PGE in ruminants is usually caused by a mixture of GINs occurring simultaneously [3, 33]. Therefore, the potential implications for studying these parasites (population biology, epidemiology, treatment efficacy etc.) offered by quantitative and multiplex PCR assays seem highly valuable. The advantages are manifold and besides simultaneous detection of multiple parasites present in a sample (use of three different dyes for the six targets means that only one third of the number of reactions must be performed to diagnose all six GINs) include high-throughput capacity, wide dynamic range and avoidance of inaccuracies related to end-point analysis. High-throughput sequencing devices are being increasingly applied to the challenge of GIN identification [34] and complex nematode community studies [35, 36, 37]. This ‘nemabiome’—type technology is still relatively expensive and requires expertise and complicated bioinformatic analysis, so is still the preserve of specialist laboratories. In contrast, real-time PCR technology is very widespread today and became common equipment of larger laboratories, allowing diagnostic procedures based on these platforms to be used relatively routinely. This fact, together with time and laboratory simplicity ensuring immediate results without bioinformatic pipelines, bringing our present multiplex real-time PCR assays for specific detection and semi-quantitative evaluation of the most prevalent GIN infections directly from faeces of domestic livestock closer to the wider scientific community.
Roeber et al. (2012) [17] developed an automated multiplexed-tandem PCR (MT-PCR), with properties very similar to our approach, with the possible exception that the individual species real-time components are carried out in parallel, in separate reactions, rather than simultaneously in the same reaction. The platform, targeting ITS2 regions of six GINs (Haemonchus spp., T. circumcincta, Trichostrongylus spp., Oesophagostomum spp., Ch. ovina, and C. curticei) was validated for European applications[22] and showed promise for routine use in practice. Although they tested direct extraction of DNA out of 0.25 g of faeces using the PowerSoil DNA purification kit (MoBio) as one of the possible methods in the MTPCR study [17], the results showed decreased detection sensitivity which led to decision of the continued use of previously extracted eggs/L3 larvae only. In comparison, we see the advantage of our approach in need for no additional equipment such as handling robot devices and in the usage of faecal samples directly to DNA extraction. The utilization of preextraction steps focused on proper homogenization and sample dissolution, and the usage of specialized kit possessing inhibitor removal technology improved the extraction process and overcame this issue in the case of our study. These provisions ensured a higher sample input, increased diagnostic sensitivity, and reduced the overall time. This was demonstrated by detecting samples' false negative by FEC (possibly in a prepatent stage of infection) and by the absence of PCR inhibition evidenced by IAC amplification.
The parasites for which the multiplex real-time PCR assays are described in this study represent, apart from A. sidemi, the most important GIN of sheep. The oligonucleotides designed for this study show sequence identity with the currently available NCBI GenBank database gene sequences of individual targets (data not shown). Therefore, even though we did not investigate the specific detection of further targeted species of the same genera (namely H. placei, T. davtiani, T. trifurcata, T. vitrinus, and T. rugatus) due to the unavailability of DNA material, it can be assumed that the respective probes will detect even these closely related species for which the present detection systems were intended [14].
Ashworthius detection was included in the presented multiplex assay for the following reasons. This hematophagous abomasal nematode is phylogenetically related and morphologically/morphometrically almost indistinguishable in immature stages from H. contortus, and poses a threat to become one of the most widespread pathogenic GINs of autochthonous European ruminants [29]. This invasive parasite originally endemic in Asiatic deer was probably introduced to Europe by sika deer in the late 19th and early 20th century and since then it has successfully spread among new hosts (such as red deer, roe deer, fallow deer, or moose) and is highly pathogenic in some e.g. European bison [38]. The susceptibility of domestic sheep to ashworthiosis has so far only been confirmed experimentally [39], whereas the first cases of natural infection in cattle were recorded in Poland by Moskwa et al. (2015) [40]. Thus, horizontal transmission of this parasite from wild ruminants to domestic livestock is probably just a matter of time in all regions where these hosts share the same pastures. Thus, having access to a reliable diagnostic tool that can detect A. sidemi in livestock was the main purpose of including this parasite in our multiplex assay.
Detection systems for members in the genus Nematodirus are typically not included in similar studies devoted to strongyle nematodes of livestock due to their distinguishability based on the eggs, which are much larger and contain distinctive dark cells. However, N. battus is an especially important pathogen in temperate climates causing significant health problems in grazing lambs in spring. For this reason, we decided to include this species in the multiplex assay aimed at faecal samples, which could be used as an alternative tool to coproscopy. During the assay’s specificity tests, we noticed a weaker non-specific amplification of T. circumcincta DNA (Table 2) with the Nematodirus system. However, this phenomenon was not recorded during tests on faecal samples, even in those with a high EPG in the Teladorsagia system (Table 6). This anomaly could be caused either by the presence of other Teladorsagia species than T. circumcincta, to which the N. battus probe did not bind, due to sequence variations or it was more likely due to the presence of an excess amount of non-specific DNAs in a complex sample, such as faeces, that might block nonspecific hybridization during the PCR reaction the same way as sheared DNAs [32, 41, 42]; such enhancement of specificity has previously been demonstrated by the usage of competing primers [43] or presence of tRNA [44]. In the present study, Nematodirus spp. was detected only in six faecal samples out of 44 examined by McMaster (Table 5). In five of these samples (No. 32, 41–44) the Nematodirus spp. eggs were identified, while in one (No. 8) N. battus was also present but only judged on a single egg. Sample No. 8 was re-extracted and tested repeatedly by multiplex real-time PCR but always with negative results. This could be as a result either of the misidentification of Nematodirus species based on a single egg (since other species than N. battus would not be detected by our assay, e.g., N. filicollis, N. helvetianus), sampling error (no eggs were present in the faecal sample used for molecular analysis), or because the amount of DNA was below the LOD for this system. The use and suitability of this detection system needs to be further evaluated during a broader study on field samples from sheep flocks.
The LOD determined by assays sensitivity tests on adult nematodes gDNAs (Table 3) cover the range of 0.5–5 pg which equates to a proportion of gDNA extracted from a single egg [11]. These results are consistent with previous studies based on other singleplex [11, 21] and multiplex real-time PCR assays [25]. To refine the data interpretation for faecal samples, the limit of 1 EPG corresponding to 234 plasmid copies was established. In the present study, this corresponded to Cq 35.51 in Haemonchus, Cq 39.75 in Teladorsagia, Cq 36.77 in Trichostrongylus, Cq 39.03 in Nematodirus, Cq 40.00 in Ashworthius, and Cq 35.70 in Chabertia detection system, which all were derived from the median values of a plasmid standard.
Based on IAC, which was amplified in all samples, there was no evidence of inhibition in the PCR following the column extraction of gDNA directly from faeces. Due to recent improvements in commercial faecal extraction kits towards purification of the sample suitable for downstream molecular-based applications, faecal inhibitors themself are no longer of the biggest concern today. Harmon et al. (2007) [18] pointed out that the factor causing the majority of the variance in Cq values between samples containing the same numbers of eggs is partly attributed to the procedure of DNA extraction, when the variation seen from amplified egg DNA exceeded that obtained with plasmids. They also suggested that these variances are corrected in multiplex reactions where intra-sample comparisons are made for each trichostrongyle target [18]. Similarly to Avramenko et al. (2015) [36], we calculated genus specific correction factors to eliminate any sequence representation bias (which could be given for example by the different target sequence features or their copy number in respective genomes). Still, the limit values 0.879–1.0 of their range points to the minimal effect of these parameters on resulting values.
Bott et al. (2009) [11] used a microscopic McMaster approach combined with molecular real-time PCR followed by melting-curve analysis optimized for extracted eggs and suggested that such a platform can be used for the semi-quantitation of target DNAs and infection intensity evaluation based on Cq values of individual targets. The reason for this consideration was that semi-quantitative evaluation together with knowledge of species/genus composition is more valuable than absolute quantification, given that the intensity of nematode infection does not have to reflect the number of eggs excreted per gram of faeces; the exception to this are nematodes with a high biotic potential (e.g., H. contortus and Oesophagostominae) [11]. Based on this knowledge and experience, we conclude that a semi-quantitative enumeration of different GIN eggs in faeces based just on ITS2 H. contortus plasmid control is feasible by our multiplex real-time PCR assays. The results of individual faecal samples (Table 6) and their comparison to FEC suggests with the moderate agreement in the evaluation of the infection severity, that genus Haemonchus is confirmed as the most pathogenic GIN in sheep [45], being widespread and highly prolific, the trigger point for diagnostic evaluation and the selection of follow-up measures, at least under European conditions. Knowledge of the representation of other species in the sample adds further diagnostic value and may influence the choice of subsequent control and treatment procedures; these may then differ from those determined just on the basis of FEC results. Testing of the 44 individual faecal samples revealed three samples, where the presence of nematodes remained undetected by the coproscopy, but this negativity turned out to be false by multiplex real-time PCR. However, this was not noted the other way around, i.e., there were no samples positive in FEC which would test negative by molecular analysis. This indicates that the molecular approach is more sensitive for the detection of eggs directly from faeces than the Concentration McMaster technique, where one egg equates to 20 EPG. These observations are basically in agreement with earlier results by Sweeny et al. (2011) [16]. Nematode composition based on LC proportionally agreed with those resulting from multiplex assays, however, a slight bias can be observed in the example of F1, where Trichostrongylus DNA was not detected but microscopically determined at 8% according to LC results. Since Teladorsagia was the second most abundant genus on this farm, this indicates a confusion of Trichostrongylus and Teladorsagia larvae during microscopical identification of L3. Usually these two genera are grouped together even with an experienced observer [27]. The high sensitivity of the present multiplex realtime PCR, together with the approach of individual animal testing, preventing bias of the result due to animals with aggregated parasites, are the required features for reliable GIN diagnostics.
Although, emphasis must be placed on proper homogenization and mechanical cell disruption of the faeces, there are several advantages of using multiplex real-time PCR assays compared to LC. The most noticeable advantage is the significant shortening in number of steps included in the procedure resulting in reduction of the whole procedure duration from 7 days on cultured L3 [46], 2 days with L1 [14], to only a few hours in the case of the present multiplex assays. Detection of specific DNA also allows GIN diagnostics on frozen faecal samples, where the vast majority of eggs rupture or are otherwise damaged causing gDNA release.
Another indisputable but little known and underestimated advantage of using biological samples directly is the presence of parasite cell-free DNA (cfDNA). CfDNA is released during disintegration of different life stages (cellular apoptosis/necrosis) and active secretion from cells and its presence (and other parasite components) within the faeces cannot be identified by microscopy [47]. The detection of cfDNA is being increasingly applied as an efficient biomarker for the accurate diagnosis of parasites occupying tissues and the circulatory system. The presence of extracellular cfDNA helps to overcome problems connected with intermittent egg shedding and thus greatly increases the sensitivity and accuracy of the semi-quantitative diagnostic assay. Its presence can also reveal the prepatent stage or the hypobiotic larvae (arrested larval development) [11], since cfDNA is released by the larvae present in the wall of the gastrointestinal tract. The theory of cfDNA and its potential to increase assay accuracy might be supported by individual faecal sample results that tested negative by microscopical FEC; in the case of sample No. 25, we assume that a low intensity of infection detected by multiplex assay (total EPG = 44) could adversely affect faecal flotation, which may be subject to a dilution error, but this was presumably not the case in the other two samples. Samples No. 31 and 34 tested negative by faecal flotation technique but highly positive by multiplex real-time PCR [25], with rankings of severe infections. In both samples, high levels of Haemonchus, Teladorsagia and Trichostrongylus were detected. A possible explanation for these results could be a prepatent period of infection with a high representation of developing larval stages. Although less likely scenarios such as some form of hypobiosis or sampling error given, e.g., by intermittent egg shedding, cannot be ruled out; the faeces for both tests were collected at the same time of the day, and we reckon that in such a heavy infection would be at least some eggs present anyway.
No molecular approach for specific DNA detection of GINs in sheep using fresh faeces has yet been implemented in standard diagnosis practice. In our opinion, the two multiplex real-time PCR assays presented in this study may allow reliable evaluation of GIN genus/species richness within European sheep flocks and thus improve diagnosis of GIN infections in ruminant livestock. Presented preliminary evidence of the assays’ ability to rapidly identify and rank nematodes according to their numerical contribution to observed FEC in mixed infections may also suggest the potential to become a powerful alternative or practical adjunct to conventional faecal egg count reduction tests (FECRT) to enable the rapid inference of which strongyle species or genera are susceptible or resistant to particular anthelmintic drugs [12, 16]. However, further evaluation and refinements of these assays needs to be carried out during further testing on real samples to refine its precision; then it could usefully supplement or replace the conventional coprological techniques for the purpose of routine diagnosis.