Predicting the effect of immunocontraceptive recombinant murine cytomegalovirus on population outbreaks of house mice (Mus musculus domesticus) in mallee wheatlands

Virally vectored immunocontraception using a modified murine cytomegalovirus ( MCMV) is being developed for the control of house mice in Australia. In this paper, we develop disease-host models using a combination of laboratory and field data. We then combine these models with a model of a previous mouse population outbreak to explore the likely effectiveness of modified MCMV for controlling mice. Models of homogeneous mixing with and without vertical/pseudovertical transmission provided reasonable fits to field serological data collected during the onset and development of a mouse population outbreak in south-eastern Australia. Both models include the high transmission rate of MCMV suggested by the data. We found no strong support for non-linear contact rates or heterogeneous mixing. When applied to a past outbreak of mice both models gave similar results and suggested that immunocontraceptive MCMV could be effective at reducing agricultural damage to acceptable levels. Successful control was still possible when lags in the development of infertility of up to 10 weeks were added to the model, provided high levels of infertility were achieved. These lags were added because mice can become pregnant just before becoming infertile - the resultant litter would not emerge for 6-7 weeks. Trade-offs between two parameters that could be altered by engineering strains of MCMV - the level of infertility in infected mice and the virus transmission rate - were explored and suggest that a variety of parameter combinations could produce successful control. Our results are encouraging for the future development of virally vectored immunocontraception control of house mice, but future work will need to consider some of the assumptions of these single-strain models.