The smell of infection – detecting infectious disease and determining mechanisms underlying the spread of disease on social networks
Social organization facilitates contacts between individuals altering the probability of parasite transmission (Perkins et al., 2009).
Consequently, social animals have evolved mechanisms to mitigate the risks of infection. One mechanism by which parasites may alter contact patterns could be through avoidance of infected individuals by uninfected based on changes in ‘smell’. Both non-infectious and infectious diseases can cause a change in volatile organic compounds (VOCs) in many species, including, livestock, wildlife and humans (Penn and Potts, 1998; Balseiro and Correia, 2006). Studies on female mice demonstrate odour-based discrimination between infected and uninfected males for a range of parasites, including, influenza virus, the protozoan, Eimeria vermiformis, and the nematode, Heligomosomoides polygyrus (Kavaliers and Colwell 1995, Penn et al. 1998; Ehman and Scott 2001). Proof of concept in our parasite-cockroach colony system has shown that we can detect distinct scents associated with infection in whole animals (Perkins, Müller et al., in prep) However, parasites may also have evolved to exploit and/or manipulate host behaviour - contact patterns - to improve transmission (Moore, 2002). As such, epidemics may themselves drive contact network patterns.
Apiculture worldwide has been threatened by the emergence of infectious diseases (including European and American foulbrood, varroa mite, and several viruses), and associated with effects ranging from reduction in honey productivity to full colony collapse. Using honey bees as a model system we aim to investigate the effect of pathogen load, virulence, mode of transmission on contact networks and the role of ‘smell’ as a mechanism that alters how epidemics progress in social networks.
By combining VOC analysis with detailed behavioural observations, we aim to determine if the odour of infection is a mechanism driving infectious disease dynamics and to tease apart the directional effects of contacts between infected and uninfected individuals – are infected individuals avoided by uninfected individuals thereby reducing parasite transmission. Alternatively, does the parasite manipulate host odour to attract uninfected individuals, so increasing parasite transmission?
We will collect headspace air samples in-situ from animals and colonies onto thermal desorption tubes and use state-of-the-art thermal desorption combined with gas chromatography – time-of-flight mass spectrometry (TD-GC-MS-TOF) for analysis of the samples. Behavioural experiments will analyse social network avoidance behaviours of infected and uninfected individuals using behavioural software such as ‘Ethovision’.
We anticipate that the findings of the project will elucidate key mechanisms in disease transmission and deliver a novel approach to monitoring infections in apiculture and potentially infections in general.