Because pathogens replicate within hosts and transmit between hosts, selection takes place on multiple levels.  There has been ongoing interest for more than two decades in trying to understand the conditions favoring the evolution of acute, highly transmissible infections, focusing on trade-offs such as the transmissibility-virulence trade-off and the invasion-persistence trade-off. Studies have shown that these types of trade-offs lead to intermediate pathogen attack rates. These earlier studies typically consider the evolution of a single trait under a defined trade-off. However, for some pathogens, the course of infection within the host is likely to be more complex, determined by more than a single dimension opening the door for more complicated strategies related to disease severity.

The protozoa Plasmodium falciparum (Pf), which causes the most severe type of malaria in humans, is one example of such a pathogen. During the course of an infection, Pf has the ability to express up to 60 different variants of surface proteins (PfEMP1) encoded by a family of var genes, which are recognized by the host immune system and which also act as virulence factors.  The temporal separation in the rise of different antigenic variants is believed to allow for an extended duration of infection.

Results/Conclusions

Here we study conditions that would favor the ordered expression of these variants by allowing each to evolve independently. We specifically ask about the order of severe vs. mild variants with a transmission model that considers both within and between host dynamics.  We consider invasion persistence trade-offs in terms of critical community size of hosts and seasonality in vector transmission.  Analytical and simulation results show that parasite fitness is higher when the variants are allowed to evolve independently.   In particular, an ordered expression of severity is favored.  We discuss the implications of this result for what constitutes a strain in malaria.