KIR specificity is protective against viruses evolving MHC-like decoys.

<p>A host population was inoculated with a wild type virus after a “burn in” period of years; we allowed for mutation of the virus 5000 years after the initial epidemic. (A) The MHC down-regulating virus (red line) is fitter than the wild type virus (blue line), infecting more individuals and resulting in a larger decrease of the total population size (grey line). Here, the host population has degenerate KIRs (with bit). (B) Population with a highly specific receptor-ligand recognition (with bits). (C) Viruses evolving MHC down-regulation and decoy molecules (black line) take over the wild type virus (blue line), infecting more individuals and resulting in a larger decrease of the total population size (grey line). In a host population with degenerate KIRs, i.e., bits, almost all individuals are infected with the virus. (D) The spread of the virus is reduced in host populations with specific KIRs, e.g., bits. (A)–(D) single representative simulations of the first 15 000 years of the simulations. (E) Ten simulations were carried out per specificity value ranging from bit to bits. At the end of each simulation (i.e. 200 000 years), the mean of the total population size over the last 100 000 years was taken as a measure of the protection level of the host population. Protection against the wild type virus (blue) was not dependent on the specificity of the KIRs. Protection against an MHC down-regulating virus (red) was slightly, yet significantly worse in simulations with specific KIRs (i.e. ), whereas protection against the virus evolving decoy molecules (black line) increased drastically in simulations with specific systems (i.e. ). (** represent <i>p</i> values and were calculated using the Mann-Whitney <i>U</i> test. The boxes represent the interquartile range, the thick horizontal lines represent the median, and the circles are outliers.).