441

Journal of the Egyptian Society of Parasitology, Vol. 36, No. 2, 2006

J. Egypt. Soc. Parasitol., 36 (2): 441 - 453

INFERENCE OF MOLECULAR PHYLOGENY OFSARCOCYSTIS FELIS (SARCOCYSTIDAE) FROM

CATS BASED ON NUCLEAR-ENCODED RIBOSOMALGENE SEQUENCES

ByHANY M. ELSHEIKHA1, DOAA M. SOLTAN2 AND

MANAL F. EL GARHY3

Department of Parasitology1, College of Veterinary Medicine,(Email: elsheika@mans.edu.eg) and Department of Parasitology 2,

Faculty of Medicine, Mansoura University, Mansoura 35516,and Department of Zoology3, Faculty of Science, Cairo

University, Egypt

Abstract

The phylogenetic position of four clinical isolates of Sar-cocystis felis was assessed using ssurRNA and ITS1 genesequences in the context of a wide array of other Sarcocystis sp.Phylogenetic reconstructions using neighbour-joining and maxi-mum parsimony methods generated identical tree topologieswith strong support values at each node. High ssurRNA seque-nce similarity (≥99%) and the resulting phylogeny demonstratedthat S. felis and S. neurona are significantly closely related toeach other. The two Sarcocystis formed a monophyletic groupdistinct from the other Sarcocystis sp., irrespective of the align-ment algorithms or tree-building method used. The absolute(100%) identity of ssurRNA sequences of sarcocysts and spo-rocysts obtained from one cat raised the question regarding thecat's role as a potential intermediate host besides its known roleas a definitive host of S. felis. On the other hand, S. felis sarco-cyst DNA sequence was found to be quite dissimilar over theITS1 region when compared to S. neurona. These findings indi-cated that using sequences from two different genetic loci pro-vided a stronger comparative basis than would have been possi-ble using either one.

442

Introduction

The genus Sarcocystis Lankester, 1882 is one of the largest(~200 sp.) and most widespread of cyst-forming coccidiangenera (Odening, 1998). Despite the global distribution and eco-nomic importance of this genus, evolutionary relationships ofsome of its members are still poorly understood. This lack ofunderstanding can be attributed to the nature of morphologicalcharacters in Sarcocystis. Sarcocystis sp. are morphologicallysimilar and most of their taxonomy was based on description ofsarcocyst structure. Genetic analysis and/or cross-transmissionexperiments are necessary to distinguish between Sarcocystis sp.(Mehlhorn et al., 1976; Tadros and Laarman, 1982). Indeed, ge-netic analysis provided more insight into this large genus andnew taxonomic schemes were proposed (Tenter et al., 2002).However, the evolutionary relationships of S. felis to otherSarcocystis sp. are still unclear. Molecular phylogeny has notbeen fully used to address this issue, as only one previous studydemonstrated that S. felis did not show a strong affinity to anyother Sarcocystis (Gillis et al., 2003). Sarcocysts of S. neuronaand S. felis have been detected in cats in the USA (Dubey et al.,2003; Gillis et al., 2003). Although both species share featurescommon to members of the genus Sarcocystis, morphologicalfeatures of their cyst wall are fundamentally different (Gillis etal., 2003). Because these two species share the cat as an inter-mediate host, it is hypothesized that the two organisms mighthave a common evolutionary history. However, it is not certainthat they really are related or have evolved convergently. Thus,a clarification of the phylogenetic position of S. felis will allowus to address this systematic issue. Here, the ssurRNA and theITS1 sequences were characterized from four S. felis isolatesfrom cats in Michigan and compared to previously reportedsequences of S. felis and other Sarcocystis sp. Furthermore, thegenetic analyses provided more insight and allowed detailed in-ferences about S. felis evolution as it compared the organismalsource of the sequence.

443

Materials and Methods

Four new sequences were obtained from cysts of three catsand sporocysts from one of these cats, which were submitted fornecropsy at Michigan State University (MSU), USA. These catswere collected as a part of a survey on S. felis cysts in cats inMichigan during the calendar year 2004. Cats were examinedfor the presence of S. felis sarcocysts in their muscles. Samplesfrom diaphragm or cardiac muscles were examined under adissecting microscope, and tissues with visible cysts were re-moved. Tissue specimens were stored frozen at -80°C or keptfresh at 4°C until DNA extraction for polymerase chain reaction(PCR) and sequencing. Fecal materials from the cat intestinewere inspected for the presence of any oocysts/sporocysts usingSheather’s sugar fecal flotation technique. The harvested floatswere examined under a light microscope at 400X magnificationeither as fresh or as a smear stained with Diamant fuchsin. Se-quence was determined for parasite materials of more than onecat to assess the degree of intraspecific variation among S. felisstrains. Sequences of the partial ssurRNA and ITS1 genes wereproved valuable to study the internal relationships and evaluatethe phylogenetic position of apicomplexan parasites (Marsh etal., 1995, 1999; Barta, 2001). Sequences of ssurRNA and ITS1genes of representative well-defined Sarcocystis sp. were obtai-ned from GenBank and included in the phylogenetic analyses tohelp determine the taxonomic position of the present materialsobtained from cats.

Isolation of DNA: Total genomic DNA was extracted fromcyst-containing tissues using a QIAamp blood extraction kit(Qiagen, Valencia, California, USA). The protocol followed wasthat suggested by the Manufacturer with some modifications.Briefly, tissues were grinded in liquid nitrogen and tissue lysiswas performed in the presence of proteinase K for 10 min at70°C. The lysed material was applied to a spin column con-taining a silica gel-based membrane and washed twice with Tris(10 mM, pH 8.0). Also, the DNA was obtained from sporocystswith a Qiagen DNeasy tissue Kit. Purified DNA was elutedfrom the columns in 200 µl of Tris and stored at 4°C until usedas template for PCR amplification. DNA concentrations of each

444

isolate for each animal were based on optical density readings ata wavelength of 260 nm.

PCR primers: The sequences used were from the internaltranscribed spacer (ITS) region of nuclear ribosomal DNA(nrDNA) and small subunit ribosomal RNA (ssurRNA) gene.These loci have become the sequences of choice for addressingphylogenetic questions within genera because they are easilyaligned and amplified, they have a high mutation rate, and theyare relatively small (Baldwin et al., 1995). JNB69/JNB70 pri-mers for the ITS1 region were previously described (Tanhauseret al., 1999). ITS1 primers were designed from the conserved 3'end of the small ribosomal subunit gene and the 5' end of thelarge ribosomal subunit gene. Amplification of ssurRNA wascarried out using the primers LSM1 (5’-GGATTCCCATCATTCCAATCAC) and LSM2 (5’CTTGTCTTAAAGATTAAGCCATGC) that amplified 496 bp of the ssurRNA gene. Primerswere synthesized by Integrated DNA technologies Inc. (Coral-ville, IA, USA).

PCR amplification and automated sequencing: Reaction andcycling conditions were identical for the DNA of both thesarcocyst and sporocyst reactions. PCR amplification of the ssurRNA and ITS1 regions was carried out in a 50 µl reaction. Eachreaction mixture contained 2.5 mM MgCl2, 10 pmoles of eachprimer, 0.2 mM each of dATP, dGTP, dCTP, dTTP, 1.25 U ofTaq polymerase and ~50 ng of template DNA. One piece of thessurRNA gene and two pieces of the ITS1 region were amplifiedand sequenced separately. Amplifications were carried out usinga PTC-100 thermocycler (MJ Research Inc.) with a heated lid.The PCR cycle employed for the amplification of both geneswas as follows: 5 min initial denaturation at 95°C followed by35 cycles of denaturation at 95°C for 1 min, primer annealing at55°C (ITS1) or 60°C (ssurRNA) for 1 min, and extension at72°C for 4 min. A final extension step of 5 minutes at 72°C wasincluded in each amplification reaction to ensure that the finalamplification products were all of full length. After amplifica-tion, the PCR products were stored at –20°C. Amplification pro-ducts were electrophoresed in 1.5% low melting point agarosegel for 1.5 to 2 hr at 100 V. The gel was then stained in 300 mlof 1× Tris acetate-EDTA (TAE)-30 µl of EtBr for 15 min and

445

destained in 1×TAE for 5 min. The presence of PCR productswas determined from observation of single, intense bands on thegel on a UV transilluminator.

PCR products were direct-purified using the QIAquick PCRPurification Kit (Qiagen). The Manufacturer's Protocol of theautomated sequencing was used to determine the partial seque-nce of ssurRNA and two regions of ITS1 gene for each isolate.DNA sequencing reactions used fluorescently labeled dideoxy-nucleotide technology (BigDye® Terminator v3.1 Cycle Seque-ncing Kit; Perkin-Elmer, Applied Biosystems, Foster City, Cali-fornia, USA) according to the Manufacturer’s instructions.Sequencing reaction products were separated and data were co-llected using an ABI 3100 capillary automated DNA sequencer(Applied Biosystems). The sequences were fully determined forboth strands of the DNA template and a consensus sequence wasobtained to ensure accuracy of the data. Sequences were editedand assembled using the Wisconsin Sequence Analysis Package(Genetics Computer Group [GCG], Madison, WI, USA). BLA-ST (www.ncbi.nlm.nih.gov/BLAST/) searches were performedto determine whether the sequences were similar to any of thepreviously published sequences of Sarcocystis. Aligned seque-nces were compared against sequences of closely relatedorganisms using the Pileup program of the (GCG) package andthe percent identities were calculated. A nexus file was createdfrom the sequences alignment using ClustalX v1.81 (Thompsonet al., 1997) using default settings. Minor adjustments weredone to the alignment by eye and indels were treated as missingdata. The newly obtained S. felis nucleotide sequences were de-posited in the Genbank database (NCBI) and were received theGenBank accession number AY656815 for partial ssurRNAsequences from sporocysts. For simplicity, ssurRNA sequencesfrom sarcocysts were not deposited because of the absolute iden-tity to ssurRNA sequences of sporocysts. Likewise, partial se-quences of two regions of the ITS1 gene were found to be iden-tical to previously published ITS1 sequences by Gillis et al.(2003), thus, not submitted to Genbank.

Molecular phylogeny: Sequences were subjected to phylo-genetic analyses using parsimony and distance analyses. All thephylogenetic reconstructions were performed using the compu-

446

ter program PAUP version 4.0b10 (Swofford, 2002) run on aPower Macintosh. For, maximum parsimony (MP) analysis, theshortest trees were found via a heuristic search with 100 repli-cate random stepwise addition and tree-bisection-reconnection(TBR) branch swapping. All uninformative characters were ex-cluded and the remaining characters were considered unorderedand equally weighed. Neighbour-joining tree (NJ) (Saitou andNei, 1987) was generated based on a Kimura 2-parameter dista-nce matrix (K2-p) (Kimura, 1980). For MP and NJ analyses,statistical support for clades in the phylogenetic tree was asse-ssed by the bootstrap method (Felsenstein, 1985) with 1000 re-plications. The bootstrap proportions of <50% were consideredas not supported, proportions between 50 & ≤70% as weaklysupported, and proportions >70% as potentially well supported(Hillis and Bull, 1993).

Results

Sarcocystis felis taxonomy:Host: Domestic cat Felis catus Linnaeus.Locality: Central-South Michigan, (4243-4279N, 8418-8486W), USA.

Voucher specimens: Histological sections of S. felis sarco-cysts from the heart and diaphragm have been deposited in theUS National Parasite Collection, Beltsville, MD 20705, USA(USNPC accession Nos. 94790, 94859 and 94891).

Microscopic examination: Floats obtained from the intestinalcontents of one cat were found to have few sporocysts (Fig. 1).The sporocysts were very similar to those of other Sarcocystissp. Definitive identification of S. felis cysts in the muscle tissueswas previously performed based on ultrastructure characteri-zation of cysts and bradyzoites.

Sequence analysis: Sarcocystis sp. confirmation was thoughtby PCR and DNA sequence analysis, which is one of the mostaccurate methods for distinguishing Sarcocystis sp. DNA wasused as a template for two PCR assays. A total of 496 bp partialsequence of ssurRNA gene was obtained from the amplifiedproducts using the LSM1 and LSM2 primers. The ssurRNA catsequence was identical to the nucleotide sequence of Sarcocystis

447

felis (GenBank accession numbers AY190080). This sequenceincluded only a 2 bp (GG) insertion at position 238-239 bp thatwas absent from S. neurona sequences AY009112 from Seaotter and AF252406 from Harbor seal. Less homology (99%sequence identity) was observed between the S. felis sequence ofthe cat and sequence of S. neurona isolate from a horse (SNU07812).

The second PCR amplified two segments of the ITS1 geneusing previously described primers (Tanhauser et al., 1999). Nodifferences were noted between either ssurRNA or ITS1 seque-nces of sarcocysts from the three cats and no differences werenoted between these sequences and sequences reported by Gilliset al., (2003). Therefore, only one sequence was used in subse-quent phylogenetic analysis to reduce computational time. Se-quences from the newly obtained S. felis showed the greatestsimilarities to S. neurona in ssurRNA gene. On the contrary, awide range of ITS1 sequence divergence between the presentmaterials and S. neurona was noticed.

Molecular phylogeny and systematics: Sequence data wereused to characterize S. felis isolates through phylogenetic analy-sis. Phylogenetic trees were built using the PAUP program. In aMP tree based on 496 bp sequence of ssurRNA gene, Sarco-cystis sp. fell into two major groups (Fig. 2). The upper group ofthe tree consists of nine Sarcocystis sp. including S. singapore-nsis, S. sinensis, S. fusiformis, S. cruzi, S. rodentifelis, S. dispel-rsa, S. lacerata and S. mucosa, as well as S. neurona from ahorse (SNU07812).

448

The lower group consisted of 2 species; S. neurona from asea otter (AY009112) & a harbor seal (AF 252406), S. felis froma cat, and Sarcocystis obtained from cats in the present study.Tree constructed by NJ analysis demons-trated the same majorgroupings.

For simplicity NJ tree was not shown except its bootstrapvalues (Fig. 2). Both MP and NJ trees suggest that S. neuronaand S. felis were more closely related to each other than to otherSarcocystis. Phylogenetic reconstructions based on partial ITS1gene sequences produced tree topology identical to thosereported by Gillis et al. (2003).

449

Discussion

Wide speculation surrounds the origin, genetic variability,and phylogenetic relationship of S. felis to other Sarcocystis. Inorder to address these issues and to gain a better understandingof taxonomy of S. felis, the phylogenetic relationship of fourpreviously undescribed clinical isolates of S. felis to otherputatively related Sarcocystis was inferred from partial sequ-ences of the ssurRNA and ITS1 genetic loci using neighbour-joining algorithm and maximum-parsimony methods of phylo-genetic reconstruction. Results of ssurRNA sequence showedthat Sarcocystis DNA obtained from cats in the present studyand S. neurona formed a distinct monophyletic group. The otherSarcocystis appeared to be distantly related to S. felis. The phy-logenetic tree supported the close evolutionary relationship of S.felis and S. neurona, thus, agreed with what have been suggestedby Gillis et al. (2003).

The construction of trees, which group together species thatexhibit similar host specificity by using the cat as an inter-mediate host, gives an insight into the evolution of S. felis. Theminor differences in ssurRNA sequences between S. felis and S.neurona confirm the close relationship between these organisms.Likewise, the phylogram topology revealed that the presentmaterial, S. felis and S. neurona isolates from harbor seal andsea otter are very closely related. These data show not only thatthe host specificity of some Sarcocystis sp. that infect cats is lessrestricted as previously thought but also that there is a complexcirculation of Sarcocystis sp. in the environment.

Ribosomal gene sequencing is a commonly used method tofind genetic polymorphisms between organisms. In this study,sequences of two regions of the ITS1 gene and the partialsequence of ssurRNA gene were compared to representatives ofSarcocystis to determine the phylogenetic position of S. felis. Byusing sequences from two different genetic markers, the authorsprovided a stronger comparative basis than would have beenpossible using either marker. The ssurRNA sequences wereeffective tools for estimating relationships among S. felis andSarcocystis sp. The absolute ssurRNA sequence similarityamong the S. felis isolates and the limited sequence variation

450

between S. felis and S. neurona shown suggested a high level ofrelatedness among the two species and a relatively short periodof time elapsed since the isolates had diverged. Thus, both spe-cies formed a well-supported clade. Indeed, several aspects ofthe phylogeny were strongly supported and seen in most of themethods used. However, ssurRNA sequences did not provideenough phylogenetic information to fully resolve relationshipsbelow the S. felis species level. The ITS region of rDNA isoften used in intra-species studies because variation is usuallyfound there. In this study, the ITS sequenced region was not alsosufficiently adequate to resolve intraspecies relationships withinS. felis isolates. ITS1 results were unreliable since statisticalsupport for the phylogeny was weak (50 and ≤70%). Further-more, it was noted unexpectedly, a wide sequence variation ofITS1 regions between S. felis and S. neurona.

Domestic cats (F. domesticus) and wild felids are importantin the epidemiology of Sarcocystis spp. because Sarcocystisinfection has been always described in cats as a definitive host(Gillis et al., 2003). Occasionally, sarcocysts have been alsoreported in muscles of domestic cats, lion, and which therebyserve as intermediate hosts. Even though Sarcocystis sporocystsare not pathogenic to the definitive host, sporocysts excreted bythem are potentially pathogenic and can cause clinical disease inthe infected intermediate hosts. Consequently, the distinctionbetween the role of cat as a definitive and/or as an intermediatehost of Sarcocystis is important from an epidemiological pers-pective. Some animals can act as a definitive and intermediatehost in the same time but not necessarily to the same Sarcocystissp. (Dubey et al., 1989).

This study described for the first time a case of concurrentinfection with sarcocysts and sporocysts of S. felis in themuscles and intestine of an adult male cat from the State ofMichigan, respectively. The four isolates of S. felis were gene-tically identical and no intraspecific variability was observedwithin the sequences. In the present study, the finding of S. felistissue cysts in the muscles as well as sporocysts in the intestineof one cat was unexpected because the tissue cysts and sporo-cysts of this species were never reported in the same cat. Thesefindings raise questions concerning the authenticity of the cat as

451

a possible definitive host of S. felis and the capability of thisparasite to infect and cause clinical illness to the cat, which waspreviously known to be non-pathogenic. It is not clear how thiscat was infected. Due to its age, the possibility of transplacentalinfection was ruled out. A common source of horizontal infec-tion to cat is via ingestion of tissue cysts. Due to factors not yetunderstood, cats seem to be much more susceptible to infectionwith the tissue cyst but not with sporocysts of Sarcocystis.Immuno-suppression was suggested as the reason for the unus-ual presence of sarcocysts in cats (Hill et al., 1988). But, in thepresent study cats were found to be in good body condition.Additionally, there was no evidence of an impaired immunesystem as serological tests done previously for antibodies tofeline immunodeficiency virus, feline leukemia virus, and felineperitonitis virus were negative. Moreover, there was no evidencefor immuno-deficiency syndrome. Whether or not these condi-tions are involved in the present case remains unknown. Basedupon the finding that cat could be an intermediate host of S.felis, it is strongly recommended that preventive measures betaken to reduce the risk of transmission of S. felis to other poten-tial definitive hosts.

Although the present study helped to clarify the phylogeneticposition of S. felis isolates from Michigan, S. felis relationship toS. neurona isolates and other Sarcocystis sp. will not be com-pletely understood unless a larger phylogenetic study of S. felisand other Sarcocystis is conducted. Many of the phylogeneticrelationships of S. felis that have been demonstrated herein canbe expected to change as a wider taxonomic sample is intro-duced. In order to unravel taxonomic problems of S. felis,further work is necessary to assess the precise phylogeneticposition and the molecular taxonomy of the felids Sarcocystis bycollecting and sequencing more isolates. Although a clearerunderstanding of the infra-relationships of S. felis isolates couldnot be determined due to the identical sequences and the smalltaxonomic sample, the potential of these genetic regions inresolving phylogenetic and taxonomic problems of S. felis wasdemonstrated in the present study.

452

Acknowledgements

This work was financially supported by a grant from theGrayson Jockey-Club Research Foundation through ProfessorLinda S. Mansfield Laboratory at Michigan State University,USA. The authors would like to thank Dr. Frances A. Kennedyand Dr. Alice J. Murphy from Michigan State University forkindly providing the specimens. Nucleotide sequences wereobtained from Gen-Bank database (www.ncbi.nlm.nih.gov).

References

Baldwin, B.G.; Sanderson, M.J.; Porter, J.M.; Wojciech-owski, M.F.; Campbell, C.S. and Donoghue, M.J. (1995): TheITS region of nuclear ribosomal DNA: a valuable source ofevidence on angiosperm phylogeny. Ann. Missouri Botan. Gar-den, 82: 247-277.Barta, J.R. (2001): Molecular approaches for inferring evolu-tionary relationships among 240 protistan parasites. Vet. Pa-rasitol., 101: 175-186.Dubey, J.P.; Hartley,W.J. and Badman, R.T.(1989): Fatalperinatal sarcocystosis in lamb. J. Parasitol., 75(6): 980-982.Dubey, J.P; Benson, J. and Larson, M.A. (2003): ClinicalSarcocystis neurona encephalomyelitis in a domestic cat follo-wing routine surgery. Vet. Parasitol., 112: 261-267.Felsenstein, J. (1985): Confidence limits on phylogenies: anapproach using the bootstrap. Evolution, 39: 783-791.Gillis, K.D.; Mackay, R.J.; Yowell, C.A.; Levy, J.K.; et al.,(2003): Naturally occurring Sarcocystis infection in domesticcats (Felis catus). Inter. J. Parasitol., 33: 877-883.Hill, J.E.; Chapman, Jr. W.L. and Prestwood, A.K. (1988):Intramuscular Sarcocystis sp. in two cats and a dog. J. Parasitol.,74: 724-727.Hillis, D.M. and Bull, J.J. (1993): An empirical test of boot-strapping as a method for assessing confidence in phylogeneticanalysis. Sys. Biol., 42: 182-192.Kimura, M. (1980): A simple method for estimating evolution-nary rate of base substitutions through comparative studies ofnucleotide sequences. J. Mol. Evol., 16: 111-120.

453

Marsh, A.E.; Barr, B.C.; Sverlow, K.; Ho, M.; Dubey, J.P.and Conrad, P.A. (1995): Sequence analysis and comparison ofribosomal DNA from bovine Neospora to similar coccidialparasites. J. Parasitol., 81: 530-535.Marsh, A.E.; Barr, B.C.; Tell, L.; Bowman, D.D.; Conrad, P.A.; Ketcherside, C. and Green, T. (1999): Comparison of theinternal transcribed spacer, ITS-1, from Sarcocystis falcatulaisolates and Sarcocystis neurona. J. Parasitol., 85: 750-757.Mehlhorn, H.; Hartley, W.J. and Heydorn, A.O. (1976): Acomparative ultrastructural study of the cyst wall of 13 Sar-cocystis species. Protistologica, 1: 451-445.Odening, K. (1998): The present state of species-systematics inSarcocystis Lankester, 1882 (Protista, Sporozoa, Coccidia). Sys.Parasitol., 41: 209-233.Saitou, N. and Nei, M. (1987): The neighbor-joining method: anew method for reconstructing phylogenetic trees. Mol. Biol.Evol., 4: 406-425.Swofford, D.L. (2002): PAUP*: Phylogenetic Analysis UsingParsimony (*and other methods). Version 4, Sinauer Associates.Sunderland, MA.Tadros, W. and Laarman, J.J. (1982): Current concepts on thebiology, evolution and taxonomy of tissue cyst-forming eimeriidCoccidia. Adv. Parasitol., 20: 293.Tanhauser, S.M.; Yowell, C.A.; Cutler, T.J.; Greiner, E.C.;MacKay, R.J. and Dame, J.B. (1999): Multiple DNA markersdifferentiate Sarcocystis neurona and Sarcocystis falcatula. J.Parasitol., 85(2): 221-228.Tenter, A.; Barta, J.R.; Beveridge, I.; Duszynski, D.W.; Me-hlhorn, H.; Morrison, D.A.; Thompson, R.C.A. and Conrad,P. (2002): The conceptual basis for a new classification of theCoccidia. Inter. J. Parasitol., 32: 505-616.Thompson, J.D.; Gibson, T.J.; Plewniak, F.; Jeanmougin, F.and Higgins, D.G. (1997): The CLUSTAL_X windows inter-face: flexible strategies for multiple sequence alignment aidedby quality analysis tools. Nucleic Acids Res., 25: 4876-4882.