are indicated. The first two residues,

with the initiation methionine, upstream the F1 primer, are highlighted in

light grey shading. The putative initiation methionine of the Ch_1281

cDNA variant is highlighted in underlined bold. N-glycosylation sites are

underlined, and potential phosphorylation sites are indicated by black

filled rectangles and dark grey shading (putative exons 3a, 3b and 3c).

Premature stop codons are indicated by asterisks. Ch_X_Y = cDNAs. X

represents the cDNA size in nucleotides and Y represents the variants A or

B of cDNAs comprising 1,349 bp (49_A or 49_B) and 1,352 bp (52_A or

52_B). Sequence data of these NWS GluCla variants were deposited with

the GenBank Data Library: KJ725365-KJ725373

Table 3 Coding SNPs described in the ChGluCla cDNAs and

number of clones analysed (number of clones with the SNP/total

number of clones analysed in the respective bioassay group)

AA substitutions

ChSusc

SNP/total

ChSusc

AA substitutions

ChIVR

SNP/total

ChIVR

N149S 1/30 Q105R 1/30

N149D 1/30 I138V 1/30

S159T 1/30 N141D 1/30

M182T 1/30 V177I 1/30

V193I 1/30 F194S 1/30

L215F 1/30 E198G 1/30

L219S 1/30 L239P 1/30

N224S 1/30 K242R* 1/45

S225G 1/30 I252T 1/30

K236E 1/30 E350G 1/30

K242R* 10/45 K398� 3/30

R242� 1/45 Y443C 1/30

I250F 1/30

Q251R 1/30

W264R 1/30

S279P 1/30

V314I 1/30

A322 V 1/30

E325 V 1/30

S333P 1/30

I427 V 1/30

T428A 1/30

A434 V 1/30

Positions according to ChGluCla amino acid sequence characterised* p value \0.05, Fisher’s exact test� Premature stop codon

Invert Neurosci

123

Gouaux 2011). For these reasons, we suggest that this gene

is an ortholog of other dipteran GluCla genes.

The features of ChGluCla cDNAs of the bioassay lar-

vae—different sizes, inferred exons and differential splice

sites, along with sequences similar to consensus splice site

motifs—agree with the occurrence of splice variants found

in GluCla orthologs of insects such as T. castaneum (Jones

and Sattelle 2007), A. mellifera (Jones and Sattelle 2006)

and D. melanogaster (Semenov and Pak 1999). As with the

fruit fly, honey bee and Tribolium GluClas, the predicted

ChGluCla alternative exons and splice sites diversify this

receptor. Evidence of this is the use of different variants of

exon 3 with amino acid substitutions and insertion that

disrupt the position of potential phosphorylation sites in the

vicinity of loop D, which may interfere in agonist binding.

Moreover, we propose that the lack of this exon, as is the

case for Drosophila GluCla isoform PE (Online Resource

2C, D), in cDNA variant of 1,281 bp could be the cause for

the differential position of start codon and for the absence

of one phosphorylation and one N-glycosylation site in the

N-terminal extracellular domain. Species-specific variants

also occur, since the putative Cochliomyia GluCla exon 3c

and those of the orthologous Drosophila and Tribolium

differ substantially in the N-terminal portion.

Another hot spot that increases variability of GluClareceptors is the M3–M4 loop. The ChGluCla shows

remarkable diversity in this domain (Fig. 2), in addition to

inferred premature end of translation that causes the loss of

potential phosphorylation and N-glycosylation sites.

Interestingly, in Drosophila GluCla isoforms, the intra-

cellular loop is highly similar to the equivalent region in

the Cochliomyia ortholog with identical stretches between

them (Online Resource 2F). We suggest that, as with its

Drosophila counterpart, the alterations in the intracellular

loop of Cochliomyia GluCla cDNAs are generated by the

use of differential splice sites and alternative exons.

These cDNAs still showed premature stop codons,

although insignificantly different between the groups.

When compared with the reference ChGluCla amino acid

sequence, these premature stop codons results in loss of

subunit domains (Fig. 2), with probable impact in the

protein function and in the insecticide binding. Several

authors have associated ML resistance in invertebrates to

viable truncated transcripts of nAchRs (Baxter et al. 2010;

Watson et al. 2010; Hsu et al. 2012) and GluCla homo-

logues (Dent et al. 2000). Although these researchers

described truncated transcripts that were artificially

induced or spinosad resistant, their work suggests that

uncommon modified forms of cys-loop LGICs bind dif-

ferently to MLs compared with the standard protein

structure. Moreover, the loss of putative N-glycosylation

and phosphorylation sites in the M3–M4 loop affects

receptor activity (Cascio 2006). This suggests that the

predicted variants of ChGluCla cDNAs with these features

could bind differently to IVM.

The clones with the R242 allele occurred in both bioassay

groups, at a significant lower frequency in the surviving

larvae. This SNP is close to the M1 domain and to L256F

mutation in GLUCLa of nematodes, involved in IVM

binding (Hibbs and Gouaux 2011) and in ML resistance

(Yamaguchi et al. 2012; Njue et al. 2004), respectively.

Furthermore, the K242R SNP is positioned in a domain (pre-

M1) associated with conformational changes that result in

channel gating and opening by agonists (Bartos et al. 2009;

Lee et al. 2009). Therefore, alterations on the M1 domain and

adjacent regions such as this SNP are probably involved in

conformational changes, which could avoid the receptor to

be open by ligands such as IVM. Since this polymorphism is

similar to pre-mRNA A-to-I editing in the equivalent posi-

tion of the Drosophila GluCla, it is probable that the same

process occurs in NWS GluCla. In order to verify this

alteration and the alternative splicing, the NWS GluClagenomic DNA sequencing needs to be carried out.

Although these alterations occurred in a specific sub-

population and were in low frequency, the two SNPs and

cDNA variants in the ChIVR group could clarify mecha-

nisms probably associated with IVM resistance in NWS.

The absence of any polymorphism clearly linked to IVM

resistance could be explained by other targets or mecha-

nisms that need to be addressed: alterations in other ion-

channel genes or the metabolic detoxification, which could

be the major type of resistance in this strain. Nevertheless,

the ChGluCla sequencing of this isolate of NWS, along

with the cDNA variants, provides a basis for character-

isation of new molecular markers that could help to track

resistant natural populations.

Fig. 3 Portion of multiple sequence alignment of GluCla protein

sequences of C. hominivorax (KF214912), C.elegans (U14524.1) and

C. oncophora (AY372756). M1 = first transmembrane domain. The

two cysteines forming the loop C (LpC) are highlighted in black

shading. The position of the L256F mutation in C. oncophora is

indicated by empty triangle while the position of the K242R SNP in

C. hominivorax is indicated by black star. Residues associated to

IVM-binding in C. elegans and those corresponding in C. hominiv-

orax and C. oncophora are highlighted in bold. The pre-M1 region is

indicated by dashed line

Invert Neurosci

123

Acknowledgments This study was funded by Grants from Conse-

lho Nacional de Desenvolvimento Cientıfico e Tecnologico and

Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (#

2011/01030-5). We would like to thank Rosangela Rodrigues for the

technical assistance and Salete Couto Campos for the help with the

bioassay and NWS strain maintenance. We are grateful to Dr. Thiago

Mastrangelo for advices in preparing the manuscript.

Conflict of interest None.

References

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic

local alignment search tool. J Mol Biol 215(3):403–410

Bartos M, Corradi J, Bouzat C (2009) Structural basis of activation of

cys-loop receptors: the extracellular-transmembrane interface as

a coupling region. Mol Neurobiol 40(3):236–252

Baxter SW, Chen M, Dawson A, Zhao JZ, Vogel H, Shelton AM et al

(2010) Mis-spliced transcripts of nicotinic acetylcholine receptor

alpha6 are associated with field evolved spinosad resistance in

Plutella xylostella (L.). PLoS Genet. doi:10.1371/journal.pgen.

1000802

Carvalho RA, Azeredo-Espin AM, Torres TT (2010) Deep sequenc-

ing of New World screw-worm transcripts to discover genes

involved in insecticide resistance. BMC Genom. doi:10.1186/

1471-2164-11-695

Cascio M (2006) Modulating inhibitory ligand-gated ion channels.

AAPS J 8(2):E353–E361

Cederholm JM, Schofield PR, Lewis TM (2009) Gating mechanisms

in Cys-loop receptors. Eur Biophys J 39(1):37–49

Cleland TA (1996) Inhibitory glutamate receptor channels. Mol

Neurobiol 13(2):97–136

Cully DF, Vassilatis DK, Liu KK, Paress PS, Van der Ploeg LH,

Schaeffer JM et al (1994) Cloning of an avermectin-sensitive

glutamate-gated chloride channel from Caenorhabditis elegans.

Nature 371(6499):707–711

Dent JA, Smith MM, Vassilatis DK, Avery L (2000) The genetics of

ivermectin resistance in Caenorhabditis elegans. Proc Natl Acad

Sci USA 97(6):2674–2679

Dermauw W, Ilias A, Riga M, Tsagkarakou A, Grbic M, Tirry L et al

(2012) The cys-loop ligand-gated ion channel gene family of

Tetranychus urticae: implications for acaricide toxicology and a

novel mutation associated with abamectin resistance. Insect

Biochem Mol 42(7):455–465

Hall TA (1999) BioEdit: a user-friendly biological sequence align-

ment editor and analysis program for Windows 95/98/NT.

Nucleic Acids Symp Ser 41:95–98

Hall M, Wall R (1995) Myiasis of humans and domestic animals. Adv

Parasit 35:257–334

Hibbs RE, Gouaux E (2011) Principles of activation and permeation

in an anion-selective Cys-loop receptor. Nature 474(7349):54–60

Hsu J-C, Feng H-T, Wu W-J, Geib SM, Mao C-H, Vontas J (2012)

Truncated transcripts of nicotinic acetylcholine subunit gene

Bda6 are associated with spinosad resistance in Bactrocera

dorsalis. Insect Biochem Mol 42(10):806–815

Huang X, Madan A (1999) CAP3: A DNA sequence assembly

program. Genome Res 9(9):868–877

Jensen ML, Schousboe A, Ahring PK (2005) Charge selectivity of the

Cys-loop family of ligand-gated ion channels. J Neurochem

92:217–225

Jones AK, Sattelle DB (2006) The cys-loop ligand-gated ion channel

superfamily of the honeybee, Apis mellifera. Invertebr Neurosci

6(3):123–132

Jones AK, Sattelle DB (2007) The cys-loop ligand-gated ion channel

gene superfamily of the red flour beetle, Tribolium castaneum.

BMC Genomics. doi:10.1186/1471-2164-8-327

Jones AK, Bera AN, Lees K, Sattelle DB (2010) The cys-loop ligand-

gated ion channel gene superfamily of the parasitoid wasp,

Nasonia vitripennis. Heredity 104(3):247–259

Kane NS, Hirschberg B, Qian S, Hunt D, Thomas B, Brochu R

et al (2000) Drug-resistant Drosophila indicate glutamate-

gated chloride channels are targets for the antiparasitics

nodulisporic acid and ivermectin. Proc Natl Acad Sci USA

97(25):13949–13954

Knipple DC, Soderlund DM (2010) The ligand-gated chloride

channel gene family of Drosophila melanogaster. Pestic Bio-

chem Phys 97(2):140–148

Kwon DH, Yoon KS, Clark JM, Lee SH (2010) A point mutation in a

glutamate-gated chloride channel confers abamectin resistance

in the two-spotted spider mite, Tetranychus urticae Koch. Insect

Mol Biol 19(4):583–591

Lee WY, Free CR, Sine SM (2009) Binding to gating transduction in

nicotinic receptors: Cys-loop energetically couples to pre-M1

and M2-M3 regions. J Neurosci 29(10):3189–3199

Marygold SJ, Leyland PC, Seal RL, Goodman JL, Thurmond J,

Strelets VB et al (2013) FlyBase: improvements to the bibliog-

raphy. Nucleic Acids Res 41:D751–D757

McCavera S, Rogers AT, Yates DM, Woods DJ, Wolstenholme AJ

(2009) An ivermectin-sensitive glutamate-gated chloride channel

from the parasitic nematode Haemonchus contortus. MolPharmacol 75(6):1347–1355

Njue AI, Hayashi J, Kinne L, Feng XP, Prichard RK (2004) Mutations

in the extracellular domains of glutamate-gated chloride channel

alpha3 and beta subunits from ivermectin-resistant Cooperia

oncophora affect agonist sensitivity. J Neurochem 89(5):

1137–1147

Omura S, Crump A (2004) The life and times of ivermectin—a

success story. Nat Rev Microbiol 2(12):984–989

Ortells MO, Lunt GG (1995) Evolutionary history of the ligand-gated

ion-channel superfamily of receptors. Trends Neurosci

18(3):121–127

Petersen TN, Brunak S, von Heijne G, Nielsen H (2011) SignalP 4.0:

discriminating signal peptides from transmembrane regions. Nat

Methods 8(10):785–786

Pierleoni A, Indio V, Savojardo C, Fariselli P, Martelli PL, Casadio R

(2011) MemPype: a pipeline for the annotation of eukaryotic

membrane proteins. Nucleic Acids Res 39:W375–W380

Rice P, Longden I, Bleasby A (2000) EMBOSS: the European

molecular biology open software suite. Trends Genet

16(6):276–277

Robertson JL (2007) Bioassays with arthropods. CRC Press, Boca

Raton

Robertson JL, Smith KC, Savin NE, Lavigne RJ (1984) Effects of

dose selection and sample size on the precision of lethal dose

estimates in dose-mortality regression. J Econ Entomol

77:833–837

Semenov EP, Pak WL (1999) Diversification of Drosophila chloride

channel gene by multiple posttranscriptional mRNA modifica-

tions. J Neurochem 72(1):66–72

Sigrist CJ, Cerutti L, de Castro E, Langendijk-Genevaux PS, Bulliard

V, Bairoch A et al (2010) PROSITE, a protein domain database

for functional characterization and annotation. Nucleic Acids

Res 38:D161–D166

Sine SM, Engel AG (2006) Recent advances in Cys-loop receptor

structure and function. Nature 440(7083):448–455

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG

(1997) The CLUSTAL_X windows interface: flexible strategies

for multiple sequence alignment aided by quality analysis tools.

Nucleic Acids Res 25(24):4876–4882

Invert Neurosci

123

Vargas ME, Espin AM (1995) Genetic variability in mitochondrial

DNA of the screwworm, Cochliomyia hominivorax (Diptera:

Calliphoridae) from Brazil. Biochem Genet 33(7–8):237–256

Vargas-Teran M, Hofmann HC, Tweddle NE (2005) Impact of

screwworm eradication programmes using the sterile insect

technique. In: Dyck VA, Hendrichs J, Robinson AS (eds) Sterile

insect technique. Springer, Netherlands, pp 629–650

Watson GB, Chouinard SW, Cook KR, Geng C, Gifford JM,

Gustafson GD et al (2010) A spinosyn-sensitive Drosophila

melanogaster nicotinic acetylcholine receptor identified through

chemically induced target site resistance, resistance gene iden-

tification, and heterologous expression. Insect Biochem Mol

40(5):376–384

Welch J, Hall MJR (2008) New World screwworm (Cochliomyia

hominivorax) and Old World screwworm (Chrysomya bezziana).

In: OIE Biological Standards Commission (ed) Manual of

diagnostic tests and vaccines for terrestrial animals. World

Organisation for Animal Health, Paris, pp 265–275

Wilkins MR, Gasteiger E, Bairoch A, Sanchez JC, Williams KL,

Appel RD et al (1999) Protein identification and analysis tools in

the ExPASy server. In: Link AJ (ed) 2-D proteome analysis

protocols. Humana Press, Totowa, pp 531–552

Wolstenholme AJ, Rogers AT (2005) Glutamate-gated chloride

channels and the mode of action of the avermectin/milbemycin

anthelmintics. Parasitology 131:S85–S95

Yamaguchi M, Sawa Y, Matsuda K, Ozoe F, Ozoe Y (2012) Amino

acid residues of both the extracellular and transmembrane

domains influence binding of the antiparasitic agent milbemycin

to Haemonchus contortus AVR-14B glutamate-gated chloride

channels. Biochem Biophys Res Commun 419(3):562–566

Invert Neurosci

123