Context dependent symbiont transmission: Part I

If ever there was a “law” in parasite ecology, it would be that macroparasites are aggregately distributed among their hosts. For instance, when Shaw and Dobson (1995) surveyed the literature, they found that 268 out of 269 macroparasite distributions were overdispersed/aggregated, so that a minority of the hosts harbored the majority of the parasites. The other options – which were not observed – were that parasites would be randomly or uniformly distributed among their hosts.

Previously, we’ve discussed some of the mechanisms that can lead to aggregation of macroparasites. Most of these mechanisms fall under a big umbrella that we can label “variation in transmission rates among hosts causes parasite aggregation.” Additionally, aggregation can occur if parasites reproduce directly on the host and the “offspring” do not all leave the host. That isn’t a typical characteristic of a macroparasite life cycle, but it is a possibility. So, we have direct reproduction and variation in transmission rates pushing macroparasite distributions away from random distributions and towards aggregated distributions (Anderson and Gordon 1982).

I should also point out that there are mechanisms working in the opposite direction to push parasite distributions from overdispersed or random distributions to more uniform or underdispersed distributions. These are density-dependent processes, such as density dependent parasite mortality, density dependent parasite transmission/emigration, and infection intensity dependent host mortality (Anderson and Gordon 1982).

To illustrate these ideas, let’s take a look at a real system. Enter Chaetogaster limnaei, an oligochaete worm that lives on the headfoot and in the mantel cavity of freshwater snails. It is unclear whether Chaetogaster is a parasite, a commensal, or a mutualist of snails. Like many symbionts, it may be that the net outcome of the symbiont-host relationship is context-dependent. But for our purposes, let’s use the term “macroparasite” when referring to Chaetogaster, because we know that these oligochaetes share an important trait with macroparasites: the oligochaetes are aggregately distributed among snail hosts (Hopkins et al. 2013). However, unlike many macroparasites, Chaetogaster reproduce directly on the host and the “offspring” do not all immediately leave the host, so Chaetogaster may be aggregated for that reason alone. But are there other processes affecting Chaetogaster distributions?

First, let’s ask whether there are any density-dependent processes that might favor random or underdispersed Chaetogaster distributions. As far as we know, the oligochaetes do not kill their typical snail hosts at any densities, so infection intensity dependent host mortality is probably not a prominent process. We also don’t know much about density-dependent parasite mortality, though the oligochaetes may have higher mortality rates at high density (Hopkins et al. 2013). Finally, Chaetogaster dispersal rates to new hosts do not appear to be influenced by Chaetogaster density, so transmission isn’t density dependent (Hopkins et al. 2015).

Before we talk about variation in Chaetogaster transmission rates, we need to discuss how Chaetogaster are transmitted, because it was previously unclear whether Chaetogaster could leave one host to disperse to another host without host contact. Hopkins et al. (2015) made tiny little leashes for snails and tethered pairs of snails so that they could or could not touch. When the snails could touch, Chaetogaster dispersed among the snails until they reached a roughly 50/50 distribution. But when the snails could not touch, Chaetogaster wouldn’t disperse. Therefore, Chaetogaster are directly transmitted…unless the snail that starts with the Chaetogaster is euthanized, in which case the Chaetogaster readily jumped ship.

In a series of subsequent experiments, Hopkins et al. (2015) showed that the roughly 50/50 Chaetogaster distribution among snail hosts only occurred when the two snails were similar. If the snails were different sizes, more Chaetogaster would disperse from large to small snails than from small to large snails. There was also variation in Chaetogaster dispersal/transmission rates depending on whether the donor or receiver snails were infected and shedding trematode cercariae – a major food source for Chaetogaster. So there was variation in Chaetogaster transmission rates, which could help explain why Chaetogaster are aggregated among snail hosts!

But why did Chaetogaster transmission rates vary with host characteristics? It is probably related to Chaetogaster fitness on hosts with different characteristics, where Chaetogaster are more likely to disperse from hosts where Chaetogaster fitness is low. However, it’s hard to measure Chaetogaster fitness, because the worms reproduce asexually. Next week, I’ll tell you about my favorite paper from 2015, which quantified fitness-based symbiont dispersal in a different system. Stay tuned!!

References:

Anderson R, Gordon D (1982) Processes influencing the distribution of parasite numbers within host populations with special emphasis on parasite-induced host mortalities. Parasitology 85:373–398.

Hopkins SR, Wyderko JA, Sheehy RR, Belden LK, Wojdak JM (2013) Parasite predators exhibit a rapid numerical response to increased parasite abundance and reduce transmission to hosts. Ecol Evol 3:4427–4438.

Hopkins SR, Boyle LJ, Sheehy RR, Belden LK, Wojdak JM (2015) Dispersal of a defensive symbiont depends on contact between hosts, host health, and host size. Oecologia 179(2):307-18.