Characterisation of P-glycoprotein-9.1 in Haemonchus contortus

The existence nematodes of veterinary importance such as Haemonchus contortus resistant to anthelmintic drugs, including the macrocyclic lactones, has become a major concern in animal health. Macrocyclic lactone resistance in H. contortus seems to be multigenic including the active efflux of these drugs by P-glycoproteins, members of the ABC transporter family, present in this parasite. The goals of the present work were to determine the activity of H. contortus P-glycoprotein 9.1 (Hco-PGP-9.1) and its interaction with the avermectins, ivermectin, abamectin, and also the milbemycin, moxidectin. Additionally, the localisation of Hco-PGP-9.1 was sought in adult worms. Hco-Pgp-9.1 was cloned and expressed in mammalian cells and its expression profile was determined at the transcriptional and protein level by qRT-PCR and Western-blot, respectively. The nematode transport activity was assessed using the tracer dye Rhodamine 123. A ligand competition assay between different macrocyclic lactones and Rhodamine 123 was used to establish whether or not there was interaction between Hco-PGP-9.1 and the avermectins (abamectin and ivermectin) or moxidectin. In addition, immunostaining was carried out to localise Hco-PGP-9.1 expression in the transgenic cells and in adult female parasites. Hco-PGP-9.1 was expressed in the cell membrane of the transfected host cells and was able to extrude Rhodamine 123. Ivermectin and abamectin, but not moxidectin, had a pronounced inhibitory effect on the ability of Hco-PGP-9.1 to transport Rhodamine 123. Antibodies raised against Hco-PGP-9.1 epitopes localised to the uterus of adult female H. contortus. These results suggest a strong interaction of the avermectins with Hco-PGP-9.1. However, possibly due to its physico-chemical properties, moxidectin had markedly less effect on Hco-PGP-9.1. Because of the greater interaction of the avermectins than moxidectin with this transporter, it is more likely to contribute to avermectin resistance than to moxidectin resistance in H. contortus. Possible over expression of Hco-PGP-9.1 in the female reproductive system in resistant worms could reduce paralysis of the uterus by macrocyclic lactones, allowing continued egg release in drug challenged resistant worms.


Background
Anthelmintic resistance (AR) in gastro-intestinal (GI) nematodes such as Haemonchus contortus, has become a serious problem for livestock industries. In small ruminants, H. contortus can significantly decrease flock production [1]. The primary means of parasite control in animal health is through anthelmintic drugs, of which, the macrocyclic lactones (MLs) are widely used, with activity against both ecto-and endoparasites [2]. Ivermectin (IVM), the first ML anthelmintic commercially developed, is a chemically modified product from Strepmomyces avermitilis fermentation [3]. The MLs have two subclasses: one corresponding to the avermectins, such as IVM, abamectin (ABA), doramectin (DOR), eprinomectin (EPR), selamectin (SEL) [4] which are used to control nematodes and some ectoparasites in animals. IVM is also used in humans. Emamectin (EMC) is used in fish farming to control sea lice [5]. The other group corresponds to the milbemycins including moxidectin (MOX) and milbimycin oxime (MO), used against parasitic nematodes in farm and companion animals [6][7][8]. The MLs act by binding irreversibly to glutamate-gated chloride channels (GluCls), which are present in invertebrates such as nematodes, producing a flaccid muscular paralysis [9]. Structurally the MLs have a macrocyclic lactone ring based in a benzofuran core [10] but have some differences, notably the attached disaccharide moiety at the C-13 of the ML ring present in the avermectins but not on the milbemycins [11]. In addition, the milbemycin, MOX has a methoxime at C-23 and an olefinic side chain at C-25, which are not present in the avermectins [2]. The use of the ML in veterinary medicine has been the anchor strategy to control parasitic pathogens but also in the case of IVM, has been applied successfully to control human parasitoses such as river blindness disease caused by Onchocerca volvulus [12]. Despite the remarkable ML efficacy, parasitic nematodes have developed resistance to these drugs, limiting their parasiticidal effect [13,14]. In large animal nematodes such as H. contortus, ML resistance has been reported in field isolates and experimentally selected strains [15][16][17][18]. ML resistance in nematodes has a genetic basis involving several genes turning into a complex process not completely understood [2]. One of the mechanisms hypothesised to be participating in ML resistance in nematodes is possibly the active transport of these drugs by the ATP-Binding-Cassette (ABC) transporters such as P-glycoprotein (PGP) [19]. ABC transporters are cellmembrane proteins that transport across the membrane, a wide range of structurally unrelated substrates, including anthelmintics such as IVM [20]. In nematodes, including H. contortus and the free-living model Caenorhabditis elegans, an up-regulation of ABC transporters that has a direct correlation with a ML resistance phenotype has been reported [21,22]. Moreover, in H. contortus some P-gps have been reported to be over-expressed in IVM field resistance strains, including Hco-Pgp-2 or A [21][22][23] and also Hco-Pgp-9 or E [23]. Recently, the H. contortus genome has been fully sequenced, describing at least 10 Hco-Pgp genes among the ABC transporters present in this parasite [24]. In H. contortus, three copies of Hco-Pgp-9 gene (Hco-Pgp-9.1, Hco-Pgp-9.2 and Hco-Pgp-9.3) have now been partially identified [24], with expression of the Hco-Pgp-9.1 transcript, previously termed Hco-Pgp-E [23], being reported to be up-regulated in ML selected and field resistant strains of H. contortus [23]. Regarding orthologs in other trichostrongylids from ruminants, an up-regulation of Tcipgp-9 in all life-cycle stages of the sheep nematode Teladorstagia circumcinta in IVM resistant isolates has been described [25]. In another helminth from cattle, Cooperia onchophora an up-regulation of the ortholog Con-pgp-9 in an IVM selected strain was also described [26].
Due to this data showing an over-expression of Hco-Pgp-9.1, and its orthologs, in H. contortus and other nematodes with veterinary importance resistant to ML, as well as the strong interaction described between Hco-PGP-2 and the avermectins, suggesting a nematode transporter involvement in ML resistance in H. contortus [27], the aim of the present work was to characterise the Hco-PGP-9.1 transport activity and its interaction with the MLs, in order to ascertain its possible involvement in the resistance to these drugs. In addition, Hco-PGP-9.1 expression and localisation in the adult stage of this parasite was investigated.

Ethics statement
Animals and standard operating procedures (SOPs) used in this research study were approved (Protocol #3845) and adhered to the guidelines from the Animal Care Committee of McGill University.

Worms
Adult H. contortus from the laboratory PF23 isolate, susceptible to the MLs [15], were used. Worms were collected from the abomasum of sheep and incubated in PBS (phosphate-buffer saline) for 2 h at 37°C.

RNA extraction
Worms were homogenised and lysed with the Trizol® reagent according to the manufacturer's instruction. The RNA pellet obtained was dried and eluted in 50 μl of RNase free water. RNA quantification was measured in a nanophotometer IMPLEN® (Westlake Village, CA, USA) at 260 nm wavelength. All extracted RNA was kept at -80°C.

Reverse transcription polymerase chain reaction (RT-PCR)
Total RNA (1 μg) was reverse transcribed using the Omniscript® reverse transcription kit following the manufacturer's protocol. Synthesised cDNA was stored at-20°C for further use.
Amplification and cloning of the full length H. contortus Hco-Pgp-9.1 cDNA sequence H. contortus P-glycoprotein-9.1 DNA sequence was from ( [28], Additional file 1). Specific primers to amplify the full-length sequence were designed using Geneious bioinformatics software, version 5.5.6 (Biomatters Ltd, Auckland, New Zealand). The cloning strategy was to ligate the Hco-Pgp-9.1 cDNA sequence and the linearised pcDNA3.1 (+) mammalian expression vector using restriction sites. To carry this out, the primers used were: forward primer 5′-aagttagggatcccCaccATGGGCTTTTTGAAGA ACGG-3′ including a specific sequence for pcDNA3.1 (+) in small case, then a BamH I restriction site (underlined in small case), after a Kozak sequence, around the start codon, both in bold [29].
Reverse primer 5′-agacctagcggccgcTCATCCCTTGGC CAATGTTTGCTT-3′, included a Not I restriction site (in bold small case), then a stop codon (underlined as above). A PCR with these primers was run. After confirmation that the expected amplicon was present, the PCR product was ligated into pGEM-T® Easy subcloning vector. Furthermore, the cloned sequence was digested individually with each restriction enzyme referred above, and ligated into pcDNA 3.1(+) mammalian expression vector using the same restriction sites.
Transformation was carried out in E. coli TOP10F' competent cells. After, positive colonies were screened by colony PCR and restriction digestion analysis. DNA extracted from positive transformants was digested with BamHI and Not I, confirming the construct ligation and these DNA plasmids were sent for sequencing at Genome Quebec Innovation Centre (McGill University, QC, Canada) to confirm the specific Hco-Pgp-9.1 full-length cDNA sequence.

Stable transfection in mammalian cells
For stable transfection the mammalian porcine cell line LLC-PK1 was used [27]. Original LLC-PK1 parental cells were seeded on 24 well plates (2 × 10 5 cells/well) and grown in 199 medium supplemented with 10 % fetal calf serum and penicillin/streptomycin (100 units/ml; 100 μg/ml, respectively). Cells close to 100 % confluency were transfected using 18 μg of plasmid DNA using Lipofectamine 2000®, following the manufacturer's instructions. In parallel, as a positive control for transfection, the Chloramphenicol Acetyl Transferase (CAT) gene was transfected into these cells. Twenty-four hours after transfection, medium was removed and replaced with supplemented 199 medium. Forty-eight hours after transfection, the cells were washed, collected and counted in order to seed them to 25 % confluency in 24 well plates under selection pressure of G418 or neomycin (Geneticin®) at 400 mg/l.
After several weeks of selection, stable transfected colonies were collected individually and transferred to new culture flasks. Each individual colony was assessed for growth rate, morphological shape and survival under selection with G418.

Expression profile characterisation
Transcript profile: qRT-PCR RNA was extracted from stable transfectants using the RNeasy extraction kit, following the manufacturer's instructions as above. Subsequently, a reverse transcription-PCR was done with the extracted RNA. Then, the synthesised cDNA was used as template for a qRT-PCR with specific primers (see Table 1) for Hco-Pgp-9.1 and the pig mdr1 (GenBank accession number XM_003130205.3). Pig GAPDH (Glyceraldehyde-3 phosphate dehydrogenase) was used as a housekeeping gene for normalisation (GenBank accession number NM_001206359.1). After serial dilution of the cDNA template, a qRT-PCR was carried out using SYBER Green in a Rotor-Gene thermal cycler (Qiagen, Hilder, Germany) with the following conditions: initial hold at 95°C for 15 min, then 40 cycles including: 94°C for 15 s; 57°C for 30 s for annealing and 72°C for 1 min final extension. Subsequently, a melting curve analysis was done for all the amplified PCR products. Primers for target and housekeeping genes were run in four replicates. After every qRT-PCR run, data were analysed and the efficiency of the reaction estimated according to [30]. Relative expression of Hco-Pgp-9.1 over the endogenous pig mdr1 was obtained based on the 2 ΔΔ Ct method.

Protein profile: membrane enrichment fraction
Hco-Pgp-9.1/LLC-PK1 transfected cells were grown in 75 cm 2 flasks close to 100 % confluency. Cells were collected with a scraper, counted and re-suspended in buffer containing protease inhibitors (1X PBS pH 7.4; 1 mM PMSF; 1 mM EGTA; 10 μg/ml Leupeptin; 1 μg/ml Pepstatin A). Cells were then washed twice with a solution containing 1X PBS + 1 mM DTT + 1 mM PMSF to a volume of 50 ml. Then the cell suspension was divided into two tubes (25 ml volume each) and centrifuged at 1000× g for 15 min at 4°C. Afterwards, each pellet was re-suspended in an appropriate volume of buffer containing 1X PBS + 1 mM DTT + 1 mM PMSF. Cells were then sonicated 3 times for 10 s using 40 % amplitude. Subsequently, cell pellets were centrifuged at 1400× g for 15 min at 4°C. Supernatants were carefully collected and combined in a single tube. This sample was ultra-centrifuged at 135 000× g for 60 min at 4°C. The supernatant was discarded and the pellet re-suspended in 2.5 ml of a solution containing 1X PBS + 1 mM DTT + 1 mM MgCl 2 . Finally, the protein concentration from isolated membranes was quantified by the BCA protein assay method and kept at-80°C. In parallel, an aliquot of whole crude membrane from H. contortus worms was isolated. Fifty H. contortus adult worms (females and males) were homogenised in buffer containing PBS 1× pH 7.4; 1 mM PMSF; 1 mM EGTA; 10 μg/ml Leupeptin and 1 μg/ml Pepstatin A. The solution was centrifuged at 45 000× g for 15 min at 4°C, the supernatant was recovered and centrifuged again at 45 000× g for 15 min at 4°C. The supernatant was carefully recovered and sonicated 3 times for 10 s with 40 % amplitude. Finally, the sample was ultracentrifuged at 135 000× g for 60 min at 4°C, the pellet was recovered and re-suspended in PBS + 1 mM DTT + 1 mM MgCl 2 + 0.5 % Triton X-100 for further protein quantitation by the BCA protein assay method. Samples were kept at-80°C until further separation on SDS-PAGE.
Hco-PGP-9.1 antibody preparation and Western blot Peptide antigens were selected from the H. contortus PGP-9.1 amino acid sequence, analysed, synthesised commercially and used to immunise rabbits by 21 st Century Biotech. Inc. (Marlborough, MA, USA) (see Additional file 1). The specificity of purified Hco-PGP-9.1 peptide epitopes was determined by ELISA and dot blot. A mixture 1:1 of both antibodies for each antigen peptides for Hco-PGP-9.1 was used for further Western-blot and immunolocalisation. Membrane protein fractions previously obtained from Hco-Pgp-9.1/LLC-PK1, CAT/LLCPK1 control cells and H. contortus worms whole crude membrane extracts, from a mixed population of female and male adult worms, were separated in a 4.5-10 % polyacrylamide gradient gel. Proteins were transferred, by the semi-dry system, to a PVDF membrane at 25 volts for 25 min. The membrane was blocked overnight at 4°C in 5 % skim milk blocking solution with 0.05 % Tween-20®. Then the membrane was washed thoroughly in TBS-0.05 % Tween-20®. The membrane was then incubated, overnight at 4°C, with anti-Hco-PGP-9.1 antibody, at 1/500 dilution.
After several washes, the membrane was incubated with anti-rabbit secondary antibody conjugated to horseradish peroxidase in a 1/5000 dilution at room temperature for 1 h. After washing, the membrane was developed using a chemiluminescence kit and X-Omat films.
Immunofluorescence assay (IFA) on Hco-Pgp-9.1/LLC-PK1 transfectants Hco-Pgp-9.1/LLC-PK1 transfectants were grown on coverslips to 70 % confluency. Cells were then fixed with 4 % paraformaldehyde (PAFOH). The cells were washed and blocked with 3 % BSA for 1 h at room temperature, and then washed and incubated with the anti-Hco-PGP-9.1 antibody, or anti-Hco-PGP-9.1 antibody pre-absorbed by both synthetic peptides (pre-absorbed negative control), at 1/50 dilution overnight at 4°C in the dark. The cells were then incubated with Alexa®Fluor 488® secondary antibody in 1/3000 dilution for 45 min at room temperature in the dark. Cells were then washed and incubated with nucleic DAPI medium for 4 min at room temperature in the dark. Following this, the cells were washed and mounted on a slide with mounting media and sealed with nail polish. After overnight incubation, stained cells were examined under an ECLIPSE Ti inverted epifluorescence microscope (Nikon, Melville, NY, USA) at λ488nm excitation and λ519nm emission. Images obtained with different dyes were merged using ImageJ software (http://imagej.nih. gov/ij/).

with Rhodamine 123 and macrocyclic lactones
In order to measure the nematode transporter activity, Hco-Pgp-9.1/LLC-PK1 transfected cells were tested with Rhodamine 123 (Rho123) [11] as substrate for efflux translocation. Cells were plated in 24 well plates in G418free medium until confluent. Then, cells were incubated in HBSS media, 1 % BSA containing 10 μM Rho123 with or without increasing concentrations of the endectocides: IVM, ABA or MOX (0.00625-20 μM). Additionally, the cyclosporin A analogue, valspodar (VSP or PSC833) (kindly provided by Novartis Animal Health, Switzerland) was used, at 5 μM, as a maximum reference inhibitor of Hco-PGP-9.1 transport function since this compound is a slow transport substrate for MDR1.
All the drugs were dissolved in DMSO and diluted in HBSS transport media with a final DMSO concentration of < 0.2 %. Hco-Pgp-9.1/LLC-PK1 transfected cells were incubated with all the reagents, including Rho123 for 2 h. After incubation, the medium was discarded and cells carefully washed three times with PBS. Finally cells were lysed in PBS/0.5 % SDS. The lysates were stored at-20°C until further analysis.

Fluorescence measurement
The intracellular accumulation of Rho123 was measured using a FLUOstar Galaxy fluorimeter (BMG LABTECH, Ortenberg, Germany). The reading parameters for Rho123 were λmax = 507 nm excitation and λmax = 529 nm emission. The inhibition of the Hco-PGP-9.1 transport function by the MLs was compared against the maximum effect produced by valspodar [11] and expressed as percent of total VSP inhibition. Each experiment was repeated using three biological replicates and each drug concentration was run in triplicate. The data generated were fitted using GraphPad Prism software (Version 6).

Localisation of H. contortus P-glycoprotein-9.1 in adult worms
The Hco-PGP-9.1 antibody mixture described above was used for immunolocalisation of Hco-PGP-9.1 expression in the H. contortus adult stage. Worms were collected and fixed in 3.7 % (w/v) paraformaldehyde in PBS at 4°C for 6-12 h. Worms were then washed three times in PBS for 15 min and stored at 4°C in 0.1 % (w/v) paraformaldehyde in PBS. Subsequently, worms were incubated overnight at 37°C in 5 % (v/v) 2-mercaptoethanol, 1 % Triton X-100, 125 mM Tris-HCl pH 6.9, and washed briefly in PBS. Fixed worms were digested for 6-20 h until complete removal of the cuticle layer, in 120 U/ml collagenase at 37°C in 1 mM CaCl 2 , 100 mM Tris-HCl pH 7 [31]. Worms were then washed in PBS and incubated for 72 h at 4°C with a 1/200-300 dilution of the primary antibody in PBS containing 0.1 % (w/v) BSA, 0.5 % (v/v) Triton X-100 and 0.05 % (w/v) sodium azide. Non-specific binding was removed after several washes in PBS/0.1 % (v/v) Triton X-100 at 4°C [32]. Worms were incubated with the secondary antibody Alexa® Fluor 633 F(ab') 2 farred fluorescein goat anti-rabbit in a 1/3000 dilution at 4°C for 15-18 h. Worms were washed in PBS containing 0.1 % (v/v) Triton X-100. Finally, worms were mounted on slides using mounting medium and examined under the confocal microscope (laser Bio-Rad Radiance®2100 MP) at λ632 nm emission and λ647 nm excitation.

Expression profile of Hco-Pgp-9.1 in transfected cells
The relative expression of Hco-Pgp-9.1 in mammalian LLC-PK1 cells is shown in Fig. 1. The Hco-Pgp-9.1 transcript was expressed at 30 fold the level of endogenous pig mdr1 present naturally in this porcine origin cell line at the second passage of these transfectants. The expression of the Hco-Pgp-9.1 transcript appeared to be decreasing by the third passage of the same culture (although the differences were not statistically significant), but the difference compared with the endogenous pig mdr1 was still large. The expression profile was also corroborated at the protein profile by Western blot. A protein band of 130 kDa molecular weight was observed from membranes extracted from Hco-Pgp-9.1/LLC-PK1 transfectant clones (Fig. 2, lanes A and C). In the whole crude membranes extracted from H. contortus adult worms (lane B), a similar 130 kDa band and a higher molecular size band close to 180 kDa were seen. The protein band of 130 kDa identified in transfected cells may correspond to the unglycosylated form of Hco-PGP-9.1, whereas the whole crude membranes from the adult stage of the parasite, may contain both unglycolysalted (130 kDa) and glycosylated (~180 kDa) forms of Hco-PGP-9.1.
The expression of this nematode transporter in the host cells was also supported by the subcellular localisation of Hco-PGP-9.1 in the plasma membrane of the transfected cells (Fig. 3a). Hco-pgp-9.1 transfected cells exposed to anti-Hco-PGP-9.1 antibody that had been pre-absorbed with the epitope peptides (Fig. 3b) did not show staining in the cell membrane. The CAT/LLC-PK1 control cells (Fig. 3c) exposed to the anti-Hco-PGP-9.1 antibody (not pre-absorbed) showed generalised fluorescence in the cytoplasm, possibly consistent with autofluorescence or an interaction with the CAT enzyme expressed in the control cells. However, these control cells did not show labelling of the plasma membrane as in the Hco-Pgp-9.1 transfected cells treated with the same antibody. A similar generalised fluorescence of the cytoplasm of CAT transfected cells was previously observed (27), using the same secondary antibody, but a different primary antibody, suggesting that this generalised fluorescence of the CAT transfected cells is nonspecific.
Transport studies of Hco-PGP-9.1 After the Hco-PGP-9.1 expression profile was assessed, Hco-Pgp-9.1/LLCPK1 transfected cells were subjected to a Rho123 translocation assay. It was found that Hco-Pgp-9.1/LLCPK1 transgenic cells accumulated less than 40 % Rho123 fluorescence compared with the control parental cells (LLC-PK1) and CAT/LLC-PK1 cells (Fig. 4), indicative of Rho123 efflux by the transfectants expressing Hco-PGP-9.1. Although Rho123 fluorescence accumulation was higher in Hco-Pgp-9.1/LLCPK1 transfectants than in positive control cells over expressing mdr1a (mouse PGP), these fluorescence means are significantly lower than in the control LLC-PK1 untransfected cells. Subsequently, a competition assay between the different MLs and the Rho123 transport by Hco-PGP-9.1 was run. A pronounced inhibitory effect by the avermectins (IVM and ABA) on Hco-PGP-9.1 Rho123 transport was found, giving a saturable effect on the inhibition of transport of this fluorophore, which was close to 80 % of the total Rho123 inhibition effect produced by VSP on Hco-PGP-9.1 (Fig. 5). In contrast, MOX caused a non-saturable inhibitory effect on the Rho123 transport by Hco-PGP-9.1. This MOX effect contrasts with the avermectin effects, for both IVM and ABA in which 0.5 μM produced a maximal inhibition of Rho123 transport corresponding to 80 % of the effect produced by VSP at 5 μM. In contrast, MOX even at 10 or 20 μM concentrations, reached only 50 % or less of the maximal VSP effect (5 μM) on the inhibition of Rho123 transport.

Localisation of Hco-PGP-9.1 in H. contortus adult stage
The same specific antibody was used to try to determine the localisation of Hco-PGP-9.1 in H. contortus adult worms. After examining the whole body, a defined signal corresponding to the reproductive system, particularly at the level of the uterus, can be found in female worms (Fig. 6a-d).

Discussion
ML endectocides represent the cornerstone of current chemotherapy to control different pathogens such as ticks, diptera and nematodes [33]. Resistance to these compounds has arisen particularly in trichostrongylids such as H. contortus, initially against IVM [34] and later against ABA and MOX [35,36]. The involvement of ABC transporters in ML resistance in nematodes has been investigated and suggests a linkage between  We have investigated the transport activity of Hco-PGP-9.1 expressed in mammalian cells and its interaction with the different MLs. An expression profile of this transporter was obtained in the transfected cells, based in its transcript and protein levels in addition to its localisation in the plasma membrane of the host cells. The expressed Hco-PGP-9.1 was probably unglycosylated in the host cells, based on a comparison of the molecular size of the band seen in Western blots and the predicted molecular size of the protein. This result is consistent with the Hco-PGP-2 expression pattern found in the same mammalian cell line and in whole crude membranes extracted from H. contortus [27]. Glycosylation is not required for functional activity of PGPs expressed in cells [40]. Both the unglycosylated Hco-PGP-9.1, and the previously characterised unglycosylated Hco-PGP-2 [27] were able to translocate Rho123.
The protein expression pattern in the porcine cells was similar to the one previously obtained for Hco-PGP-2 [27] except that the transcript level of the Hco-pgp-9.1 transgene appeared to decrease by the third passage (the transcription was not very stable). This is probably because the Hco-Pgp-9.1 gene, for transfection into the LLC-PK1 cells, was not codon optimised for the host pig cells as was Hco-Pgp-2 [27].
When the Hco-PGP-9.1 transport function was assessed, using the fluorophore probe Rho123, a pronounced reduction of this tracer dye in Hco-Pgp-9.1/ LLC-PK1 transfectants was determined, indicative of an active transport by Hco-PGP-9.1. Previous work had pointed out the good substrate characteristics of Rho123 to test nematode transporter activity [28][29][30][31][32][33][34][35][36][37][38][39][40][41]. As shown in Fig. 5, the avermectins (ABA and IVM) produced a marked inhibitory effect on the Rho123 efflux by the Hco-PGP-9.1 transporter. In mammalian models the interaction of the avermectins with ABC transporters, has been described with ABA [42] and IVM inhibiting PGP activity [43] and MRP (Multi-drug Resistant Protein) activity [44]. This is comparable with the corresponding effect that the avermectins produced on Rho123 transport inhibition by Hco-PGP-2 [27] and by mouse PGP [11], confirming a strong interaction of the avermectins for both nematode and mammalian ABC transporters. On the other hand, MOX displayed a lower inhibitory effect on the Rho123 transport by Hco-PGP-9.1, implying a lesser interaction with this transporter. These observations are consistent with previous work describing the interaction of these compounds with Hco-PGP-2 [27], where the avermectins produced a saturation of the Rho123 transport activity compared with the inferior effect of MOX on Rho123 transport. These differences in the interaction with Hco-PGP-9.1, between the avermectins and MOX, are likely to be due to structural differences between the ML subgroups. Avermectins, in contrast to MOX, have a di-(or mono-) saccharide moiety attached to the C-13 of the macrocyclic ring, whereas MOX is protonated at this position [2,45]. This olaeadrose moiety, in IVM and ABA, is thought to enhance the binding of the avermectins to the transporter [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27]. On the other hand, MOX, in addition to the absence of sugar groups attached to the macrocyclic ring, has other structural features, notably the methoxime group at the level of the C-23 of the macrocyclic ring, which may reduce its interaction with nematode transporters. Another aspect which may contribute to this distinction in the interaction of the MLs with Hco-PGP-9.1, is a physicochemical characteristic that differs between the avermectins and MOX. In general, MLs are lipophilic drugs that need to cross membranes to reach their targets inside the parasite [20]. In this respect, MOX is more lipophilic and has a higher octanol/water coefficient (logp) than the avermectins [2], perhaps being retained by the lipid bilayer of the cell membrane. MOX has a longer half-life than the avermectins inside the host due to its longer residency in fat tissue [46,47]. Recently, a nematode PGP from the equine helminth Cylicocylus elongates, termed Ceg-Pgp-9 (ortholog of Ce-Pgp-9), was expressed in yeast and its interaction, in a competition assay between the antifungical ketoconazole and MLs, characterised. A strong interaction with the avermectins (IVM and EPR), and a lesser interaction with MOX, was also found [48].
Our second objective was to try to localise the expression of Hco-PGP-9.1 in H. contortus adult worms. In the free-living nematode C. elegans, its paralogue Cel-Pgp-9, is predicted to be expressed in the pharynx and intestine [49]. However in H. contortus, there may be three paralogs of the Cel-Pgp-9, namely Hco-Pgp-9.1, Hco-Pgp-9.2 and possibly Hco-Pgp-9.3 [24][25][26][27][28]. It was reported that a differential transcript profile exists between Hco-Pgp-9.2 and Hco-Pgp-9.3 on the one hand, which were reported to be expressed more abundantly in the L3 stage, and Hco-Pgp-9.1, which was more abundant in the adult worm [28]. The primary antibodies were raised to epitopes that may be shared, at least between Hco-PGP-9.1 and Hco-PGP-9.2 (see Additional file 1). As a full sequence is unavailable for putative Hco-PGP-9.3, we are unable to assess whether the antibodies may also detect this putative P-glycoprotein. The anti-PGP-9.1 antibody labelled the uterus of the female reproductive system. If there are three paralogues of Pgp-9, each may have a different function and possible expression profile in the parasite. Adult female worms possess a developed reproductive tract, including the uterus, which may constitutively express this nematode transporter. The actual function of Hco-PGP-9.1, possibly in the uterus, is unknown. From the literature it is well established that in higher eukaryotes such as mammals, ABC transporters such as PGP are highly expressed in the apical epithelium membranes including the blood-brain-barrier [50,51] and placenta [52,53], modulating xenobiotic passage to the central nervous system or from mother to fetus. It was found that mdr1a (-/-) (ABCB1 knockout) adult mice had significantly greater susceptibility to IVM neurotoxicity than wild-type mice [50]. In the CF-1 mouse strain carrying a spontaneous mutation on the mdr1a gene, due to the absence of PGP expressed in the placenta barrier, IVM induces a congenital disorder similar to cleft palate in mouse pups [54]. In addition, at the level of the placenta, the active transport of lipophilic molecules such as progesterone, an important female hormone during pregnancy, has been described [55]. Our finding of possible Hco-PGP-9.1 expression in the uterus suggests analogous functions to ABCB1 in mammals, including transport of xenobiotic products from uterine tissue, or recruitment of some lipid molecules to complete ova maturation before release from the female worm through the vulva.
In the context of ML resistance in parasitic nematodes such as H. contortus, the apparent localisation of Hco-PGP-9.1 could represent an interesting discovery. It is well known in the human filarial nematode O. volvulus, that IVM produces a temporary arrest of microfilaria release from the female worms [56]. In another filarial species, the heartworm of dogs, Dirofilaria immitis, and in the pig nematode Oesophagostomum dentatum, IVM interferes with embryogenesis in the adult female worms [57,58]. Recently, a high expression of GluCl receptors, such as the avr-14 gene, has been reported in the reproductive system of female and male worms of Brugia malayi [59]. The localisation of this GluCl may explain the IVM effects on reproductive tissues. A similar effect of MLs on the reproductive system of trichostrongylid nematodes has not been described. However, our data could suggest a role in which Hco-PGP-9.1, apparently expressed in the uterus of H. contortus, may be transporting these compounds, thereby protecting this tissue from the paralysing effect of avermectins on the uterine muscles and the release of nematode eggs from the female worm.
Complementary to the possible role of Hco-PGP-9.1 in ML resistance in H. contortus, there is evidence that paralogues of Hco-Pgp-9.1, expressed in other veterinary nematodes, may also be linked to avermectin resistance [25,26]. The differences in the interaction of Hco-PGP-9.1 between the avermectins and MOX, may also explain the differences observed in the development of resistance to these drugs. ABA, IVM and MOX are among the most common MLs used in livestock to control gastro-intestinal helminths [60]. Different studies on small ruminant farms, have described a reduction in efficacy (<95 %, at the recommended dose rate), mainly to the avermectins whereas MOX was still effective or showed a higher efficacy against trichostrongylids, including H. contortus [61][62][63]. Nonetheless resistance to MOX can arise in gastro-intestinal parasites following MOX selection pressure [64,65]. A possible contribution to the earlier and more common appearance of resistance to the avermectins could be the active efflux of these compounds by nematode transporters, such as