INTRODUCTION
Anthelmintic resistance has become an increasingly serious problem for helminth parasite control in nematode parasites of livestock and horses throughout the world (Kaplan, Reference Kaplan2004; Wolstenholme et al. Reference Wolstenholme, Fairweather, Prichard, von Samson-Himmelstjerna and Sangster2004). Anthelmintic resistance has also appeared, or seems likely, in some nematode parasites of humans (De Clerq et al. Reference De Clercq, Sacko, Behnke, Gilbert, Dorny and Vercruysse1997; Albonico et al. Reference Albonico, Bickle, Ramsan, Montresor, Savioli and Taylor2003, Reference Albonico, Engels and Savioli2004; Osei-Atweneboana et al. Reference Osei-Atweneboana, Eng, Boakye, Gyapong and Prichardin press). In addition to nematode parasites, anthelmintic resistance has been reported in trematode parasites of animals and humans (Fenwick and Webster, Reference Fenwick and Webster2006; Brennan et al. Reference Brennan, Fairweather, Trudgett, Hoey, McCoy, McConville, Meaney, Robinson, McFerran, Ryan, Lanusse, Mottier, Alvarez, Solana, Virkel and Brophy2007). Available biological methods for assessing the level and extent of anthelmintic resistance in nematode parasites of livestock have been reviewed (Coles et al. Reference Coles, Jackson, Pomroy, Prichard, von Samson-Himmelstjerna, Silvestre, Taylor and Vercruysse2006). Many of the biological methods for measuring anthelmintic resistance in animals are not readily available for assessing the situation in parasites of humans, for a variety of reasons. For instance, the available biological assays have limitations in sensitivity and cost, and frequently do not provide a rapid output.
Alternative approaches are thus needed. Drug resistance has a genetic basis and it should be possible, provided we understand enough about the mechanisms of drug resistance or have reliable markers that are genetically linked to resistance mechanisms, to develop DNA or other molecular markers which could be used for resistance assessment. The desirability, feasibility and possible methods of using DNA markers for anthelmintic resistance have been outlined (Prichard, Reference Prichard2001; von Samson-Himmelstjerna and Blackhall, Reference von Samson-Himmelstjerna and Blackhall2005). However, the task of developing panels of anthelmintic resistance markers for different drugs, for all of the economically or socially important helminth species in animals and humans, is a formidable task. This is particularly so because we do not fully understand how different anthelmintics work and have an even poorer understanding of the mechanisms of resistance to most anthelmintics in different helminth species. In order to increase the rate of progress for developing reliable molecular markers for anthelmintic resistance, von Samson-Himmelstjerna and Blackhall (Reference von Samson-Himmelstjerna and Blackhall2005) proposed that an international consortium be formed to foster a collaborative effort amongst scientists and others interested in maintaining parasite control in animals and humans to undertake the research needed to develop the desired panels of markers. Arising from this call, CARS, the Consortium for Anthelmintic Resistance SNPs (single nucleotide polymorphisms), was formed in October, 2005 and held its first workshop in August, 2006.
THE INITIAL CHALLENGE AND OPPORTUNITIES
Limitations of resources will force us to keep expectations modest and realistic and perhaps to choose amongst the following options for resistance marker development: either a panel of markers for multiple drug classes in one species, or a panel for one drug class in multiple species. If resistance mechanisms are reasonably well understood (e.g. benzimidazole resistance; see von Samson-Himmelstjerna et al. in this special issue), a panel of SNPs for a few major species is feasible. Since mechanisms of resistance to other anthelmintics are less well understood, a panel of SNPs for one parasite species may be a more realistic goal. Each option has its pros and cons. The former would be more feasible technically, but the latter would have a broader global utility. Many of the loci that might be targeted for SNP discovery and association surveys are currently known only from Haemonchus contortus. Finding these loci in a range of other species could take significant effort, but some of this work is already underway.
The goal of establishing a panel of resistance markers is to be able to detect the presence of anthelmintic resistance in field populations of parasites. To do this, SNPs on the panel should preferably, but not necessarily, be mechanistically associated with the resistance phenotype. To find associated and informative SNPs we need candidate genes and experimental access to both affected and unaffected populations. The number of such SNPs needed from each targeted locus will depend on the genetic structure of the populations used. For example, does a species have a single global population, considering the amount of gene flow that occurs even between continents? Will any particular SNP associated with resistance in Australia also be associated with resistance in Europe, or even be present in Europe at a detectable frequency? How much linkage disequilibrium (LD – the non random association of loci) exists in parasite species? We know very little about the population and genetic structure of any parasitic helminth. Will these limitations impact our choice for a SNP panel?
Association studies require the availability of two populations, one that exhibits a particular phenotype and one that does not. This requirement will limit our choices of species and drug classes. Finding informative SNPs associated with resistance would be difficult in a species/drug class combination for which resistance is not well developed in at least some populations from different parts of the world. Being able to detect the first appearance of resistance in a species would be ideal, and might be possible with sufficient ‘blind’ (noninformative) SNPs distributed evenly across candidate loci. Parasitic nematodes, however, generally appear to possess an extraordinary amount of genetic variation, possibly reflecting their large population sizes. LD may then be correspondingly low across the genome. A high density of SNPs can overcome this problem, but a useful panel must necessarily restrict the total number of SNPs. Whether to use natural populations and/or artificially generated strains for SNP discovery is also a question that must be addressed. Many laboratory strains and field isolates already exist, but others might need development before SNP discovery and characterization could be addressed. These strains and isolates are very useful for identifying candidate loci, but they are laborious to produce, impacting on how many species could be included in a potential one-drug-class/multiple-species panel. A database of available populations would be very valuable.
Another issue that needs to be addressed early is the choice of candidate genes to include in the preliminary search for informative SNPs. Taking macrocyclic lactones (MLs) as an example, which loci should be included? Various studies point to multiple mechanisms of ML resistance even within a single species. The process of SNP discovery and characterization will help clarify which loci actually contribute to resistance, or at least which loci are associated with resistance, but we lack the resources to survey every candidate locus, particularly if the gene has not yet been isolated. Also, different species may have different mechanisms of ML resistance, leading to the inclusion of different candidate loci on a multiple-species panel. And then there is the P-glycoprotein problem (see Prichard and Roulet, in this special issue) – so many genes in the family to find and choose from, an uncertain prevalence of increased copy number, and the need to find SNPs associated with increased copy number?
Most researchers investigating the development of SNP markers for anthelmintic resistance work with H. contortus. It is the most studied trichostrongylid species of ruminants and a genome sequence project is underway for it. Potential resistance markers developed in H. contortus could thus be tested for possible application in other parasites of livestock. From this perspective, combined with its broader utility, a panel for one drug class/multiple species could be more appealing. Questions to be considered in this context include: how many and which species should be targeted? Which drug class(es), and which loci should receive priority? What populations are available for SNP discovery, and how many SNPs would be manageable? Would individual panels for each species (for one drug class) be better than one universal panel for all species, without significantly increasing the amount of work?
Also, a variety of technical issues will arise during the developmental of SNP panels. These issues include choosing methodologies for SNP discovery, determination of allele-frequencies, species specificity of PCR amplification, point mutations versus changes in copy number (e.g. P-glycoproteins), the complexity of multigene families, SNP selection and the association of SNPs with functional effects, working with pooled DNA samples, multiplexing analyses, genotyping platforms, and selection of appropriate techniques for studying populations.
The amount of significant information that could come from this kind of project is potentially large: gene discovery and characterization, identification of loci associated with resistance in different species/populations, markers of resistance for epidemiological studies and parasite management, basic population genetics (neglected in research on helminths), discovery of means to overcome, in part, anthelmintic resistance, identification of potential new targets for chemotherapeutic drug or vaccine discovery, and much more.
FIRST CARS WORKSHOP
The first CARS Workshop was held in Glasgow prior to the ICOPA XI Congress and was attended by scientists from around the world involved or interested in developing molecular markers for anthelmintic resistance in important helminth parasites of animals and humans. The objectives were to: (1) Bring together scientists interested in collaborative research on this topic; (2) Promote the formulation of collaborative research proposals; (3) Help identify research topics suitable for collaborative research on anthelmintic resistance mechanisms and marker; (4) Support funding applications/apply for funding on anthelmintic resistance markers; (5) Help coordinate SNP related research activities in anthelmintic resistance; (6) Establish a database of relevant nematode/helminth populations available globally; and (7) Organize future meetings of interested scientists, industry partners, potential user groups and sponsors.
The workshop considered progress in helminth genome projects and how the new genome information can be used to find relevant SNPs for anthelmintic resistance (see Gilleard and Beech, in this special issue). This was followed by a consideration of alternative strategies for developing panels of anthelmintic resistance markers (see von Samson-Himmelstjerna et al. in this special issue), and a consideration of bioinformatics resources and methods useful in developing panels of drug resistance markers. Presentations on the major different anthelmintic classes in which resistance occurs in nematodes followed. SNP markers for benzimidazole resistance in veterinary nematodes are considered by von Samson-Himmelstjerna et al. (in this special issue), while the need for and progress in developing markers for benzimidazole resistance in human parasitic nematodes is also addressed (see Prichard, in this special issue). Attention is then given to levamisole/pyrantel resistance, by first considering the mode of action of levamisole and pyrantel and how this could help in understanding levamisole and pyrantel resistance (see Martin and Robertson, in this special issue). This has been complimented by new data on the identification of levamisole resistance markers in H. contortus using a cDNA-AFLP approach (see Neveu et al. in this special issue). It must be recognized that developing markers for resistance to macrocyclic lactone (ML) anthelmintics is complex, as we do not fully understand either the full mechanism of action of the MLs or the range of resistance mechanisms to MLs in nematodes. The MLs exert powerful effects at ligand-gated chloride channels and an appraisal of the involvement of these channels in ML resistance and prospects for developing molecular markers based on them has been reviewed (see McCavera et al., in this special issue). However, in addition to the possible involvement of ligand-gated chloride channels in ML resistance, other mechanisms and potential markers, such as ABC transporters and β-tubulin, may also be involved (see Prichard and Roulet, in this special issue).
The CARS Consortium established a number of Work Groups to coordinate efforts with the different anthelmintic groups, to collect information on parasite isolates that may be useful for this work, to evaluate genetic methods and resources that could validate or facilitate the effort to develop genetic markers for anthelmintic resistance, and to increase public access to the work of the Consortium, hold future workshops and plan group research applications. Further information and contact addresses can be found at http://consortium.mine.nu/cars/. The effort to develop molecular markers for anthelmintic resistance in important parasites of animals and humans will be continued by the CARS initiative, and be integrated with efforts to develop new anthelmintics for use in humans, recently launched as the Helminth Initiative by the World Health Organization.
ACKNOWLEDGEMENTS
We thank Pfizer Animal Health, Fort Dodge Animal Health, Bayer HealthCare AG, Novartis, MLA, Australia and Merial for financial support and McGill University, Stiftung Tierärztliche Hochschule, Hannover, and the University of Glasgow for other support.
