Disease ecologists routinely compare transmission models and estimate important parameters with single time step infection experiments. These experiments often involve exposing hosts to parasites for a brief interval, and then diagnosing infections following a relatively long incubation period. However, this common experimental design can create problems for model fitting and analysis. For example, the true exposure a host experiences (number of parasites it directly encounters in a transmission experiment) can often be difficult to quantify. Furthermore, a long waiting period between exposure and diagnosis can result in a biased estimate of transmission if experimental individuals die before they are diagnosed. Here I report experimental and simulation-based analyses, respectively, of two distinct methods aimed at addressing these pitfalls for the human parasite, Schistosoma mansoni, and its intermediate snail host, Biomphalaria glabrata. First, I tested several dyes for their ability to label parasites without impacting infectivity to allow for tracking of parasites that invaded hosts vs. remained in the environment. Second, I developed a mathematical approach to estimate transmission parameters in the face of high mortality virulence during the incubation period.
Results/Conclusions
I successfully labeled and visualized invading parasites. Schistosomes were successfully labeled by BODIPYTM FL C12: 100% of parasites were fluorescent after exposure to 100nM BODIPYTM FL C12 for 45 minutes in the dark. No other dyes investigated reliably labeled schistosomes. Furthermore, BODIPYTM FL C12 showed no impact on infectivity (t= 0.7738, p=0.4825). In addition, accounting for mortality during the incubation period improved estimates of transmission parameters. Simulations showed that tracking individuals that died in the pre-patent period in a separate class resulted in a less-biased estimation of ß when compared to excluding these individuals from analysis or including them in either the infected or uninfected classes. This result was most pronounced at high parameter values for virulence and/or background death. These methodological improvements will enable investigation of mechanistic transmission models in heterogeneous snail populations and communities, and they could enhance understanding of transmission in a variety of other systems.