Molecular evidence of host specificity in Buxtonella-like ciliates in mandrills (Mandrillus sphinx) and related primates

Abstract

Balantioides coli (formerly Balantidium coli) is a ciliate protozoan causing balantidiasis in humans and non-human primates (NHPs), yet it remains underrecognized in zoological settings. Morphologically similar ciliates described as Buxtonella-like organisms have been reported in NHPs and are generally considered asymptomatic. In this study, we report symptomatic infection in mandrills (Mandrillus sphinx) and investigate the prevalence and genetic diversity of Balantioides spp. and Buxtonella-like ciliates across multiple NHP species housed in Polish zoological gardens. Faecal samples were collected from 70 individually identified NHPs representing 25 species across five facilities between October 2023 and September 2025 and examined using coproscopic methods and molecular analyses. While standard coproscopy detected ciliate cysts in 5.7% of samples, PCR identified infections in 18.6% of individuals, demonstrating substantially higher sensitivity. Balantioides coli was detected in several NHP species, whereas a Buxtonella-like organism was found exclusively in seven mandrills, five of which exhibited gastrointestinal symptoms. Phylogenetic analyses revealed a distinct lineage associated with mandrills, suggesting host specificity and previously unrecognised pathogenic potential. These findings challenge the assumption of asymptomatic carriage of Buxtonella-like ciliates and underscore the importance of molecular diagnostics for accurate detection and assessment of clinically relevant protozoan infections in captive NHP populations.

Introduction

The organism traditionally known as Balantidium coli and later Neobalantidium coli is currently classified as Balantioides coli, reflecting ongoing instability in its taxonomy and nomenclature¹. Although the disease caused by this small protozoan organism is often overlooked in urban areas, particularly in highly developed countries, it remains a relevant contemporary issue^1,2. Balantidiasis occurs predominantly in tropical and subtropical regions and is especially common in areas where pigs are raised^1,2. While pigs are the primary hosts, the organism may also infect a wide range of other mammalian species, including humans and non-human primates^1–20.

In both humans and non-human primates, the clinical presentation and course of infection vary considerably and depend on numerous host and environmental factors^{1,2,5,11,15,16,19,21}. The infection may range from asymptomatic carriage to severe, potentially fatal disease, particularly in immunocompromised individuals. Typical symptoms include diarrhoea and abdominal discomfort^{2,16,19,21–23}. If left untreated, infection may progress to general weakness, apathy, and other nonspecific systemic signs^2,5.

The parasite exhibits a relatively simple life cycle confined to the gastrointestinal tract, with a primary association with the large intestine^1,2,5. It exists in two morphological forms that transition between one another: the cyst, which is excreted in the faeces and subsequently ingested orally, and the trophozoite, which emerges following excystation and resides in the host’s intestine. Within the intestinal environment, trophozoites undergo encystation to produce new cysts, which are shed again in faeces and dispersed into the environment, where they may be ingested by another (or the same) host^1,2,11.

For diagnostic purposes, during periods of active shedding, cysts can be identified in stained direct smears prepared from fresh faecal material. Owing to their size, which exceeds that of most protozoa, they can be detected relatively easily, even by less experienced observers. The large trophozoites may reach up to 200 μm in length, while the slightly smaller cysts may measure up to 70 μm¹. In addition to their considerable size, their characteristic morphological features include a clearly visible macronucleus and micronucleus, and in some cases, cilia along the surface. Whenever feasible, the presence of the protozoan should be further confirmed using molecular methods^6,12,15,16.

A closely related ciliate protozoan, Buxtonella sp., is commonly found in cattle, where it causes similar gastrointestinal symptoms. Its life cycle is likewise restricted to the gastrointestinal tract^24,25. There are reports describing fatal cases of buxtonellosis in young cattle²⁴.

Despite these established diagnostic parameters, identifying these ciliates in non-traditional hosts remains challenging, as evidenced by a recent clinical case in a zoological setting.

An interesting case has arisen at one of our collaborating zoological facilities, where a troop of five mandrills (Mandrillus sphinx) presented with recurrent diarrhoea and slightly liquid stools. The animals had been exhibiting these symptoms for several months. The local veterinary team conducted repeated parasitological examinations using a flotation method, but no parasites were detected. Following these negative results, the animals were treated with charcoal and antibiotics; however, neither intervention produced lasting improvements. According to the zookeepers, the mandrills appeared weaker, displayed persistently loose stools, and seemed generally less active and more apathetic.

As part of a project conducted by our team, fresh faecal samples were collected immediately after defecation from all five mandrills housed in the enclosure. A coproscopic examination performed the following day revealed Balantioides-like cysts in four of the five samples (Fig. 1). However, subsequent molecular analysis demonstrated that the detected ciliate was not Balantioides coli as initially expected, but a Buxtonella-like organism based on BLAST comparison²⁶.

Fig. 1.

Buxtonella-like cyst in mandrill faecal sample. Representative microscopic image showing a Buxtonella-like cyst observed in a faecal smear from a mandrill (Mandrillus sphinx). The cyst displays characteristic size, shape, and internal structures consistent with ciliates morphologically similar to Balantioides spp. and Buxtonella spp.

The strong morphological resemblance between the cysts of both taxa initially hindered accurate identification, and importantly, the presence of a Buxtonella-like species in these mandrills was unexpected, particularly as it appeared to be associated with clinical symptoms - contrary to previous reports suggesting a predominantly asymptomatic course¹². In light of this unexpected and symptomatic detection, a broader investigation involving a larger number of primate species was subsequently initiated.

Results

Balantioides or Buxtonella organisms were detected using standard coproscopic methods in only four of the 70 samples, yielding a prevalence of 5,7%. Among these methods, the direct stained smear provided the most reliable visualisation of the parasite.

All of the samples underwent molecular analysis (PCR), which indicated the presence of Balantioides in 14 of the 70 faecal samples visualised in the electrophoresis. The results are summarised in Table 1. Upon sequencing, one of the samples that initially tested positive via electrophoresis did not yield sequences that could be aligned or identified, instead showing either fungal DNA or an uncharacterized ciliate. Therefore, the final prevalence of the parasite was 18.6%, more than three times higher than that observed with standard coproscopic methods. All other positive samples demonstrated high similarity to either Balantioides coli (Neobalantidium coli) or a Buxtonella-like organism. Specifically, Balantioides coli was identified in Pan troglodytes, Saimiri sciureus and Colobus guereza. In contrast, the Buxtonella-like organism was detected exclusively in seven Mandrillus sphinx individuals. What is more, only five individuals of the Mandrillus sphinx species presented symptoms; other non-human primates with either B. coli or Buxtonella-like organisms did not present any symptoms. In the case of five symptomatic mandrills, no other parasite was detected (including helminths, or most commonly detected protozoans, including Giardia sp. and Entamoeba histolytica) besides Blastocystis sp. known for mostly asymptomatic invasions^27,28.

Table 1.

PCR-positive samples subjected to sequencing and BLASTn comparison. Summary of samples that tested positive by PCR and were subsequently sequenced, with corresponding BLASTn results used for taxonomic identification of Balantioides spp. and Buxtonella-like organisms.

Lp.	Species	Zoological garden	BLAST comparison after sequencing and GenBank Accession No	% of similarity
1.	Pan troglodytes	ZOO Wrocław (Wroclaw, Poland)	Balantioides coli MZ676845	99.8%
2.	Pan troglodytes	ZOO Wrocław (Wroclaw, Poland)	Balantioides coli OM423706	100%
3.	Pan troglodytes	ZOO Wrocław (Wroclaw, Poland)	Neobalantidium coli JQ073354	99.8%
4.	Saimiri sciureus	ZOO Wrocław (Wroclaw, Poland)	Neobalantidium coli JQ073354	99.8%
5.	Colobus guereza	ZOO Wrocław (Wroclaw, Poland)	Fungi	-
6.	Mandrillus sphinx	Gdański Ogród Zoologiczny (Gdansk, Poland)	Buxtonella-like sp. JQ073338	99.3%
7.	Mandrillus sphinx	Gdański Ogród Zoologiczny (Gdansk, Poland)	Buxtonella-like sp. JQ073338	99.4%
8.	Mandrillus sphinx	Gdański Ogród Zoologiczny (Gdansk, Poland)	Buxtonella-like sp. JQ073338	99.4%
9.	Colobus guereza	Gdański Ogród Zoologiczny (Gdansk, Poland)	Balantioides coli MT252079	99.8%
10.	Mandrillus sphinx	Śląski Ogród Zoologiczny (Chorzow, Poland)	Buxtonella-like sp. JQ073338	99.8%
11.	Mandrillus sphinx	Śląski Ogród Zoologiczny (Chorzow, Poland)	Buxtonella-like sp. JQ073338	99.8%
12.	Mandrillus sphinx	Śląski Ogród Zoologiczny (Chorzow, Poland)	Buxtonella-like sp. JQ073338	99.8%
13.	Mandrillus sphinx	Śląski Ogród Zoologiczny (Chorzow, Poland)	Buxtonella-like sp. JQ073338	99.8%
14.	Mandrillus sphinx	Śląski Ogród Zoologiczny (Chorzow, Poland)	Buxtonella-like sp. JQ073338	99.8%

Several of our sequences (PX746709–PX746713) were placed within the Balantioides sp. clade. Notably, sequence PX746713, derived from Pan troglodytes, grouped closely with sequences from pigs (Sus scrofa). A significant number of the newly generated sequences (PX746702–PX746708) formed a highly supported, monophyletic sub-clade within the Buxtonella-like group. Importantly, all these sequences were isolated from Mandrillus sphinx and showed high similarity to existing GenBank sequences from other cercopithecids, such as Cercocebus agilis and Papio hamadryas. The intraspecific similarity within the Buxtonella-like Mandrillus lineage was 99.5%-100%, while Buxtonella-like sequences derived from Cercopithecus agilis differed by approximately 5% from the Mandrillus clade (Fig. 2).

Fig. 2.

Phylogenetic relationships among Balantioides and Buxtonella-like ciliates detected in non-human primates. Bayesian phylogenetic tree illustrating the relationships between Balantioides spp., Buxtonella spp., and Buxtonella-like organisms identified in this study and reference sequences obtained from public databases. Sequences generated in this study are highlighted within the tree. Star symbols indicate nodes with posterior probability support > 95%.

Discussion

Although often considered a neglected disease, the present study, together with previous reports from other facilities and countries, confirms the presence of Balantioides spp., Buxtonella spp., and Buxtonella-like organisms in non-human primates^2,4–20,29. Prevalence estimates for these ciliates vary widely: while some surveys report low infection levels around 2–15%^4,7,20,29, others document infection rates exceeding 40%^4,8,13. Clinical outcomes are equally variable. In some instances, infected animals remain asymptomatic^14,16,19. In contrast, other reports describe severe gastrointestinal manifestations, including dysentery and even fatal cases of balantidiasis in non-human primates^{6,14,16,19,21–23}. Importantly, the existing literature does not describe symptomatic infections attributed specifically to Buxtonella-like organisms^12–14, indicating that the symptomatic cases identified in the present study may represent the first documented occurrence of a clinically relevant Buxtonella-like infection in non-human primates. These findings collectively argue that, although sometimes overlooked as commensals, these ciliates warrant closer attention.

Numerous reports have described what was presumed to be balantidiasis in non-human primates^{6,16,19,21–23}. However, the absence of molecular diagnostic tools meant that all mild, severe cases, and asymptomatic cases were frequently attributed to Balantioides coli^{4,7–10,13,18,19,22,29}. More recent studies, however, have revealed that organisms morphologically similar to Balantioides coli under standard coproscopic examination may in fact represent a genetically distinct taxon closely related to Buxtonella sp. (Buxtonella-like protozoan)^6,12–14. While some authors have proposed microscopic features to differentiate these organisms^9,14,18,29, others contend that reliable discrimination is difficult or impossible based on morphology alone¹². Some studies explicitly classify these organisms as Buxtonella sp^20,25,29., whereas others, relying solely on flotation, sedimentation, or direct smears, suspect the presence of Buxtonella sp. or Buxtonella-like but cannot differentiate it with certainty due to morphological overlap^13,29. Thus, molecular approaches are essential for distinguishing this organism^12–14.

A particularly compelling line of evidence arises from host distribution patterns. Whereas Balantioides coli has been reported across multiple primate families and genera, the Buxtonella-like and Buxtonella sp. organisms appears restricted to only four genera^6,12–14. Notably, Mandrillus sp., Cercocebus sp., Macaca sp. and Papio sp. are among the most closely related primates according to current phylogenies^30,31. The clustering of infections exclusively within this genetically cohesive group suggests a degree of host specificity that is unusual for classical Balantioides sp. or Buxtonella sp. This pattern provides a biologically plausible argument that the organism historically labelled as “Buxtonella-like” may represent a distinct species adapted to a narrow clade of Old-World Primates rather than a variant or misidentified form of known taxa. What is more, a rather separate lineage closely related to each other is restricted to Mandrillus sphinx species. Such host specificity strongly supports the hypothesis that a new species may be involved. However, given that our analysis is based on a single molecular marker, these findings should be interpreted as hypotheses requiring confirmation through a multilocus molecular approach and detailed morphological analyses.

Following the initial detection of a Buxtonella-like organism in four of the five symptomatic mandrills described above, a broader study was undertaken across a larger primate cohort. The findings also highlight methodological challenges that limit detection accuracy. Coproscopic detection is influenced by sample condition: fresh samples or those preserved under cooling, rather than frozen, improve the likelihood of identifying these organisms³². Among the methods used, direct stained smears provided the clearest visualization of both Balantioides spp. and Buxtonella-like ciliates.

Several limitations of the study should be considered. Only small amounts of faecal material were available for microscopic examination, and only 20 of 76 samples were supplied fresh or cool-preserved, which likely influenced the coproscopic detection rate. Expanding the sample volume and increasing the proportion of unfrozen samples would likely improve detection sensitivity, capture a broader spectrum of genetic diversity, and help clarify whether the Buxtonella-like organism indeed represents a novel species. Given that this organism has been detected only within a narrow group of Old-World primates to date, its zoonotic potential appears limited; however, confirmation will require studies involving larger and more diverse host populations. The study of free-living mangabeys conducted by Pafčo et al.¹³ revealed the presence of a Buxtonella-like ciliate in non-human primates. Still, it did not detect it in the human population living in close contact with these NHPs, whereas other parasites were identified in both species, suggesting active transmission. This, together with the current finding that the parasite occurs in only a very narrow group of NHPs, argues against substantial zoonotic potential; however, as also suggested by Yan et al.²⁵, further studies are required to address this question more definitively.

This study documents the detection and characterization of ciliate protozoans, specifically Balantioides coli and a Buxtonella-like organism, in non-human primates (NHPs) housed in Polish zoological gardens. A symptomatic synchronus occurrence in mandrills (Mandrillus sphinx) led to the identification of a Buxtonella-like ciliate, previously assumed to be asymptomatic, demonstrating for the first time its association with clinical gastrointestinal disease in NHPs.

Our findings highlight several key points: (1) Diagnostic challenges: Standard coproscopic methods detected only a fraction of infections (5.7%), whereas molecular techniques, particularly polymerase chain reaction (PCR) and sequencing, revealed a nearly threefold higher prevalence (18.6%). This highlights the greater sensitivity and more accurate detection provided by molecular tools, which is essential for distinguishing between morphologically similar taxa.; (2) Novelty and host specificity: The Buxtonella-like organism was exclusively detected in Mandrillus sphinx individuals, five of which were symptomatic, and absent in other primate species. Phylogenetic analyses and host distribution patterns suggest that this organism may represent a distinct lineage or potentially novel species adapted to a narrow clade of Old-World primates; (3) Neglected but clinically relevant: While often overlooked in zoological and wildlife health studies, these ciliates can be associated with significant gastrointestinal symptoms and general health impacts in susceptible hosts. Recognition of their presence is important for animal health monitoring and management; (4) Zoonotic potential: Based on current evidence, including host restriction and absence in human contact populations, the zoonotic risk appears limited, though further investigation in larger and more diverse host populations is warranted.

In conclusion, this study emphasises the importance of combining morphological and molecular approaches for accurate parasite detection, documents the first symptomatic occurrence of a Buxtonella-like ciliate in mandrills, and highlights the need to reassess the clinical and epidemiological relevance of neglected protozoans in captive NHP populations.

Methods

Study population and sample collection

For this study, faecal samples were collected from non-human primates (NHPs) housed in five zoological gardens across Poland (Gdański Ogród Zoologiczny, Śląskie Zoo, ZOO Wroclaw, Nowe Zoo Poznań and Ogród Zoobotaniczny w Toruniu), with sampling conducted from October 2023 to September 2025. In total, 70 samples were obtained from 70 individual animals, representing 25 species. All of the individuals are listed in Table 2.

Table 2.

Non-human primate species included in the study. Overview of the non-human primate species sampled across participating zoological institutions, including the number of individuals examined per species.

Lp.	Latin species name	English name	Number of individuals
1.	Alouatta caraya	Black Howler Monkey	4
2.	Ateles geoffroyi vellerosus	Black-handed Spider Monkey	3
3.	Callithrix jacchus	Common Marmoset	1
4.	Cebuella pygmaea	Pygmy Marmoset	1
5.	Cercocebus torquatus	Red-capped Mangabey	1
6.	Colobus guereza	Eastern Black-and-white Colobus (Guereza)	4
7.	Colobus polykomos	Western Black-and-white Colobus	4
8.	Erythrocebus patas	Patas Monkey	5
9.	Hylobates lar	Lar Gibbon / White-handed Gibbon	1
10.	Hylobates pileatus	Pileated Gibbon	1
11.	Lemur catta	Ring-tailed Lemur	7
12.	Macaca fuscata	Japanese Macaque	8
13.	Macaca nigra	Crested Black Macaque	1
14.	Mandrillus sphinx	Mandrill	8
15.	Nomacus leucogenys	Northern White-cheeked Gibbon	1
16.	Pan troglodytes	Common Chimpanzee	3
17.	Papio cynocephalus	Yellow Baboon	1
18.	Pongo pygmaueus	Bornean Orangutan	1
19.	Saguinus imperator	Emperor Tamarin	2
20.	Saguinus oedipus	Cotton-top Tamarin	2
21.	Saimiri sciureus	Common Squirrel Monkey	1
22.	Semnopithecus entellus	Northern Plains Gray Langur / Hanuman Langur	2
23.	Trachypitecus auratus auratus	Javan Langur	4
24.	Varecia varegiata	Black-and-white Ruffed Lemur	3
25.	Varecia rubra	Red Ruffed Lemur	1

Samples were collected from animals immediately following defecation to preclude cross-contamination. To facilitate subsequent statistical analysis, pertinent data for each individual were recorded at the time of collection. This information included the animal’s species, sex, age, social group size, and details of its most recent antiparasitic treatment, including the date administered and the substance used. Any known comorbidities or other relevant clinical information were also noted.

If the samples could be collected by the researchers from the zoological garden and examined within 72 h after defecation, they were kept in cooling conditions (+ 2 to + 8 Celsius degrees). If collection and examination could not be completed within this period, each sample was placed in a sterile container and frozen at -20 °C. Samples remained frozen without chemical preservatives, such as ethanol or formalin, until they were processed in the laboratory.

Sample collection and macroscopic examination

Each faecal sample was subjected to macroscopic examination before processing. The consistency, colour, and structure of each sample were recorded, along with any abnormalities that might indicate concomitant disease. Microscopic analysis was performed on each sample afterwards. The methods included direct stained smear and the modified flotation method^33,34.

For the smear, a small amount (approximately 2 × 2 mm) of faeces was placed on a microscopic slide, mixed with a drop of 0,9% saline solution for better visualisation and a drop of 3–5% iodine solution. It was examined under the microscopic slide after placing a cover slide on the top³⁴.

A sample of 2 g of preserved stool was weighed using a portable scale. 28 ml of tap water (or bottled still water) was added. The suspension was stirred and sieved through a 0,8 mm aperture mesh. The solution was then placed in plastic probes, closed and centrifuged for 5–6 min. The flotation solution was prepared using water, sodium chloride and saccharose. After the removal of the supernatant, the flotation solution (specific gravity 1,3 mg/ml) was added, and a cover slide was placed in the top of the filled to the top probe³³. The microscopical examination was done after 5–10 min. Presence of any visible parasite form was noted.

DNA extraction, PCR amplification and sequence analysis

Genomic DNA was isolated from approximately 200 mg of each faecal sample using a commercial extraction kit - EURx GeneMATRIX Stool DNA Purification Kit - following the manufacturer’s protocol. The purified gDNA was stored at -20 °C until required for molecular analysis. The starters coding the ITS region (i.e., ITS1–5.8s-rRNA–ITS2 region) of rDNA were chosen based on the literature^6,17,25. Therefore, the starters B5D (5’GCTCCTACCGATACCGGGT’3) and B5RC (5’GCGGGTCATCTTACTTGATTTC’3) were selected for a simple PCR reaction. The reactions were performed using the T100 Thermal Cycler (Bio-Rad). For each reaction, distilled water served as a negative control, used to detect possible reaction contamination³⁵. Polymerase chain reaction was carried out using an initial denaturation at 94 °C for 10 min, followed by 30 amplification cycles consisting of denaturation at 94 °C for 1 min, annealing at 60 °C for 1 min, and extension at 72 °C for 1 min. A final extension step was performed at 72 °C for 5 min, then the reaction was cooled to 4 °C^6,17. Electrophoresis in 1,0% agarose gel, stained was used for the PCR products. A commercial purification kit was used to purify the selected products, which were visualised during electrophoresis. The products were then stored at + 4 °C²⁹. Positive PCR products were sequenced bidirectionally using standard Sanger sequencing. Raw sequence data will be edited and assembled into consensus contigs using DNA Baser (Heracle BioSoft SRL). The resulting sequences were used for parasite identification and further analyses.

The PCR sequences were compared with GenBank sequences using BLASTn, and alignments were refined with the ClustalW algorithm. Selected sequences from GenBank, together with our own sequences, were utilised to determine species through phylogenetic analysis. This analysis was conducted using a Bayesian inference approach (MrBayes 3.2), applying the GTR + INVGAMMA substitution model with 2 million replicates³⁶.

Author contributions

Weronika Hildebrand conceived and designed the study, coordinated and conducted the research, prepared the manuscript, and compiled the tables and figures; the author also acquired the photographic documentation presented in Fig. 1. Authors Michał Załuski, Mirosław Kalicki, Wojciech Paszta and Tomasz Jóźwik provided biological material and essential data on the animals, enabling the execution of the study. Kamila Kot contributed to laboratory procedures and sample processing. Grzegorz Zaleśny performed the data analyses, including phylogenetic analyses, contributed to manuscript writing and critical revision of the text, and supervised the overall research process and manuscript preparation. All authors reviewed and approved the final manuscript.

Funding

The article is part of a PhD dissertation titled “Chronic stress and parasitic infections in captive non-human primates in Poland”, prepared during Doctoral School at the Wrocław University of Environmental and Life Sciences. The APC/BPC is financed by Wrocław University of Environmental and Life Sciences.

Data availability

The datasets (DNA sequences) generated during the current study are available in the GenBank repository (https://www.ncbi.nlm.nih.gov/genbank/), under the following accession numbers: PX746702–PX746713.

Declarations

Competing interests

The authors declare no competing interests.

Ethics declaration

All procedures were conducted in accordance with applicable local and national regulations for animal research. Approval from a local ethical committee was not required, as the study was entirely non-invasive and involved no direct contact with the animals; all samples were collected without handling or interfering with the animals.