Molecular identification of four Sarcocystis species in cattle from Lithuania, including S. hominis, and development of a rapid molecular detection method

Background Six Sarcocystis species are known to use cattle (Bos taurus) as the intermediate host, two of which, S. hominis and S. heydorni, are zoonotic. There is a need for a method that will enable rapid identification of the Sarcocystis species in cattle. Methods The diaphragm muscles of 102 cattle from Lithuania were examined for the presence of Sarcocystis spp., using two different methods for species identification. Individual sarcocysts were isolated from squash preparations of the diaphragm muscle under the light microscope, followed by genetic characterisation of excised cysts using sequence analysis of the 18S rRNA (18S rRNA) and cytochrome c oxidase subunit I (cox1) genes. The same cattle muscle samples were digested and species-specific PCR analyses targeting cox1 were developed to identify the Sarcocystis isolates to the species level. Results Under the light microscope, sarcocysts were detected in 87.3% of animals, and Sarcocystis infection was verified in all digested samples. Three species, namely S. cruzi (n = 20), S. bovifelis (n = 23) and S. hirsuta (n = 6), were identified by DNA sequence analysis of isolated sarcocysts. Based on sequence analysis of cox1, the level of genetic variability depended on Sarcocystis species and geographical location. Four Sarcocystis species, S. cruzi (96.1%), S. bovifelis (71.6%), S. hirsuta (30.4%) and S. hominis (13.7%), were confirmed in the digested samples. In individual samples, the most common finding was two species of Sarcocystis (44.1%), followed by three species (26.5%), a single species (24.5%) and four species (4.9%). Conclusions Although examination of tissue preparations under the light microscrope did not detect any sarcocysts belonging to S. hominis, this species was identified in the digested samples subjected to a cox1-specific PCR analysis. These results demonstrate the need for effective molecular diagnosis techniques to detect Sarcocystis spp., which may be present at a lower prevalence and not detectable among the limited number of sarcocysts identified individually under the light microscope.

life-cycle and endogenous sporulation. These parasites form sarcocysts mainly in the muscle tissues of the intermediate hosts and develop oocysts and sporocysts in the small intestine of the definitive hosts [1].
Three Sarcocystis species, S. suihominis, S. hominis and S. heydorni, are known to infect humans as definitive hosts [2][3][4]. The latter two species can be found in the muscles of cattle (Bos taurus), while humans may become infected by consuming raw/undercooked beef containing mature sarcocysts of S. hominis and S. heydorni. There has been much debate about the validity and classification of several Sarcocystis species from cattle, as well as their specificity to the intermediate host [5][6][7][8]. Current consensus is that that cattle may harbour up to six Sarcocystis species, namely S. bovifelis, S. bovini, S. cruzi, S. heydorni, S. hirsuta and S. hominis [4,[8][9][10][11]. As morphological analysis is not commonly sufficient to identify Sarcocystis species in cattle [12][13][14], the need for molecular methods is increasing. Sarcocystis spp. from cattle have been molecularly characterised at the genes for 18S rRNA (18S rRNA), 28S rRNA (28S rRNA) and cytochrome c oxidase subunit I (cox1) and at nuclear rDNA internal transcribed spacer 1 (ITS1) [8,10,11,15]. Databases contain many 18S rRNA gene sequences of the genus Sarcocystis used for species identification [16][17][18]. However, studies have revealed that cox1 is the preferable target to differentiate taxonomically related Sarcocystis spp. whose intermediate hosts are such ruminants as cattle, sheep, goats, deer and others [19].
In Lithuania, the prevalence of Sarcocystis infection ranges from 44.9 to 98.1%, and infection has been recorded in six different muscle types of cattle [20]. However, the parasite species in the area under investigation have not yet been identified. In the study reported here, we isolated individual sarcocysts from squash preparations of muscle and then characterised the sarcocysts based on sequence analysis of 18S rRNA and cox1. We also report out development of a diagnostic technique to identify Sarcocystis species in cattle based on tissue digestion and PCR analysis using species-specific primers targeting cox1.

Samples and microscopic examination of sarcocysts
In 2017-2018, the diaphragm muscles of 102 cattle were collected from farmers and meat processing plants from all over Lithuania. The prevalence and density of Sarcocystis infection were evaluated in methylene blue-stained muscle specimens by counting sarcocysts per gram of sample according to Kirillova et al. [21]. The morphological analysis of sarcocysts was performed in freshly squashed preparations as described by Prakas et al. [22]. Sarcocysts were isolated from muscle fibres under a Nikon ECLIPSE 80i light microscope (Nikon Corp., Tokyo, Japan), at 40× or 100× magnification, and morphologically differentiated according to size and shape of sarcocysts, as well as the structure of the sarcocyst wall. Overall, 49 sarcocysts excised from the muscle tissues were separately preserved in 70% ethanol for molecular analysis.

Molecular analysis of sarcocysts
Genomic DNA was extracted from individual sarcocysts using the QIAamp ® DNA Micro Kit (Qiagen, Hilden, Germany) tissue protocol according to the manufacturer's recommendations. Sarcocystis species were identified by sequence analysis of 18S rRNA and cox1. The SarCF/ SarDR primer pair was applied for 18S rDNA sequence amplification [23], while the SF1/SR9 primer pair was used to amplify cox1 sequences [19,24]. The PCR conditions and product evaluation were as described previously [25]. The samples were sequenced directly with the 3500 Genetic Analyzer (Applied Biosystems, Thermo Fisher Scientific, Foster City, CA, USA) using the same forward and reverse primers as for the PCR. The obtained sequences were compared with those of various Sarcocystis spp. by Nucleotide BLAST (megablast option). A total of 213 cox1 and 126 18S rRNA sequences were selected and aligned using the MUSCLE algorithm [26] included in the MEGA7 software [27]. TOPALi v2.5 software [28] was used to select a nucleotide substitution model with the best fit to the aligned sequence datasets and to construct phylogenetic trees under Bayesian inference. The intraspecific genetic variation indices, including the number of haplotypes (h), the number of segregating sites (S), haplotype diversity (Hd) and nucleotide diversity (π), were calculated for whole dataset and separate samples using DnaSP v6 [29]. Genetic differentiation for Sarcocystis spp. sample pairs was assessed with Φ ST (i.e. the proportion of genetic diversity due to differences among populations) based on the Tamura-Nei distance [30] with Arlequin v. 3.5.2.2 [31]. The statistical significance of Φ ST was tested by 10,000 permutations and at the 0.05 confidence level.

Trypsin digestion
Digestion of cattle diaphragm muscles was performed according to the optimised protocols of Verma et al. [32] and Chiesa et al. [33]. Following trypsin digestion of 20-g samples each in 50 ml of digestion solution (trypsin 1:250, 2.5 g/l phosphate buffered saline [PBS]) for 16 h at 37 °C in an incubator with stirring, the suspension was centrifuged for 5 min at 7000 rpm, following which the pellet was resuspended in 5 ml PBS and centrifuged for 3 min at 5000 rpm. The resulting pellet was resuspended in 5 ml PBS, and 300 µl of the suspension was used for DNA extraction.

Molecular analysis of digested samples
Genomic DNA was extracted from the digested samples using the GeneJet Genomic DNA Purification Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania) according to the manufacturer's protocol. The resulting DNA samples were kept frozen at − 20 °C until used as templates for PCR amplification of cox1.
Primers targeting cox1 were designed with the help of Primer3Plus [34]. The primers designed for S. cruzi, S. bovifelis, S. bovini, S. heydorni, S. hirsuta and S. hominis are listed in Table 1. Each PCR was performed in a final volume of 25 μl containing 0.5 μM of each primer, dNTP mix (0.2 mM of each), 10× DreamTaq buffer with 20 mM MgCl 2 , 1.25 U DreamTaq polymerase (Thermo Fisher Scientific Baltics), 1 μg genomic DNA and nuclease-free water. The cycling conditions began with one cycle at 95 °C for 5 min, followed by 40 cycles at 94 °C for 45 s, 56 °C for 45 s and 72 °C for 45 s, and ending with one cycle at 72 °C for 10 min. PCR products were visualised by 1% agarose gel electrophoresis. The selected PCR products were purified using the GeneJET PCR Purification Kit (Thermo Fisher Scientific Baltics) according to the manufacturer's recommendations. Sequencing of the purified PCR fragments was performed, and the obtained sequences were compared by BLAST analysis. Differences in the prevalence of the identified Sarcocystis species were evaluated using Chi-squared test.

Morphological examination of sarcocysts
Examination of methylene blue-stained squash preparations of diaphragm muscles from cattle reared in Lithuania revealed the presence of sarcocysts in 87.3% (89/102) of the samples collected. The number of sarcocysts per gram of muscle ranged from one to 79 (mean 10.5/g muscle, median 5.0/g muscle). Three types of sarcocysts were detected under the light microscope. Type I sarcocysts were thread-shaped, about 800 × 75 μm in size (n = 58) and had thin 3-µm-long hair-like protrusions; these sarcocysts were characterised as being S. cruzi-like. Type II sarcocysts were spindle-or fusiform-shaped, about 1600 × 130 μm in size (n = 40) and had 5-to 6-µm-long finger-like protrusions; these cysts were characterised as being S. bovifelis-and S. bovini-like. Type III sarcocysts were thread-shaped, about 2600 × 500 µm in size (n = 2) and had about 7-µm-long finger-like protrusions; these cysts were considered as being S. hirsuta-like.
The molecular analysis revealed that cysts belonging to types I-III were sarcocysts of S. cruzi (n = 20), S. bovifelis (n = 23) and S. hirsuta (n = 6), respectively. In some cases, sarcocysts morphologically assigned to S. hirsuta were identified as S. bovifelis or vice versa. In two cases, fragments of sarcocysts appeared to be smooth; however, based on molecular methods these cysts were assigned to S. cruzi.

Molecular results on isolated sarcocysts
The Phylogenetic analysis showed that the 18S rRNA fragment used in this study was suitable for the discrimination of Sarcocystis spp. forming cysts in the muscles of cattle (Fig. 1a). It should be noted that four species (S. bovifelis, S. bovini, S. hominis and S. hirsuta) characterised on the basis of sarcocysts with finger-like protrusions [8] were grouped into one cluster. The same topology was shown by the phylogenetic tree generated by sequence analysis of cox1 (Fig. 1b).
The genetic comparison of cox1 sequences of S. bovifelis, S. cruzi and S. hirsuta obtained in the present study are presented in Table 2. The highest sequence differences were found within S. cruzi. The 18S rRNA region used in this study was characterised by minimal intraspecific genetic variability among the analysed Sarcocystis species. Therefore, S. bovifelis, S. cruzi and S. hirsuta from different geographical areas were compared only for cox1. The highest overall intraspecific genetic variability was detected in S. cruzi, while a similar level of overall genetic variation was calculated in S. bovifelis and S. hirsuta (Table 3). Differences in the level of genetic variation were observed between samples of the same species. Relatively low variation was established for S. bovifelis from Lithuania and S. hirsuta from New Zealand, and relatively high variation was noticed for S. cruzi from China. The overall Φ ST values showing genetic differentiation between populations were high for S. hirsuta (Φ ST = 0.401, P < 0.001) and S. cruzi (Φ ST = 0.255, P < 0.001) and medium for S. bovifelis (Φ ST = 0.141, P < 0.01).

Molecular results of digested samples
Cox1 fragments amplified using the GaBfEF/GaB-fER, GaCrEF/GaCrER, GaHiEF/GaHiER and GaHoEF/ GaHoER primer pairs were examined by DNA electrophoresis. To ascertain whether the used primers were species-specific, three PCR products amplified with each primer pair were purified and sequenced. Primer-annealing sequences were not included in the BLAST analysis. The sequence identity values are given in Table 2. DNA of S. bovifelis, S. cruzi and S. hirsuta extracted from individual sarcocysts were further used as positive or negative controls in the analysis of the digested samples. The control DNA sample of S. hominis was isolated from the muscular tissue of cattle in Italy [11] and subjected to 18S rRNA amplification and sequencing using the SarAF/ SarBR and SarCF/SarDR primer pairs [23]. The obtained 1796-bp-long 18S rRNA sequence (MT792481) showed 99.4-99.7% similarity with S. hominis (JX679470-JX679471, KF954731) and up to 97.3% similarity with other Sarcocystis spp.
Sarcocystis spp. were detected in all 102 beef diaphragm samples examined using the newly developed molecular identification technique based on muscle digestion and species-specific PCR analysis targeting cox1. The Lithuanian cattle analysed in this study were most frequently infected with S. cruzi, which was identified in 98 samples (96.1%) (Fig. 2a). The prevalence of S. cruzi in the analysed cattle was statistically significantly higher than that of S. bovifelis (71.6%; χ 2 = 22.59, P < 0.0001), the presence of which was confirmed in 73 samples. The third most common species was S. hirsuta (30.4%), detected in 31 samples. The prevalence of S. hirsuta was statistically significantly lower than of S. bovifelis (χ 2 = 34.60, P < 0.0001) and higher than that of S. hominis (13.7%; χ 2 = 8.24, P < 0.01), which was detected in 14 samples. Although the molecular technique did not identify S. hominis among the individual sarcocysts excised from the diaphragm samples, it was identified in the digested samples. These results show that individual cattle may be infected up to four species of Sarcocystis at any one time (Fig. 2b). The most common finding was the presence of two species of Sarcocystis (45 samples, 44.1%); three species were detected in 27 samples (26.5%), a single species was identified in 25 samples (24.5%) and four species were identified in five samples (4.9%). In cases of a single species being detected, S. cruzi was identified in 21 samples and S. bovifelis was confirmed in four samples. When two species of Sarcocystis were observed in the

Discussion
In the present study sarcocysts isolated from the diaphragm muscles of cattle were attributed to three morphological types, with molecular methods identifying the types as S. cruzi, S. bovifelis and S. hirsuta. Of the six Sarcocystis species known to use cattle as an intermediate host, two species, S. cruzi and S. heydorni, can be relatively easily differentiated by morphological characteristics [4]. Sarcocysts of S. cruzi are distinguished by   [4,8,9,35]. During the course of this study, morphological examination of fragments of two sarocysts released from muscle fibres appeared to be smooth and to be similar to S. heydorni; however, based on molecular methods, these were identified as S. cruzi. In these cases, the protrusions of S. cruzi were likely torn off during manipulation of the sarcocysts with preparation needles. Moreover, sarococyts belonging to S. bovifelis were sometimes misidentified as S. hirsuta and vice versa. Sarcocysts of S. hominis were not found during the light microscopy study, but this species was confirmed in 13.7% of digested samples using molecular methods. These results suggest that the isolation of individual sarcocysts is ineffective when the prevalence of infection of the species is low and, therefore, that the digestion protocol may be preferable in epidemiological studies. The 18S rRNA fragment used in the present study showed sufficient variability for the discrimination of Sarcocystis spp. found in cattle (Fig. 1a). Studies involving population genetic research on Sarcocystis species from cattle are scarce. One study demonstrated a low genetic variability among various isolates of S. cruzi from different geographical regions [36]. In the present study, analysis of cox1 sequences implied that the level of genetic variability depended on the Sarcocystis species and geographical area. Also, a relatively high genetic differentiation between the samples of compared species was revealed.
This study established that the GaBfEF/GaBfER, GaCrEF/GaCrER, GaHiEF/GaHiER and GaHoEF/  (Fig 2a). The prevalence and abundance of Sarcocystis species may depend on the type of the muscle being analysed. For example, S. heydorni tissue tropism was noted in cattle from China [35], with the highest prevalence (9.7%) recorded in skeletal muscles, followed by 3.4% prevalence in the oesophagus, 2.5% in the diaphragm and 0.1% in the tongue muscles; however, S. heydorni was absent in the heart muscle. This zoonotic species has also been identified in cattle from the Netherlands (1%) and Turkey [4,10]. It should be noted that due to the different morphological, serological or molecular methods used in studies, it is difficult to compare the distribution of Sarcocystis species identified in cattle. Nevertheless, several trends in the prevalence of Sarcocystis species can be indicated. Most studies carried out in cattle reported the highest prevalence of S. cruzi employing canids as definitive hosts [10,14,15,18,37,[40][41][42][43][44]. By contrast, a relatively low prevalence of S. hirsuta using felids as definitive hosts was often recorded [14,15,37,41,45]. Research performed in cattle from the Netherlands and Italy showed S. hirsuta infection rates of < 2% [10,40]. By contrast, using PCR-restriction fragment length polymorphism, S. hirsuta infection in Iran was recorded as reaching 58.9% [46], whereas microscopy analysis noted a prevalence of S. hirsuta as high as 94.0% in Brazil [47]. Sarcocystis hirsuta has been identified in different geographic regions [8,10,14,15,40,44,45]. Felids are also known to be definitive hosts of S. bovifelis and S. bovini [8,48]; however, there is a lack of knowledge of the geographical distribution of these two Sarcocystis species. It should be pointed out that S. bovini has not been identified in Europe yet; S. bovifelis and S. bovini have been confirmed in Argentina, and only S. bovini has been found in New Zealand [8] and in beef from New Zealand imported to Japan [15]. In addition, based on comparison of 18S rRNA sequences S. bovini has also been found to be common in China [8]. Differences in the distribution of Sarcocystis species in cattle across geographical regions are possibly related to the abundance of definitive hosts.
We detected from one single Sarcocystis species up to four Sarcocystis species simultaneously in individual samples, with the highest prevalence noted for two species concurrently in one animal (Fig 2b). Two species forming sarcocysts in cattle, namely S. bovifelis and S. bovini, were (re)described only in 2016 [8]. Therefore, there is a lack of reliable data on the number of Sarcocystis species in individual bovine samples. However, mixed infections of different Sarcocystis species in cattle is a common occurrence [10,37,44]. Further studies are needed to reveal the rate of combinations of Sarcocystis species in individual samples from different countries.
The high level (> 40%) of S. hominis infection reported in cattle from countries such as Italy and Iran [33,46] is most likely overestimated due to the confusion in identifying S. hirsuta and S. bovifelis/S. bovini. The majority of the 18S rRNA sequences formerly presented in GenBank as S. hominis actually belonged to S. bovini, S. bovifelis or S. sinensis infecting the water buffalo [8]. There is a lack of detailed S. hominis molecular data [49] and the first cox1 sequence of this species became available in the GenBank database only in 2018. The prevalence of S. hominis infection in different countries based on molecular methods is poorly studied. A 0.2% prevalence of S. hominis was detected in China [50], 0.5% in Argentina [42], 3.5% in Iran [44], 6.2 % in Germany [37] and 12.5% in the Netherlands [10]. The molecular technique established in this study is an important step in the development of zoonotic S. hominis diagnosis in both cattle and humans.

Conclusion
Based on molecular methods, the presence of S. cruzi (96.1%), S. bovifelis (71.6%), S. hirsuta (30.4%), and S. hominis (13.7%) was confirmed in Lithuanian cattle meat for the first time. A rapid, accurate and relatively cheap technique was developed for the identification of Sarcocystis species in cattle meat. Further research on S. hominis epidemiology in various regions is needed.