Morphological and molecular characterization of a Sarcocystis bovifelis-like sarcocyst in American beef

Abstract

Background

Parasites in the apicomplexan genus Sarcocystis infect cattle worldwide. Assessing the economic importance of each such parasite species requires proper diagnosis. Sarcocystis cruzi, a thin-walled species, infects virtually all cattle. The prevalence of the other thin-walled parasite, Sarcocystis heydorni, remains less well established. The remaining six species all have thick (> 3 µm) cyst walls (Sarcocystis hirsuta, S. hominis, S. bovifelis, S. bovini, S. sigmoideus, and S. rommeli). Thick-walled sarcocysts often induce inflammation in striated muscles (causing bovine eosinophilic myositis), leading to condemnation of carcasses at slaughter. One of these, S. hirsuta, can be seen macroscopically and lead to condemnation of beef. Two Sarcocystis species, S. hominis and S. heydorni, are zoonotic. Although S. hominis has been reported as prevalent in Europe, the occurrence of thick-walled species in the US remains poorly known. Here, for the first time to our knowldge, we characterize a thick-walled Sarcocystis species from a sample of beef from a local grocery store in Maryland. By morphological and genetic criteria, it closely, but not perfectly, resembles parasites previously ascribed to S. bovifelis.

Methods

Beef samples were examined for Sarcocystis infection, using acid-pepsin digestion to search for bradyzoites, microscopically by compression between a glass slide and coverslip, by histology of paraffin embedded sections stained with hematoxylin and eosin, and by transmission electron microscopy (TEM). Molecular characterization was attempted employing genetic markers: 18S rRNA, 28S rRNA, cox1, ITS1, gapdh1, ron3, and rpoB.

Results

Molecular evaluation revealed 100% identity with S. bovifelis-like sarcocysts from naturally infected cattle from Germany and Argentina; although the condition of the frozen material precludes complete characterization by TEM, we noted morphological features which differed from the S. bovifelis originally described from experimentally infected cattle from Germany.

Conclusions

A novel Sarcocystis species is described from beef from the USA but not named until further evaluation.

Graphical Abstract

Background

Sarcocystis infections in cattle are ubiquitous worldwide, attributable to an array of parasite species that are difficult to differentially diagnose. These species differ in their prevalence and pathogenicity. Proper diagnosis of Sarcocystis spp. is important to assess the economic and public health importance of each. Currently, there are eight named species infecting cattle: Sarcocystis hirsuta, S. cruzi, S. hominis, S. bovifelis, S. heydorni, S. bovini, S. rommeli, and S. sigmoideus [1, 2]. Canids are definitive hosts for S. cruzi, the species most pathogenic to cattle. Two species, S. hominis and S. heydorni, employ humans as their definitive host. Cats are definitive hosts for S. bovifelis, S. hirsuta, and S. rommeli. The definitive hosts for S. bovini and S. sigmoideus are unknown. Of these Sarcocystis species, S. hirsuta sarcocysts can become macroscopic and can lead to condemnation of beef on aesthetic grounds [3]. Additionally, Sarcocystis infections have been linked to an inflammatory condition of striated muscles termed bovine eosinophilic myositis (BEM). Cattle affected by BEM appear clinically normal. Diagnosis of BEM at slaughter occurs when inspecting the carcass surface or once the carcass has been divided into prime cuts or quarters. The etiology of BEM is uncertain, because the condition has not been reliably induced experimentally in cattle. Of these Sarcocystis species, the full endogenous life cycle stages are known only for S. cruzi [4]. The life cycle of S. bovifelis is partly known, but key studies were performed 4 -5 decades ago, before the advent of DNA sequencing necessary to confirm the identity of the etiological agent [5].

Among these eight Sarcocystis species, S. cruzi and S. heydorni are morphologically distinct because their sarcocysts are microscopic and have thin walls (< 1 µm thick). Sarcocysts of the remaining six Sarcocystis species have thick walls (> 3 µm). Differential diagnosis among those with thick walls is important given the zoonotic potential of some species and the economic costs imposed by BEM. Whereas in Europe, thick-walled and zoonotic species have been documented at considerable prevalence, little is known of the presence of thick-walled sarcocysts in the USA.

A survey for Sarcocystis species in American beef is in progress at the Animal Parasitic Diseases Laboratory (APDL), United States Department of Agriculture (USDA), Beltsville, Maryland. This survey involves testing beef for Sarcocystis bradyzoites in muscle digests, searching for sarcocysts in unstained muscle squashes, searching for sarcocysts histologically, and characterizing Sarcocystis species via DNA sequencing and transmission electron microscopy (TEM). As part of this study, we identified thick-walled sarcocysts in histological sections of one beef sample purchased from a grocery store. The sarcocyst wall appeared different from those previously described from beef; hence, the present detailed investigation was undertaken to characterize this thick-walled sarcocyst in the American beef. Collectively, our data identify a parasite corresponding closely, but not perfectly, to prior genetic and morphological descriptions for S. bovifelis. The available materials, once frozen, preclude comprehensive ultrastructural characterization. The molecular data provide a reference for future studies seeking to determine the prevalence and identity of this thick-walled species.

Methods

Sample

The beef sample was a Top Round package weighing around 500 g bought on July 23, 2023, from a local grocery store in Maryland. After trimming, a 100 g sample was tested for Sarcocystis. Fifty grams of muscle was ground in a meat grinder and digested in acid pepsin for 1 h; the digest was examined microscopically for Sarcocystis bradyzoites as previously described [6]. No sarcocysts were found in 30 muscle squashes examined microscopically, but (per our standard protocol) 50 g of muscle was stored at − 80 °C for future investigation.

Histologic evaluation

Pieces of muscle were fixed in 10% buffered formalin and processed for histological preparation. Paraffin-embedded sections were stained with hematoxylin and eosin and examined microscopically. All sarcocysts found in 4–6 pieces of paraffin-embedded muscle (each measuring ~ 1 cm × 1 cm, enough to fill ~ 4 × 2 cm area of a histological slide) were examined at 100 × and 1000 × magnifications.

Examination of sarcocysts from frozen, unfixed muscle samples

Frozen sample of beef was thawed at room temperature, squashed between glass slides and coverslip, and examined microscopically. Sarcocysts were photographed, numbered, and (after removing the coverslip) either saved in PBS (cyst numbers) for DNA extraction or fixed in commercial Trump’s fixative (2% paraformaldehyde, 2.5% glutaraldehyde in 100 mM phosphate buffer pH 6.8). Unfixed bradyzoites mechanically released from sarcocysts were also photographed.

Transmission electron microscopy

Small pieces of infected beef in Trump's fixative were sent by air to the Institute of Parasitology, University of Bern, Switzerland, for TEM examination as described [2].

DNA isolation and amplification

Genomic DNA was extracted from seven thick-walled, microscopically confirmed sarcocysts (#7, #8, #9, #10, #25, #26, #28) using the Qiagen DNeasy^® Blood and Tissue Kit (Hilden, Germany) following the manufacturer’s instructions. PCR amplification of highly conserved regions of 18S rRNA, 28S rRNA, a mitochondrial cytochrome c oxidase subunit 1 (cox1) and the highly variable internal transcribed spacer-1 (ITS1), rhoptry neck protein 3 (ron3), which encodes a protein that is part of rhoptry organelle, glyceraldehyde-3-phosphate dehydrogenase 1 (gapdh1), and a highly conserved housekeeping gene, rpoB, which encodes the beta-subunit of RNA polymerase, was performed using Sarcocystis-specific primers designed during the study (Table 1) and primers designed using isolates of Sarcocystis sigmoideus as reference (for 18S: 310F-CGGGTAACGGGGAATTAGGG and 840R-CGTGCAGCCCAGAACATCTA; for cox1: 204F-TGTCGAATGTGGTGCGGTAT and 218R-AGTACCTCCCAGGCTGAACA). The 13 µl PCR mix consisted of a 2 μl DNA template, 6 μl of Platinum Hot Start PCR Master mix (Invitrogen, USA), 0.5 μl of 10 pmol/μl of each primer (IDT, USA) (Table 1), and 4 μl of molecular grade water. After initial denaturation at 94 °C for 3 min, 35 cycles were performed consisting of denaturation at 94 °C for 30 s, annealing at 60 °C for 30 s, and elongation at 68 °C for 20 min; terminal elongation incubated products at 68 C for 5 min. The PCR products were analyzed on a 2% agarose gel, and size was estimated by comparison with the 100-bp Plus DNA Ladder. The obtained PCR products were purified using the ExoSAP method [7]. The final purified PCR products were sent for sequencing to Psomagen (Rockville, MD, USA) for direct sequencing on an ABI 3500xl Genetic Analyzer (Applied Biosystems™).

Table 1 Accession numbers for various gene markers of Sarcocystis bovifelis infections in American beef

Trimmed and annotated sequences belonging to 18S, 28S, cox1, ITS1, gapdh1, ron3, and rpoB were aligned using the Clustal W 2.1 alignment tool [8] implemented within Geneious Prime 2023.0.4 (www.geneious.com) and were submitted to GenBank (Table 1). Species identity was verified using the Basic Local Alignment Search Tool (BLAST) [9]. The final edited sequence length of each gene marker and their GC content appear in Table 1.

The web server Guidance 2 [10] was used to align and remove ambiguously aligned positions in each gene. The sequences were aligned with the MAFFT algorithm under the options Max-Iterate: 1000 and Pairwise Alignment Method: -localpair. Positions with a score < 0.93 were removed as well as positions with > 25% missing data. A model test was performed using jModelTest 2.1.7 [11] to determine the most suitable nucleotide substitution model, according to the Bayesian information criterion (BIC). Phylogenetic relationships were reconstructed under the maximum likelihood (ML) criteria. ML analyses were performed with the program IQ-TREE version 1.6.12 [12]. The analyses were run with the options -m MFP -b 1000 -nt 5 (i.e., ModelFinder + tree reconstruction + 1000 non-parametric bootstrap replicates + tree topology test). The model selected based on the BIC criterion was (18S = K2 + G, lowest BIC = 3405.9; 28S = K2 + G, lowest BIC = 2256.7; cox1 = K2 + G, lowest BIC = 9803.7; ITS1 = HKY, lowest BIC = 2815.4, rpoB = T92 + G, lowest BIC = 3446.1 and gapdh1 = T92 + I, lowest BIC = 4849.7). The nucleotide frequencies were also estimated for each gene.

Results

Light microscopy

In the original histological slide, six thick-walled sarcocysts were detected in two of six muscle pieces. Thereafter, an intensive search was made to find thick-walled sarcocysts. An additional 80 pieces of formalin-fixed samples of muscle, embedded in 16 paraffin blocks, were processed. From these, a total of two sarcocysts were detected in 1 of the 16 histological slides. These 17 paraffin blocks were cut a second time, yielding only 8 cysts. A third recut of 17 blocks revealed only 1 cyst. Thus, a total of 17 thick-walled sarcocysts were detected in 51 histological sections. One sarcocyst was thin-walled and appeared to be S. cruzi (excluded from further study).

The thick-walled sarcocysts in H&E-stained histological sections were 38–400 × 26–75 µm. The sarcocyst wall was 2.8–5.6 µm thick and contained villar protrusions that were upright to sloping; their tips were thickened (Fig. 1).

Fig. 1

A, B Two mature sarcocysts in histological sections of beef. Note thick cyst walls (cw), straight to slopping villar protrusions (vp) densely staining villar protrusion ends (arrowheads). Hematoxylin-eosin stain

Examination of sarcocysts from frozen, unfixed samples

From several grams of frozen muscle, a total of 44 sarcocysts were found (3 of which appeared thin-walled; resembling S. cruzi, they were not studied further). All the remaining 41 thick-walled sarcocysts were < 1 mm long, except one that was 1860 µm long and 43 µm wide. The sarcocyst wall was 3.4–6.7 µm, depending upon the flattening of the cyst and angle measured (Fig. 2). Free bradyzoites were 12–14 × 3–4 µm.

Fig. 2

A, B Two mature sarcocysts in muscle compressed between a glass slide and cover slip. Note variability of the thickness of cysts (cyst marked 9 is slender compared with cyst marked 2) and increased density of tips of villar protrusions (arrowheads). Unstained

TEM

Six sarcocysts were studied by TEM. Although sarcocysts were sub-optimally preserved because the specimens had been frozen, the sarcocyst wall structures were preserved. The thickness of the sarcocyst wall varied with the plane of the section (Fig. 3). The ground substance was indistinct, < 0.5 µm thick, and devoid of granules or tubules (Fig. 4). The villar protrusions (vp) were elongated, mostly sloping, up to 5.8 µm long (Fig. 5). The vp were broad at the base, up to 1.8 µm in tangentially cut section, and tapered distally. The vp tips were often forked or had hook-like structures (Figs. 4, 5). The vp were lined with fine microtubules that criss-crossed at the base and extended no more than one-half length of the vp (Figs. 4, 5).

Fig. 3

Transmission electron micrographs of sarcocyst #38. A, B Note variability of the appearance of villar protrusions (vp) and the thickness of the cyst wall (cw). The ground substance layer (gs) is thin and continues into the interior of the sarcocyst as septa (se). Bradyzoites (br) and metrocytes (me) are juxtaposed with the gs

Fig. 4

Transmission electron micrographs of two sarcocysts. A Longitudinal section of a villar protrusion (vp). Note the ground substance layer (gs) is thin and devoid of granules. The vp has a broad base and forked, tapered tip (arrow). The microtubules (mt) in vp arise at the base of gs; they are devoid of granules and extend up to half of the length of the vp. Note part of degenerated bradyzoite (br) at the base of gs. Cyst #34. B Higher magnification of the base of villar protrusion to show absence of vesicles and the origin of microtubules (mt). Cyst #38

Fig. 5

Transmission electron micrographs of two sarcocysts. A Longitudinal section of a villar protrusion (vp) (opposing arrowheads). Cyst #38. B Sections through villar tips (vp) to show hooks at the tips (arrows). Note absence of microtubules at the villar tips. Cyst #13

Molecular analysis

The selected genetic markers (18S, 28S, cox1, ITS1, gapdh1, ron3, and rpoB) were successfully amplified and sequenced. The obtained sequences under study showed nearly 100% identity to S. bovifelis described from Argentina, Lithuania, and Kazakhstan. No published reference exists for rpoB in S. bovifelis; our sequence showed < 99% identity to that reported for Sarcocystis falcatula (Table 2). Moreover, the primers intended to specifically amplify S. sigmoideus amplified a product from these seven cysts, but the resulting sequence matched that expected for S. bovifelis.

Table 2 Homogeneity of various gene marker sequences of thick-walled Sarcocystis spp. infecting muscles of cattle available in the GenBank with Sarcocystis bovifelis isolated during the study

Phylogenetic analysis

The final phylogenetic analyses included 19 taxa and 1856 positions for 18S, 13 taxa and 417 positions for 28S, 19 taxa and 1041 positions for cox1, 9 taxa and 1066 positions for ITS1, 12 taxa and 783 positions for rpoB, and 4 taxa and 846 positions for gapdh1 from the lineage of Sarcocystidae and different isolates of Sarcocystis neurona, Toxoplasma gondii (for gapdh1 gene) and Sarcocystis wenzeli (for rpoB gene) as outgroups. The phylogenetic trees for the six loci were reconstructed and analyzed based on maximum likelihood and Bayesian criteria for the positioning of the species under study in the lineage of Sarcocystis species infecting cattle, primarily the thick-walled sarcocysts. The trees reconstructed only differ in the position of a few low-supported branches showing close homology between the Sarcocystis species infecting cattle (Fig. 6). The tree based on all the genetic markers utilized during the study included all available data from species infecting cattle (i.e. S. bovini, S. bovifelis, S. cruzi, S. heydorni, S. hirsuta, S. hominis, S. rommeli, and S. sigmoideus).

Fig. 6

Phylogenetic relationships of Sarcocystis bovifelis with various Sarcocystis species infecting cattle inferred from different genetic markers (A = 18S rRNA, B = 28S rRNA, C = cox1, D = ITS1, E = rpoB and F = gapdh1) under maximum likelihood criterion (ModelFinder + tree reconstruction + 1000 non-parametric bootstrap replicates + tree topology test). Branch supports are indicated near the corresponding nodes. The highlighted part (blue) shows the S. bovifelis cluster; pink shows the thick-walled cyst cluster, and green part shows the thin-walled cyst cluster with highest nodal support. Species in bold are the ones obtained during the study

Apparently, the phylogenetic tree for each genetic marker distinguished, with strong bootstrap support, thick- from thin-walled species of Sarcocystis infecting cattle (Fig. 6), consistent with previous findings [13,14,15].

For the 18S and cox1 genes, there were two clusters, one thick- and the other thin-walled with high consensus support showing the tendency of tissue specificity for Sarcocystis species. The sequences isolated from all seven cysts clustered with different isolates of S. bovifelis described from Argentina based on the best match (which has been called as “S. bovifelis cluster” in the tree) in a single clade with high bootstrap values. This cluster also includes other thick-walled Sarcocystis species including S. bovini, S. rommeli, S. hirsuta, and S. hominis, the other cluster being thin-walled consisting of S. cruzi and S. heydorni.

For other genes (28S, ITS1 and rpoB), due to less data availability of S. cruzi and S. heydorni at the NCBI, only thick-walled cluster and the “S. bovifelis cluster” could be studied.

Discussion

Based on molecular characterization, the thick-walled Sarcocystis species in the present study most closely resembled S. bovifelis, but with some differences (SNP’s or single nucleotide polymorphism). Some of this confusion and uncertainty may never be resolved because of inadequate original description of S. bovifelis. Therefore, it is important to document the basis for the original description of S. bovifelis.

In 1975, Heydorn et al. [16] proposed a new nomenclature for naming Sarcocystis species in livestock. When naming S. bovifelis [16], all cattle Sarcocystis species transmitted by cats were then attributed to one species, S. bovifelis. Its description was based on characterizing sarcocysts in experimentally infected cattle [17]. For this, six weaned calves were orally inoculated with sporocysts from the feces of eight cats that had been fed the esophagi of naturally infected cattle in Germany (details were translated into English [5]). Transmission electron microscopic observations were based on two calves necropsied 98 and 60 days post inoculation (p.i.). Relevant details are restated here. The villar protrusions (vp) on the sarcocyst wall were 3.8–5.4 long; the original description makes no mention of thickening of the villar tips [17]. The vp were sloping, broader at the base, and tapered distally, but the dimensions were not specified. Microtubules were present throughout the vp. Additionally, vesicles were stated to be present at the base of the vp [17]. However, of the five illustrations of vp [17], four show sloping vp but no vesicles at the base. One image (Fig. 8 of [17]) shows a sarcocyst with vp that have straight protrusions with flattened tips and the presence of vesicles at the base of vp. Better quality illustrations of the cyst wall of S. bovifelis from the same experiment were provided by Mehlhorn et al. [18]. The vp were broader at the base and tapered distally. In all three illustrations, no vesicles are evident at the base of vp [18]. No archived specimens remain from the original study.

The authors of the initial description [16] submitted specimens of sarcocysts of S. bovifelis and other species to several individuals and institutions, including Beltsville, Maryland, USA. Re-examination of the calf necropsied at 98 days p.i. revealed a 5.5-µm-thick sarcocyst wall but no thickening of the villar tips (Fig. 5C, D of [5]). An unpublished electron micrograph from the German study [17] provided to us is produced here (Fig. 7). It shows sloping vp but no vesicles at the vp base. It is likely that more than one Sarcocystis species was present in calves experimentally infected by Gestrich et al. [17]; the sporocysts were derived from feces of several cats that were fed naturally infected beef and thus could have been derived from more than one Sarcocystis species.

Fig. 7

Transmission electron micrograph of a sarcocyst in muscle of a calf fed sporocysts from the feces of naturally infected cats in Germany from Gestrich et al. [17]. Note microtubules extend from the base to end of villar protrusions and the absence of any vesicles at the base of villar protrusions

The only other study of the life cycle of a cat-transmitted Sarcocystis is that of Dubey [5, 19]. These parasites were isolated from muscle of an experimental cow in Bozeman, Montana, USA, and subsequently serially passed in cats and calves [19]. Newborn calves were inoculated with sporocysts from experimentally infected cats, and calves were killed at intervals to study parasite development through ensuing life cycle stages, documenting development of schizonts and sarcocysts. Only a few mature sarcocysts were present in experimentally infected calves. By light microscopy, the sarcocyst wall resembled that of [17] and villar tips were not thickened; ultrastructure of sarcocysts was not studied [5]. There are no archived paraffin blocks for molecular studies.

Gjerde [15] molecularly characterized 45 S. bovifelis-like sarcocysts from naturally infected beef from Argentina and Germany using six genes. These sarcocysts were 1–3 mm long, much longer than those reported from experimentally infected cattle; this report lacked accompanying TEM description. Gjerde [15] also named another species, S. bovini, whose sarcocysts morphologically resembled S. bovifelis by light microscopy but which differed molecularly; again, no TEM accompanied that description. These findings are relevant for the discussion of the Sarcocystis species in the present study.

Features of sarcocysts of different Sarcocystis species by TEM are summarized in Table 3. Sarcocystis hirsuta and S. hominis sarcocysts are morphologically distinct. The ground substance layer in sarcocysts of S. hirsuta and S. hominis is thick, and their vp are also distinctive. The vp of S. hirsuta have a small stalk at the base and vp are expanded laterally. The vp of S. hominis are cylindrical. Sarcocystis rommeli sarcocysts have prominent vesicles at the base of vp; these are absent in other species (Table 3). The microtubules in vp in the species in the present study are absent in the 1/3 distal portions of vp, and there are no vesicles at the base of vp. Naming of the new species here is deferred until optimally fixed specimens are available for ultrastructural description.

Table 3 Comparison of cyst wall structures of thick-walled sarcocysts of cattle by transmission electron microscopy

Thus, we document the occurrence of a thick-walled species of parasite in American beef that most closely resembles genetic signatures previously reported for S. bovifelis but which differ, ultrastructurally, from the material initially employed in the description of S. bovifelis. The available materials, once frozen, preclude definitive ultrastructural characterization. The genetic data, however, may prove useful in discerning the prevalence and identity of the thick-walled parasite herein identified.