African Journal of Biotechnology Vol. 1 (2), pp. 39-45, December 2002

ISSN 1684-5315  © 2002 Academic Journals  

The diagnosis of trypanosome infections: applications of novel technology for reducing disease risk

Picozzi, K.¹, Tilley, A.¹, Fèvre, E.M.¹, Coleman, P.G.¹, Magona, J.W.², Odiit, M.², Eisler, M.C³ and Welburn S.C.^1*

¹Centre for Tropical Veterinary Medicine, Royal (Dick) School of Veterinary Studies, The University of Edinburgh, Easter Bush, Roslin, Midlothian, Scotland, EH25 9RG, UK

²Livestock Health Research Institute, PO Box 96, Tororo, Uganda

³University of Glasgow Veterinary School, Bearsden Road, Glasgow, Scotland

*Corresponding author; tel: +44 131 650 6228; fax:  + 44 131 7348; e-mail: sue.welburn@ed.ac.uk

Accepted 12 November 2002

Reliable DNA based methodologies to determine prevalence of trypanosome species in domestic livestock have been available for over 10 years.  Despite this, they are rarely used to generate baseline data for control operations for these diseases in the field.  Rather, such operations tend to rely on data which can be generated using low technology methods such as direct observation of parasites by light microscopy.  Here we show the pitfalls of relying on such low tech methodology which, although simple in its application, can provide inaccurate and inadequate data on which to base control methodologies. Our analysis of 61 cattle selected for trypanosome carrier status by either microscopy, low PCV or poor condition score, showed that 90% were infected with trypanosomes while 84% of the total were infected with T. brucei. Diagnosis by PCR on buffy coat preparations on Whatman^® FTA^® matrices was the most sensitive methodology relative to the gold standard, whereas microscopy was the least sensitive.

Keywords: sleeping sickness, human African trypanosomiasis, epidemiology, chemotherapy, PCR.

Human African trypanosomiasis (HAT) or sleeping sickness, a disease thought to have been largely conquered during the 1960’s, has re-emerged as a serious public health problem over large swathes of sub-Saharan Africa (Welburn et al., 2001).  It has been estimated that 300,000–500,000 people are currently infected and 100,000 deaths are caused each year by the disease (Cattand et al., 2001).  HAT is strictly a problem for rural communities in Africa since it is dependent for transmission on tsetse.  The disease persists in areas which have suffered most acutely from general breakdown of infrastructure, including health care (Louis, 2001).

The cause of human sleeping sickness in eastern Africa is Trypanosoma brucei rhodesiense, a zoonotic parasite, which is mainly a non-pathogenic parasite of livestock and wild bovids (Heisch et al., 1958; Onyango et al., 1966) but which results in sleeping sickness when transmitted to humans.  Control of T. b. rhodesiense HAT is complicated by the fact that T.b. rhodesiense co-exists in domestic livestock with the morphologically identical Trypanosoma brucei brucei which is not pathogenic to humans.  Furthermore, in the regions where HAT prevails, several other trypanosome species, including T. vivax and T. congolense, are prevalent which affect the health of cattle and other livestock.  Taken as a group, the trypanosomiases of livestock are responsible for severe losses in the agricultural sector (Kristjanson, 1999), and together with the human disease burden imposed by human sleeping sickness (Odiit et al., personal communications), the trypanosomiases more generally form a very significant group of parasitic infections.

Quantifying the risk presented to rural peoples by T.b. rhodesiense infected livestock and wild animals is essential in determining the extent to which control activities are required and the control methodology which is appropriate. Of particular importance to public health is this new possibility of rapidly identifying animals that are serving as reservoirs for human infective trypanosomes.  The risk of spreading sleeping sickness is closely related to the movement of cattle populations, and the screening and treatment of livestock to prevent disease spread has been advocated (Fèvre et al., 2001). This was until recently impossible, as there was no means of safely differentiating this human infective parasite from morphologically identical T. b. brucei.  Recently, however, it was observed that a strain of T. b. rhodesiense belonging to the Busoga zymodeme (noted for its persistence over time in southeast Uganda) had a gene (serum resistance associated gene, SRA), co-expressed with the variant antigen gene, which was shown experimentally to confer human serum resistance (Xong et al., 1998).  This gene has been shown to unequivocally differentiate T. b. brucei from T. b. rhodesiense in southeast Uganda.  SRA has been found in all human isolates of T. b. rhodesiense identified in southeast Uganda to date and moreover can confirm the human infectivity status of trypanosomes isolated from animals. SRA is not present in those samples previously designated as T. b. brucei by restriction fragment length polymorphism analysis (Welburn et al., 2001b; Welburn and Odiit, 2002).

Although these molecular diagnostic tools are proving extremely useful for elucidation of sleeping sickness and animal trypanosomiasis epidemiology, they are also of practical use in terms of better targeting drug use in animals to improve animal health. Perhaps most importantly, the molecular tools allow the identification of individual reservoir hosts carrying human-infective T. b. rhodesiense parasites. Despite this, however, more traditional technologies are still in widespread use, particularly microscopy-based diagnosis (reviewed in Uilenberg, 1998).

In this paper, we compare different methodologies used to prepare parasite material in the field for the eventual assessment of trypanosome-infection status of the animals, and thus for the risk posed by those animals to disease in cattle-keeping populations.  The more accurate the diagnosis of infections can be, the more targeted interventions can become and the greater the reduction in the risk of human disease.  Selective treatment of infected animals in this way results in a holistic approach to disease control, concurrently combating livestock and human disease agents, and resulting in an overall improvement in the livestock and human health of rural populations.

Sampling

In 2001, as part of a large survey of trypanosome infections in the cattle population in the sleeping sickness focus in Soroti district, eastern Uganda (Fevre et al., 2001), 150 animals were subjected to clinical examination and screened for trypanosomes by two standard field methods, the Buffy Coat technique, BCT (Murray et. al., 1977) and using the haematocrit centrifugation technique, HCT (Woo, 1970). Animals were selected for the study according to either a  positive  identification  of  carrier  status  by microscopy, a low PCV or a poor observed condition score (Nicholson & Butterworth, 1986) following Duvallet et al. (1999) who showed that 50% of parasitologically negative animals with a haematocrit value under 25% were subsequently identified as trypanosome positive using PCR.  Ten millilitres of blood was withdrawn into a heparinised vaccutainer from the jugular vein [Solano et al. (2002) showed heparin was less inhibitory to PCR reactions than EDTA].

Sample preparation for PCR in the field

Three methods of sample preparation were assessed for sensitivity analysis for PCR determination of T. brucei: direct DNA preparation using DNAzol; application of whole blood to Whatman FTA cards; application of buffy coat preparations to FTA cards.

Direct DNA preparation

DNA was directly isolated from 1ml of bovine blood using DNAzol^® BD Reagent (Chomczynski et al., 1997).  DNAzol “is a complete, non-toxic and ready to use reagent for the isolation of genomic DNA from various biological sources”.  Whole blood was mixed at a ratio of 1:2 with DNAzol^® BD prior to the addition of isopropanol.  The sample was shaken vigorously and left at room temperature for 5 minutes.  Precipitated DNA was isolated by centrifugation at 6000 g for 6 minutes and stored at -20°C prior to further processing.

Whole blood application to FTA^® Cards

One hundred microlitres of whole blood was applied directly onto FTA^® Cards (Whatman^®) which were allowed to dry thoroughly prior to storage at room temperature.

Buffy coat preparations onto FTA^Ò Cards

One ml whole blood was centrifuged at 10,000 g for 5 minutes.  100ml of buffy coat was withdrawn and applied directly on the FTA^®  Cards.

Preparation of DNA for PCR analysis

DNAzol^® BD lysate: The DNA pellet was washed with 0.5ml of reagent buffer, mixed to ensure complete dispersal, and centrifuged at 2000g for 4 minutes.  The supernatant was removed and the pellet washed with 1ml of 95% ethanol. DNA was recovered by centrifugation and solubilised in 66ml of 8mM NaOH, the solution neutralised after 5 minutes with the addition of 33ml of HEPES.

FTA^® Card: Two mm discs were cut from the sample-saturated cards and prepared according to the manufacturers instructions.  Briefly, the discs were washed twice in FTA purification reagent to remove any PCR inhibitors from the sample, and traces of FTA buffer were removed by two further washes in TE buffer.  The discs were air dried, then transferred to PCR tubes where they were used to seed the amplification reaction.

Trypanosome species specific PCR

DNA prepared using DNAzol, and bound to FTA matrix (from blood of buffy coat source) were assessed for the presence of the trypanosomes which may be present in domestic cattle; namely Trypanozoon, Duttonella and Nannomonas (four types of Trypanosoma congolense).  The primers used were supplied by Sigma (see Table 1).

    Table 1. Primer sequences and amplification product size.

PCR reaction conditions and amplification protocols  

Standard PCR amplifications were carried out in 25ml reactions mixtures containing the final concentrations, 10 mM TrisHCL pH 8.3, 50 mM KCl, 1.5 mM MgCl₂, 200 mm of each of the 4 deoxynucleoside triphosphates and 1 Unit of REDTaq DNA polymerase (Sigma).  Genomic template preparations of 1 ml DNAzol purified DNA or 1x2mm FTA card discs were amplified.   Specific PCR conditions in terms of primer concentrations and amplification programmes are listed on Table 2, a final elongation step of 5 minutes was incorporated into each programme. PCR products were separated by electrophoreses in a 1.5% (w/v) agarose gel containing 0.5 mg/ml ethidium bromide and visualised by ultraviolet light.

Table 2. Species specific PCR conditions and programmes.

Defining any trypanosome species infection

It was assumed that each test was 100% specific (i.e. identification by microscopy or detection or amplification of DNA primers was indicative of trypanosome presence). Therefore, the gold standard, or true infection status of each of the 61 cows, was defined as positive if at least one of the four diagnostic tests were positive.

Defining T. brucei species infection

Because of the difficulty of distinguishing trypanosome species with microscopy, the gold standard, or true T. brucei species infection status for each of the 61 cows, was defined as positive if a least one of the three DNA tests were positive.

Statistical analysis

For each of the four tests, the sensitivity (and 95% confidence intervals) was calculated relative to the gold standard status, for both infection with any trypanosome species and infection with T. brucei species. Differences between the sensitivity of the different tests were examined using the c² test. The kappa statistic (and 95% confidence intervals), which gives a measurement of the degree of agreement between each of the four tests and the gold standard, was also calculated, again for infection with any trypanosome infection and infection with T. brucei species. The value of kappa lies between 0 and 1, with 1 meaning perfect agreement between tests and 0 meaning that the association is no better than expected from chance alone. All statistical analyses were performed in SPSS (release 10.0.5).

 

Abstract

Introduction

Materials and Methods

Results

Discussion

References

A total of 61 cows were diagnosed using the four tests. A summary of the results is shown in Table 3. Overall, 55 (90.1%) of the cows had a trypanosome infection with 51 infected with T. brucei species. The sensitivities of the four different tests, for any trypanosome infection and T. brucei species specifically, relative to the gold standards (defined above) are shown in Figure 1. Statistical comparisons of the test sensitivities are shown in Tables 4a (for any trypanosome species detection) and 4b (for T. brucei species detection). The measurements of agreement between each of the four diagnostic tests and the defined gold standard are shown in Figure 2.

Table 3. Summary of positive diagnostic results. 

*includes one cow that was identified as positive for T. brucei species with microscopy but negative on all three PCR test. The cow was shown by PCR to have a T. vivax infection.

Figure 1. Sensitivity of the four diagnostic tests relative to the gold standard.

Figure 2. Measure of agreement (kappa statistic) between each of the four diagnostic tests and the gold standard.

Table 4a. Comparison of the test sensitivities for any trypanosome species-detection using the c² test

 

	Microscopy	DNAzol	Cards-blood	Cards-buffy coat
Microscopy	_	c2=7.14 P=0.008	c