Pathogenic animal trypanosomes affecting livestock have represented a major constraint to agricultural development in Africa for centuries, and their negative economic impact is increasing in South America and Asia. Chemotherapy and chemoprophylaxis represent the main means of control. However, research into new trypanocides has remained inadequate for decades, leading to a situation where the few compounds available are losing efficacy due to the emergence of drug-resistant parasites. In this review, we provide a comprehensive overview of the current options available for the treatment and prophylaxis of the animal trypanosomiases, with a special focus on the problem of resistance. The key issues surrounding the main economically important animal trypanosome species and the diseases they cause are also presented. As new investment becomes available to develop improved tools to control the animal trypanosomiases, we stress that efforts should be directed towards a better understanding of the biology of the relevant parasite species and strains, to identify new drug targets and interrogate resistance mechanisms.

The first step in studying the epidemiology of a disease is the accurate identification of the pathogen. Traditional reliance on morphological identification has given way to the use of molecular methods for the detection and identification of pathogens, greatly improving our understanding of epidemiology. For the African tsetse-transmitted trypanosomes, the growth of PCR methods for identification of trypanosomes has led to increased appreciation of trypanosome genetic diversity and discovery of hitherto unknown trypanosome species, as well as greater knowledge about the number and type of trypanosome infections circulating in mammalian hosts and vectors. Sequence data and phylogenetic analysis have provided quantitative information on the relatedness of different trypanosome species and allowed the new trypanosome genotypes discovered through the use of species identification methods in the field to be accurately placed in the phylogenetic tree.

The antigenic relationships of 7 stocks and 7 clones of Trypanosoma vivax from East and West Africa were compared by immune lysis. Sera from goats infected with different stocks and clones of T. vivax, collected on days 40 and 80 after infection, were used in the immune lysis test with homologous and heterologous stocks and clones of trypanosomes. Sera from infected cattle were included to compare stocks and clones from Kenya. The parasites that were used as antigen in the immune lysis tests were collected from infected mice when variable antigen type (VAT) homogeneous populations were used, from goats for infection with stocks and clones from Nigeria, The Gambia and Uganda, and from cattle for Kenyan stocks. Reciprocal cross-reactivity between sera and parasites was found between all the stocks and clones from Nigeria and The Gambia with the exception of one clone from Nigeria that was not recognized by antisera to a clone from The Gambia. There was also cross-reactivity between a stock and clone from Uganda and stocks and clones from Nigeria and The Gambia. Sera from goats infected with stocks and clones from Nigeria, The Gambia and Uganda recognized parasite populations that were homogeneous for one VAT (ILDat 1.2) of the rodent infective stock from Nigeria. Some antisera to West African stocks recognized another stable variant from a Ugandan stock adapted to rodents (ILDat 2.1), indicating that these VATs were expressed in the repertoires of the heterologous stocks. There was no cross-reaction between stocks from Nigeria, The Gambia or Uganda with Kenyan stocks. A stock from Galana (Kenya) and Bamburi (Kenya) showed reciprocal cross-reactivity. Two other Kenyan stocks, from Kilifi and Likoni, also showed cross-reactivity by immune lysis but showed no antigenic relationship with the other Kenyan stocks.

Sera from goats and cattle that were infected with two Trypanosoma vivax clones (ILDat 1.2 and ILDat 2.1) derived from different stocks were analysed for antibody activity against the variable surface glycoproteins (VSGs) of the infecting clones by enzyme-linked immune assays (ELISA) and immune lysis. To obtain purified VSG, lysed trypanosomes were separated on dodecyl sulphate-polyacrylamide gels. The gels were copper stained and the VSG protein band was excised from the gel. After destaining, the proteins were electroeluted from the gel slices and used as antigens in ELISA. High titres of IgM and IgG1 antibodies and lytic antibodies against the VSG of the infecting clone were detected. The IgG1 response appeared about 4 days later than the IgM response. IgG2 antibodies were only detected in goats and cattle that were infected with ILDat 1.2. Two goats and two calves that were infected with ILDat 1.2 showed recurrent peaks in lytic activity and of IgM and IgG1 antibody activity to the VSG of the infecting variable antigenic type (VAT). Two goats that were infected with ILDat 2.1 showed a similar pattern, but in two other goats there was a recurrent peak only in the IgM class. Recurrent peaks of antibody activity to the VSG of ILDat 1.2 and ILDat 2.1 were not detected in the sera of goats that had been inoculated with irradiated trypanosomes or that had been infected with an unrelated T. vivax clone. The recurrence of antibody peaks against the VSG of infecting VATs suggests that trypanosomes with completely or partially identical surface determinants reappear during T. vivax infection of ruminants.

Teneral Glossina morsitans centralis were fed on the flanks of African buffalo, N'Dama or Boran cattle infected with Trypanosoma vivax IL 2337. The infected tsetse were maintained on goats and on day 25 after the infected feed, the surviving tsetse were dissected to determine the infection rates. The mean mature infection rates (% ± S.E.) in the tsetse fed on buffalo, N'Dama and Boran cattle were 34·3 ± 9·9, 33·7 ± 13·4 and 58·9 ± 7·1, respectively. Logistic regression analysis indicated that infection rates in the labrum and hypopharynx of the tsetse were significantly lower when fed on the infected buffalo or N'Dama than Boran cattle. Similarly, the risk of infection was significantly lower in male than female tsetse. When teneral G. m. centralis, G. pallidipes, G. p. gambiensis, G. brevipalpis and G. longipennis were fed simultaneously on either the buffalo cow, the N'Dama bull or the Boran steer infected with T. vivax IL 2337, the mature infection rates were higher in the two morsitans group than the two fusca group tsetse, whilst G. p. gambiensis was relatively refractory to the infection, irrespective of the host species on which they fed. Logistic regression analysis indicated that the infection rates in the labrum and hypopharynx were significantly different amongst the five tsetse species for each of the three infected host animals. Nevertheless, the trypanotolerant African buffalo and N'Dama cattle may serve as reservoirs of T. vivax infection as can trypanosusceptible Boran cattle.

An incomplete ring of electric nets was placed around oxen which were either uninfected, infected with Trypanosoma vivax, or infected with T. congolense. The numbers of fed and unfed Glossina pallidipes caught on the nets were used to estimate the attractiveness of the oxen to tsetse, and the feeding success of the tsetse on the oxen. Oxen infected with T. congolense attracted more G. pallidipes than the other groups of oxen. Taking into consideration daily variation in the abundance or activity of the flies, oxen infected with T. congolense were about 70% more attractive to G. pallidipes than were uninfected oxen or oxen infected with T. vivax. The latter two groups mostly attracted high numbers of G. pallidipes on days when the flies were especially abundant or active. The feeding success of G. pallidipes declined with increase in the rate at which oxen made anti-fly movements. Taking this movement rate into consideration, the feeding success of G. pallidipes on oxen infected with T. congolense was approximately 60% greater than on uninfected oxen or oxen infected with T. vivax. It is suggested that vasodilation induced by T. congolense may account for the difference in feeding success. The level of parasitaemia of T. congolense or T. vivax was not found to affect either the attractiveness of oxen or the feeding success on oxen. There was significant daily variation in the mean fat content of male G. pallidipes caught around the oxen but no effect of mean daily fat content on the proportion of males that fed. The mean haematin content of fed male G. pallidipes was positively correlated with the packed cell volume of the ox on which they fed.

The characterization of four Trypanosoma vivax isolates from Colombia in South America showed that although minor phenotypic differences existed between them, these parasites are antigenically related and belong to a single serodeme. Characterization by isoenzyme assay, karyotyping and DNA probe analysis, showed the Colombian isolates to be more similar to the West African than to Kenyan T. vivax. There was, however, little serological cross-reactivity between South American and African groups of T. vivax. Although the T. vivax isolates from Colombia were pathogenic for dairy calves which showed the typical sign of progressive emaciation, these parasites failed to infect mice or tsetseand could not be cultivated as bloodstream forms in vitro.This study represents initial attempts to establish the phenotypic and serological diversity amongst T. vivax isolates from South America.

Trypanosome infection of mammalian hosts leads, within days, to a strong early response against a small, distinct number of parasite proteins. One of these proteins is the variable surface glycoprotein (VSG). Most of the others are apparently non-variable, intracellular trypanosome proteins. Two of these antigens I2 and I17 are now characterized at the molecular level. Both exhibit a highly repetitive amino acid sequence organization, but they show no sequence similarity either to each other or to any other proteins known to date. Preliminary serological analyses indicate that both allow the early, sensitive and specific detection of infections with different species of trypanosomatids, making them interesting candidates for the development of diagnostic tools for trypanosomiasis detection.

Trypanosoma evansi and Trypanosoma vivax have shown a very high immunological cross-reactivity. Anti-T. vivax antibodies were used to monitor changes in the T. evansi intracellular Ca2+ concentration ([Ca2+]i) by fluorometric ratio imaging from single parasites. A short-time exposure of T. evansi parasites to sera from T. vivax-infected bovines induced an increase in [Ca2+]i, which generated their complete lysis. The parasite [Ca2+]i boost was reduced but not eliminated in the absence of extracellular Ca2+ or following serum decomplementation. Decomplemented anti-T. evansi VSG antibodies also produced an increase in the parasite [Ca2+]i, in the presence of extracellular Ca2+. Furthermore, this Ca2+ signal was reduced following blockage with Ni2+ or in the absence of extracellular Ca2+, suggesting that this response was a combination of an influx of Ca2+ throughout membrane channels and a release of this ion from intracellular stores. The observed Ca2+ signal was specific since (i) it was completely eliminated following pre-incubation of the anti-VSG antibodies with the purified soluble VSG, and (ii) affinity-purified anti-VSG antibodies also generated an increase in [Ca2+]i by measurements on single cells or parasite populations. We also showed that an increase of the T. evansi [Ca2+]i by the calcium A-23187 ionophore led to VSG release from the parasite surface. In addition, in vivo immunofluorescence labelling revealed that anti-VSG antibodies induced the formation of raft patches of VSG on the parasite surface. This is the first study to identify a ligand that is coupled to calcium flux in salivarian trypanosomes.

In this study, we addressed the phylogenetic and taxonomic relationships of Trypanosoma vivax and related trypanosomes nested in the subgenus Duttonella through combined morphological and phylogeographical analyses. We previously demonstrated that the clade T. vivax harbours a homogeneous clade comprising West African/South American isolates and the heterogeneous East African isolates. Herein we characterized a trypanosome isolated from a nyala antelope (Tragelaphus angasi) wild-caught in Mozambique (East Africa) and diagnosed as T. vivax-like based on biological, morphological and molecular data. Phylogenetic relationships, phylogeographical patterns and estimates of genetic divergence were based on SSU and ITS rDNA sequences of T. vivax from Brazil and Venezuela (South America), Nigeria (West Africa), and from T. vivax-like trypanosomes from Mozambique, Kenya and Tanzania (East Africa). Despite being well-supported within the T. vivax clade, the nyala trypanosome was highly divergent from all other T. vivax and T. vivax-like trypanosomes, even those from East Africa. Considering its host origin, morphological features, behaviour in experimentally infected goats, phylogenetic placement, and genetic divergence this isolate represents a new genotype of trypanosome closely phylogenetically related to T. vivax. This study corroborated the high complexity and the existence of distinct genotypes yet undescribed within the subgenus Duttonella.

The taxonomic and phylogenetic relationships of Trypanosoma vivax are controversial. It is generally suggested that South American, and East and West African isolates could be classified as subspecies or species allied to T. vivax. This is the first phylogenetic study to compare South American isolates (Brazil and Venezuela) with West/East African T. vivax isolates. Phylogeny using ribosomal sequences positioned all T. vivax isolates tightly together on the periphery of the clade containing all Salivarian trypanosomes. The same branching of isolates within T. vivax clade was observed in all inferred phylogenies using different data sets of sequences (SSU, SSU plus 5.8S or whole ITS rDNA). T. vivax from Brazil, Venezuela and West Africa (Nigeria) were closely related corroborating the West African origin of South American T. vivax, whereas a large genetic distance separated these isolates from the East African isolate (Kenya) analysed. Brazilian isolates from cattle asymptomatic or showing distinct pathology were highly homogeneous. This study did not disclose significant polymorphism to separate West African and South American isolates into different species/subspecies and indicate that the complexity of T. vivax in Africa and of the whole subgenus Trypanosoma (Duttonella) might be higher than previously believed.

Trypanosoma evansi and Trypanosoma vivax are the most extensively distributed trypanosomes responsible for diseases in livestock. Western blot and indirect immunofluorescence assays revealed a high immunological cross-reaction between these two parasites. An antigen with an apparent molecular mass of 64 kDa (p64), which exhibited cross-reactivity with T. vivax, was purified to homogeneity from a Venezuelan isolate of T. evansi. This antigen is glycosylated, contains a glycosyl-phosphatidylinositol anchor and appeared to be localized through the cell except in the nucleus, indicating that it could primarily be confined to the parasite surface. These results, together with its relative abundance and apparent molecular weight, suggest that p64 probably corresponds to the soluble form of a variable surface glycoprotein from T. evansi. Anti-p64 polyclonal antibodies, raised on mice, recognized a 53 kDa polypeptide band from a Venezuelan isolate of T. vivax on Western blots. Additionally, sera obtained from naturally infected animals also recognized p64, suggesting its potential use as a diagnostic reagent. Mild acid treatment only slightly decreased the immunorecognition of p64, suggesting its potential use as a diagnostic reagent. Mild acid treatment only slightly decreased the immunorecognition of p64, demonstrating that another relevant cross-reacting epitope, different than the inositol-1,2-cyclic phosphate of the cross-reacting determinant, must exist in p64. To date, p64 represents the first antigen isolated and partially characterized from T. evansi.

A two-year survey was undertaken to determine the trypanosomosis risk of Egbe, a town in Kogi state of Nigeria, and its suitability for cattle raising. Tsetse flies were trapped using biconical traps and examined for trypanosome infection, and blood samples from cattle and goats were examined by blood film and haematocrit centrifugation technique (HCT) for trypanosome parasites. No infection was detected from the 152 Glossina palpalis palpalis and 52 G. tachinoides dissected over the period. Blood film examination of 181 cattle and 14 goats also gave negative results. Examination by HCT of 21 cattle from a local abattoir showed 14.3% Trypanosoma vivax infection, but the 10 resident cattle and 14 goats similarly examined were negative. The average age of the caught flies, determined by mean wing fray values (for males) and ovarian analysis (for females), was between 11 and 14 days, which is too low to sustain an infective population. Coupled with the absence of G. morsitans submorsitans, the most important vector of animal trypanosomosis in Nigeria, and the disappearance of large game from Egbe, our results would point to the area being a low trypanosomosis risk one.

Epidemiological models are used to analyse 8 published data sets reporting age–prevalence curves for trypanosome infections of the tsetse fly Glossina pallidipes. A model assuming a fixed maturation period and a rate of infection which is independent of fly age is adequate for Trypanosoma vivax-type infections, explaining 98% of observed variance in prevalence by site and age, allowing that the rate of infection may be site dependent. This model is not adequate for T. congolense-type infections and the fit can be improved by allowing (i) the rates of infection to decline with age (although non-teneral flies remain susceptible), (ii) a fraction of resistant flies, which may vary between sites, (iii) increased mortality of infected flies and (iv) variation in the maturation period. Models with these features can explain up to 97% of observed variance. Parameter estimates from published experimental data suggest that all may contribute in practice but that (i) and/or (ii) are likely to be the most important.

Studies were conducted during the dry period February to May 1986 in River Shelliekha, Bahr El Arab fly belt, Sudan, to obtain baseline data on catch and age compositions, reproductive status and trypanosome infection rates of trapped and fly round samples of Glossina morsitans submorsitans. The objective was to assess the tsetse situation before the establishment of trypanosomiasis treatment stations at the northern limits of the fly belt. Trapped tsetse included significantly higher proportions of teneral and non-teneral females and a lower male:female ratio. Insemination rates were over 98% in both samples. In any one group, the frequency of pregnancy with egg predominated, followed by the second, first and lastly the third instar larva. There were significant differences between the two groups of females in the proportions of nullipars and pregnancy with any one of the larval instars. Abortion was the predominant reproductive abnormality and no relationship could be found between abortion rate and the sampling method, age or trypanosome infection rate of females. Age compositions were similar in the two samples, save for age categories 0 and 1 for females and two and four for males. Flies were infected with Trypanosoma vivax and T. congolense only, though T. brucei infections could also be diagnosed in livestock.

The infection rates and transmission characteristics of sterile male Glossina morsitans centralis, G. austeni and G. tachinoides for Trypanosoma vivax, T. congolense and T. brucei were studied. Teneral males were irradiated with 10 krad in a 137caesium source under ambient conditions. Five groups of teneral males of each species were sterilised by irradiation with 10 krad caesium source under ambient conditions and then fed for 3 days, together with five groups of normal males of their respective species, on three goats infected with T. vivax, T. congolense or T. brucei. The mean sterility rates were 98, 93 and 90% for G. m. cenlralis, G. austeni and G. tachinoides respectively. The mean percentage mature infection rates in the irradiated as against the controls of G. m. centralis, G. austeni and G. tachinoides were respectively 27.3 ± 9.5/27.7 ± 11.1, 4.0 ± 1.4/4.0 ± 2.6 and 14.8 ± 7.0/12.2 ± 5.0 for T. vivax; 24.6 ± 5.2/30.3 ± 4.3, 0.7/0.3 and 0.3/0.3 for T. congolense; and, 22.7 ± 2.6/23.4 ± 3.1, 1.1/0.2 and 0.7/1.7 for T. brucei. The transmission rates and frequency to A/J mice were very high and did not differ between the sterile and normal males. There was also no difference in the pattern of the diseases in the goats challenged by sterile or sexually fertile males of the three tsetse species. These results indicate that releases of sterile male tsetse in tsetse control programmes will potentially increase the trypanosomiasis risk in the affected area. Hence appropriate measures should be integrated in any SIR method of tsetse control campaign in order to provide adequate protection to man and his associated livestock.