Final Report: Predicting and monitoring climate change in insects: from genes to distribution shifts

Investigators: Ary Hoffmann (University of Melbourne) (chief investigator)

Carla Sgro (Monash University)

Scott Ritchie (James Cook University)

Craig Williams (University of South Australia)

Michael Kearney (University of Melbourne)

Peter Ryan (Queensland Institute of Medical Research)

Report date: December 2010

Summary of application: We need methods to effectively predict the risk of invasions of health/agricultural pest organisms, and the risk of extinction of threatened organisms, if we are to mitigate future changes in climate. We will use insects from threatened habitats and Aedesmosquitoes (vectors of dengue fever) to implement a new approach for predicting the future distribution of key species with climate change. Our approach links genetic markers and GIS data with powerful new mechanistic models that allows us to predict the impact of both evolutionary potential and changed human behaviour on the future distribution of species.

Commonwealth Environment Research Facilities – Significant Project


Summary of the major activities undertaken by the organisation

We tested more than 50 genetic markers for their involvement in climatic adaptation in a variety of ways. Using RNAi techniques, several markers were identified that affected cold and desiccation resistance, including heat shock genes, starvin, ebony and frost. We have shown that there has been a selective sweep around markers in the rainforest restricted species of Drosophila and we have identified several polymorphisms and non-functional genes. Several of the polymorphisms were linked to trait variation. We also used selection experiments and DNA hybridization to associate several genes with variation in desiccation resistance. Finally, the heritability of stress resistance was investigated in multiple Drosophila species to test the notion that genetic variation and stress sensitivity are linked. This component of the project contributed to 8 papers and two workshops, involved two postdoctoral researchers and also supported 3 students.

We developed trait-based mechanistic models that use energy balance equations to predict how climate will directly affect organisms under different climate scenarios. Physiological data were collected to fit parameters to the models and a novel evolutionary component was developed to investigate how evolved changes in a key trait (egg desiccation) would influence predicted distributions under climate change. We also collected data on the thermal properties of various types of water storage receptacles and modeled their suitability as breeding sites. Population comparisons were completed to investigate local adaptation and genetic variation in the mosquitoes. This extended to investigations of how different populations performed in terms of fecundity, mating success, and locomotion in various temperature regimes. Such work has helped us to understand the extent of local adaptation.

An existing mechanistic model was used to investigate the extinction processes currently working against Aedesaegypti populations in Australia. This work provided a cross-validation for the trait-based mechanistic models (mentioned above), highlighted the important environmental factors limiting current distribution of this species, and enabled a risk assessment of future distribution expansion. This work provided further evidence that egg longevity (mediated by desiccation resistance) is a key trait limiting the distribution of this species.

We also undertook household surveys for dengue mosquito immatures (larvae/pupae) and contemporary water storage container habitats in representative areas of Brisbane. These surveys provided importantcontextual information on recent changes in human behaviour, largely in response to changing weather patterns that may indirectly affect the abundance and distribution of Aedesaegypti in Queensland. The explosion in the abundance of peridomestic water storage containers in houses in south east Queensland (currently in 65% of houses), compared with their almost complete absence from houses fifteen years ago, represents an important change in the potential receptivity of south east Queensland (pop 2.8 million) to the re-introduction of dengue vectors.

An outline of any demonstration/communication activities undertaken (as specified in Item 1 of the Schedule)

Three workshops were run as part of the project. A workshop entitled “dengue and mosquito vectors in southern Queensland and other areas: what is the likely threat and what can we do about it?” was run in Brisbane in 2010 and attracted a variety of council workers, Queensland Health officials, and environmental managers. A workshop entitled “genetic translocations and adaptation under climate change; theory and practise” was run in Melbourne in 2010. The workshop attracted participants from the Department of Sustainability and Environment (Victoria), Department of Conservation and Land Management (Western Australia), CSIRO, and several universities and NGOs. A workshop entitled “genetic approaches to climate change adaptation” was run in Canberra in 2009. This workshop was well attended by Commonwealth Government DEWHA representatives.

A total of 11 publications have so far been produced during the project (more papers pending). A highlight of the project was the first demonstration of a link between trait heritability, species distributions and physiological limits published in Science which provides a new understanding of the genetic basis of evolutionary limits and the identification of species lineages that are particularly susceptible to climate change. They also include a major review published in Nature Reviews Genetics on genetic markers for climate adaptation, and several papers on candidate genes identified through association studies and gene knockout assays. The project also produced an important paper published in Functional Ecology where mechanistic models were first used to predict the likely effects of climate change on the dengue vector Aedesaegypti in northern Australia and for the first time incorporated evolutionary components into species distribution prediction. Two follow up papers highlight the importance of local knowledge in predicting species responses and the dangers posed by tyres in propagating mosquito populations under extreme conditions. The candidate gene list and distribution changes have both been presented on line and the mosquito distribution work generated substantial interest from the print and television media.

The results of the household surveys in representative areas of Brisbane have been communicated broadly to local and state governments, particularly Brisbane City Council which is the largest Local Government in Australia (Population 1,052,458). Based on the results from the CERF Project, Brisbane City Council have agreed to undertake broad surveys across all of Brisbane's 158 Statistical Local Areas (total of 10,000 households to be surveyed). This work will identify high risk areas for the establishment of dengue mosquitoes, which can then form the basis of a prioritised approach to monitoring of climate related range expansions in medically important insect species.

The benefits and outcomes of the Activity as a whole

The project has provided a new and sophisticated approach to understanding the likely effects of climate change in the distribution of pest species. This is the first time that mechanistic models have been applied to a pest species and the resulting models demonstrate the potential for dengue vectors to spread to Darwin and to southern locations as long as breeding containers are available. The project has also developed new methods for species distribution modelling incorporating evolution that should be equally applicable to other pest species and species of conservation concern. The project has highlighted the thermal buffering capacity of tyres as containers with the potential to support mosquito populations even in unfavourable areas, and it has demonstrated for the first time the presence of local adaptation in mosquito populations suggesting adaptive shifts.

The project provides a major breakthrough in how adaptive limits are regarded in species. This is the first demonstration that genetic limits underlie ecological limits in species, forcing lineages to be restricted to moist environments from which they cannot easily expand. The results raise the issue that vulnerability of species can be identified from species genomes, an idea that remains to be tested through genomic comparisons in the future. These results have been presented to numerous forums and stimulated discussion around physiological limits and the genomic basis of biodiversity. The adaptive markers identified in the project have been widely reported in the literature and should help to stimulate the use of adaptive markers for routine monitoring of populations. The markers are already being used in the identification of pollution effects.

The project has identified several issues of concern for health authorities in dealing with mosquitoes. Water storage containers are a risk to produce Ae. aegypti mosquitoes, and regulatory efforts should be made to ensure they are "mosquito proof". Furthermore, local government should consider public education and inspections of domestic rain water storage systems to prevention the potential re- emergence of Ae. aegypti in some urban areas, particularly the large population centres in south east Queensland (total population 3 million) where Ae. aegypti are currently absent.

The degree to which the Activity has effectively achieved its objectives (each objective should be discussed separately)

The project successfully achieved all its objectives. We pointed out that to manage species responses to climate change, it is crucial that we identify species most at risk from climate change before they become threatened by extinction, and then implement appropriate management strategies. We carried out extensive heritability experiments to test the notion that evolutionary potential can be limited in species. We also developed a series of genetic markers for monitoring the early impacts of climate change on species distributions and abundances.

We aimed to link the traits of the organism (rather than its distribution) to GIS data to predict their distribution. We showed how mechanistic models can be developed to predict the likely spread of the dengue fever transmission vector, Aedesaegypti, under climate change. The project identified key mosquito traits that determine its distribution and evaluated the potential for evolution to change these traits and further expand the Ae. aegypti distribution. The research program led to the discovery of genes indicative of adaptive potential well beyond the experimental system proposed.The approach developed in this proposal serves as a template for predicting changes in disease vectors and agricultural pests. It also serves as a template for rapid genetic approaches in screening threatened species.

The surveys in south east Queensland highlighted the important role that human adaptation strategies may play in future distribution of medically important insects. In south east Queensland, local government mosquito control programs do not routinely target mosquitoes in and around houses. However, these mosquitoes, commonly referred to as ‘container breeders’, can impact on human health and lifestyle. Previously, natural habitats, mainly bromeliad axils, garden accoutrements and discarded items were the most productive habitats for container mosquitoes in south east Queensland. This project has identified contemporary household water storage patterns in Brisbane over two consecutive Winter and Summer periods (2008-2009), including an explosion in the numbers of water storage containers. Now, almost one in two houses have a rainwater tank and in some suburbs up to a third of householders are collecting water in buckets, drums and other containers. This phenomenon is a direct result of water shortages and public policies to encourage house water conservation and storage. Intentionally filling (garden accoutrements) or harvesting (water storage containers) of water provides suitable, year round, habitat for container breeding mosquitoes, potentially increasing the receptivity of urban Brisbane to dengue vectors. This empirical information, combined with the mechanistic models that have been developed as part of this project, will provide a sound basis for the development of public health policy to manage these threats.

The appropriateness of the approaches used in the development and implementation of the Activity

The approaches we used in the project were adequate to tackle the main aims of the project. The heritability studies across species were completed successfully and were adequate to test the hypothesis that genetic variance can limit adaptive responses. The DNA hybridization and marker association approaches we adopted to identify candidates worked well and many new candidates were subsequently tested for functional significance in RNAi assays. We also successfully extended the markers to other species although progress in this area was perhaps slower than we might have hoped for mainly associated with PCR problems.

The mechanistic model for predicting the historical distribution and future spread of Aedesaegypti was successfully developed and implemented, and the use of existing models enabled an investigation of extinction processes in this species. We identified key life history traits limiting distributions and tested for genetic variation in these traits as well as adaptive divergence. We showed that there was some evidence of divergence among populations in the expected direction. Our microclimatic measurements of containers were successful in highlighting containers of concern for mosquito surveillance particularly as climate variability increases. The survey results were successful in demonstrating the potential risks posed by rainwater tanks as a reservoir for mosquito populations although additional surveys are required to establish risks across Brisbane. Based on the results from the CERF Project, Brisbane City Council have agreed to undertake broad surveys across all of Brisbane's 158 Statistical Local Areas (total of 10,000 households to be surveyed).

Papers

Hoffmann, A. A. Willi, Y. (2008) Detecting genetic responses to environmental change. Nature Reviews Genetics 9:421-432

Kellermann, V. van Heerwaarden, B. Sgrò, C.M. Hoffmann, A.A. (2009). Fundamental evolutionary limits in ecological traits drive Drosophila species distributions. Science 325: 1244-1246.

Kearney, M., Porter, W.P., Williams, C., Ritchie, S., Hoffmann, A.A, (2009) Integrating biophysical models and evolutionary theory to predict climatic impacts on species’ ranges: the dengue mosquito Aedesaegypti in Australia.Functional Ecology 23: 528-538.

McKechnie, S.W. Blacket, M. Song, S.Rako, L. Caroll, X. Johnson, T.K.Jensen, L.T.Lee, S.F. Wee, C.W. Hoffmann, A.A.(2010).A clinally-varying promoter polymorphism associated with adaptive variation in wing size in Drosophila. Molecular Ecology 19: 775-784

Colinet, H. Lee, S.F., Hoffmann, A.A. (2010). Temporal expression of heat shock genes during cold stress and recovery from chill coma in adult Drosophila melanogaster.FEBS Journaldoi:10.1111/j.1742-4658.2009.07470.x.

Mitchell, K.A. Hoffmann, A.A. (2010). Thermal ramping rate influences evolutionary potential and species differences for upper thermal limits in Drosophila. Functional Ecology24: 694-700

Colinet, H. Hoffmann, A.A. (2010) Gene and protein expression of Drosophila Starvin during cold stress and recovery from chill coma Insect Biochemistry and Molecular BiologyBiology 40: 425-8

Colinet, H. Lee, S.F. Hoffmann, A. A. (2010). Knocking down expression of Hsp22 and Hsp23 by RNA interference affects recovery from chill coma in Drosophila melanogaster. Journal of Experimental Biology 213:4146-4150.

Colinet, H. Lee, S. F. Hoffmann, A.A. (2010). Functional characterization of the Frost gene in Drosophila melanogaster: importance for recovery from chill coma. PLOS One 5: e10925.

Hoffmann, A.A. (2010) A genetic perspective on insect climate specialists.Australian Journal of Entomology 49: 93-101.

Mitchell, K.A. Sgrò, C.M. Hoffmann, A.A. (2011). Phenotypic plasticity in upper thermal limits is weakly related to Drosophila species distributions. Functional Ecology (early view)