Gebäude 35, 22, 23
Tel.+49 30 838 62310 Fax.+49 30 838 62323
Tsetse flies inhabit 10 million km2 of subsaharan Africa, transmitting Human African Trypanosomoses (HAT) and African Animal Trypanosomoses (AAT). Public health services of most African countries are not able to reach the affected rural communities. Besides, trypanocides often are inefficient and vaccinations are unavailable. Thus, various means of vector control remain for disease management. In order to avoid unreasonable interventions against tsetse, decision support tools help defining the most efficient control strategies: trypanosomosis risk assessment and profound knowledge on local tsetse populations and their behaviour. Large-scale risk surveys and tedious serological laboratory analyses are too expensive at the community-level. That is why the objective of this work was rationalizing trypanosomosis risk assessment and improving current tsetse analysis methods.
Chapter 1 provides a literature review on trypanosomosis epidemiology, tsetse biology, physiology, control means and methods for risk assessment and bloodmeal analysis.
Chapter 2 deals with the application of a tsetse challenge formula that simplified relative AAT risk estimation in 2 villages of the Sikasso region in southeast Mali. During 6 months tsetse were trapped at animal watering sites, followed by microscopic examination of the flies for trypanosome infection rates and by PCR analysis of tsetse bloodmeals. Bloodmeals were identified by species-specific cytochrome b primers that amplified vertebrate mitochondrial DNA and by sequencing unidentifiable samples. The outcome of the field study revealed that Glossina morsitans submorsitans had vanished, while Glossina palpalis gambiensis (Gpg) and Glossina tachinoides (Gt) were still present in this area with 369 and 105 caught tsetse, respectively. Further, it became obvious that the tsetse were unevenly distributed with catches of 2-152 flies per trap with the majority in direct proximity of watering places while being absent from distances of 20 metres and onwards from a river. Trypanosome infection rates of the flies varied between 0% and 33.3% depending on the trapping location. The analysis of 120 bloodmeals revealed cattle and humans as main hosts while 2 samples showed crocodile DNA. The tsetse challenge of the 2 villages differed significantly with 6 days vs. 77 days that had to be spent by cattle at the watering site in order to contract AAT. The obtained value could in both cases be linked to the trypanosome prevalence of nearby cattle herds.
Further analysis of tsetse deriving from 20 traps in 4 villages revealed unexpected differences between the 279 analysed Gt and Gpg. Gt demonstrated no host preference whatsoever because their feeding pattern comprised in equal shares humans, cattle and surprisingly mixed meals of both. Multiple host feeding, yet rarely been described in tsetse research, did occur significantly less often in Gpg (p<0.05). Gpg showed a preference for humans over cattle (66.5% and 10.3%, respectively). The infection rate also differed with Gt being 3-fold more likely to be infected with trypanosomes (18.5%) than Gpg (5.5%). Therefore, chapter 3 contains a logistic regression analysis of the factor mixed bloodmeal towards the factors species, infection, hunger stage and sex. The statistics demonstrated that multiple host feeding was not linked to high infection rates or age but that it positively correlated with female sex in Gt and fully engorged Gpg. It is then discussed how multiple feeding possibly impacts trypanosomosis transmission mechanisms, assuming a higher vectorial competence of Gt compared to Gpg.
Although PCR has proven more sensitive than serological methods, the development of MALDI TOF MS (matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry) has become a more rapid tool for routine microbial diagnostics. Insects have rarely been specified by proteomic means, so chapter 4 consists of a proteomic database construction for the tsetse species G. morsitans morsitans, G. pallidipes, G. austeni, G. palpalis gambiensis and G. brevipalpis based by MALDI TOF MS. Lab-reared flies were analysed as entire insects and dissected, obtaining their head, wings, legs, thorax and abdomen. After a simple protein extraction, 60 mass spectrum peak patterns were created as reference spectra. The following principle component and cluster analysis confirmed that each body part was suitable for exact speciation. Evaluation of the database by crosschecking with newly extracted isolates resulted in a composite correlation index that demonstrated reliable tsetse speciation. Dendrograms drawing on peak similarity showed that G. brevipalpis stood consistently apart from the other species, confirming genomic findings that suggested their sister group status. As expected, tsetse of the morsitans group tended to cluster, with the exception of G. austeni that did not show consistent affinities to any of the 3 groups reflecting uncertainties about their group status in recent tsetse taxonomy literature. So, the constructed database apparently displayed genomic findings at the protein level and it proved to be a rapid and accurate tool for tsetse species determination.
The results are discussed in chapter 5. It could be demonstrated that a simplified risk assessment formula is able to provide AAT risk trends. This will be useful for planning future vector interventions more rationally, making it available for community-based projects. Thereby, species-specific PCR proved more efficient for bloodmeal analysis than serological methods. Still, obtaining the host preference remains the most laborious tsetse parameter, making it the limiting factor to a more time-efficient risk evaluation. Since rapid MALDI-based diagnostics at the species-level could be established, extending the database is warranted for high-throughput proteomic tsetse identification at the population-level, trypanosome diagnostics and bloodmeal analysis.