In Trichotropis Cancellata (Gastropoda)

A Centenary Research Grant Report by Erika V. Iyengar, Cornell University

Although symbiotic interactions are common, it is rare that the performance of the symbionts can be measured within and outside the interaction. Studies of this nature, however, are the only way that the importance of the interaction to each participant and possibly the evolutionary trajectory of the interaction can be determined. Individuals of the marine snail Trichotropis cancellata can either suspension feed on their own or steal food from tube-dwelling polychaete worm hosts (kleptoparasitism).

When parasitizing, the snail extends its pseudoproboscis through the feeding tentacles of the worm and into the worm's mouth. Using cilia on the pseudoproboscis, the snail draws food from the suspension-feeding worm's mouth soon after the worm ingests the particles. Because this symbiotic interaction is facultative (each organism can exist outside the interaction), the performance of each member can be evaluated separately and together.

Benefits of parasitism to Trichotropis cancellata

My experiments have shown that all Trichotropis cancellata, regardless of size, grow faster if allowed toparasitize the tube-dwelling worm Serpula columbiana than if restricted to suspension feeding. Larger female snails have a higher fecundity. Thus, kleptoparasitism is likely to be evolutionarily advantageous compared with suspension feeding as the parasites attain larger sizes faster.

Trichotropis cancellata can parasitize multiple tube-dwelling polychaete species, including Serpula columbiana (family Serpulidae), Pseudopotamilla ocellata,Schizobranchia insignis, Eudistylia vancouveri (all three in the family Sabellidae), and Sabellaria cementarium (family Sabellaridae). These hosts vary greatly in size, and were predicted to differ in the resources they provide to snail parasites. However, in the summer T. cancellata grows at the same rate on all worm hosts studied (Eudistylia,Serpula, Pseudopotamilla, and Schizobranchia). The snails grow significantly faster on worms than on brachiopods or empty serpulid tubes (the latter was used as a control, restricting snails to suspension feeding).

Brachiopods and empty serpulid tubes promote the same rate of growth in snail parasites, despite laboratory observations that the snails appear to position their shells to take advantage of the brachiopod feeding currents. In the winter, T. cancellata parasites grew faster on the sabellids Eudistylia and Schizobranchia than on the serpulid Serpula.. Although a treatment using Sabellaria cementarium as a host was created, none of the snails remained on this host species throughout the experiment. This result was surprising, given that the snails have been observed to parasitize sabellarids in the laboratory and naturally occur on this host. I conclude that Sabellaria cementarium is not as beneficial a host for the snails as the other polychaete species used in experiments (Eudistylia, Schizobranchia, and Serpula). Worm size was an insignificant predictor of snail growth (both in the summer and winter experiments).

Impact of the snail on the host

In laboratory observations of Trichotropis cancellata parasitizing Serpula, Pseudopotamilla and Schizobranchia, the snail removes food from the host worm more than 50% of the time that the host has its tentacles extended and is feeding. This implies that the presence of a snail parasite can have profound impacts on the worm host, as potentially a large proportion of the worm's captured food is diverted to the snail. Indeed, growth experiments using the worm host Serpula columbiana demonstrate that parasitized worms grow significantly slower than unparasitized worms.

Snail host preference

Trichotropis cancellata chooses to reside on live Serpula columbiana significantly more often than on empty S. columbiana tubes in subtidal choice racks (with interspersed replicates of the two treatments). When given the choice between live Serpula and live Schizobranchia insignis, equal numbers of snails choose each of these host species. However, the prevalence of infection is higher in the Serpula than the Schizobranchia. Thus, multiple infections of the same host are less likely to occur on Serpula than on Schizobranchia. Y-maze experiments indicate that the snails do not appear to be able to chemically detect hosts at a far distance; an equal number of snails chose plain seawater as chose live hosts (Serpula columbiana or Schizobranchia insignis).

Evolution of parasitism within the family

The long lower lip of all trichotropids makes it likely other members are parasites. To investigate the evolution of parasitism, trichotropids were observed in southeast Alaska (Juneau, Sitka, Ketchikan, and Craig), southwest Alaska (Kasitsna Bay) and British Columbia (eastern side of Vancouver Island). Trichotropis cancellata steals food from tube-dwelling polychaetes throughout the entire range studied. In Kasitsna Bay, T. cancellata parasitized a new host, a holothuroid (likely Eupentacta quinquesemita). T. conica also parasitizes tube-dwelling polychaetes in Alaska and British Columbia. Whether this is truly a separate species from T. cancellata is currently under investigation. Over 70 T. insignia were collected in Kasitsna Bay, but none were ever positioned to allow parasitism. Therefore, I conclude that T. insignia is solely a suspension feeder. These results indicate kleptoparasitism has likely evolved at least twice within the family Capulidae, as some Capulus are also known to be kleptoparasites.

I would like to thank The Society for awarding me the Centenary Research Grant, which helped ensure a success.

Erika V. Iyengar, Cornell University

Fig. 2. Host worms of the snail kleptoparasite Trichotropis cancellata. The first three are in the family Sabellidae and have parchment-like tubes, worms in the family Serpulidae have hard white calcareous tubes, and worms in the family Sabellaridae have tubes made of sand grains. The mouth in all of these species is in the centre of the crown of feeding tentacles. None of these species can leave their tubes. In Puget Sound, Washington, the most common host of Trichotropis cancellata is Serpula columbiana.

A: Pseudopotamilla ocellata

B: Eudistylia vancouveri

C: Schizobranchia insignis

D: Serpula columbiana

E: Sabellaria cementarium.