A systems biology approach to infectious disease transmission: linking individuals populations and ecosystems

Did you ever wonder (and wish) that mosquitoes would bite your dog instead of you? Ecological theory tells us that animals should evolve to specialize on diets that most increase their survival and reproduction. In the case of blood-feeding insects, this evolution may explain why some species bite only humans, whereas others prefer domesticated or wild animals. Evidence shows that the survival and reproduction of most blood-feeding insects, including mosquitoes, bed bugs, fleas, and biting flies, depends on the species of vertebrate that they bite. The survival of parasites within these insects, many of whom cause severe disease in humans and animals, is also influenced by vertebrate species choice; with the blood of some animals enhancing parasite growth, and others blocking it. Given this is the case, what would happen if the range of animal hosts available to blood-feeding insects was suddenly changed on a landscape scale, and how would this change impact the spread of human and animal disease? This experiment is occurring in nature, where the variety of animal species available to blood-feeding insects is rapidly changing as human populations expand; reducing the availability of other vertebrates in favour of themselves and their livestock. The downstream impacts of this narrowing of vertebrate species choice are not yet known; particularly with respect to how it will impact the biodiversity and abundance of biting insects, and the spread of the diseases they carry. Understanding how these changes in vertebrate host species composition influence the survival, population growth and parasite transmission potential of biting insects is the central aim of this project. Using this information to formulate and test mathematical models that predict how insect abundance and disease risk will change in response to specific human-induced land-use changes is my ultimate goal. In Africa, a continent inflicted by the world's deadliest insect-borne diseases; changes in land-use practices are starting to take place. Large-scale urbanization and intensive agricultural development are not yet common, but will become increasingly so within the next decade. Thus here more than anywhere, study of the impact of vertebrate species diversity on blood-feeding insects has the potential to provide pre-emptive solutions to detrimental ecological and epidemiological consequences associated with changes in land use. I will conduct a series of novel laboratory analyses, biologically realistic behavioural assays, and intensive field collections in order to test whether the survival and reproduction of the common African mosquitoes Anopheles gambiae s.s. and An. arabiensis is influenced by the availability of host species that they prefer (humans and cows) relative to those that are secondary (goats, chickens and dogs). I will test whether the ability of these two mosquito species to compete with one another is determined by the relative abundance of their preferred host species (humans for An. gambiae, and cows for An. arabiensis). Finally I will investigate whether the development of malaria parasites within these mosquitoes is influenced by the type of blood (human, cow, goat, chicken or dog) that they consume when they are infected. Information gathered in the experiments will be used to formulate a mathematical model of mosquito and parasite population growth as a function of host species composition. I will use this model to examine how skewing the composition of vertebrate hosts towards humans (as expected under urbanization), or to a mixture of livestock and humans (as expected under agricultural expansion) will influence mosquito abundance, biodiversity, and malaria transmission intensity. Finally, these models will be fit to actual landscapes within a malaria-endemic region of east Africa to predict how current and forecasted changes in land-use activities could impact Anopheline population dynamics and human health.

To predict the behaviour of complex biological systems such as organisms, populations and ecosystems; the links between genes, physiology, organism fitness, and population dynamics must be identified. Few biological systems have been sufficiently characterized to permit this, with a rare exception being human infectious diseases that have been studied extensively at the molecular, organismal, and population level. Here I will combine the rich multi-scale data available for human malaria parasite with new experiments that address linkages betsween processes occurring at different scales to evaluate the ability of biological properties associated with individual resource use to inform population, community and epidemiological dynamics. I will begin with lab experiments that test how the fitness of individual mosquito vectors and their parasites depends on the vertebrate species from which they take blood. Field studies will follow to examine how vertebrate behaviour influences the feeding success and mortality of individual mosquitoes; and how intra- and interspecific competition between larvae of vectors An. gambiae and An. arabiensis are influenced by the vertebrate host choice of their parents. These individual-based behavioural and fitness data will be used to parameterize a spatially-explicit demographic model of mosquito and parasite dynamics. Within a Geographic Information System, I will simulate known vertebrate host distributions from 6 Tanzanian sites; and examine the model's ability to predict the rank-order of independently-collected, hierarchical epidemiological properties of each site including vector survival, reproduction, abundance, species composition, and finally parasite transmission. This integrative approach uniting experiments, field observation, and mathematical modelling will provide a unique opportunity to test our ability to predict the dynamics of a complex infectious disease system as a function of lower-order biological phenomena.