Trait-based disease ecology seeks to provide a causal and broadly predictive framework of disease dynamics. Using general functional trait axes (i.e., slow – fast/quick phenotypic axes), trait-based disease ecology has predicted the identity of pathogen reservoirs and disease progression in plants and animals. Here, we move beyond bivariate correlations between these functional trait axes and host-pathogen interactions by developing a functional-trait based structural equation model to identify why individual hosts vary in pathogen load and their tolerance of infection. The model identifies direct and indirect influences of host resource acquisition, allocation, and storage traits on within-host pathogen titer and the subsequent impact of infection on host growth. We tested the model by conducting controlled greenhouse experiments using six California grass species and an aphid vectored generalist plant pathogen (barley yellow dwarf virus PAV; BYDV), and factorially manipulating supplies of both soil nitrogen and phosphorous. BYDV destroys phloem cells, which lowers host acquisition and allocation of growth limiting resources, particularly carbon. This subsequently reduces plant growth, fecundity, and survivorship, although the effects vary widely among individual plants and host species.
Results/Conclusions
Prior to infection, slow return individuals had a greater root mass fraction (i.e., root:total biomass ratio) compared to fast/quick-return individuals. This indicated higher pre-infection resource stores in slow-return individuals. Root mass fraction prior to infection, however, did not explain BYDV reductions of final host mass. Relative virus titer was greater in fast/quick return individuals compared to slow return individuals. After controlling for virus titer, the final biomass of infected fast/quick return individuals were less negatively affected by infection. Compared to slow returns, infected fast/quick return individuals were better able to increase both photosynthetic capacity (i.e., carbon acquisition) and post-infection root mass fraction (i.e., carbon stores). These results suggest that fast/quick return individuals support larger virus population sizes yet are more tolerance of infection. Despite lower pre-infection resource stores and larger pathogen loads, fast/quick-return individuals reduce the cost of infection (i.e., depleting resource stores) by increasing their acquisition of growth limiting resources.