INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Babesia spp. are intraerythrocytic hematozoa which occur in a wide variety of mammalian hosts and are transmitted by various species of ticks. Certain species have emerged as significant human pathogens in recent years, including the rodent species Babesia microti in North America, the bovine species Babesia divergens in Europe, and as-yet-unnamed species, referred to as WA1, MO1, and TW1, found in the western United States, south-central United States, and Taiwan, respectively (14, 15, 29, 32, 36, 37, 39). The spectrum of infection in humans includes subclinical to fatal infection, the latter occurring most commonly in individuals who are immunocompromised, most notably secondary to asplenia.

Recently we observed fatal hemolytic crises in two profoundly immunosuppressed baboons (Papio cynocephalus) undergoing experimental stem cell transplantation (one allogeneic and one autologous) (2). Greater than 50% of their erythrocytes were found to be infected with typical Babesia-like organisms. Both baboons received blood transfusions from two donors, one of which was shown to be parasitemic, with very rare and delicate intracellular hematozoa most suggestive of Entopolypoides macaci, a Babesia-related parasite described in African and Asian nonhuman primates and, rarely, in humans (8, 12, 13, 23, 24, 26). Detection of these parasites in animals with intact and abrogated immune systems raised important questions regarding their characterization and origins, whether they represented variants of the same species, their prevalences in colony animals, and mechanisms of transmission. The potential effects on research programs of such acute and chronic infections in colony animals are significant. Also, the increasing use of baboon xenografts in humans makes the identification of such infections prior to transplantation mandatory to prevent possible transmission of a zoonotic agent (10).

In this paper we present preliminary findings on the morphologic characterization, prevalence, and phylogenetic affiliation of a piroplasm found to be enzootic in baboons in our breeding facility, some of which originated from a second breeding facility in the United States while others were from foreign sources.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Animals studied. The two baboons in which hematozoa were first noted had undergone total-body irradiation followed by stem cell transplantation (one autologous and one allogeneic); both animals were subsequently treated with cyclosporine and were noted to be profoundly immunosuppressed (2). Sixty-five additional baboons (P. cynocephalus) were subsequently examined as part of this study. Twenty-three of these animals were born and raised in the Medical Lake, Wash., breeding facility administered by the Regional Primate Research Center (RPRC) located at the University of Washington in Seattle. Another 26 originated from the Southwest Foundation for Biomedical Research in San Antonio, Tex., and were shipped directly to the RPRC without spending time at the Medical Lake breeding facility; animals at both breeding facilities were held in outdoor enclosures and were known to comingle. Sixteen animals were imported from Africa (13 from Ethiopia, 2 from Kenya, and 1 from an unspecified African location) and shipped to the RPRC in Seattle. All animals were housed and cared for in our facilities in accordance with accepted standards (Association for Assessment and Accreditation of Laboratory Animal Care) for laboratory animal care.

Parasite characterization by serologic methods. Air-dried blood smears from the index case were forwarded to the U.S. Department of Agriculture (Don Knowles), Pullman, Wash., for examination with fluorescein-tagged monoclonal antibodies raised against Babesia bovis, Babesia equi, Babesia caballi, and Babesia bigemina in a direct fluorescent-antibody assay. Sera from six smear-positive and six smear-negative baboons were forwarded to the Centers for Disease Control and Prevention (Marianna Wilson) for analysis of reactivity against B. microti antigens in an indirect immunofluorescent assay (IFA) (4). Additional sera were tested for reactivity to the antigen of WA-1, also in an IFA, at the University of California, Davis (Patricia Conrad) (36).

Genomic DNA preparation and amplification of nss rDNA. Whole-blood DNA from a baboon chronically infected with E. macaci was extracted with the Isoquick DNA extraction kit (Orca Research Inc., Bothell, Wash.). The precipitated DNA was resuspended in 50 µl of 10 mM Tris buffer, pH 8.0 (100× stock from Sigma, St. Louis, Mo.). Piroplasm-specific DNA was amplified with 18sA and 18sC universal primers, which are directed against highly conserved portions of the eukaryotic nuclear small-subunit ribosomal DNA (nss rDNA) and which have been used successfully to detect other Babesia-like DNA sequences, such as WA1 (5, 29, 39). Five microliters of extracted DNA was used in a 50-µl PCR mixture. The reaction mixture contained 10 mM Tris (pH 8.3), 1.5 mM MgCl₂, 50 mM KCl, 0.001% gelatin, 200 mM (each) deoxyribonucleoside triphosphate (Boehringer Mannheim, Mannheim, Germany), 10% glycerol (Sigma), 0.5% Tween 20 (Sigma), 0.25 U of Ampli-Taq polymerase (Perkin-Elmer/Roche, Branchburg, N.J.), and 50 pmol of each of the primers. Amplification and subsequent isolation and visualization of the PCR product were performed as described previously (5).

Direct sequencing of PCR products. PCR products were prepared for sequencing as previously described (17). Five hundred base pairs of the PCR product were directly sequenced multiple times with both of the PCR primers (18sA and 18sC) and internally located piroplasm consensus sequence primers. Sequencing was performed with a model 373 or 377 automated DNA sequencing instrument (Perkin-Elmer Applied Biosystems Division, Foster City, Calif.). Overlapping sequences were aligned with Assemblylign sequence assembly software (Oxford Molecular, Campbell, Calif.) to generate a contiguous consensus sequence.

Phylogenetic analysis. The contiguous consensus sequence obtained for the baboon babesial species was used as data input to search the GenBank sequence database, using the FASTA algorithm (11) to find related sequences. B. bovis (accession no. L19077), Babesia canis (L19079), B. divergens (U16370), B. equi (Z15105), Babesia gibsoni (L13729), B. microti (U09833), Babesia rodhaini (M87565), Theileria annulata (M64243), Theileria buffeli (Z15106), Theileria parva (AF013418), Theileria sergenti (AB000271), Theileria taurotragi (L19082), and WA1 (L13730) were used in the alignment. The sequences were aligned by using the Pileup program of the Wisconsin package (11). Phylogenetic analysis of the alignment was performed as described previously (17) with the Molecular Evolutionary Genetics Analysis program, version 1.01 (20), to make a Jukes-Cantor distance measurement and perform a neighbor-joining analysis with 500 bootstrap replicates. The Phylogenetic Analysis Using Parsimony program, version 3.1.1 (38), was used to confirm the order observed by the neighbor-joining analysis (using a branch-and-bound algorithm with 100 bootstrap replicates).

Nucleotide sequence accession number. The recovered partial nss rDNA sequence has been deposited in GenBank under accession no. AF081465 (Babesia sp. strain PB-1 from P. cynocephalus).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Morphological characterization. In the acute clinical infection seen during hemolytic crisis, we observed numerous intraerythrocytic ring-shaped and pyriform trophozoites; merozoites forming pairs and tetrads, the latter displaying occasional Maltese cross arrangements; and multiple infections of single erythrocytes (Fig. 1A). Individual trophozoites measured on average 1.8 × 3.5 µm and contained a small, dense chromatin mass usually oriented toward the pointed end. Hemoglobin breakdown products were not present, and the host erythrocyte was not stippled or enlarged. Greater than 50% of erythrocytes were parasitized by the seventh day following the onset of acute hemolytic crisis, and extracellular forms were frequently seen. Morphologically, this parasite resembled a typical Babesia sp. organism and was consistent with a description in older literature of Babesia pithici, a species found infecting cercopithicine monkeys and baboons (Table 1) (34). Whether the parasitemia represented recrudescence of a preexisting infection or was acquired via transfusion from the known positive animal could not be determined.

View larger version (118K): [in this window] [in a new window]   FIG. 1.   (A) Giemsa-stained thin blood film from an immunosuppressed baboon in severe hemolytic crisis secondary to fulminant Babesia infection. Note the presence of single intracellular ring-shaped trophozoites, typical piroplasms, and merozoites forming a tetrad or Maltese cross (arrow); (B) Giemsa-stained thin blood film of a healthy baboon displaying very rare accolé forms (arrow) consistent with E. macaci; (C and D) Wright's-stained thin blood film from a baboon which had undergone splenectomy 30 days previously and had developed spontaneous parasitemia with E. macaci. Numerous delicate ring trophozoites, accolé forms (arrow in panel C), a tetrad (arrow in panel D), and erythrocytes containing multiple forms are seen. Bar = 12 µm.

Prevalence studies. Blood smear examination of 65 animals revealed an overall infection prevalence with E. macaci of 31%: 30% (9 of 23) of animals born at the breeding facility at Medical Lake were positive; 35% (9 of 26) of animals shipped to the RPRC in Seattle from Southwest Foundation, and which were not held in the breeding facility at Medical Lake, were positive; and 12% (2 of 16) of the animals imported from Africa were positive. Thirteen of these 16 African animals originated from Ethiopia, all of which were smear negative; while records are incomplete, at least one of the 2 positive animals originating from Africa came from a supplier in Kenya.

Serologic characterization. Blood smears from the index case were tested with fluorescein-labelled species-specific monoclonal antibodies directed against bovine (B. bovis and B. bigemina) and equine (B. equi and B. caballi) species of Babesia in a direct immunofluorescence procedure; no positive reactions were observed. Sera from six baboons positive by thick- and/or thin-smear examination and six smear-negative baboons were tested against B. microti antigen in an IFA: all reactions were negative, although sera from two smear-positive baboons gave titers of 1:16, which were considered to be inconclusive. The presence of this equivocal reactivity could not be correlated with the degree of parasitemia due to the rarity of parasites seen in blood smears. IFA of sera from smear-positive baboons with antigen of strain WA-1 was also negative.

Analysis of nss rDNA sequences. Phylogenetic analyses were performed on sequences obtained from the nss rDNA segments amplified by PCR from baboon blood samples. The region chosen for these analyses was an ~500-bp 5' portion of the nss rDNA determined previously to contain both alignable well-conserved regions and variable regions, and it was consequently considered to be an appropriate representative sequence (7, 29). Thirteen other related sequences were included in the analyses as listed in Materials and Methods. B. bovis, B. canis, and B. divergens are more distantly related and were used as outgroups (Fig. 2). The analyses indicate that PB-1 (for Papio-Babesia-1) is most closely related to B. microti (97.9% sequence similarity). Both the neighbor-joining and branch-and-bound parsimony analyses supported the clustering of PB-1 and B. microti in most of the bootstrap replicates (99 and 96%, respectively). Both analyses also supported the branch order shown in Fig. 2.

View larger version (10K): [in this window] [in a new window]   FIG. 2.   Neighbor-joining analysis showing evolutionary relationships of 13 Babesia and Theileria species with PB-1; B. bovis, B. canis, and B. divergens represent outgroups. PB-1 appears to be most closely related to B. microti, with a sequence similarity of 97.9%. The percentages of 500 bootstrap replications in which designated groupings of species appeared are noted above the branches.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

An incidental finding of lethal babesiosis in two immunocompromised baboons led to the discovery of enzootic chronic babesiosis in feral baboons of African origin and in baboons born and raised in two breeding colonies maintained in North America. The recognition of this parasitic infection raised questions as to the nature of the etiologic agent, origin, mechanism(s) of transmission, prevalence, and significance to animal and, possibly, human health. Based on the phylogenetic, serologic, and epidemiological studies presented here, and on a review of pertinent literature, we suggest that the genus Entopolypoides is synonymous with that of Babesia and that the morphological variations seen are occurring within a single species as a function of changes in host immune status.

Among described genera of apicomplexan hematozoa known to infect nonhuman primates (Plasmodium, Hepatocystis, Entopolypoides, and Babesia), morphological characteristics of the organisms seen in smears from two baboons which experienced fulminant hemolytic crises secondary to induced immunosuppression are typical for members of the genus Babesia (33). While reference was made by Ross in 1905 to the presence of B. pithici in Cercopithecus and Cercocebus monkeys and baboons, no further reports of babesiosis naturally occurring in baboons have been made, making the status of this species uncertain (34).

Unlike the typical babesial pyriforms seen during acute infection, blood stage forms seen in chronically infected, but otherwise healthy, baboons were significantly similar to the published descriptions of E. macaci. This species was originally described as a Babesia-like parasite from M. irus monkeys in Java but differed from typical Babesia spp. by consisting of small rings and amoeboid forms and by lacking the common occurrence of pyriform trophozoites (23). Hawking found a similar Entopolypoides parasite in a high proportion of Cercopithecus primates of African origin held at the Delta Regional Primate Research Center in Covington, La., and carefully described a variety of morphological forms which encompasses those we have seen in both acutely and chronically infected baboons (13). Subsequently, Kuntz and Moore found feral baboons in Kenya infected with a small, intraerythrocytic parasite which they also called E. macaci (21). In most naturally occurring Entopolypoides infections, parasitemias remained low and pathogenic effects were slight or unnoticeable unless the animal had undergone splenectomy, in which case high parasitemias usually did result but, again, with few apparent side effects (8, 12, 13, 16, 24). These published clinical observations of African and Asian primates closely parallel our experiences with baboon infections. Two cases of human infection with Entopolypoides spp. have also been reported, with both individuals suffering from hepatic dysfunction and one from asplenia, conditions known to predispose to severe piroplasmosis (42). This is of particular interest should these parasites be related to the Babesia spp. present in baboons.

On the basis of published biological, clinical, and morphological similarities between Babesia spp. and Entopolypoides spp., the taxonomic validity of the latter has been questioned, with Levine speculating that the two are synonymous (12, 13, 22). The data we present in support of synonymy comes from several sources. Clinically, the high prevalence of asymptomatic infection seen in colony baboons along with recrudescence following immunosuppression appears to be typical of natural infection with E. macaci in macaques, cercopithicine monkeys, and baboons. Most importantly, the phylogenetic analysis of a ribosomal gene sequence of Entopolypoides parasites recovered from the blood of a chronically infected baboon has been shown to be closely related to those of other species of Babesia, and specifically to B. microti (Fig. 2). Furthermore, sera from acutely and chronically infected baboons all reacted strongly in an IFA to the antigen harvested from the peripheral blood of a splenectomized baboon chronically infected with E. macaci, suggesting that the two morphological forms are the same, or closely related, antigenically (1, 3). While the variety of morphological forms of piroplasms seen in blood from acutely and chronically infected baboons thus appears to represent one species, whether they represent the previously described species, B. pithici and E. macaci, respectively, cannot be readily answered due to insufficient published information.

Additional analyses of phylogenetically meaningful molecules (e.g., nss rDNA) of Entopolypoides parasites known to infect macaques, Cercopithecus monkeys, and humans and their comparison with the PB-1 sequence would prove especially helpful in clarifying whether these parasites are identical to each other and to PB-1 or whether they represent different species. The use of such phylogenetic studies has been an important adjunct in the understanding of human infections caused by different species in the family Babesiidae; morphological analyses alone appear to be incapable of making such distinctions (14, 15, 18, 19, 28-30, 39).

The finding of 31% of baboons overall infected with PB-1 on the basis of blood smear evaluations and the presence of the parasite in animals born and raised in three different geographic locations, including 12% of those directly imported from Africa, points to the origin of the parasite in the animals' ancestral homelands, with continued, effective, transmission occurring in breeding colonies in North America. This is consistent with the findings of Moore and Kuntz, who found 13% of baboons surveyed in central Kenya positive for E. macaci (24). While this preliminary epidemiological data does not permit us to more specifically address what mechanisms of transmission may be involved, either horizontal (mechanical or vectoral) transmission, vertical (maternal-fetal) transmission, or a combination of the two is possible. In all described life cycles of Babesia and Theileria spp., transmission is nominally accomplished through the bite of an appropriate, infected tick vector. While tick transmission is undoubtedly involved in the animals' natural habitats, the geographic distances between the two U.S. breeding facilities, differences in local tick vectors, absence of ticks or other ectoparasites on any animal, and the continuous indoor caging of some animals suggest that additional mechanisms of transmission should be considered. Congenitally acquired infections are known to occur with some Babesia spp. infecting equine and human hosts (9, 43). E. macaci has also been detected in five baby baboons born to dams imported to the United States from Kenya, supporting the concept of vertical transmission as an epidemiological component of the infection (24).

The rarity of trophozoites in the peripheral blood of otherwise healthy baboons demonstrated that estimation of the prevalence of PB-1 infection based upon thick- and thin-smear evaluations may grossly underestimate the actual prevalence of infection in colony animals; the incidental finding of babesial recrudescence in immunosuppressed baboons further confirmed the presence of latent babesiosis. The lack of clinical symptoms and rarity of parasitemia in many B. microti-seropositive humans in areas of endemicity suggest that serologic assays may be a more accurate indicator of infection than smear evaluation (4, 18, 30, 31). Molecular methods of detection have also demonstrated that piroplasm-specific DNA may be recovered from individuals with previously unrecognized infection, revealing that a chronic carrier state, previously described in animals, may occur in humans as well (18, 29, 35). Considering this likelihood, we are currently investigating the use of a serologic assay as a more sensitive indicator for detecting Babesia infections in baboons (1, 3). The use of such an assay will be important in the management of our baboon breeding colony and may also provide additional epidemiological data from which mechanisms of transmission may be inferred and, ultimately, proven.

The finding of an animal model of babesiosis which, through experimental manipulation, closely mimics acute and chronic babesiosis in humans may offer unique opportunities to study pathophysiology, diagnostic approaches, therapeutic interventions, and the clinical outcomes of interactions with other simultaneously transmitted tick pathogens (e.g., Borrelia burgdorferi and Ehrlichia spp.) (19, 27). Also, the enzooticity of babesiosis in baboons and its phylogenetic similarity to B. microti, a recognized human pathogen, mandates screening of donor animals prior to xenotransplantation to prevent the accidental introduction of a potentially lethal blood parasite into an immunocompromised host (10).