Characterisation of milbemycin-, avermectin-, imidazothiazole- and benzimidazole-resistant Teladorsagia circumcincta from a sheep flock

Papers Characterisation of milbemycin-, avermectin-, imidazothiazole- and benzimidazole-resistant Teladorsagia circumcincta from a sheep flock N. D. Sargison, F. Jackson, D. J. Wilson, D. J. Bartley, C. D. Penny, J. S. Gilleard Selection for milbemycin resistance in a population of Teladorsagia circumcincta was examined in a sheep flock in which a lack of persistence of an oral dose of.2 mg/kg moxidectin against T circumcincta had previously been identified. A faecal egg count reduction test also showed resistance to benzimidazole, levamisole and avermectin anthelmintic groups. Bioassays were used to compare the moxidectin-resistant T circumcincta with another previously characterised benzimidazole-, levamisole- and ivermectin-resistant (MTci5) strain that had been isolated from a sheep flock in the same region in south-east Scotland and with an anthelmintic-susceptible (MTci3) strain of T circumcincta. The mean ED value (the concentration of drug required to prevent per cent of eggs from hatching) obtained for thiabendazole in egg hatch assays was higher in the moxidectin-resistant T circumcincta than in the ivermectin-resistant MTci5 strain. The inclusion of the cytochrome p4 inhibitor piperonyl butoxide in larval feeding inhibition assays increased the level of ivermectin resistance in vitro in the ivermectin- and moxidectin-resistant populations, but not in the ivermectin-susceptible MTci3 strain of T circumcincta. DURING recent years, the management of nematode parasites to minimise their effects on the productivity of sheep in the UK has become complicated by climate change, farm economics and agricultural policies, parasite evolution, and the emergence of resistance to all three currently marketed anthelmintic action groups (Sargison and others 21, 27, Yue and others 23, Bartley and others 26, Veterinary Laboratories Agency [VLA] 28). Under these circumstances, it is essential that all of the available anthelmintic drug groups are used in an effective and responsible manner with the aim of sustaining their practical efficacy, while ensuring adequate nematode parasite control. The milbemycin and avermectin anthelmintic groups share a common mechanism of action, but the molecular structure of moxidectin differs from that of the avermectin anthelmintics, affording greater potency and higher lipid solubility, and therefore persistence (Pankavich and others 1992). This persistent activity against infective Veterinary Record (21) 166, 681-686 N. D. Sargison, BA, VetMB, PhD, DSHP, DipECSRHM, FRCVS, D. J. Wilson, MA, BVMS, DSHP, DipECSRHM, MRCVS, C. D. Penny, BVM&S, CertCHP, DBR, DipECBHM, MRCVS, Large Animal Practice, Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush Veterinary Centre, Roslin, Midlothian EH25 9RG F. Jackson, BSc, PhD, D. J. Bartley, BSc, PhD, Department of Parasitology, Moredun Research Institute, Pentlands Science Park, Bush Loan, Penicuik, Midlothian EH26 PZ doi: 1.1136/vr.b4853 J. S. Gilleard, BVSc, PhD, DipEVPC, MRCVS, Department of Comparative Biology and Experimental Medicine, Faculty of Veterinary Medicine, University of Calgary, 33 Hospital Drive NW, Calgary, Alberta, Canada E-mail for correspondence: neil.sargison@ed.ac.uk Provenance: not commissioned; externally peer reviewed larval stages makes moxidectin useful for the management of teladorsagiosis and haemonchosis whenever it is necessary to graze lambs on fields that are heavily contaminated with infective nematode larvae, and for the strategic treatment of ewes when they are turned on to pastures contaminated with infective nematode larvae after lambing. Side resistance has been shown between milbemycin and avermectin drugs (Shoop and others 1993), but moxidectin at the recommended oral dose rate of.2 mg/kg has been shown to be effective in removing ivermectin-resistant Teladorsagia circumcincta from infected sheep (Bartley and others 24), indicating that the drug might be effective for quarantine anthelmintic treatment of introduced sheep (Sargison and others 25) (sequential administration of a full dose of levamisole is usually recommended). These practical applications for moxidectin underpin the importance of ensuring that the drug is used responsibly with the aim of preserving its efficacy against sheep nematode parasites. Reduced prophylactic activity in sheep against parasitic larval stages of ivermectin-resistant T circumcincta has been demonstrated in the UK following an oral dose of.2 mg/kg moxidectin in both critical efficacy (Bartley and others 24) and field (Sargison and others 25) studies. Under certain conditions, this reduced persistent activity of moxidectin against T circumcincta may confer a further level of tail selection for resistance by allowing the establishment of avermectin-resistant larvae while continuing to prevent the establishment of larvae that are not resistant (Rendell and others 26). Under these circumstances, the use of moxidectin to control avermectin-resistant nematodes suppressively may be considered to be imprudent. The macrocyclic lactone anthelmintics act on ligand-gated ion channels, including glutamate-gated chloride channels (GluCl) (Blackhall and others 1998b) and GABA-gated chloride channels (GABACl) (Blackhall and others 23). Glutamate and GABA are neuro transmitters that regulate muscle contractions, thus regulating activities such as feeding, locomotion and reproduction (Yates and others 23). However, the mechanisms responsible for the development of resistance are poorly understood. Variations in allele frequencies between resistant and susceptible Haemonchus contortus have been May 29, 21 Veterinary Record

shown in a putative subunit GABACl gene (Prichard 21), but causal associations with ivermectin resistance have not been demonstrated. A functional single nucleotide polymorphism (SNP) mutation has been shown at codon 256 (L256F) in a GluCl subunit gene of the cattle nematode parasite Cooperia oncophora, associated with ivermectin resistance (Njue and Prichard 24). However, this mutation does not appear to be consistently present in ivermectin-resistant C oncophora and has yet to be found in any other parasitic nematodes. In addition to the ligand-gated chloride channels, resistance mechanisms are also thought to involve a mutation in a P-glycoprotein (P-gp) gene (Blackhall and others 1998a, Kerboeuf and others 23), leading to more rapid drug clearance from the nematode parasite. Some mutations that are involved with xeno biotic drug regulatory mechanisms may also be involved with multiple anthelmintic drug resistance. Furthermore, it is possible that the selection for such drug regulatory mechanisms using macrocyclic lactone anthelmintics may also confer some resistance to other anthelmintic drug groups with different modes of action, including new groups such as the amino-acetonitrile derivative drugs (Kaminsky and others 28). A SNP mutation resulting in a single amino acid substitution from tyrosine to phenylalanine at position 2 (F2Y) of the polypeptide encoded by the isotype 1 b-tubulin gene is considered to be functionally significant (Kwa and others 1995) in conferring benzimidazole resistance in H contortus (Kwa and others 1994) and T circumcincta (Elard and others 1996). Other resistance mechanisms in the isotype 1 b-tubulin gene (Silvestre and Cabaret 22), other tubulin genes (Lubega and others 1994), and genetic pathways involving cell membrane efflux pumps or drug metabolism and excretion are also implicated in benzi midazole resistance (Kerboeuf and others 23, Kotze and others 26). Selection for the isotype 1 b-tubulin F2Y SNP mutation has been demonstrated following repeated ivermectin treatment of sheep infected with H contortus, showing that the use of macrocyclic lactone anthelmintics may select for benzimidazole resistance (de Lourdes Mottier and Prichard 28). Sensitive molecular markers are required to elucidate the population genetics of resistance to all anthelmintic drug groups, in order to provide a basis for advice about their responsible use. In the long term, it is possible that genome-wide comparisons between ivermectin- susceptible and -resistant nematodes could reveal polymorphisms linked to a variety of genes. Where these polymorphisms are linked to genes involved with the known mechanisms of action of the anthelmintic, such as ligand-gated chloride channels, they will become obvious candidates for conventional molecular biological study. However, there is growing evidence to suggest the involvement of other genes, for example, those concerned with xenobiotic clearance and regulatory mechanisms, which may not be immediately obvious as candidates for further study. The preliminary characterisation of a moxidectin-resistant field population of T circumcincta, which was identified in a flock in south-east Scotland with a previous history of ivermectin resistance, is described in this paper. It is hoped that the results may help to elucidate the involvement of mechanisms of anthelmintic resistance associated with drug targets or with resistance mechanisms that prevent the drug from reaching its target. Materials and methods Flock history with reference to anthelmintic efficacy During May 25, the routine monitoring of nematode parasite control, involving four to five weeks post-anthelmintic drench faecal worm egg counts (FWECs) and coprocultures for larval identification, highlighted the reduced persistence of moxidectin against Teladorsagia species in a large upland sheep flock of approximately 12 crossbred and terminal sire ewes in south-east Scotland. This problem was first identified in a group of greyface ewes that had been treated orally with.2 mg/kg moxidectin (Zermex.1 per cent oral drench; Fort Dodge Animal Health) after lambing with the aim of managing their periparturient rise in faecal nematode egg output. Similar findings of reduced persistence of.2 mg/kg moxidectin against T circumcincta have been reported in association with the presence of ivermectin and multiple anthelmintic drug resistance (Vickers and others 21, Sargison and others 25). Thus, although this initial observation was not followed up with a faecal egg count reduction test (FECRT), it nevertheless indicated the probable presence of macrocyclic lactone-resistant nematodes on the farm, and highlighted the need to adopt a responsible approach to reduce the risk of further selection for resistance. In summer 27, nematode control using suppressive anthelmintic treatments was attempted in a group of approximately 1 terminal sire ewes and their lambs on the same farm, with the owner s objective of achieving maximal growth rates in the ram lambs. The ewes were treated with oral moxidectin after lambing in January, when they were given access, for six to eight weeks, to fields that had been intensively grazed by sheep during the previous autumn. The ewes and lambs were then all treated with oral moxidectin at the beginning of March and moved on to pasture that was considered to be moderately contaminated with nematode larvae, on the basis of its previous grazing history. The lambs were weaned in mid-may, but not treated with anthelmintic until late June, when, following the identification of suboptimal growth rates and observation of intermittent signs of diarrhoea, the ram lambs were treated with oral moxidectin. Oral moxidectin treatments were repeated in mid and late July, resulting in three oral moxidectin treatments during a period of five weeks. After the third oral moxidectin treatment the lambs were moved to a different field, which had recently been grazed by another group of sheep. Despite this suppressive anthelmintic control strategy, the lambs achieved suboptimal growth rates and signs of diarrhoea persisted. The pooled faecal trichostronglylid egg count (FWEC) of six lambs, 17 days after the last oral moxidectin treatment, was epg. This count supported a diagnosis of parasitic gastroenteritis, but did not provide conclusive evidence of drug inefficacy, because of uncertainty about correct administration of the anthelmintic, the drug expiry date and storage of the product used. Field investigation of anthelmintic efficacy A FECRT was performed, beginning in mid-august, 22 days after the last oral moxidectin treatment (Wilson and Sargison 27). Forty lambs were weighed and randomly allocated to four treatment groups of 1 animals. Faecal samples were collected from each animal before drenching on the basis of the heaviest in each group with at least 5 mg/kg liveweight fenbendazole (Panacur 2.5 per cent oral suspension; Intervet UK), 7.5 mg/kg liveweight levamisole (Nilzan Gold; Schering-Plough Animal Health),.2 mg/kg liveweight ivermectin (Oramec drench; Merial Animal Health) or.2 mg/kg liveweight moxidectin (Cydectin.1 per cent oral drench for sheep; Fort Dodge Animal Health). The anthelmintic products had all been correctly stored and were used before their expiry dates. Further faecal samples were collected from each animal seven days (levamisole group) and 14 days (benzimidazole and macrocyclic lactone groups) after dosing, and trichostrongyle egg counts were performed on each sample using the modified McMaster technique, where each egg counted represents epg (Ministry of Agriculture, Fisheries and Food 1986). Pooled faecal material taken on the day of treatment and from each of the treated groups post-treatment was cultured for 14 days at approximately 2 C to provide infective larvae for identification (Ministry of Agriculture, Fisheries and Food 1986, Van Wyk and others 24). The faecal egg count reduction for each treatment group and 95 per cent confidence intervals were calculated using the arithmetic mean data along the lines recommended by the World Association for the Advancement of Veterinary Parasitology (Coles and others 1992), but the inclusion of an untreated control group was considered to be impractical. Resistant populations were defined more accurately by calculating reductions in FWECs for each nematode genus (McKenna 1996, 1997). In vitro characterisation of nematodes from the terminal sire flock Rectal faecal samples were collected and pooled from a group of lambs that had been treated with.2 mg/kg moxidectin 14 days previously, and from some larger lambs from the same group that had not been treated with an anthelmintic recently. The faeces were cultured for 1 days at approximately 2 C to yield third-stage larvae (L3), which were identified morphologically as per cent Teladorsagia species and 1 per cent Trichostrongylus species. A seven-month-old worm-free donor male lamb was then dosed in October (day 1) with 15, L3 harvested from the coprocultures (subsequently referred to as the field Veterinary Record May 29, 21

TABLE 1: Results of faecal egg count reduction tests on a field population of parasitic nematodes from a sheep flock in south-east Scotland Treatment group Mean (sd) pretreatment FWEC (epg) Mean (sd) posttreatment FWEC (epg) population). A harness and faecal collection bag was fitted from day 25 to enable daily L3 coprocultures to be set up, to facilitate monitoring of FWECs and to ensure that sufficient faeces could be collected as a source of eggs and first-stage larvae (L1) for egg hatch assays (EHAs) and larval feeding inhibition assays (LFIAs), respectively. The donor lamb was euthanased on day 134, and T circumcincta recovered from the abomasum were stored in absolute ethanol for future molecular study as required. Contemporary T circumcincta donor lambs that had been infected at approximately the same time with characterised reference strains were available throughout the study to enable comparison with the field population. These lambs had been infected with 1, L3 of a strain of T circumcincta (MTci5 [Moredun triple-resistant isolate]) derived from a field benzi midazole-, imidazothiazole- and ivermectin-resistant in vivo population from a sheep flock in south-east Scotland (Sargison and others 21, Bartley and others 25), and with a field benzimidazole-, imidazothiazole- and ivermectin-susceptible in vivo strain (MTci3), derived from a field anthelmintic-susceptible population submitted to the Central Veterinary Laboratory during the 19s. A contemporary donor lamb infected with an anthelminticsusceptible in vivo strain of Trichostrongylus vitrinus (MTvi1) was also available. The results of EHAs and LFIAs using this donor were used to crudely estimate discriminatory drug concentrations in the field population that could have been attributable to Trichostrongylus species, thereby enabling calculation of values for T circumcincta. For example, the number of eggs in a drug concentration replicate of the EHA that could have been Trichostrongylus species (n) was calculated using the formula: n=a x b x c, where a is the total number of field population eggs and L1 in the drug concentration replicate, b is the percentage of Trichostrongylus species identified in coprocultures of the field population, and c is the number of eggs as a percentage of the total number of eggs and L1 in the corresponding drug concentration replicate of the EHA for T vitrinus (this approach is intrinsically inaccurate, in part because it assumes that the drug sensitivity and behaviour of the MTvi1 strain of T vitrinus is the same as that of the Trichostrongylus species in the field population, but was nevertheless adopted to provide a useful, crude estimation of discriminatory drug doses attributable to T circumcincta). The numbers of Oesophagostomum and/or Chabertia species L1 were estimated on the basis of their distinctive morphology, being broader and having longer tails. Post-treatment efficacy (%) (95% CI) Post-treatment efficacy for Teladorsagia (%) Benzimidazole (n=1) 9 (521) 11 (855) ( 124-31).6% Teladorsagia 1% Teladorsagia Levamisole (n=1) 5 (3) 1 (15) 83 (63-92) 81.6% Teladorsagia 1% Teladorsagia Ivermectin (n=1) 7 (759) 3 () 51 ( 26-81) 47.6% Teladorsagia 99.3% Teladorsagia.7% Trichostrongylus Moxidectin (n=1) 7 (453) 1 (158) 82 (58-92).6% Teladorsagia 1% Teladorsagia CI Confidence interval, FWEC Faecal worm egg counts Egg hatch assays Nematode eggs were extracted from faeces collected per rectum using a standard sieving and saturated saline flotation method. The eggs were washed by three cycles of resuspension in water and centrifugation, and then titrated to yield about 1 eggs/1 µl of water. Serial dilutions were prepared of 1,,, 2, 15, 1 and 5 µg thiabendazole (Sigma)/ml dimethyl sulfoxide (DMSO). An aliquot of 1 µl of each test drug dilution, 18 µl of distilled water and 1 µl of water containing approximately 1 eggs was then added in sequence to the wells of 24-well plates, giving a dilution factor of 1/2 for thiabendazole, with drug dilutions of.5,.3,.2,.1,.75,.5 and.25 µg/ml. Each drug dilution was used in two wells and DMSO controls were included. The assays were incubated for 48 hours at 22 C, after which the contents of each well were killed, stained and fixed with a drop of helminthological iodine (2 g potassium iodide and g iodine, made up to ml with distilled water). The numbers of L1 and eggs in each well were then counted to obtain the percentage of eggs hatched to L1 in different concentrations of thiabendazole. These values were used to plot dose-response curves, from which egg development (ED) values were calculated by Probit analysis using Minitab v 13 software. ED values were used as universal estimates of the concentration of drug required to prevent per cent of the eggs from hatching; benzimidazole resistance is generally deemed to be present if more than per cent of the eggs hatch in.1 µg/ml thiabendazole (Hunt and Taylor 1989). EHAs were performed on days 36, 38, 42 and 52 using eggs extracted from faeces that had been collected from the rectum of the donor lambs infected with the field population, the benzimidazole-resistant MTci5 strain and the benzimidazole-susceptible MTci3 strain of T circumcincta. An EHA was also performed using eggs extracted from the faeces of the benzimidazole-susceptible T vitrinus donor. EHAs were performed on day 57 in the presence of verapamil and piperonyl butoxide to assess whether or not the cytochrome p4 (piperonyl butoxide) (Kotze and others 26) and P-glycoprotein (vera pamil) (Beugnet and others 1997) inhibitor drugs affected egg hatching in the presence of thiabendazole, and to elucidate any differences in the biochemical mechanisms of benzimidazole resistance between the field population and ivermectin-resistant and -susceptible strains of T circumcincta. An 18.75mM solution of verapamil was prepared by adding 18.75 µl of.5m stock solution (.5 g of verapamil hydrochloride [Sigma] in 2 ml of DMSO) to 31.25 µl of DMSO. An aliquot of 2.66 µl of 18.75mM verapamil was then added to each 2 ml replicate of the EHA to give a final concentration of 25µM. A 9.3mM solution of piperonyl butoxide was prepared by adding 9.3 µl of.1m stock solution (32 µl piperonyl butoxide [Sigma] in 1 ml of DMSO) to.7 µl of DMSO. An aliquot of 2.66 µl of 9.3mM piperonyl butoxide was then added to each 2 ml replicate of the EHA to give a final concentration of 12.5µM. Paired EHAs were performed without the inclusion of piperonyl butoxide or verapamil (the 2 ml wells would have contained about.5 per cent DMSO, while those containing the inhibitor drugs would have contained about.63 per cent DMSO). ED values were calculated by Probit analysis using Minitab v 13 software. Larval feeding inhibition assays Nematode eggs were extracted from fresh faecal samples, washed and incubated in petri dishes in 25 ml of tap water for about 24 hours at 22 C to yield L1. The L1 were then cleaned by transferring the contents of the petri dish to mini Baermann apparatus, consisting of 25 µm nylon mesh held over the end of sections of hard plastic tubing, and placed within the wells of a six-well plate. The Baermann setups were incubated at 22 C for approximately one hour, before removing the cylinders and mesh and transferring the contents of the wells of the six-well plates to hard centrifuge tubes, filled to the 1 ml mark. The contents were then mixed, larvae in 1 µl aliquots were counted, and the contents were titrated so that each 1 µl contained approximately 1 L1. Serial dilutions were prepared of ivermectin (Sigma) in DMSO. Microcentrifuge tubes, 1.5 ml in size (Eppendorf), were labelled with the concentration of the ivermectin test dilution and arranged on a tray in two replicated May 29, 21 Veterinary Record

(a) 1 Percentage of eggs hatched 2 1 (b) 1 Percentage of eggs hatched Field population MTci5 MTci3.1.2.3.4.5 Concentration of thiabendazole (μg/ml) 2 1 Moxidectin-resistant T circumcincta MTci5 MTci3.1.2.3.4.5 Concentration of thiabendazole (μg/ml) Fig 1: Mean (se) egg hatching in egg hatch assays comparing (a) the field population or (b) estimated Teladorsagia circumcincta within the field population with benzimidazole-resistant (MTci5) and benzimidazole-susceptible (MTci3) strains of T circumcincta (a) 1 Percentage of L1 fed 2 1.1.1.1.1.1.1 (b) 1 Percentage of L1 fed 2 1 Concentration of ivermectin (µg/ml) Field population control Field population + piperonyl butoxide MTci5 control MTci5 + piperonyl butoxide MTci3 control MTci3 + piperonyl butoxide.1.1.1.1.1.1 Concentration of ivermectin (µg/ml) Field population control Field population + verapamil MTci5 control MTci5 + verapamil MTci3 control MTci3 + verapamil Fig 2: Mean (se) L1 feeding in larval feeding inhibition assays comparing the effects of (a) piperonyl butoxide and (b) verapamil on the moxidectin-resistant population, the ivermectin-resistant (MTci5) and ivermectin-susceptible (MTci3) strains of Teladorsagia circumcincta rows. Aliquots of 1 µl of ivermectin test solution, 13 µl of tap water and 1 µl of tap water containing about 1 L1 were then pipetted into each tube in turn. The racks of tubes were left to stand for about two hours in a microbial extraction cabinet, with the lids open, during which time preprepared vials of frozen fluorescein isothiocyanate (FITC)-labelled lyophilised Escherichia coli were thawed. An aliquot of 7 µl of FITC-labelled E coli was then pipetted into each tube, before they were sealed, mixed and incubated for 18 hours at 22 C. After approximately 18 hours the microcentrifuge tubes were removed from the incubator and centrifuged for approximately three minutes at g. An aliquot of 1 µl including the pellet containing L1 was then transferred from the bottom of the tube on to a microscope slide and examined under an inverted fluorescent microscope fitted with a blue filter. Fed L1 with a fluorescent pharynx and gut and unfed L1 were counted to determine the percentage fed at each ivermectin dilution and to produce dose-response curves from which larval feeding inhibition (LFI) values were calculated by Probit analysis using Minitab v 13 software. LFI were used as universal estimates of the concentration of drug at which and 99 per cent of the L1, respectively, do not feed (Álvarez-Sánchez and others 25). LFIAs were performed on days 38, 53 and 19 using L1 hatched from eggs extracted from faeces that had been collected from the donor lambs infected with the field population, the benzimidazoleresistant MTci5 strain and the benzimidazole-susceptible MTci3 strain of T circumcincta. A LFIA was also performed using L1 hatched from eggs extracted from the faeces of the benzimidazole-susceptible T vitrinus donor. LFIAs were performed on days 58 and 88 in the presence of verapamil and piperonyl butoxide to assess whether or not the metabolic inhibitor drugs affected L1 feeding in the presence of ivermectin, and to elucidate any differences in the biochemical mechanisms of ivermectin resistance between the field population and the ivermectinresistant and -susceptible strains of T circumcincta. An aliquot of 2 µl of 18.75mM verapamil was added to each 1.5 ml replicate of the LFIA to give a final concentration of 25µM. An aliquot of 2 µl of 9.3mM piperonyl butoxide was added to each 1.5 ml replicate of the LFIA to give a final concentration of 12.5µM. LFI values were calculated by Probit analysis using Minitab v 13 software. Results Field investigation of anthelmintic efficacy The FECRT showed that the field population was resistant to benzimidazole, imidazothiazole, and both ivermectin and moxidectin macro cyclic lactone anthelmintics. The faecal egg count data, calculated post-treatment efficacies, coproculture results and the reductions in FWEC for the resistant Teladorsagia species are shown in Table 1. In vitro characterisation of moxidectin-resistant nematodes from the terminal sire flock The mean FWEC of the field population donor lamb between days 28 and 17 was 66 epg (range 24 to 84 epg). Coprocultures of faeces collected between days 36 and 66 yielded 76 per cent Teladorsagia species, 18 per cent Trichostrongylus species and 6 per cent Oesophagostomum and/ or Chabertia species. The dose-response curves and ED values for the EHAs confirmed the in vitro benzimidazole susceptibility of the MTci3 strain of T circumcincta (mean [sd] ED.63 [.12] µg/ml thiabendazole) and the high level of in vitro benzimidazole resistance in the MTci5 strain Veterinary Record May 29, 21

TABLE 2: Mean ED values comparing the effects of piperonyl butoxide and verapamil on the moxidectin-resistant field population, and the benzimidazole-resistant (MTci5) and benzimidazole-susceptible (MTci3) strains of Teladorsagia circumcincta Control (inhibitors not included) ED (µg/ml thiabendazole) Piperonyl butoxide Verapamil Moxidectin-resistant population.352.272.281 MTci5 T circumcincta.7.6.233 MTci3 T circumcincta.72.82.68 TABLE 3: Mean LFI values (µg/ml ivermectin) comparing the effects of piperonyl butoxide and verapamil on the moxidectin-resistant population, and the ivermectinresistant (MTci5) and ivermectin-susceptible (MTci3) strains of Teladorsagia circumcincta Control (inhibitors not included) Piperonyl butoxide Verapamil LFI LFI 99 LFI LFI 99 LFI LFI 99 Moxidectin-resistant population.1.18.35.55.16.35 MTci5 T circumcincta.2.3.2.3.8.22 MTci3 T circumcincta.11.18.5.9.1.19 of T circumcincta (ED.266 [.87] µg/ml thiabendazole). The doseresponse curves and ED values showed that the field population was more resistant in the EHA (ED.352 [.87] µg/ml thiabendazole) than the MTci5 strain of T circumcincta. This increased level of resistance became even more apparent when the estimated dose-response curve for the moxidectin-resistant T circumcincta in the field population was compared with the ivermectin-resistant and partially moxidectinsusceptible MTci5 strain of H contortus (Fig 1). L1 feeding in drug-free replicates of the LFIA was consistently poorer in the field population than in the MTci3 and MTci5 strains of T circumcincta. Nevertheless, there was no difference in LFI or LFI 99 values between the ivermectin-susceptible MTci3 strain (LFI.11 [.13] µg/ml ivermectin; LFI 99.18 [.22] µg/ml ivermectin) and the ivermectin-resistant MTci5 strain (LFI =.2 [.1] µg/ml ivermectin; LFI 99 =.3 [.1] µg/ml ivermectin) and field populations of T circumcincta (LFI.1 [.11] µg/ml ivermectin; LFI 99.18 [.17] µg/ml ivermectin). Effect of inhibitors of the xenobiotic response on the resistance phenotype Inclusion of 25µM verapamil or 12.5µM piperonyl butoxide in the EHA had no effect on the dose-response curves or ED values for the field population or for the benzimidazole- and ivermectin-susceptible and -resistant MTci3 and MTci5 strains of T circumcincta (Table 2) (the ED values for the controls were similar to those reported in the characterisation study, but differed because different contemporary controls were used). Inclusion of verapamil in the LFIA reduced L1 feeding in ivermectin-free control replicates and in replicates containing lower concentrations of ivermectin for the field population and the MTci5 and MTci3 strains (Fig 2). The dose-response curves show that inclusion of the cytochrome p4 inhibitor piperonyl butoxide increased the level of ivermectin resistance in the ivermectin- and moxidectin-resistant field population and in the ivermectin-resistant MTci5 strain of T circumcincta, but not in the ivermectin-susceptible MTci3 strain. The doseresponse curves are shown in Fig 2, and the LFI values for LFIAs in the presence of piperonyl butoxide and verapamil are shown in Table 3. Discussion Multiple resistance in T circumcincta to the benzimidazole, imidazothiazole and avermectin anthelmintics was confirmed by the low treatment efficacies and 95 per cent confidence intervals. The postmoxidectin treatment efficacy of per cent observed in this case is significant, because it demonstrates the failure of the drug to remove adult female T circumcincta from their host, providing clear evidence of milbemycin resistance. Similar findings of milbemycin resistance have been reported in various countries in the southern hemisphere in avermectin-resistant isolates of T circumcincta (for example, Hughes and others 24) and H contortus (for example, Le Jambre and others 25). The apparent failure of oral moxidectin at a dose rate of.2 mg/kg to remove adult T circumcincta from infected sheep is cause for concern, because the sustainable control of nematode parasites on many UK sheep farms involves its strategic use as part of an integrated parasite control programme. While continued economic sheep production can be achieved in the face of multiple anthelmintic resistance, the failure of moxidectin in this case will necessitate the adoption of increasingly complex evasive and integrated nematode parasite control strategies. Accurate in vitro characterisation of the moxidectin-resistant population was hindered by the fact that the donor lamb was infected with a mixed population of T circumcincta, Trichostrongylus species (probably T vitrinus) and Oesophagostomum species. This situation arose due to practical considerations that were identified at the time, but, as is often the case in the investigation of field problems, were unavoidable. The milbemycin-resistant T circumcincta has subsequently been isolated by infection of a worm-free donor lamb with the field population and selection using moxidectin treatment. This isolate will enable further characterisation and will provide a valuable resource for future studies of anthelmintic resistance. A clear difference was seen in the EHA dose-response curves for the field benzimidazole-, levamisole-, ivermectin- and moxidectinresistant T circumcincta and the MTci5 benzimidazole-, levamisole- and ivermectin-resistant population of T circumcincta, accompanied by a higher ED value for the moxidectin-resistant field population. These results would be consistent with the upregulation of an existing resistance mechanism such as the tyrosine to phenylalanine point mutation at position 2 (F2Y) of the polypeptide encoded by the isotype 1 b-tubulin gene (Kwa and others 1994, 1995, Elard and others 1996), or involvement of an additional resistance mechanism in the moxidectinresistant population, influencing egg hatching in the presence of thiabendazole. However, the different levels of benzimidazole resistance may merely characterise differences between the different populations and strains. The inclusion of metabolic inhibitory drugs in the EHA was not helpful in elucidating the biochemical pathway involved. No differences were seen in the LFIAs between the moxidectinresistant population, the ivermectin-resistant MTci5 strain of T circumcincta and the ivermectin-susceptible MTci3 strain of T circumcincta. This could be because the resistance phenotype that governs L1 pharyngeal pumping in the LFIA differs from the resistance phenotype in parasitic stages of T circumcincta in vivo. Alternatively, it might have been caused by problems of generally poor L1 feeding that are frequently encountered using this assay, particularly during the winter months. Verapamil and piperonyl butoxide were included in the bioassays because they are classically used as biochemical inhibitors of P-glycoproteins (Beugnet and others 1997) and cytochrome p4 enzymes (Kotze and others 26), respectively. Therefore, their ability to modulate the resistance phenotype might implicate xenobiotic responses that result in drug inactivation or efflux. The inclusion of verapamil drastically reduced the level of egg hatching in the EHA. Verapamil has previously been shown to enhance the toxicity of thiabendazole in H contortus eggs (Beugnet and others 1997, Kerboeuf and others 1999). The inclusion of verapamil also reduced the level of L1 feeding in the LFIA. However, these effects involved the anthelminticresistant and -susceptible T circumcincta to similar extents, suggesting that they may have been due to toxic effects of verapamil itself, such as direct action as a calcium channel blocker inhibiting larval feeding, rather than, or in addition to, inhibition of P-glycoprotein-regulated cellular drug efflux mechanisms. Nevertheless, the evidence for the involvement of P-glycoproteins and other ATP-binding cassettetransporter mechanisms for ivermectin resistance in parasitic nematodes is growing (Prichard and Roulet 27, Blackhall and others 28). May 29, 21 Veterinary Record

The inclusion of piperonyl butoxide in the LFIA increased the level of ivermectin resistance in vitro in the ivermectin- and moxidectinresistant field population and the MTci5 strain of T circumcincta, but not in the ivermectin-susceptible MTci3 strain of T circumcincta. Although piperonyl butoxide is generally considered an inhibitor of cytochrome p4 activity, its precise effects in simple experiments at the whole organism level are difficult to predict. For example, piperonyl butoxide is known to induce the expression of certain cytochrome p4 genes in Drosophila melanogaster (Daborn and others 22), and so in some circumstances could enhance the activity of particular enzymes (Kotze and others 26, Willoughby and others 27). Hence, the enhancement of the resistant phenotype in the macrocyclic lactone-resistant isolates, but not in the susceptible MTci3 strain, provides intriguing circumstantial evidence for the potential involvement of xenobiotic responses. This work has characterised a field isolate of T circumcincta that is resistant to all of the current broad-spectrum classes of anthelmintic. The detailed phenotypic characterisation of such field isolates is an essential first step for further molecular and genomic studies aimed at identifying the underlying mechanisms of anthelmintic resistance, because it could enable a challenge hypothesis for any anthelmintic resistance markers that are found. Acknowledgements The Moredun Research Institute receives financial aid from the Scottish Government. References ÁLVAREZ-SÁNCHEZ, M. A., PÉREZ-GARCÍA, J., BARTLEY, D., JACKSON, F. & ROJO-VÁZQUEZ, F. A. (25) The larval feeding inhibition assay for the diagnosis of nematode anthelmintic resistance. Experimental Parasitology 11, 56-61 BARTLEY, D. J., DONNAN, A. A., JACKSON, E., SARGISON, N., MITCHELL, G. B. B. & JACKSON, F. (26) A small scale survey of ivermectin resistance in sheep nematodes using the faecal egg count reduction test on samples collected from Scottish sheep. Veterinary Parasitology 137, 112-118 BARTLEY, D. J., JACKSON, F., JACKSON, E. & SARGISON, N. (24) Characterisation of two triple resistant field isolates of Teladorsagia from Scottish lowland sheep farms. Veterinary Parasitology 123, 189-199 BARTLEY, D. J., JACKSON, E., SARGISON, N. & JACKSON, F. (25) Further characterisation of a triple resistant field isolate of Teladorsagia from a Scottish lowland sheep farm. Veterinary Parasitology 134, 261-266 BEUGNET, F., GAUTHEY, M. & KERBOEUF, D. (1997) Partial in vitro reversal of benzimidazole resistance by the free-living stages of Haemonchus contortus with verapamil. Veterinary Record 141, 575-576 BLACKHALL, W. J., LIU, H. Y., XU, M., PRICHARD, R. K. & BEECH, R. N. (1998a) Selection at a P-glycoprotein gene in ivermectin- and moxidectin-selected strains of Haemonchus contortus. Molecular and Biochemical Parasitology 95, 193-21 BLACKHALL, W. J., POULIOT, J. F., PRICHARD, R. K. & BEECH, R. N. (1998b) Haemonchus contortus: selection at a glutamate-gated chloride channel gene in ivermectinand moxidectin-selected strains. Experimental Parasitology, 42-48 BLACKHALL, W. J., PRICHARD, R. K. & BEECH, R. N. (23) Selection at a g-aminobutyric acid receptor gene in Haemonchus contortus resistant to avermectins/milbemycins. Molecular and Biochemical Parasitology 131, 137-145 BLACKHALL, W. J., PRICHARD, R. K. & BEECH, R. N. (28) P-glycoprotein selection in strains of Haemonchus contortus resistant to benzimidazoles. Veterinary Parasitology 152, 11-17 COLES, G. C., BAUER, C., BORGSTEEDE, F. H. M., GEERTS, S., KLEI, T. R., TAYLOR, M. A. & WALLER, P. J. (1992) World Association for the Advancement of Veterinary Parasitology (WAAVP) methods for the detection of anthelmintic resistance in nematodes of veterinary importance. Veterinary Parasitology 44, 35-44 DABORN, P. J., YEN, J. L., BOGWITZ, M. R., LE GOFF, G., FEIL, E., JEFFERS, S. & OTHERS (22) A single p4 allele associated with insecticide resistance in Drosophila. Science 297, 2253-2256 DE LOURDES MOTTIER, M. & PRICHARD, R. K. (28) Genetic analysis of a relationship between macrocyclic lactone and benzimidazole anthelmintic selection on Haemonchus contortus. Pharmacogenetics and Genomics 18, 129-1 ELARD, L., COMES, A. M. & HUMBERT, J. F. (1996) Sequences of beta-tubulin cdna from benzimidazole-susceptible and -resistant strains of Teladorsagia circumcincta, a nematode parasite of small ruminants. Molecular and Biochemical Parasitology 79, 249-253 HUGHES, P. L., MCKENNA, P. B. & MURPHY, A. (24) Resistance to moxidectin and abamectin in naturally acquired Ostertagia circumcincta infections in sheep. New Zealand Veterinary Journal 52, 22-24 HUNT, K. R. & TAYLOR, M. A. (1989) Use of the egg hatch assay on sheep faecal samples for the detection of benzimidazole resistant nematodes. Veterinary Record 125, 153-154 KAMINSKY, R., DUCRAY, P., JUNG, M., CLOVER, R., RUFENER, L., BOUVIER, J. & OTHERS (28) A new class of anthelmintics effective against drug-resistant nematodes. Nature 452, 176-1 KERBOEUF, D., BLACKHALL, W., KAMINSKY, R. & VON SAMSON- HIMMELSTJERNA, G. (23) P-glycoprotein in helminths: function and perspectives for anthelmintic treatment and reversal of resistance. International Journal of Antimicrobial Agents 22, 332-346 KERBOEUF, D., CHAMBRIER, P., LE VERN, Y. & AYCARDI, J. (1999) Flow cytometry analysis of drug transport mechanisms in Haemonchus contortus susceptible or resistant to anthelmintics. Parasitology Research 85, 118-123 KOTZE, A. C., DOBSON, R. J. & CHANDLER, D. (26) Synergism of rotenone and piperonyl butoxide in Haemonchus contortus and Trichostrongylus colubriformis in vitro: potential for drug synergism through inhibition of nematode detoxification pathways. Veterinary Parasitology 136, 275-282 KWA, M. S. G., VEENSTRA, J. G. & ROOS, M. H. (1994) Benzimidazole resistance in Haemonchus contortus is correlated with a conserved mutation at amino acid 2 in b-tubulin isotype 1. Molecular and Biochemical Parasitology 63, 299-3 KWA, M. S., VEENSTRA, J. G., VAN DIJK, M. & ROOS, M. H. (1995) Beta-tubulin genes from the parasitic nematode Haemonchus contortus modulate drug resistance in Caenorhabditis elegans. Journal of Molecular Biology 246, -51 LE JAMBRE, L. F., GEOGHEGAN, J. & LYNDAL-MURPHY, M. (25) Characterization of moxidectin resistant Trichostrongylus colubriformis and Haemonchus contortus. Veterinary Parasitology 128, 83- LUBEGA, G. W., KLEIN, R. D., GEARY, T. G. & PRICHARD, R. K. (1994) Haemonchus contortus: the role of two beta-tubulin gene subfamilies in the resistance to benzimidazole anthelmintics. Biochemical Pharmacology 47, 15-1715 MCKENNA, P. B. (1996) Potential limitations of the undifferentiated faecal egg count reduction test for the detection of anthelmintic resistance in sheep. New Zealand Veterinary Journal 44, 73-75 MCKENNA, P. B. (1997) Further potential limitations of the undifferentiated faecal egg count reduction test for the detection of anthelmintic resistance in sheep. New Zealand Veterinary Journal 45, 244-246 MINISTRY OF AGRICULTURE, FISHERIES AND FOOD (1986) Manual of Veterinary Parasitological Laboratory Techniques. HMSO NJUE, A. I. & PRICHARD, R. K. (24) Genetic variability of glutamate-gated chloride channel genes in ivermectin-susceptible and -resistant strains of Cooperia oncophora. Parasitology 129, 741-751 PANKAVICH, J. A., BERGER, H. & SIMKINS, K. L. (1992) Efficacy of moxidectin, nemadectin and ivermectin against an ivermectin-resistant strain of Haemonchus contortus in sheep. Veterinary Record 1, 241-242 PRICHARD, R. K. (21) Genetic variability following selection of Haemonchus contortus with anthelmintics. Trends in Parasitology 17, 445-453 PRICHARD, R. K. & ROULET, A. (27) ABC transporters and beta-tubulin in macrocyclic lactone resistance: prospects for marker development. Parasitology 134, 1123-1132 RENDELL, D. K., RENTSCH, T. E., SMITH, J. M., CHANDLER, D. S. & CALLINAN, A. P. L. (26) Evidence that moxidectin is a greater risk factor than ivermectin in the development of resistance to macrocyclic lactones by Ostertagia spp in sheep in south eastern Australia. New Zealand Veterinary Journal 54, 313-317 SARGISON, N. D., JACKSON, F., BARTLEY, D. J. & MOIR, A. (25) Failure of moxidectin to control benzimidazole-, levamisole- and ivermectin-resistant Teladorsagia circumcincta in a sheep flock. Veterinary Record 156, 15-19 SARGISON, N. D., JACKSON, F., BARTLEY, D. J., WILSON, D. J., STENHOUSE, L. J. & PENNY, C. D. (27) Observations on the emergence of multiple anthelmintic resistance in sheep flocks in the south-east of Scotland. Veterinary Parasitology 145, 65-76 SARGISON, N. D., JACKSON, F. & SCOTT, P. R. (21) Multiple anthelmintic resistance in sheep. Veterinary Record 149, 778-779 SHOOP, W. L., HAINES, H. W., MICHAEL, B. F. & EARY, C. H. (1993) Mutual resistance to avermectins and milbemycins: oral activity of ivermectin and moxidectin against ivermectin-resistant and susceptible nematodes. Veterinary Record 133, 445-447 SILVESTRE, A. & CABARET, J. (22) Mutation in position 167 of isotype1 betatubulin gene of trichostrongylid nematodes: role in benzimidazole resistance? Molecular and Biochemical Parasitology 12, 297- VAN WYK, J. A., CABARET, J. & MICHAEL, L. M. (24) Morphological identification of nematode larvae of small ruminants and cattle simplified. Veterinary Parasitology 119, 277-6 VLA (28) Suspected macrocyclic lactone resistance in common sheep nematodes. Veterinary Record 163, 673-676 VICKERS, M., VENNING, M., MCKENNA, P. B. & MARIADASS, B. (21) Resistance to macrocyclic lactone anthelmintics by Haemonchus contortus and Ostertagia circumcincta in sheep in New Zealand. New Zealand Veterinary Journal 49, 11-15 WILLOUGHBY, L., BATTERHAM, P. & DABORN, P. J. (27) Piperonyl butoxide induces the expression of cytochrome P4 and glutathione S-transferase genes in Drosophila melanogaster. Pest Management Science 63, 3-8 WILSON, D. & SARGISON, N. (27) Anthelmintic resistance in Teladorsagia circumcincta in sheep in the UK. Veterinary Record 161, 535-536 YATES, D. M., PORTILLO, V. & WOLSTENHOLME, A. J. (23) The avermectin receptors of Haemonchus contortus and Caenorhabditis elegans. International Journal for Parasitology 33, 1183-1193 YUE, C., COLES, G. C. & BLAKE, N. (23) Multiresistant nematodes on a Devon farm. Veterinary Record 153, 4 Veterinary Record May 29, 21