Open Access Research

Bovine fasciolosis at increasing altitudes: Parasitological and malacological sampling on the slopes of Mount Elgon, Uganda

Alison Howell1, Lawrence Mugisha23, Juliet Davies1, E James LaCourse1, Jennifer Claridge4, Diana JL Williams4, Louise Kelly-Hope1, Martha Betson1, Narcis B Kabatereine5 and J Russell Stothard1*

Author Affiliations

1 Disease Control Strategy Group, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L3 5QA, United Kingdom

2 College of Veterinary Medicine, Animal Resources & Biosecurity, Makerere University, Kampala, Uganda

3 Conservation & Ecosystem Health Alliance (CEHA), P.O. Box 34153, Kampala, Uganda

4 Veterinary Parasitology, Institute of Infection and Global Health, School of Veterinary Science, University of Liverpool, Liverpool, L69 7ZJ, United Kingdom

5 Vector Control Division, Ministry of Health, Kampala, Uganda

For all author emails, please log on.

Parasites & Vectors 2012, 5:196  doi:10.1186/1756-3305-5-196


The electronic version of this article is the complete one and can be found online at: http://www.parasitesandvectors.com/content/5/1/196


Received:14 April 2012
Accepted:14 August 2012
Published:7 September 2012

© 2012 Howell et al.; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

To clarify the extent and putative transmission zone of bovine fasciolosis on the slopes of Mount Elgon, Uganda, conjoint parasitological and malacological surveys, inclusive of inspection of animals at slaughter, were undertaken at increasing altitudes.

Results

A total of 239 cattle were sampled across eight locations ranging in elevation from 1112-2072 m. Faecal material was examined for presence of Fasciola eggs and sera were tested by ELISA for antibodies against Fasciola antigens. Bolstering this, 38 cattle at slaughter from 2 abattoir sites at 1150 m and 1947 m were inspected; in addition, wild buffalo stool (n = 10) opportunistically picked within Mount Elgon National Park (MENP) at 3640 m was examined. By faecal egg detection, prevalence of Fasciola gigantica at low (<1500 m) and high (>1500 m) altitude sites was 43.7% (95% CI 35.4-52.2) and 1.1% (95% CI 0.0-6.0), respectively, while by ELISA was much higher, low altitude - 77.9% (95% CI 69.7-85.4) and high altitude - 64.5% (95% CI 51.3-76.3). The decline in prevalence with increasing altitude was corroborated by abattoir sampling. Thirty seven aquatic habitats, ranging from 1139-3937 m in altitude were inspected for freshwater snails, 12 of which were within MENP. At lower altitudes, Lymnaea (Radix) natalensis was common, and often abundant, but at higher altitudes became much rarer ceasing to be found above 1800 m. On the other hand, Lymnaea (Galba) truncatula was found only at altitudes above 3000 m and within MENP alone. The snail identifications were confirmed by DNA analysis of the ribosomal 18S gene.

Conclusions

Active infections of F. gigantica in cattle are common in lower altitude settings but appear to diminish with increasing elevation. This is likely due to a growing paucity of intermediate hosts, specifically populations of L. natalensis for which a natural boundary of 1800 m appeared. Although F. hepatica was not encountered, the presence of several populations of L. truncatula at elevations over 3000 m point towards a potential transmission zone within MENP should this parasite be introduced.

Background

Fasciolosis, caused by infection with the liver fluke Fasciola, can cause significant economic losses in African livestock [1,2]. The complex nature of the lifecycle and epidemiology of this snail-borne disease presents challenges for predictive mapping at the herd-level, as well as disease management and animal husbandry at the individual-level [3]. Fasciola gigantica and Fasciola hepatica can infect a wide variety of domesticated animals, wildlife and people [4-9]. Thus the disease-endemic zone can be difficult to define from parasitological data alone and so consideration of the distribution of associated snail intermediate hosts can be important [10]. F. gigantica is the most common liver fluke in sub-Saharan Africa, being adapted to warmer conditions [11] likely due to the widespread distribution of its intermediate host Lymnaea (Radix) natalensis [12]. On the other hand owing to a more limited distribution of its intermediate host Lymnaea (Galba) truncatula [12], F. hepatica can exist in zoonotic foci which are more restricted to cooler regions of Africa, including Kenya, Ethiopia and Tanzania [1,10,13]. Nonetheless, actual or potential overlap of both types of fasciolosis can occur especially where snail-habitats converge, for example, with increasing altitude as in the highlands of Ethiopia [14] or perhaps in upland zones of eastern Uganda, as yet to be fully explored. In the Mount Elgon area of Uganda, fasciolosis is poorly studied as there is no systematic veterinary or medical disease surveillance system.

Cattle are Uganda’s most economically important livestock species with an estimated population of 11 million [15]. The majority are either indigenous Zebu or Sanga, with less than 5% being imported ‘exotic’ breeds, mainly Friesians [16]. Livestock production is hampered by many disease constraints of which fasciolosis is considered the most important helminth infection [17]. The Mount Elgon region consists of predominately rural subsistence farmers covering a zone between 1000-2300 m in altitude rising towards an important wildlife reserve, the Mount Elgon National Park (MENP). Encompassing a total area of some 1,145 km2, MENP commences at 2300 m and extends to 4321 m at Wagagi Peak. Within the park, a number of herds of wild ruminants are known including buffalo, antelope and elephant but illegal cattle trading routes, from Uganda to Kenya and vice versa, traverse throughout. However, with increasing Uganda Wildlife Authority (UWA) foot patrols servicing an increasing hiking and camping tourism, illegal cattle trafficking has declined in recent years.

Like elsewhere in Uganda, the lowland areas of Mount Elgon are known to be endemic zones for F. gigantica with reports documenting the prevalence of F. gigantica at 54.7% in cattle [16,18,19]. A contemporary situational analysis, however, is yet lacking. From a malacological perspective, there has been no update to the formal snail surveys conducted by Georg Mandahl-Barth and by Hubendick in their general treatise on Ugandan freshwater snails and Lymnaea, respectively over 50 years ago [12,20]. Both Lymnaea natalensis and Lymnaea mweruensis have been reported from the area with the latter species now considered synonymous with Lymnaea truncatula, as collected by C.C. Cridland from Sasa River Camp at 2900 m (now within MENP). Such upland areas, like those in neighbouring Tanzania, are thought suitable zones for the transmission of F. hepatica, for example, L. truncatula being recently found at 2712 m & 2720 m with identifications confirmed by DNA analysis of the ribosomal 18S [10]. Though F. hepatica has yet to be encountered in natural transmission cycles in Uganda, it has been known from earlier reports within UK-imported cattle [21].

Clearly defining such local zones of transmission in eastern Uganda is also important for further modelling of the suitability of habitats elsewhere in East Africa. Various authors have designed models based on climate and intermediate host presence to predict the prevalence of Fasciola spp. [14,22]. However, areas that appear broadly similar in terms of climate can have very different snail populations due to variations in micro-climate and local aquatic factors, e.g. water pH and conductivity. This limits the accuracy of such climatic models, and localised parasitological and malacological data are still required for prediction of actual disease zones or outbreaks [22,23]. In many countries, signalment of cattle condition [24] and subsequent meat inspection provides an opportunity to monitor the incidence of fasciolosis, also allowing access to adult worms enabling morphological identification [23,25]. However, it is not able to detect past infections in those animals that have either been treated or developed immunity and self-cured. A suitable immunological test could fill this gap, and also detect pre-patent infections, but presently this is only available for assaying antibody titres in cattle to excretory/secretory (ES) antigens of F. hepatica [26]. With this assay, heterologous reactions to F. gigantica are likely but as yet not known, however, serological testing should be an interesting adjunct in revealing putative transmission zones.

Using a combination of parasitological sampling, bolstered by experimental serology, our study aimed to investigate the occurrence of fasciolosis in bovids at low and high altitude areas on the slopes of Mount Elgon and also assessed animal condition (i.e. body signalment). The parasitological surveys were complemented with a conjoint malacological appraisal in an attempt to better define the actual or potential disease transmission zone of these parasites.

Methods

Study area and design

A preliminary visit to the Mount Elgon study area took place in March 2011 by the corresponding author when local arrangements were made with the District Administration Offices for future surveys in June-July and entry into MENP. In addition, a selection of 20 freshwater habitats was identified and spot-surveyed for freshwater snails for later comparison within the year. In June-July 2011, a cross-sectional study was conducted; sampling of cattle centred around six main sites of varying altitudes: Mbale (1150 m), Sironko (1155 m), Bududa (1268 m), Sipi (1856 m), Kapchorwa (1947 m) and Kween (2072 m). In addition, cattle were also sampled from larger roving herds in two sites between Kapchorwa (1112 m) and Ngenge (1468 m), Figure 1. Sample size calculations were performed using WinEpiscope 2.0 (University of Edinburgh, UK). For 80% power, 95% confidence, an estimated prevalence of 40% at the low altitude (<1500 m) sites [18,27] an estimated prevalence of 20% at the high altitude (>1500 m) sites (assumed to be lower based on absence of lymnaeids found in a preliminary study in March 2011) a sample size of 80 cattle was required from high and low altitude sites.

thumbnailFigure 1. Lymnaeid snails found at various sites in the Mount Elgon region. Sites are coded by altitude (background map redrawn from Google, 2011 http://maps.google.co.uk/maps webcite).

At each of the six main sites, at least 30 cattle were sampled, a maximum of two per owner in order to avoid bias due to multiple animals from the same herd having similar risk factors. Where possible, stool and blood specimens were taken from each animal at the time of tethering. At four of these sites, convenience sampling was employed, with local community leaders mobilising farmers to bring their animals to a central point. At two of the sites this method was unsuccessful and individual households around a central point were visited. From the two large roving herds in between Kapchorwa and Ngenge, a random selection of 10 cattle from each were sampled. Additionally, during a three-day trek on foot to altitudes of between 3000-4000 m in MENP, freshly deposited buffalo stool samples (n = 10) were opportunistically picked from the ground at an elevation of 3900 m at a unique grazing site adjacent to a hot spring.

Signalment of cattle

The data on signalment of each cow were recorded as follows: Age category - calf (approximately 2 weeks to 6 months), sub-adult (6 months to 18 months), or adult (18 months plus); sex; breed - Friesian, local (zebu type) or hybrid; and body condition score (BCS) graded through 1 to 5 classifications according to Roche [24].

Faecal egg detection

Faecal samples were obtained per rectum or taken from the ground if seen to have been directly produced. Approximately 3 g of stool was thoroughly mixed into 250 ml of bottled mineral water containing 0.5% Tween-20 (Sigma-Aldrich, UK) before filtration-concentration; two methods were each used for the isolation of eggs from faeces. ‘Individual stool’ was analysed using Flukefinder® kit (see http://www.flukefinder.com webcite) and in an attempt to confirm these individual findings ‘pooled stool of 10 animals’ was subjected to standard coprological filtration with a descending pore series of 3 large-diameter metal sieves (at 425, 125, and 32 microns respectively) following [10]. Faecal eggs, from liver flukes or paramphistomes, were finally collected in a 10 cm glass petri-dish stained with a few drops of 10% methylene blue solution, viewed and counted under the dissecting microscope at x40 magnification [28].

Blood sampling and serology

Blood was obtained from an ear vein and harvested into sterile 10 ml plastic syringes and allowed to clot in a 1.5 ml eppendorf tube. Serum was then separated by centrifugation, and heat-inactivated by incubation at 56°C for 30 min, as required for importation licensing by the Department for the Environment, Food and Rural Affairs (DEFRA), UK. After heat-inactivation samples were stored in liquid nitrogen before transportation to the UK. ELISA was performed according to [26,29] with the following minor modifications: The concentration of F. hepatica E/S antigens used to coat the ELISA plate was 1 mg/ml; the concentration of monoclonal anti-bovine IgG used was 1:70,000 (this was first optimised by a checkerboard titration); and 20 minutes following the addition of TMB substrate, 100 μl of stopping solution (0.5 M HCL) was added to each well prior to reading. The results are given as the mean of the optical density (OD) obtained from duplicate samples expressed as a percentage of the strong positive control (PP), with PP of 15 or above considered a positive result [26,29].

Liver necropsy

Livers obtained from cattle slaughtered at Mbale (n = 30) and Kapchorwa (n = 8) abattoirs, at 1150 m and 1947 m, were inspected for flukes by cutting open the main bile ducts into the liver parenchyma. Adult flukes were identified morphologically based on size and shape [30,31]. Faecal and blood samples were taken from the large intestine and mesenteric vein respectively and processed as described above for ELISA.

Malacological sampling

In total 37 freshwater sites, ranging in altitude from 1139 m to 3937 m above sea level, were selected and surveyed for aquatic snails. Sites were chosen to include a variety of streams, marshes and pools to cover as wide an area as possible, within easy reach of vehicular access, with the exception of the sites above 3000 m within MENP that were visited on foot. Using collecting sieves and snail scoops, two people surveyed each site for 10 minutes and all collected snails were counted. If there were different types of habitat within each location, for example, slow/fast flowing water within streams or drainage ditches, these were all surveyed. The presence and numbers of each species of snail were recorded according to field identification keys of Brown [32]. To later confirm the identification of encountered lymnaeids, a selection of snails was placed in 70% ethanol for DNA analysis. Spot-site water chemistry readings were taken for pH, conductivity, total dissolved salt and temperature from each of the different habitats using a handheld water meter (Hanna H1-9816-6; VWR, UK) to investigate ecological associations.

DNA-based snail identification

Genomic DNA was extracted from a total of 16 snails representative of L. natalensis (n = 8) and L. truncatula (n = 8) using the DNeasy Blood and Tissue Kit (QIAGEN, Germany). A 450 base pair region of the nuclear ribosomal 18S was amplified by PCR with the primers 18SLYMFOR 5′ AGTAGTCATATGCTTGTCTCAAAGATTAAGCCA and 18SLYMREV, 5′ TGCGCGCCTCTGCCTTCCTTGGATGTGGTAGCCGT, following Stothard [33]. Amplification products were purified using the QIAquick PCR Purification Kit (QIAGEN, Germany) and sequenced using the ABI PRISM™ BigDye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, UK). Sequencing chromatograms were produced by the DNA Sequencing Facility, The Natural History Museum, London, UK and analyzed using DNASTAR’s Lasergene Sequence Analysis Software (Madison, USA). Compiled sequences were aligned and also compared with other lymnaeids on GenBank. The putative secondary structure of the variable V2 E10-1 helix was investigated using the RNAfold web server (http://rna.tbi.univie.ac.at/ webcite) and compared to that described by Stothard and Bargues & Mas-Coma [33,34]. To investigate known restriction site variation, amplification products were also digested, separately, with either mspI or cfoI enzymes and subsequently separated by PAGE and ethidium bromide staining according to [33], and photographed with a Gel Doc E2 Imager (BioRad, UK).

Statistical analysis

For raw prevalence data, exact binomial confidence intervals were calculated for the cattle surveys using Stata v. 9 (2007, Texas, 77845). Other tests were performed using PASW v. 18 (2010, SPSS Inc, Chicago). Poisson regression was used for multivariate association with snail data and recorded environmental variables. Firstly, parameters with close correlations (>0.8) were excluded from being modelled together. Secondly, each parameter was entered separately and the model with the smallest difference between scaled deviance and degrees of freedom taken as being the most representative.

Results

The signalment data (i.e. body condition) recorded for each inspected animal is shown in Table 1. Many of the cattle were often subjected to zero grazing, especially in peri-rural settlements, whereby animal holders provided on-site food to tethered animals.

Table 1. Biometric data assessing signalment of cattle sampled for fluke infection at high (n = 145) and low (n = 94) altitudes [95% confidence intervals are shown]

Faecal egg count analysis

Faecal samples obtained from 233 cattle were processed individually with testing on-site near the point of collection. From the high altitude population, only one cow was found to be egg positive for Fasciola, with only a single egg observed in this animal’s faeces. In the low altitude population, 43.7% (95% CI 35-52) of cattle tested were egg positive for Fasciola by Flukefinder®, Table 2. Faecal egg counts ranged from 1 to 43 eggs per gram (epg) of stool, with the majority being low numbers of eggs (<10). The results from the pooled stool samples were in all cases qualitatively the same as the Flukefinder® results with the exception that eggs were not found at altitudes above 1500 m. All of the buffalo samples were found negative for eggs of Fasciola. Eggs of paramphistomes were often seen with prevalence estimated by Flukefinder® to be 74% (95% CI = 66.4-81.2%) and 58% (95% CI = 47.8-69.1) at low and high altitude sites, respectively. Faecal egg counts for amphistomiasis ranged from 1 to over 250, with low altitude cattle showing a higher intensity of excreted eggs. Of the buffalo faecal samples, no egg of Fasciola was seen but 6 out of 10 were positive for amphistome eggs. There were no significant relationships between liver fluke infection and either BCS, breed, sex, or paramphistome infection.

Table 2. Prevalence of Fasciola infections as detected by faecal egg detection, ELISA and gross liver inspection at abattoir [95% confidence intervals are shown in brackets]

Serological analysis

Owing to resource constraints and sampling spoiling, a sub-set of animals were tested by ELISA, finding that 64.5% (95% CI = 51.3-76.3) of cattle at high altitude were positive for Fasciola spp. whereas 77.9% (95% CI = 69.7-85.4) of cattle at low altitude were positive, Table 2. This was a statistically significant difference, however, as the high ELISA positive rate was not confirmed by egg detection in the high altitude samples (i.e. confirmation of active infections), only egg detection results are used for subsequent analysis and further discussion.

Liver necropsy

Livers from 30 animals originating at low altitude and 8 from high altitude were inspected. The proportion of animals infected with liver fluke at the low altitude sites was 48.4% (95% CI 30.2-66.9). At the high altitude sites, none (95% CI 0.0-37.9) of the animals were found to be infected, Table 2. Owing to time constraints within slaughter houses, numbers of flukes found were not recorded, but in the majority of animals less than 10 flukes were found. Based on body size and shape, all flukes found were confidently identified as F. gigantica, there was no evidence of F. hepatica-like worms.

Malacological surveys

Lymnaea natalensis was found at 12 sites ranging in altitude from 1139 m to 1770 m. By contrast L. truncatula was found at only 5 sites, all above 3500 m. There was no geographical overlap between the two species, with the zone between 1800 m and 3500 m being a lymnaeid-free zone, Figure 1.

Correlation analysis showed significant correlation of snail numbers with altitude and water temperature for both species. Dissolved salt concentration and electrical conductivity showed significant correlation with L. truncatula numbers only, Figure 2.

thumbnailFigure 2. Correlations between number of snails found at a site, and the mean temperature, pH, conductivity and concentration of the water and elevation. The strongest correlations are for temperature and altitude.

Using Poisson regression, altitude, pH and conductivity appeared to be predictors of L. natalensis and L. truncatula with contrasting associations, Table 3. The location and numbers of snails collected alongside the presence of other aquatic gastropod genera is shown in Table 4.

Table 3. Results of Poisson regression on collected snail and aquatic habitat data

Table 4. Location and occurrence of snails at each of the 37 collection sites

DNA-based snail identification

RFLP examination and DNA sequence analysis of the 450 bp PCR fragment confirmed the presence of L. natalensis and L. truncatula by sequence homology with other lymnaeid 18S accessions GenBank, Figure 3. No RFLP variation was observed for mspI digestion between the samples, all having a 3-banded restriction profile of fragments of 300 bp, 100 bp and 50 bp which is named as profile mspI type 1 [33]. In contrast, two cfoI restriction types were observed, Figure 3a, with L. natalensis having a 3-banded profile while L. truncatula exhibited a 2-banded profile. These restriction sites were mapped onto the sequenced region and DNA variation was located in the V2 E10-1 helix region, Figure 3b. Of note is that the mspI restriction site (CCGG) at position 201 is absent in these Uganda L. natalensis samples owing to a C to A mutation, Figure 3c. The cfoI restriction site (GCGC) at position 208 is intact in L. natalensis and combined with the cfoI restriction site at 161 gives rise to a 3-banded RFLP profile, previously identified as cfoI type 2 [33]. In L. truncatula, the C to G mutation leads to a loss of this cfoI site at position 208, hence a 2-banded RFLP profile is observed, previously recognised as cfoI type 1 [33]. This sequence variation is fully described in the alignment schematic of Figure 3b, with the putative secondary structure of the E10-1 helix shown in Figure 3c. Of note is the stem region of L. natalensis which is 3 bp longer than that of L. truncatula and conforms to that previously described [34,35]. No other sequence variation was found within the samples.

thumbnailFigure 3. Molecular characterization of Lymnaea. A) RFLP analysis of 18S region with cfoI. B) Schematic RFLP map of the 18S region with corresponding DNA sequence variation within the E10-1 helix. C) Putative secondary structure of the E10-1 helix illustrating the RFLP variation and nucleotide differences.

Discussion

Taken together the parasitological and malacological surveys have shown that fasciolosis, resultant from F. gigantica, was widespread in cattle at lower altitudes in the Mount Elgon area, however, defining where the actual transmission zone ends is, however, problematic; especially so given the existence of suitable intermediate snail hosts up to 1800 m. Moreover, the confirmation of several populations of L. truncatula within the MENP, a single population being reported over 50 years ago [20], raises some potential concerns for transmission of F. hepatica should this parasite ever be introduced into this area.

In terms of general disease, the body condition signalment data gathered about the cattle revealed that there were a significantly greater proportion of cattle classified as ‘underweight’ at lower altitude, see Table 1. Animals at high altitude were often subjected to zero-grazing, being regularly fed on plantain leaves and were used for milk production, whereas at lower altitudes animals were free-grazing and used for meat production. Total cattle numbers in the high altitude area are estimated to be in excess of 95,000 whereas around Mbale they are thought to be slightly less in just under 65,000 animals (LM, personal communication). Information about the prevalence of specific signs of fluke related morbidity would be required in order to determine the extent of pathology due to fasciolosis and hence the relative importance of the parasite in the local cattle population [36,37]. Although useful for the confirmation of patently infected animals, several authors have found the sensitivity of liver inspection at the abattoir to be only 63-71% [23], thus the conditions in the two abattoirs were not conducive for ‘gold standard’ necropsy data but the available evidence can be discussed in the light of our conjoint parasitological and malacological findings, experimental serology (which might be confounded by past-infection status), and the clear ecozonation of the Mount Elgon area as shown in Figure 4 in terms of local vegetation.

thumbnailFigure 4. A map derived from remotely sensed data of the changing ecozones within the study region based on GlobCover project land cover map and ground observations. Survey sites are shown as black dots. The Uganda-Kenya border is shown.

Observations associated with increasing altitude

Within the cattle sampled at altitudes between 1000 m and 1500 m, F. gigantica was shown to be present by all diagnostic methods used. The egg-patent prevalence by Flukefinder® was 44%, Table 2. This is in agreement with the earlier findings [18], however, between the altitudes of 1500 m and 3000 m only a single Fasciola egg positive cow was identified so it can be firmly considered that F. gigantica worms were not present in significant numbers, despite inferences from ELISA which is semi-quantitative for F. hepatica[29], at these elevations. This decline is concurrent with the growing scarcity of local L. natalensis populations, ceasing at 1800 m. Perhaps the presence of such snail hosts is a useful environmental indicator of the potential for disease transmission which can be provided relatively quickly during field-excursions as freshwater snails with dextral shell without an operculum are readily distinctive.

A key finding in this study is the dichotomy between the presence of L. natalensis at sites below 1500 m and L. truncatula at sites above 3000 m. Inspection for snails at the Sasa River Camp within MENP did not reveal the presence of lymnaeids, unlike during the 1950s where L. truncatula (mweruensis) was encountered [20]. The analysis of snail DNA very clearly confirmed the latter species likely of European origin [34,35]. The precise reasons for this lower altitudinal occurrence of L. truncatula are open to conjecture but could include longer-term climatic changes in the area restricting the suitable habitat to higher elevations with general climate warming. In other studies, L. natalensis shedding Fasciola have been found at altitudes of 2667 m in Kenya [38]. This perhaps indicates that factors other than altitude and the typically associated temperature are likely to be having an effect. There were wide variations in water temperature, water chemistry, vegetation (see Figure 4) and soil type.

Although altitude and temperature follow a fairly predictable trend across a wide area of East Africa, other factors such as snail habitat are much more variable, this study identified many sites where no snails were found, despite having similar temperatures to nearby sites where snails were plentiful. The apparent ‘random’ focalisation of snails in aquatic habitats is known likely to be due to cryptic micro-habitat associations and to the vagrancies of population colonisation [32]. In the regression model, for example, the input factors were a poor predictor for actual snail numbers. More generally, the presence of F. gigantica at altitudes above 1500 m and the absence of F. hepatica at altitudes above 1200 m conflicts with general model-based assumptions on temperature alone [14,22]. Micro-climates suitable for snails to survive can occur in otherwise hostile environments, but may be hard to find without more exhaustive sampling [39]. Furthermore, the ability of snails to aestivate, differences in annual transmission patterns and the timing of surveys have significant bearings upon the findings but there were no obvious differences in presence/absence of snails recorded in the March and June-July surveys.

Within the MENP and at altitudes above 3000 m, only L. truncatula was found alongside an enigmatic Bulinus population at high altitude, (see Table 4), tentatively identified of the B. truncatus/tropicus group [32]. Despite the presence of L. truncatula, there was no evidence of Fasciola spp. infection in the sampled buffaloes although paramphistomes were common. It could be concluded that local conditions are not suitable for the transmission of F. hepatica or perhaps that this parasite has yet to be introduced to this area. The MENP also extends into Kenya, where other livestock and people live together at higher altitudes. As F. hepatica is known to exist within natural transmission cycles in the neighbouring highlands of Kenya [40] this could be a potential route for the introduction of F. hepatica into the Ugandan highland region. Conversely, there was no evidence that L. truncatula was present at sites at lower altitudes where cattle were kept. Nonetheless, this snail species can be present at much lower altitudes in other countries [14], hence there may be some potential for this lymnaeid to establish in the 1500-2500 m zone, so further sampling should not neglect this possibility.

Implications for control and disease surveillance

From the available evidence, no community-based intervention is currently needed for management of fasciolosis in domestic cattle at altitudes above 1500 m in this Mount Elgon area. At low altitude, however, future interventions based upon de-worming are clearly worthwhile in addition to the collection of local information on farming practices, economic impact and animal trafficking. The latter is especially important with future cattle re-stocking planned from this area to central regions of Uganda (Lira/Kitgum/Gulu) following the cessation of civil insecurities. The inspected herds at Ngenge, for example, have been ear-marked for this restocking programme.

In terms of future disease surveillance the detection of populations of L. truncatula in several sites frequented by UWA patrols and hiking tourists, raises some concerns of the safety of environmentally drawn water. Although the existence of F. hepatica has yet to be proven in this area, it would be advisable to raise awareness of fluke-borne diseases in general. More broadly, we are presently PCR screening the collected snails for evidence of fluke infections in an attempt to investigate these more cryptic aspects of parasite transmission.

Conclusions

From parasitological sampling and observations at slaughter, infections of F. gigantica in cattle are common in lower altitude settings but appear to diminish with increasing elevation. This is most likely due to a growing paucity of L. natalensis within the environment, with a natural boundary of approximately 1800 m where no further populations of L. natalensis were found. Whilst F. hepatica was not encountered during these surveys, the presence of L. truncatula at elevations over 3000 m point towards a potential transmission zone within MENP should this parasite be introduced to this part of East Africa. Greater vigilance of this parasite within imported cattle, and possibly within local people, should therefore be encouraged.

Competing interests

The authors have no competing interests.

Authors’ contributions

The work reported here forms part of the MSc research projects of AH and JD under the joint supervision of EJLaC and JRS. Fieldwork was carried out by AH, LM, JD, EJLaC, NBK and JRS. Analysis of serological samples was assisted by JC and DJLW. Multivariate analysis was undertaken by AH with the help of LK-H, especially in spatial mapping techniques. JD, MB and JRS undertook the analysis of lymnaeid DNA analysis. All authors contributed to the production of the manuscript and revision, with JRS as final guarantor. All authors read and approved the final version of the manuscript.

Acknowledgements

We would like to thank the field team from Vector Control Division who assisted during the surveys and local staff within the District Veterinary Offices, also the participation of the farmers who brought their animals for inspection. We especially thank Andrew Ruggiana for his assistance in the field with really expert cattle handling.

References

  1. Abunna F, Asfaw L, Megersa B, Regassa A: Bovine fasciolosis: coprological, abattoir survey and its economic impact due to liver condemnation at Soddo municipal abattoir, Southern Ethiopia.

    Trop Anim Health Pro 2010, 42:289-292. Publisher Full Text OpenURL

  2. Mungube EO, Bauni SM, Tenhagen BA, Wamae LW, Nginyi JM, Mugambi JM: The prevalence and economic significance of Fasciola gigantica and Stilesia hepatica in slaughtered animals in the semi-arid coastal Kenya.

    Trop Anim Health Pro 2006, 38:475-483. Publisher Full Text OpenURL

  3. Mas-Coma SM, Valero A, Bargues MD: Fasciola, lymnaeids and human fascioliasis, with a global overview on disease transmission, epidemiology, evolutionary genetics, molecular epidemiology and control.

    Adv Parasitol 2009, 69:41-146. PubMed Abstract | Publisher Full Text OpenURL

  4. Hammond JA: Infections with Fasciola spp. in wildlife in Africa.

    Trop Anim Health Pro 1972, 4:1-13. Publisher Full Text OpenURL

  5. Hammond JA: Human infection with the liver fluke Fasciola gigantica.

    Trans R Soc Trop Med Hyg 1974, 68:253-254. PubMed Abstract | Publisher Full Text OpenURL

  6. Hammond JA, Sewell MMH: Pathogenic effect of experimental infections with Fasciola gigantica in cattle.

    Br Vet J 1974, 130:453-465. PubMed Abstract OpenURL

  7. Issia L, Pietrokovsky S, Sousa-Figueiredo J, Stothard JR, Wisnivesky-Colli C: Fasciola hepatica infections in livestock flock, guanacos and coypus in two wildlife reserves in Argentina.

    Vet Parasitol 2009, 165:341-344. PubMed Abstract | Publisher Full Text OpenURL

  8. Torgerson PR, Claxton J: Epidemiology and control. In Fasciolosis. Edited by Dalton JP. CABI publishing, Oxford, UK; 1999:113-150. OpenURL

  9. Walker SM, Johnston C, Hoey EM, Fairweather I, Borgsteede FHM, Gaasenbeek CPH, Prodohl PA, Trudgett A: Potential role of hares in the spread of liver fluke in the Netherlands.

    Vet Parasitol 2011, 177:179-181. PubMed Abstract | Publisher Full Text OpenURL

  10. Walker SM, Makundi AE, Namuba FV, Kassuku AA, Keyyu J, Hoey EM, Prodohl P, Stothard JR, Trudgett A: The distribution of Fasciola hepatica and Fasciola gigantica within southern Tanzania - constraints associated with the intermediate host.

    Parasitology 2008, 135:495-503. PubMed Abstract | Publisher Full Text OpenURL

  11. Wamae LW, Hammond JA, Harrison LJS, Onyango-Abuje JA: Comparison of production losses caused by chronic Fasciola gigantica infection in yearling Friesian and Boran cattle.

    Trop Anim Health Prod 1998, 30:23-30. PubMed Abstract | Publisher Full Text OpenURL

  12. Hubendick B: Recent Lymnaeidae: Their variation, morphology, taxonomy, nomenclature and distribution.

    Kungl. Svenska Vetensk. Akad. Handl 1951, 3:1-223. OpenURL

  13. Kanyari PWN, Kagira JM, Mhoma JRL: Prevalence of endoparasites in cattle within urban and peri-urban areas of Lake Victoria Basin, Kenya with special reference to zoonotic potential.

    Sci Parasitol 2010, 11:171-178. OpenURL

  14. Malone JB, Gommes R, Hansen J, Yilma JM, Slingenberg J, Snijders F, Nachtergaele F, Ataman E: A geographic information system on the potential distribution and abundance of Fasciola hepatica and F. gigantica in east Africa based on Food and Agriculture Organization databases.

    Vet Parasitol 1998, 78:87-101. PubMed Abstract | Publisher Full Text OpenURL

  15. UBOS:

    The National Livestock Census Report 2008. 2009. OpenURL

  16. Magona JW, Mayende JSP: Occurrence of concurrent trypanosomosis, theileriosis, anaplasmosis and helminthosis in Friesian, Zebu and Sahiwal cattle in Uganda.

    Onderstepoort J Vet 2002, 69:133-140. OpenURL

  17. Fabiyi JP: Production lossess and control of helminths in ruminants of tropical regions.

    Int J Parasitol 1987, 17:435-442. PubMed Abstract | Publisher Full Text OpenURL

  18. Ogambo-Ongoma AH: Fascioliasis survey in Uganda.

    Bull Epizoot Dis Afr 1972, 20:35-41. PubMed Abstract OpenURL

  19. Otim CP, Ocaido M, Okuna NM, Erume J, Ssekitto C, Wafula RZO, Kakaire D, Walubengo J, Okello A, Mugisha A, Monrad J: Disease and vector contraints affecting cattle production in pastoral communities of Ssembabule district.

    Uganda Liverstock Research for Rural Development 2004, 16:article 35.

    http://www.lrrd.org/lrrd16/5/otim16035.htm webcite

    OpenURL

  20. Mandahl-Barth G: The freshwater molluscs of Uganda and adjacent countries.

    Ann. Mus. Congo Zool 1954, 32:1-206. OpenURL

  21. Nshangano WBO: Annual Report of the Animal Health Research Centre (Entebbe).

    In Incidence of helminthiasis in exotic cattle imported into Uganda Edited by Oteng AK. 1969.

    1970 pp.ii + 96 pp

    OpenURL

  22. Stensgaard AS, Jorgensen A, Kabatereine NB, Rahbek C, Kristensen TK: Modeling freshwater snail habitat suitability and areas of potential snail-borne disease transmission in Uganda.

    Geospat Health 2006, 1:93-104. PubMed Abstract | Publisher Full Text OpenURL

  23. Khaitsa ML, Hammond JA, Opuda-Asibo J: Use of meat inspection records in veterinary planning.

    Bull Epizoot Dis Afr 1994, 42:317-326. OpenURL

  24. Roche JR, Friggens NC, Kay JK, Fisher MW, Stafford KJ, Berry DP: Body condition score and its association with dairy cow productivity, health, and welfare.

    J Dairy Res 2009, 92:5769-5801. Publisher Full Text OpenURL

  25. Bennema SC, Ducheyne E, Vercruysse J, Claerebout E, Hendrickx G, Charlier J: Relative importance of management, meteorological and environmental factors in the spatial distribution of Fasciola hepatica in dairy cattle in a temperate climate zone.

    Int J Parasitol 2011, 41:225-233. PubMed Abstract | Publisher Full Text OpenURL

  26. Salimi-Bejestani MR, McGarry JW, Felstead S, Ortiz P, Akca A, Williams DJL: Development of an antibody-detection ELISA for Fasciola hepatica and its evaluation against a commercially available test.

    Res Vet Sci 2005, 78:177-181. PubMed Abstract | Publisher Full Text OpenURL

  27. Coyle T: Liver fluke in Uganda.

    Bull Epizoot Dis Afr 1956, 4:47-55. OpenURL

  28. Valero MA, Perez-Crespo I, Periago MV, Khoubbane M, Mas-Coma S: Fluke egg characteristics for the diagnosis of human and animal fascioliasis by Fasciola hepatica and F. gigantica.

    Acta Trop 2009, 111:150-159. PubMed Abstract | Publisher Full Text OpenURL

  29. Salimi-Bejestani MR, Cripps P, Williams DJL: Evaluation of an ELISA to assess the intensity of Fasciola hepatica infection in cattle.

    Vet Rec 2008, 162:109-111. PubMed Abstract | Publisher Full Text OpenURL

  30. Kendall SB: Relationships between the species of Fasciola and their molluscan hosts.

    Adv Parasitol 1970, 8:251-8. PubMed Abstract OpenURL

  31. Periago MV, Valero MA, Panova MMas-Coma S: Phenotypic comparison of allopatric populations of Fasciola hepatica and Fasciola gigantica from European and African bovines using a computer image analysis system (CIAS).

    Parasitol Res 2006, 99:368-378. PubMed Abstract | Publisher Full Text OpenURL

  32. Brown DS: Freshwater snails and their medical importance. 2nd edition. Taylor & Francis, London, UK; 1994. OpenURL

  33. Stothard JR, Bremond P, Andriamaro L, Loxton NJ, Sellin B, Sellin E, Rollinson D: Molecular characterization of the freshwater snail Lymnaea natalensis (Gastropoda : Lymnaeidae) on Madagascar with an observation of an unusual polymorphism in ribosomal small subunit genes.

    J Zool 2000, 252:303-315. Publisher Full Text OpenURL

  34. Bargues MD, MasComa S: Phylogenetic analysis of lymnaeid snails based on 18S rDNA sequences.

    Mol Biol Evol 1997, 14:569-577. PubMed Abstract | Publisher Full Text OpenURL

  35. Bargues MD, Mangold AJ, MunozAntoli C, Pointier JP, MasComa S: SSU rDNA characterization of lymnaeid snails transmitting human fascioliasis in South and Central America.

    J Parasitol 1997, 83:1086-1092. PubMed Abstract | Publisher Full Text OpenURL

  36. Charlier J, De Meulemeester L, Claerebout E, Williams D, Vercruysse J: Qualitative and quantitative evaluation of coprological and serological techniques for the diagnosis of fasciolosis in cattle.

    Vet Parasitol 2008, 153:44-51. PubMed Abstract | Publisher Full Text OpenURL

  37. Lotfollahzadeh S, Mohri M, Bahadori SR, Dezfouly MRM, Tajik R: The relationship between normocytic, hypochromic anaemia and iron concentration together with hepatic enzyme activities in cattle infected with Fasciola hepatica.

    J Helminthol 2008, 82:85-88. PubMed Abstract | Publisher Full Text OpenURL

  38. Preston JM, Castelino JB: Study of the epdiemiology of bovine fascioliasis in Kenya and its control N-trilylmorpholine.

    Br Vet J 1977, 133:600-608. PubMed Abstract OpenURL

  39. Smith G, Wilson RA: Seasonal variations in the microclimate of Lymnaea truncatula habitats.

    J Appl Ecol 1980, 17:329-342. Publisher Full Text OpenURL

  40. Ogambo-Ongoma AH: The incidence of Fasciola hepatica in Kenya cattle.

    Bull Epizoot Dis Afr 1969, 17:429-431. PubMed Abstract OpenURL