Mathematical analysis of a mouse experiment suggests little role for resource depletion in controlling influenza infection within host

How important is resource depletion (e.g. depletion of target cells) in controlling infection within a host? And how can we distinguish between resource depletion and other mechanisms that may contribute to decline of pathogen load or lead to pathogen clearance? In this paper we examine data from a previously published experiment. In this experiment, mice were infected with influenza virus carrying a green fluorescent protein reporter gene, and the proportion of lung epithelial cells that were influenza infected was measured as a function of time. Three inoculum dose groups - 10^4 PFU, 10^6 PFU and 10^7 PFU - were used. The proportion of cells infected was estimated to be about 21 (95% confidence interval: 14-32) fold higher in the highest dose group than in the lowest dose group with the middle dose group in between. We show that this pattern is highly inconsistent with a model where target cell depletion is the principal means of controlling infection, and we argue that such a pattern constitutes a reasonable criterion for rejecting many resource depletion models. A model with an innate interferon response that renders susceptible cells resistant fits the data reasonably well. This model suggests that target cell depletion is only a minor factor in controlling natural influenza infection.