Authors Lisette van Lieshout (corresponding author), Johanna M. de Gruijter, Jaco J. Verweij, Eric A. T. Brienen and A. M. Polderman, Department of Parasitology, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands. Fax: 31 71 526 6907; E-mail: E.A.van_Lieshout@lumc.nl Michael Adu-Nsiah and Michael Haizel, Wildlife Division, Accra, Ghana. Robin B. Gasser, Department of Veterinary Science, The University of Melbourne, Melbourne, Australia.
In northern Togo and Ghana, human infection with the parasitic nematode Oesophagostomum bifurcum is of major health importance. Elsewhere, oesophagostomiasis is considered a zoonotic infection, non-human primates being the natural host. We examined 349 faecal samples of the olive baboon, mona monkey and black and white colobus monkey from two geographically distinct areas in Ghana, outside the region endemic for O. bifurcum in humans. Using both microscopy and species-specific PCR, we found a high prevalence of O. bifurcum (75–99%) in olive baboons and mona monkeys. The majority of the test-positive faecal samples contained large numbers of larvae after copro-culture (>100). No O. bifurcum was detected in the faeces of the black and white colobus monkeys. Observational studies on the behaviour of the non-human primates, focusing on defecation, food consumption and the sharing of habitat with the local human population, indicated favourable conditions for zoonotic transmission. Given that no human infection with O. bifurcum has been reported from either study area, the present findings support the hypothesis that O. bifurcum from humans in the north of Ghana, and O. bifurcum from olive baboons and/or mona monkeys are distinct.
Human infection with Oesophagostomum bifurcum (Nematoda: order Strongylida) is thought to be a rare zoonosis, but it is endemic and of major human health importance in northern Togo and Ghana (Polderman et al. 1991, 1999). In these countries, at least a quarter of a million people are infected with this geo-helminth, and in some villages the prevalence of infection is almost 70% (Pit et al. 1999; Yelifari et al. 2005). Infection with O. bifurcum causes significant disease due to the formation of granulomata and caseous nodules in the wall of the large intestine, produced by encysted larvae.
Non-human primates are the natural host of O. bifurcum and many species of monkeys and baboons are known to be infected with Oesophagostomum (Orihel & Seibold 1972). It could be assumed that monkeys serve as reservoir hosts in the area endemic for human oesophagostomiasis, but in northern Ghana and Togo the number of non-human primates has been greatly reduced and the remaining numbers are so small that they are now unlikely to play a major role in the local transmission of disease in humans (Polderman & Blotkamp 1995). Elsewhere in Ghana there are still areas where large groups of non-human primates live in close association with humans. For instance, in Mole National Park (MNP) (Northern Region, central Ghana), numerous olive baboons live in and around the human settlements, and in the villages of Baobeng Fiema (Brong Ahafo Region, central Ghana), mona monkeys and black and white colobus are consistently present. Also, both of these places represent tourist attractions and are frequently visited by international travellers. However, to date no human infection with O. bifurcum has been reported in these areas.
While the prevalence and distribution of human oesophagostomiasis in northern Ghana have been studied extensively (Krepel et al. 1992; Pit et al. 1999; Storey et al. 2000; Ziem et al. 2004; Yelifari et al. 2005), it is unclear to what extent the non-human primates in this country harbour O. bifurcum and serve as a potential reservoir for human infections. In the present study, faecal samples from olive baboons from MNP, and from mona monkeys and black and white colobus monkeys from Baobeng Fiema were examined for the presence of O. bifurcum by microscopy and species-specific PCR. Furthermore, observations were made of the behavioural interaction between the non-human primates and the villagers, focusing particularly on patterns of defecation and food consumption by the monkeys and baboons.
Materials and methods
Study areas and design
The present study was conducted at two widely separated sites in Ghana, namely the village Yipala, situated in MNP (Northern Region), and the two neighbouring villages Baobeng and Fiema in the Baobeng Fiema Monkey Sanctuary (BFMS) (Brong-Ahafo region) (Figure 1). At these sites, three species of non-human primates were studied, the olive baboon (Papio anubis) in MNP and the mona monkey (Cercopithecus mona) and the black and white colobus monkey (Colobus vellerosus) in BFMS (Teichroeb et al. 2003). Surveys in MNP were carried out from October 1999 till October 2000, with an interruption between February and April and in August 2000. In BFMS the study period ranged from October 1999 until January 2000. The surveys were performed on a monthly basis, with a time period of 7–15 consecutive days spent at each of the study sites. During these monthly periods, direct observations and faecal sample collection of the monkeys and baboons took place.
Yipala is situated at the top of the escarpment in the South East corner of MNP (9°35′ N and 2°26′ W). The ∼800 inhabitants of the village are families of the employees of MNP. In the village there is a primary school, a health post, training facilities for the National Wild Life Department, and a small hotel for (international) visitors. MNP is Ghana's largest (∼4000 km2) and most popular game reserve, known for its elephant herds. It consists largely of open savannah woodland. Up to 100 species of mammals have been reported in this park, including five species of non-human primates (P. anubis, Erythrocebus patas, Ceropithecus aethiops, Colobus polykomos and Galago senegalensis). In and around Yipala, the olive baboon is the predominant species of non-human primate.
BFMS is located in the district of Nkoranza, ∼200 km south of MNP and 20 km north of the town of Nkoranza (7°43′ N and 1°42′ W). The area (∼1.9 km2) is flat with a gentle slope, and consists mainly of partially degraded forest (Fargey 1991). The sanctuary encompasses two villages, namely Baobeng (∼1000 inhabitants) and Fiema (∼2000 inhabitants), which are situated <1 km from each other. The community belongs to the Brong–Ahafo region and mainly lives from small farming activities. Both villages have a tradition of religious worshipping the two local species of monkeys, i.e. the mona monkey and the black and white colobus. In the 1970's the sanctuary was founded to safeguard the monkeys and to stimulate tourism (Bartussek 1998).
Sample and data collection
Non-human primates were observed on a daily basis, and their behaviour was noted in a descriptive qualitative matter. These observational studies focused on places of sleeping, defecation, and food collection of the animals, and physical interaction with human activities.
Faecal samples were collected, mostly on a daily basis, by picking fresh droppings from the ground at sleeping places of the non-human primates. In FBMS, faecal samples were also collected from places where the monkeys scavenged for food. All sampling was carried out under the supervision of a professional game warden.
Faecal sample examination
Within 24 h of collection, a duplicate culture of each faecal sample was carried out to identify third stage larvae (L3) of O. bifurcum (Polderman et al. 1991). In brief, 2 g of faeces were mixed with an equal amount of ‘vermiculite’ and placed on filter paper in a Petri dish under moist conditions for 1 week. Culture fluid was collected and 100 μl of sediment was microscopically examined. L3s were identified morphologically using published keys and description (Chabaud & Larivière 1958; Blotkamp et al. 1993).
Faecal DNA samples were also subjected to specific PCR for the detection of O. bifurcum. For this purpose samples were frozen within 24 h after collection and subsequently transported to the Netherlands. DNA isolation and species-specific PCR were performed as described recently (Verweij et al. 2000).
Observations in the mole national park
In the MNP, two groups (A and B) of baboons were observed in and around the village of Yipala. Group A consisted of ∼15 olive baboons, while group B had ∼70 animals. During most of the day, group B was joined by a small number (n = 2–5) of patas monkeys. Although both groups moved through the entire study area, group A was mostly seen around the primary school, while group B was regularly spotted within the centre of the village of Yipala (see Figure 2a) and the nearby garbage disposal area. Furthermore, individual members of both groups were frequently found around the hotel, searching for food and drinking water from the swimming pool. Both groups attempted to steal food from the local people. Apart from what they obtained from these ‘charges’ (including yam, cassava, tomato ketchup, bread and bananas), olive baboons fed on grass, leaves, fruit, and chicken or guinea fowl. During the night, the baboons slept in trees in distinct areas of the park. No major fluctuations in the daily pattern of activities of both groups were detected during most of our study period, with the exception of the month of July when the number of visits to the village area was reduced significantly. Droppings from the baboons were mostly detected around their sleeping places, the garbage disposal area, the primary school, the hotel, and on the tracks between these places. In spite of the regular presence of non-human primates in the village, no faeces were detected in or around the houses of the local people. Other species of monkeys in the area included C. aethiops, C. polykomos and G. senegalensis. However, these species were not seen in or around the village.
Observations in Baobeng Fiema
Approximately 400–500 mona monkeys inhabited the BFMS. They lived in some 17 groups, each consisting of 15–60 individuals. Although, all groups visited the villages now or then, most dwelled at their fringes. Two groups of mona monkeys were regularly found in the centre of the villages. In particular one small group of around 20 animals was detected many times within the village of Baobeng, mostly early in the morning or late in the afternoon (see Figure 2b). In the villages, the mona monkeys stole food from the storehouses and from the courtyards, where people prepared and cooked their food. Also, they drank water from pots and pans used by the local people. Sometimes, the monkeys were fed by the inhabitants of the village or by tourists. No particular defecation sites were detected. Droppings from mona monkeys were seen at numerous locations, on the edge of the forest and the villages, between the houses of Baobeng–Fiema and in places where monkeys sat down to feed. In the forest outside of the village, the mona monkeys were mostly seen in the trees at a low canopy height, and sometimes on the ground. There the mona monkeys fed on fruits and leaves.
The exact number of black and white colobus monkeys in BFMS could not be determined, but is estimated to be several hundred. This species of monkey was mostly observed in the trees in the forest, but at a much higher canopy level compared with that frequented by the mona monkeys. They fed mainly on leaves. With the exception of a single occasion, this species of monkey was not seen within or around the villages. Also, no droppings from black and white colobus monkeys were found in the village, only in the forest area.
In total, 349 faecal samples from non-human primates were collected in the MNP and FBMS and examined for O. bifurcum (Table 1). Microscopic examination after coproculture was performed on 279 of these samples. The results showed that 92% and 75% of the faecal samples from olive baboons and mona monkeys, respectively, contained L3s of O. bifurcum. The numbers of L3s obtained from the cultures were high (i.e. >100 L3 per 100 μl culture sediment) in 55% and 61% of the test-positive samples from olive baboon and mona monkey, respectively. No L3s of O. bifurcum were detected in the faecal samples from the black and white colobus monkey.
Table 1. Outcome of 349 non-human primate faecal samples specifically diagnosed for O. bifurcum by coproculture (n = 279) and/or species-specific PCR (n = 125)
* Mona monkeys showed a significant lower rate of O. bifurcum infection compared to olive baboons, either determined by coproculture (χ2 = 11.2, P = 0.001) or species-specific PCR (χ2 = 24.9, P < 0.001).
Mona monkey *
Black and white colobus
In MNP, faecal samples were collected over a period of 13 months. The prevalence of infection with O. bifurcum in the samples from olive baboons ranged from 71% (July) to 100% (several other months), without a seasonal pattern being evident. Also in the BFMS, no significant variation in prevalence of infection was detected among faecal samples collected during the 4 months (data not shown).
From 125 faecal samples, DNA was isolated and subjected to the species-specific PCR assay. The PCR results were in agreement with the coproculture results, in that they showed high numbers of O. bifurcum positive samples of olive baboons or mona monkeys but no infections in the black and white colobus (Table 1). For 55 samples, both coproculture and PCR were performed. In the O. bifurcum-specific PCR, 40 (95.2%) of the 42 coproculture-positive samples tested positive. Also, 6 of the 13 samples in which no L3s of O. bifurcum were detected by microscopy were test-positive using the specific PCR.
The present study showed O. bifurcum to be highly present in stool samples from olive baboons and mona monkeys in the MNP and BFMS. Given that the microscopic findings were confirmed by species-specific PCR does not support that we were looking at other morphologically similar species of Oesophagostomum, such as Oesophagostomum colombianum, Oesophagostomum dentatum, Oesophagostomum quadrispinulatum. These species could be present in the environment (i.e., water, soil or dust), or on vegetation (i.e., grass and vegetables) in the MNP and BFMS, transmitted by other local animals, such as warthogs or goats.
Stool samples examined could not be traced back to individual animals and it cannot be excluded that more than a single faecal sample per individual animal had been collected and examined. Consequently, the determined prevalences are just rough indications. When counting the number of L3 larvae in the culture sediment, much greater numbers were found than in studies of humans (Pit et al. 1999; Yelifari et al. 2005), suggesting high infection intensities (i.e. worm burden). Our findings indicate that free-living, potentially infective, larval stages develop and survive successfully throughout the year under the prevailing environmental and ecological conditions. Consequently, the homologous transmission of O. bifurcum to the olive baboons and mona monkeys seems to be highly effective. On the other hand, the aboreal (i.e. living in trees) black and white colobus monkeys appeared not to be at risk of this geo-helminthic infection.
To date, no case of human infection with O. bifurcum has yet been reported in the study area. Approximately 90% of all inhabitants of MNP (n = ∼700) and a selection of individuals from BFMS (n = ∼100) have been examined extensively thus far by larval culture and/or PCR for the presence of nematodes. All stool samples collected were shown to be test-negative for O. bifurcum, whereas N. americanus infections were commonly detected. Large-scale treatments with anti-helminthics, including albendazole are not known to have occurred in these regions in recent years (Polderman, unpublished observations).
The absence of Oesophagostomum in humans in the MNP and BFMS is remarkable for three reasons. First, as mentioned, the conditions for transmission seem to be highly favourable. Second, the observations of the animal behaviour made in the present study indicate a close interaction between human and non-human primates through the sharing of the same habitat. The intensity of contacts seems difficult to quantify. Nonetheless, the frequency of defaecation by non-human primates in, or in the close vicinity of, the human compounds, and the mere co-existence of some 3000 peoples and over 400 mona monkeys, on a surface of no more than about 1.9 km2 are impressive indications for the closeness of the monkey-man interaction. Third, humans are perfectly suitable hosts for O. bifurcum, as demonstrated by the high prevalence of human infections, just a few 100 km to the north of MNP and BFMS (Yelifari et al. 2005). If transmission to humans can efficiently take place in northern Ghana, it is still an enigma why it does not occur in the MNP and BFMS, where the parasite is so abundantly present and human exposure to the simian infection of O. bifurcum seems to take place. In a recent morphological study, adult O. bifurcum isolated from human, olive baboon and several monkey species were shown to differ in various morphological characters, including worm length and width (de Gruijter, unpublished observations). Furthermore, genetic studies have detected population variation within O. bifurcum from human and non-human primates (Gasser et al. 1999; de Gruijter et al. 2002), and have shown that O. bifurcum from human, mona monkey, patas monkey and olive baboon are genetically distinct (de Gruijter et al. 2004, 2005). To further prove that O. bifurcum from non-human primates in Ghana is not transmissible to humans, cross-infection should be conducted. Such an experimental cross-infection study has been performed in the past, indicating poor susceptibility of laboratory monkeys to O. bifurcum of humans (Eberhard et al. 2001). However, in this study control infections (i.e. isolations from non-human primates) were not included. Evidence on the infectivity of larvae of the non-primate-host strain to humans could not be gathered from experiments; it is the present circumstantial evidence that suggests such transmission cannot (or not easily) take place.
Overall, the findings of the present study suggest that the parasite infecting the baboons and monkeys in the MNP and BFMS are unable to infect humans. Together with the previous morphological and genetic findings, the observations support the hypothesis that O. bifurcum from human and non-human primates represent biologically distinct ‘strains’ or species, which infect distinct primate hosts. Consequently, a zoonotic reservoir is probably not relevant when planning elimination of human infections.
Collection of data and material would have been impossible without the tremendous efforts of the field team workers Nemo de Groot, Martin B.A. Heemskerk, Lennart van Kranenburg, Nellemieke van Dinteren, Gosse Flootman, Willem van de Ven and Marjolein Kleppe. We also acknowledge the local wildlife division staff, in particular Samuel Amponsah and Antony N. Dassah, for their help and guidance.
The authors are grateful to the communities of Yipala, Fiema and Baobeng for their hospitality to the field team and participation in the study. Mr. Daniel Akowah is particularly acknowledged for his guidance at FBMS. We thank Gilbert Dery and Von Asighri (Parasitic Diseases Research Center, Tamale) and Juventus Ziem (University for Development studies, Tamale) for logistical assistance and Coby Blotkamp (LUMC, Leiden) for training and advising the field team. Financial support was provided by the Dutch Foundation for the Advancement of Tropical Research WOTRO-NWO and the Oesophagostomum Research Intervention Project in Northern Ghana.