Evaluation of changes in drug susceptibility and population genetic structure in Haemonchus contortus following worm replacement as a means to reverse the impact of multiple-anthelmintic resistance on a sheep farm.

A population of Haemonchus contortus that was highly resistant to benzimidazoles and avermectin/milbemycins with a subpopulation that was resistant to levamisole, was replaced with a susceptible laboratory isolate of H. contortus in a flock of sheep. The anthelmintic susceptibility and population genetics of the newly established population were evaluated for 3.5 years using in vivo, in vitro, and molecular methods. Successful replacement of the resistant population with a susceptible population was confirmed using phenotypic and genotypic measurements; larval development assay indicated full anthelmintic susceptibility; albendazole treatment yielded 98.7% fecal egg count reduction; pyrosequence genotyping of single nucleotide polymorphisms in positions 167 and 200 of the isotype-1 beta tubulin gene were present at 0.0 and 1.7%, respectively; microsatellite genotyping indicated the background haplotype was similar to the susceptible isolate; and haplotypes of the isotype-1 beta tubulin gene were similar to the susceptible isolate. To sustain the susceptibility of the new population, targeted selective treatment was implemented using albendazole. Surprisingly, within 1.5 years post-replacement, the population reverted to a resistant phenotype. Resistance to albendazole, ivermectin, and moxidectin was confirmed via fecal egg count reduction test, larval development assay, and pyrosequencing-based genotyping. Targeted selective treatment was then carried out using levamisole. However, within one year, resistance was detected to levamisole. Population genetics demonstrated a gradual change in the genetic structure of the population until the final population was similar to the initial resistant population. Genetic analyses showed a lack of diversity in the susceptible isolate, suggesting the susceptible isolate had reduced environmental fitness compared to the resistant population, providing a possible explanation for the rapid reversion to resistance. This work demonstrates the power of combining molecular, in vitro, and in vivo assays to study phenotypic and genotypic changes in a field population of nematodes, enabling improved insights into the epidemiology of anthelmintic resistance.

Using population genetics to examine relationships of Dirofilaria immitis based on both macrocyclic lactone-resistance status and geography.

Prevention of infection with canine heartworm (Dirofilaria immitis) is based on the compliant administration of macrocyclic lactone (ML) drugs. Resistance to ML drugs is well documented in D. immitis; however, there remains a paucity of information on the spatial distribution and prevalence of resistant isolates. This project aims to improve understanding of ML-resistance by using a population genetic approach. We developed a large panel of microsatellite loci and identified 12 novel highly polymorphic markers. These 12, and five previously published markers were used to screen pools of microfilariae from 16 confirmed drug-susceptible, 25 confirmed drug-resistant, and from 10 suspected drug-resistant field isolates. In isolates where microfilarial suppression testing indicated resistance, Spatial Principal Component Analysis (sPCoA), Neighbor Joining Trees and Bayesian clustering all revealed high genetic similarity between pre- and post-treatment samples. Somewhat surprisingly, the Neighbor Joining tree and sPCoA generated using pairwise Nei's distances did not reveal clustering for resistant isolates, nor did it reveal state-level geographic clustering from samples collected in Georgia, Louisiana or Mississippi. In contrast, Discriminant Analysis of Principle Components was able to discriminate between susceptible, suspected-resistant and resistant samples. However, no resistance-associated markers were detected, and this clustering was driven by the combined effects of multiple alleles across multiple loci. Additionally, we measured unexpectedly large genetic distances between different passages of laboratory strains that originated from the same source infection. This finding strongly suggests that the genetic makeup of laboratory isolates can change substantially with each passage, likely due to genetic bottlenecking. Taken together, these data suggest greater than expected genetic variability in the resistant isolates, and in D. immitis overall. Our results also suggest that microsatellite genotyping lacks the sensitivity to detect a specific genetic signature for resistance. Future investigations using genomic analyses will be required to elucidate the genetic relationships of ML-resistant isolates.

Modelling anthelmintic resistance by extending eggCounts package to allow individual efficacy.

The same anthelmintic treatment can have variable efficacy on individual animals even if the parasite population is homogenously susceptible. An extension of the R package eggCounts is proposed to take individual efficacy into account using a Bayesian hierarchical model. A simulation study is conducted to compare the performance of five different methods on estimating faecal egg count reduction and its uncertainty interval. Simulation results showed the individual efficacy model offered robust inference to two different data simulation procedures with low root mean squared error on the reduction estimate and appropriate uncertainty estimates. Different methods were used to evaluate the anthelmintic resistance in a dataset from USA with sheep and cattle faecal egg counts, where a strong anthelmintic resistance was detected. Open-source statistical tools were updated to include the proposed model.

Comparison of fecal egg counting methods in four livestock species.

Gastrointestinal nematode parasites are important pathogens of all domesticated livestock species. Fecal egg counts (FEC) are routinely used for evaluating anthelmintic efficacy and for making targeted anthelmintic treatment decisions. Numerous FEC techniques exist and vary in precision and accuracy. These performance characteristics are especially important when performing fecal egg count reduction tests (FECRT). The objective of this study was to compare the accuracy and precision of three commonly used FEC methods and determine if differences existed among livestock species. In this study, we evaluated the modified-Wisconsin, 3-chamber (high-sensitivity) McMaster, and Mini-FLOTAC methods in cattle, sheep, horses, and llamas in three phases. In the first phase, we performed an egg-spiking study to assess the egg recovery rate and accuracy of the different FEC methods. In the second phase, we examined clinical samples from four different livestock species and completed multiple replicate FEC using each method. In the last phase, we assessed the cheesecloth straining step as a potential source of egg loss. In the egg-spiking study, the Mini-FLOTAC recovered 70.9% of the eggs, which was significantly higher than either the McMaster (P = 0.002) or Wisconsin (P = 0.002). In the clinical samples from ruminants, Mini-FLOTAC consistently yielded the highest EPG, revealing a significantly higher level of egg recovery (P < 0.0001). For horses and llamas, both McMaster and Mini-FLOTAC yielded significantly higher EPG than Wisconsin (P < 0.0001, P < 0.0001, P < 0.001, and P = 0.024). Mini-FLOTAC was the most accurate method and was the most precise test for both species of ruminants. The Wisconsin method was the most precise for horses and McMaster was more precise for llama samples. We compared the Wisconsin and Mini-FLOTAC methods using a modified technique where both methods were performed using either the Mini-FLOTAC sieve or cheesecloth. The differences in the estimated mean EPG on log scale between the Wisconsin and mini-FLOTAC methods when cheesecloth was used (P < 0.0001) and when cheesecloth was excluded (P < 0.0001) were significant, providing strong evidence that the straining step is an important source of error. The high accuracy and precision demonstrated in this study for the Mini-FLOTAC, suggest that this method can be recommended for routine use in all host species. The benefits of Mini-FLOTAC will be especially relevant when high accuracy is important, such as when performing FECRT.

Motility in the L3 stage is a poor phenotype for detecting and measuring resistance to avermectin/milbemycin drugs in gastrointestinal nematodes of livestock.

Motility is a commonly used in vitro phenotype for assessing anthelmintic activity of candidate compounds, and for detecting anthelmintic resistance in nematodes. Third-stage larvae (L3) of parasitic nematodes are commonly used in motility-based assays because L3 are simple to obtain and can remain viable in storage for extended periods. To improve the measurement of motility of microscopic stages of nematodes, our laboratory developed the Worminator, which quantitatively measures motility of parasites. Using the Worminator, we compared the dose-response characteristics of several avermectin/milbemycin (AM) compounds using L3 from both AM-susceptible and AM-resistant Cooperia spp. (abamectin, doramectin, eprinomectin, ivermectin, moxidectin) and Haemonchus contortus (eprinomectin, ivermectin, moxidectin). Concentrations tested with the Worminator ranged from 0.156 to 40 µM. Differences in EC50 between AM-susceptible and AM-resistant isolates of Cooperia spp. and Haemonchus contortus were small, with resistance ratios ranging from 1.00 to 1.34 for Cooperia spp., 0.99 to 1.65 for Haemonchus contortus. Larval migration inhibition assays were conducted using the same isolates and were equally ineffective for detection of resistance with resistance ratios less than 2.0. These results contrast with those of the Larval Development Assay where we obtained a resistance ratio of 16.48 using the same isolates of Haemonchus contortus. Moreover, even at the highest concentration tested (40 µM), 100% inhibition of motility was never achieved and EC50 for Worminator assays were more than 100× higher than peak plasma levels achieved in vivo following treatment. These data demonstrate that dose-response characteristics for inhibition of motility in L3 of gastrointestinal nematodes of livestock do not significantly differ for AM-susceptible and AM-resistant isolates. These data challenge the suitability of motility as a phenotype for detecting and measuring resistance to AM drugs in gastrointestinal nematodes of livestock.

Utilization of composite fecal samples for detection of anthelmintic resistance in gastrointestinal nematodes of cattle.

Recent reports indicate that anthelmintic resistance in gastrointestinal nematodes of cattle is becoming increasingly prevalent worldwide. Presently, the fecal egg count reduction test (FECRT) is the only means available for detection of resistance to anthelmintics in cattle herds at the farm level. However, the FECRT is labor and cost intensive, and consequently is only rarely performed on cattle farms unless for research purposes. If costs could be reduced, cattle producers might be more likely to pursue drug resistance testing on their farms. One approach to reducing the cost of the FECRT, is the use of composite fecal samples for performing fecal egg counts (FEC), rather than conducting FEC on fecal samples from 15 to 20 individual animals. In this study FECRT were performed on 14 groups of cattle using both individual and composite FEC methods To measure how well the results of composite sampling reproduce those of individual sampling, Lin's Concordance Correlation Coefficient was utilized to describe both the linear relationship between methods and the slope and y-intercept of the line relating the data sets. There was little difference between the approaches with 98% agreement in mean FEC found between methods Mean FEC based on individual counts ranged between 0 and 670.6 eggs per gram of feces, indicating that the results of this study are applicable to a wide range of FEC levels. Standard error of the mean FEC and range of FEC are reported for each group prior to and following treatment to describe the variability of the data set. There was greater than 95% agreement in drug efficacy between individual and composite sampling methods, demonstrating composite sampling is appropriate to evaluate drug efficacy. Notably, for all groups tested the efficacy calculated by composite sampling was within the 95% confidence interval for efficacy calculated using individual sampling. The use of composite samples was shown to reduce the number of FEC required by 79%. These data demonstrate that pooling fecal samples from a group of cattle and then performing repeated FEC on that composite sample yields results very similar to performing individual FEC on those same animals, while substantially reducing the cost of performing a FECRT as compared to individual fecal samples. Furthermore, we have developed suggested methods for using composite samples in a FECRT, provided a cost comparison for this methodology, and described potential issues associated with the use of composite samples that must be considered.