Comparative demography elucidates the longevity of parasitic and symbiotic relationships

Abstract

Parasitic and symbiotic relationships govern vast nutrient and energy flows, yet controversy surrounds their longevity. Enduring relationships may engender parallel phylogenies among hosts and parasites, but so may ephemeral relationships when parasites colonize related hosts. An understanding of whether symbiont and host populations have grown and contracted in concert would be useful when considering the temporal durability of these relationships. Here, we devised methods to compare demographic histories derived from genomic data. We compared the historical growth of the agent of severe human malaria, Plasmodium falciparum, and its mosquito vector, Anopheles gambiae, to human and primate histories, thereby discerning long-term parallels and anthropogenic population explosions. The growth history of Trichinella spiralis, a zoonotic parasite disseminated by swine, proved regionally specific, paralleling distinctive growth histories for wild boar in Asia and Europe. Parallel histories were inferred for an anemone and its algal symbiont (Exaiptasia pallida and Symbiodinium minutum). Concerted growth in potatoes and the agent of potato blight (Solanum tuberosum and Phytophthora infestans) did not commence until the age of potato domestication. Through these examples, we illustrate the utility of comparative historical demography as a new exploratory tool by which to interrogate the origins and durability of myriad ecological relationships. To facilitate future use of this approach, we introduce a tool called C-PSMC to align and evaluate the similarity of demographic history curves.

Keywords: demography; domestication; genomics; parasitism; symbiosis.

Figure 1.

PSMC plots comparing the demographic histories of hosts and their putative parasites/symbionts. The vertical axis is scaled to 10⁴. The time period associated with human agriculture (most recent 12 000 years) is shaded. Solid lines represent hosts, whereas dotted lines represent their parasites or symbionts. (a) A. gambiae mosquito and two Pl. falciparum parasite populations appear to have grown and contracted in sync with human, but not gorilla, populations over the last 500 000 years. Demography estimated from each of two pseudo-diploid malaria genomes illustrated (see electronic supplementary material, methods). (b) Both Chinese T. spiralis isolates (red dashed lines) mirror the demographic history of sympatric members of their host species (solid red lines), while North American T. spiralis (blue dashed lines) was more similar to wild boar and pig populations from Europe (solid blue lines). (c) The algal symbiont S. minutum has tracked the demography of its known host, the anemone E. pallida, rather than the coral Acropora digitifera, at least since the LGM (approx. 20 ka). (d) Phytophthora infestans, the agent of potato blight, only shows concerted growth or decline with potatoes since domestication.

Figure 2.

Comparison of minimum average slope difference for the natural host (e.g. Pl. falciparum × human; n = 13) versus plausible (e.g. Pl. falciparum × gorilla; n = 12) and implausible (e.g. S. minutum × human; n = 56) alternative host–symbiont combinations. Error bars represent the 95% confidence interval of all comparisons. (Online version in colour.)

Figure 3.

Comparisons of average slope difference between unmodified curves (solid colour) and curves that have been randomized with respect to time (faded). Symbiont plots were aligned to their purported host plots (Western T. spiralis versus Western S. scrofa; Asian T. spiralis versus Asian S. scrofa; Pl. falciparum versus human; A. gambiae versus human; S. minutum versus Exaiptasia pallida; Ph. infestans versus potato) and the average difference in slope direction was calculated. The 99.9% confidence interval is shown in association with the randomized curves, based on the variance among randomization replicates (n = 5). Randomizing slopes with respect to time significantly worsened the match for every ecologically relevant pairing of species. (Online version in colour.)