Molecular identification of Sarcocystis halieti in the muscles of two species of birds of prey from Spain

Background Members of the genus Sarcocystis are protozoan parasites characterized by a prey–predator two-host life-cycle. Sarcocysts are formed in the muscles or central nervous system of the intermediate host (IH), while sporocysts develop in the small intestine of the definitive host (DH). Various birds of prey have been confirmed to be DH for Sarcocystis spp. Three Sarcocystis species, S. wobeseri, S. halieti and S. falcatula, have been identified in the muscles of birds of prey, of which the latter are known to be pathogenic and can cause encephalitis in various birds. The aim of this study was to identify Sarcocystis spp. in the muscles of birds of prey from Spain. Methods Between 2019 and 2020, muscle tissue samples taken from 59 birds of prey admitted to the Wildlife Recovery Centre in Ilundain (Navarra, Spain) were examined for the presence of Sarcocystis spp. Sarcocysts in fresh squashed samples were morphologically characterized under the light microscope (LM). Sarcocystis spp. were identified by means of 28S ribosomal RNA and internal transcribed spacer 1 sequence analysis. Results Microscopic examination of squashed tissue samples stained with methylene blue revealed the presence of sarcocysts in three of the 59 (5.1%) birds examined. Only one sarcocyst type was observed under the LM. Sarcocysts were thread-like (1050–2160 × 130–158 μm) and had a thin (0.7–1.4 μm) and smooth cyst wall. Septa divided the cysts into compartments filled with banana-shaped (5.9 × 1.7 μm) bradyzoites. On the basis of DNA sequence results, S. halieti was identified in the western marsh harrier (Circus aeruginosus) and the black kite (Milvus migrans) for the first time. Sarcocysts of S. halieti were shorter and wider compared to those observed in the great cormorant (Phalacrocorax carbo) and the herring gull (Larus argentatus). According to current knowledge, S. halieti may infect birds belonging to four different orders: Suliformes, Charadriiformes, Strigiformes and Accipitriformes. Conclusions This is the first report of S. halieti in the western marsh harrier and the black kite as IH. So far, little research has been conducted on birds of prey as IH for Sarcocystis spp. These results indicate that further studies combining morphological, histopathological, and molecular methods are required. Graphical abstract

Raptors are indicators of biodiversity and environmental health, and are are recognized in ecosystems as top predators and scavengers, and as flagship species [3]. Various birds of prey (eagles, hawks, falcons and owls) have been examined as possible DH [1]. A number of investigations have also looked at birds of prey as IH of Sarcocystis spp. [4][5][6][7][8][9][10][11]. The pathogenic species Sarcocystis falcatula may cause encephalitis in the free-ranging great horned owl (Bubo virginianus) [6], the golden eagle (Aquila chrysaetos) and the bald eagle (Haliaeetus leucocephalus) [7]. Likewise, an undescribed Sarcocystis sp. causing encephalitis has been detected in an immature northern goshawk (Accipiter gentilis atricapillus) from Minnesota [4]. Recently, Sarcocystis wobeseri was identified in the pectoral and cardiac muscles of the white-tailed sea eagle (Haliaeetus albicilla) [10]. Also, Sarcocystis halieti was detected in the brain and muscle tissue of a juvenile free-ranging little owl (Athene noctua) [11]. Thus, to date sarcocysts of three Sarcocystis spp., S. falcatula, S. halieti and S. wobeseri, have been recorded in the brains and tissues of birds of prey [6,7,10,11]. In another study, three morphological types of sarcocysts were detected in the Eurasian buzzard (Buteo buteo) and the long-eared owl (Asio otus), with one of the sarcocysts in the owl identified as S. otus [12]. However, this latter species is considered to be invalid [1].
In this article, we describe the molecular identification of S. halieti in the muscles of birds of prey from Spain.

Methods
Between 2019 and 2020 tissue samples of the leg muscles of 59 birds of prey (Accipitriformes, Falconiformes and Strigiformes) were examined for Sarcocystis spp. (Table 1). The birds had been admitted to the Wildlife Recovery Centre in Ilundain (Navarra) (Spain). The samples were taken by the Center's veterinary staff while carrying out their routine diagnostic protocol for the cause of death of the birds, which were either brought to the Center as dead specimens or died there. This center is under the jurisdiction of the Government of Navarra and is managed by a public company, GAN-NIK. Muscle samples were kept frozen (− 20 °C) until studied for morphological detection of the sarcocysts. The prevalence of sarcocysts and infection intensity were evaluated in methylene blue-stained muscle samples, as previously described [13].
Muscle samples of infected birds were delivered to the Laboratory of Molecular Ecology, Nature Research Centre, Vilnius, Lithuania for detailed morphological and molecular analysis. The morphological characterization of sarcocysts and bradyzoites was performed in freshsquashed samples. Sarcocysts with a small amount of host tissue were excised using two preparation needles, transferred to a drop of water on a microscope slide and measured under a light microscope (LM) at ×40-×1000 magnification.
Genomic DNA was isolated from individual sarcocysts using the GeneJET Genomic DNA Purification Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania). Partial 28S rDNA was amplified using the KL-P1F/KL-P2R primer pair [14], and the complete internal transcribed spacer 1 (ITS1) region was amplified using the SU1F/5.8SR2 primer pair [15]. The thermocycling conditions of the PCRs were as described previously [13]. Visualization, purification, and sequencing of PCR products were carried out using a previously described protocol [16]. The sequences obtained in this study were compared with those of various Sarcocystis spp. using the nucleotide BLAST program (megablast option) [17]. The multiple alignment was conducted using the MUSCLE algorithm loaded in MEGA7 software [18]. Selection of a nucleotide substitution model and phylogenetic analysis under Bayesian inference were carried out using TOPALi v2.5 [19].
Sarcocysts were detected in methylene-blue stained muscle samples from the leg muscles of one of the black kites (Milvus migrans); however, they were not observed in the freshly squashed samples not stained with methylene blue. Therefore, the muscle sample of this bird was digested with pepsin according to the modified protocol of Dubey et al. [1]. Specifically, 5 g of leg muscle tissue was cut into small pieces and suspended in 15 ml of saline solution (0.9%). The suspension was then homogenized in a commercial blender at top speed for 2 min with breaks. The homogenate was transferred into a 150-ml flask and 15 ml of digestion solution was added (pepsin, 0.26 g; NaCl 0.5 g; water up to 15 ml; and 37% HCl to pH 1.1). The entire contents of the flask were incubated at 37 °C for 2 h and the suspension was used for DNA extraction. Genomic DNA was extracted as described above. External PCR primers were SU1F/5.8SR2 [15], and internal primers GsShalF1 (5′-GAT AAT TGA CTT TAC GCG CCA TTA C-3′) and GsShalR1 (5′GTG CAC ATC CAT ATA TGC TCA TTC T-3′) were designed specifically for this study.
The first run of a nested PCR assay was conducted as described in [13]. The second run of a nested PCR assay was carried out in a final volume of 12

Results
Sarcocysts were detected in tissue samples stained with methylene blue examined under the LM in three of the 59 (5.1%) birds of prey: one western marsh harrier (Circus aeruginosus; isolate CaEs1) and two black kites (Milvus migrans; isolates MmEs1 and MmEs2). Two sarcocysts (isolates CaEs1 and MmEs1) were excised from fresh samples and subjected to amplification and sequencing of 28S rRNA and ITS1. Sarcocysts were not found in a fresh muscle sample from a single black kite (isolate MmEs2); therefore, this sample was digested and Sarcocystis DNA was amplified using a nested PCR targeting the ITS1 region. From three to seven sarcocysts were observed in 1 g of methylene blue-stained muscle sample. Under the LM, one type of sarcocyst was observed. Sarcocysts seemed to be thread-like, 1560 × 143 μm (range: 1050-2160 × 130-158 μm; n = 6) in size, with a thin (0.7-1.4 μm), apparently smooth cyst wall (Fig. 1a). The sarcocysts were clearly divided by septa into compartments filled with mature banana-shaped bradyzoites, 5.9 × 1.7 μm (range: 4.8-7.1× 1.3-2.1 μm; n = 45) in size (Fig. 1b).
The obtained 830-bp ITS1 sequence from the western marsh harrier (Circus aeruginosus) (isolate: CaEs1) was 100% identical with S. halieti from the great cormorant  HM125052). The ITS1 sequences obtained from two black kites (MmEs1 and MmEs2) displayed 97.0% similarity. In the ITS1 phylogenetic tree, the sequences from the black kites and the western marsh harrier were placed in one cluster together with those from S. halieti and Sarcocystis sp. from the Chilean skua (Fig. 2). It should be noted that the sequence from the black kite (MmEs1) formed a sister branch to the other S. halieti sequences. The 1488-bp 28S rRNA sequence from the black kite (MmEs1) differed by one to two single nucleotide polymorphisms (SNPs) from those of S. halieti (JQ733512, MF946610, MH130210) and by seven SNPs from those of S. columbae (HM125053), while the 1508-bp 28S rRNA sequence from the western marsh harrier (CaEs1) demonstrated 99.3-100% identity with that from S. halieti. Thus, on the basis of the molecular examination, S. halieti was identified in two black kites and a single western marsh harrier.

Discussion
The results of the present study reveal new IH for S. halieti, which we identified in the black kite and the western marsh harrier for the first time. A juvenile little owl was recently diagnosed with granulomatous encephalitis and muscular sarcocysts caused by S. halieti [11]; earlier studies had detected S. halieti in the great cormorant [20] and the herring gull [21]. Our results extend the body of knowledge on S. halieti specificity for the IH and indicate that this species is able to form sarcocysts in birds belonging to at least four different orders: Accipitriformes (present study), Charadriiformes [21], Strigiformes [11] and Suliformes [20]. Other avian Sarcocystis spp. (i.e. S. calchasi, S. columbae, S. falcatula and S. wobeseri) can also form sarcocysts in IH belonging to different orders [1,10,21,22]. The development of molecular methods and expansion of the diversity of the host species examined had led to the detection of the known Sarcocystis spp. in different bird orders [22]. Such investigations are of particular importance in terms of pathogenic species. It should be noted that highly pathogenic Sarcocystis spp., such as S. neurona, S. canis, S. felis, S. calchasi and S. falcatula, are multi-host adapted [1], as exemplified by the recent detection of S. halieti-associated encephalitis in a little owl from Germany [11]. Therefore, further investigation of the pathogenesis of S. halieti in various birds is required and extensive histopathological studies of this pathogenic species are recommended.
Sarcocysts of S. halieti detected in the muscles of birds of prey seemingly differed morphologically from those previously described in other IH. For comparative purposes, sarcocysts of S. halieti isolated from the leg and neck muscles of the great cormorant were very long, up to 6.5 × 0.1 mm [20], whereas sarcocysts excreted from the leg muscles of the herring gull were from 3960 μm to 7930 μm in length and from 43 μm to 128 μm in width [21]. Sarcocysts identified in the leg muscles of black kites and the western marsh harrier were, in comparison, shorter and wider (1050-2160 × 130-158 μm). Sarcocystis halieti sarcocysts of various shapes may be associated with diverse types of host anatomical structure. In terms of the distribution of muscle forces, accipitrids, falconids and strigiforms tend to possess greater proportions of distally inserted digital flexor musculature (53-64%, on average) [23]. On the other hand, the size of sarcocysts might depend on the contraction of the muscle fibers and the amount of pressure applied to the cover slip [1].
In the present study, sarcocysts were observed in three of the 59 (5.1%) birds of prey from Spain that were studied, with the help of methylene blue staining. A similar low prevalence of infection was detected in raptorial birds from Germany (3/79, 3.8%) [12] and Australia (5/38, 13.2%) [24] by histological examination. Based on pepsin digestion, significantly higher infection rates (52/114, 45.6%) were reported in birds of prey from the USA [5]. Whereas when examined by an immunofluorescence antibody test, eight of 72 (11.1%) raptorial birds from Brazil were found to be positive for Sarcocystis spp. [25]. Altogether, previous studies suggest that the prevalence of Sarcocystis spp. infection in raptorial birds varies depending on the method used for examination of the tissue samples.
Due to a lack of published data from comprehensive microscopic examinations, it is difficult to compare the morphology of the sarcocysts of S. halieti identified in the present study with those observed in other birds of prey in previous studies. Based on LM examination, two types of sarcocysts were reported in bald eagles from the USA [26]. The first type of sarcocyst was microscopic, had a thin cyst wall with spines and contained bradyzoites 5 × 1 µm in size; the second type (type II) was immature and had a 2-µm-thick striated cyst wall [26]. These sarcocysts are not similar to those observed in our study. However, type II sarcocysts were detected in the Eurasian buzzard [12]; these measured 694-1850 × 42-235 µm, had a seemingly smooth cyst wall and resembled those of S. halieti. A histological study detected thin-walled (0.5 µm) sarcocysts with a smooth surface and no visible protrusions in the cardiac muscle of the white-tailed sea eagle from Norway [27]; the length of the sarcocysts was not determined, but the diameter of the largest cyst measured was 40 µm. Subsequently, S. wobeseri was identified in the muscles of the white-tailed sea eagle from the UK [10]. Based on current knowledge, the sarcocysts of S. halieti and S. wobeseri are morphologically indistinguishable [21]. Lastly, thin-walled (≤ 1 μm) and thick-walled (2-4 μm) sarcocysts were detected in the muscles of raptors from the south-eastern USA [8]; the thin-walled sarcocysts might represent those of S. halieti, S. wobeseri or of an otherwise not yet confirmed Sarcocystis spp.
The most detailed morphological examination of sarcocyst structure occurs using transmission electron microscopy (TEM) [1]. To date, only a few studies have used TEM to describe sarcocysts in raptorial birds [7,12]. Based on TEM, the sarcocyst of Sarcocystis sp. found in the Eurasian buzzard and having a thin (up to 1.2 µm) and wavy cyst wall [12] is similar to that of S. halieti [20]. However, sarcocysts of S. halieti under the transmission electron microscope are very similar to those of S. calchasi, S. corvusi, S. fulicae, S. columbae, S. lari and S. wobeseri [1,20,28]. Also, four Sarcocystis species (S. columbae, S. halieti, S. lari and S. wobeseri) were identified in the herring gull [21]. Hence, molecular methods are needed for decisive discrimination of Sarcocystis spp. using birds as IH.
The most recent studies on Sarcocystis spp. in birds of prey have focused on the diagnosis of this apicomplexan genus using muscle digestion and subsequent nested PCR [9] or an immunofluorescence antibody test [25]. These methods are relatively sensitive to the detection of Sarcocystis spp. However, even with these methods, morphological characteristics of sarcocysts, which are important to achieve phenotypic diagnosis of Sarcocystis spp., cannot be determined. In summary, studies on the role of birds of prey as IH of Sarcocystis spp. are fragmentary, and it is difficult to compare the results obtained by different morphological and molecular methods.
Eight Sarcocystis spp., namely S. accipitris, S. alectoributeonis, S. calchasi, S. columbae, S. cornixi, S. halieti, S. lari and S. turdusi, using birds as IH are transmissible by birds of prey [1,27,29]. The white-tailed sea eagle and the Eurasian sparrow hawk (Accipiter nisus) (family Accipitridae) have been confirmed to be DH of S. halieti [27,29]. In the present study, two bird species also belonging to family Accipitridae were shown to be IH of S. halieti. Hence, there is a possibility that accipitrids might be infected with S. halieti through cannibalism. Such diet behavior is common among raptorial birds, especially among members of the Accipitridae family (65.5%) [30]. Future studies on the transmission of S. halieti are needed.

Conclusions
In the present study, S. halieti was identified in the black kite and the western marsh harrier from Navarra (Spain) by means of 28S rDNA and ITS1 sequence analysis. This is the third Sarcocystis species to be detected in the muscles of birds of prey. Studies on Sarcocystis spp. from birds of prey are fragmentary. Therefore, further combined morphological, histopathological and molecular