Despite the importance of host species choice of mosquito vectors to the epidemiology and control of malaria, our understanding of the ecological and evolutionary factors that drive the host species preference in these vectors is very limited. My PhD thesis aimed to experimentally investigate the potential ecological and evolutionary determinants of the host species choice of the African malaria vectors Anopheles arabiensis and An. gambiae s.s, which are amongst the most highly specialized and efficient malaria vectors in the world, and identify a control strategy that reduces their anthrophily. I used a unique semi-field system where these vectors were able to interact naturally with hosts of different species to establish whether their fitness depends on type of host species, they encounter and feed upon. My initial prediction was that highly host-specific feeding behaviour of these vectors is a product of natural selection whereby mosquito fitness is highest on their naturally preferred host types. This prediction was met in An. arabiensis, whose feeding success and lifetime egg production was predicted to be higher on their naturally preferred bovid hosts. However, I did not detect any association between the preference of An. gambiae s.s for humans and their lifetime reproductive success, although they obtain larger blood meals and survived longer on these naturally preferred human hosts. These findings suggest the role of host species on mosquito fitness varies between vector species. I then evaluated whether the host species-specific fitness of malaria vectors may be attributed to intrinsic defensive behaviours and haematological properties that make some host species being more beneficial than others. My initial prediction was that mosquito feeding success and fitness would be the highest in the absence of host defensive behaviours and, more specifically, that the least defensive host species would be the most highly preferred in nature. I have found that the feeding success (probability of obtaining a blood meal) of An. arabiensis is greater on host species with least effective defenses (e.g. bovids). However, this association was not apparent for anthrophilic An. gambiae s.s. Surprisingly, I found that the subsequent fitness (blood meal size and survival) of both vector species was generally greater on hosts who were free to exhibit defensive behaviours than those whose behaviours were restricted. These findings suggest that natural physical defensive behaviours made by hosts including humans may not impose strong fitness costs to malaria vectors. Therefore, I conclude that if natural host defensive behaviours shape the host species preference of malaria vectors they do so by influencing the probability of acquiring a blood meal but not the value of the blood meal if obtained. I also assessed whether the nutritive value of host blood, as determined by haematological properties of packed cell volume (PCV) and haemoglobin concentration (Hb), could explain variation in fitness of malaria vectors on different host species. I found that the PCV and Hb of host species that are commonly encountered by malaria vectors in their natural environments vary significantly. I further found that the variation in these haematological properties influence the feeding success (e.g. blood intake rate) of the anthrophilic An. gambiae s.s but not the An. arabiensis. Anopheles gambiae s.s obtain full blood meal faster on hosts with low and medium levels of PCV. Surprisingly, these haematological traits were predicted to have opposite effects on the survival of both vector species. The survival of An. gambiae s.s was positively correlated with host PCV, but negatively correlated with their Hb. In contrast, the survival of An. arabiensis was predicted to be positively correlated with host Hb, but negatively related with PCV. Overall, there was no clear evidence that haematological properties of the host species preferred by these mosquito vectors are optimal for their fitness. I then extended my investigations to a laboratory investigation to measure the impact of host species diversity on the fitness of An. gambiae s.s throughout their life. Under these conditions, I found that An. gambiae s.s had similar fitness after either feeding on a uniform (human-only) or mixed host species. These findings indicate that the blood composition of different species may be unlikely to reduce the fitness of An. gambiae s.s My PhD thesis also experimentally measured the impact of using simple intervention (e.g. an untreated bed net) on reducing the fitness of malaria vectors that acquire from human hosts. I found that the lifetime reproductive output of An. arabiensis on protected human was significantly lower than on bovid hosts. In contrast, the use of untreated nets by humans reduced survival of anthrophilic An. gambiae s.s, but the reduction was not predicted to be sufficient to significantly reduce the total lifetime reproductive output of these mosquitoes on human hosts than on animal alternatives. These findings suggest that the widespread use of simple untreated net may generate selection pressures for An. arabiensis to maintain their feeding on bovid hosts and to a lesser extent for An. gambiae s.s to reduce their anthrophily. The findings of my PhD research have implications for the epidemiology and control of malaria. I found that host species and their intrinsic properties may influence aspects of the feeding success, blood meal size and survival of malaria vectors which are the key determinants of malaria transmission intensity. I further demonstrate that selectively protecting humans with untreated nets may generate selection pressures for malaria vectors to reduce their anthrophily and consequently the transmission intensity of malaria. These findings suggest integrating existing interventions (e.g., use of untreated and insecticide treated bed nets) with environmental management that increases availability of an alternate host species (e.g. zooprophylaxis) may generate selection pressures for An. gambiae s.s to reduce their anthrophily, and An. arabiensis to maintain their feeding on alternative animal hosts (zoophily). Overall, I discuss the impacts of host species choice and intrinsic host factors on the fitness of African malaria vectors, the impacts of intervention on their fitness and their potential to select for a host shift, and the implications to epidemiology and control of malaria. I finally highlight gaps in the knowledge of the evolution of host species choice in malaria vectors where more research is required.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:532711 |
| Date | January 2010 |
| Creators | Lyimo, Issa |
| Publisher | University of Glasgow |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://theses.gla.ac.uk/2585/ |
Page generated in 0.0025 seconds