COS 2-10
Phylogenetic host specificity in sponge-dwelling social snapping shrimps Synalpheus

Monday, August 10, 2015: 4:40 PM
302, Baltimore Convention Center
Solomon Chak, Biological Sciences, Virginia Institute of Marine Science, The College of William & Mary, Gloucester Point, VA
J. Emmett Duffy, Tennenbaum Marine Observatory Network, Smithsonian Institution, Washington, DC

Intimate biological associations are responsible for a large part of biodiversity, especially in the tropics. However, the nature and the strength of such associations, and the evolutionary processes maintaining them, are not always clear. Snapping shrimps in the genus Synalpheus are obligate sponge-dwellers in which eusociality has independently evolved several times. Strong host fidelity and limited host availability are hypothesized to be the main drivers of the evolution of eusociality in Synalpheus. Although eusocial species were found to have higher host range (number of host species) than other species, these studies did not consider that closely related shrimp species might live in closely related host sponges. Therefore, counting multiple closely related hosts as different host may over-estimate the host range of Synalpheus. 

We reconstructed a multi-locus phylogeny of sponges used by Synalpheus species and tested for patterns of cophylogeny with an established Synalpheus phylogeny. Further, from the sponge phylogeny we estimated the phylogenetic host range for each shrimp species as the effective number of phylogenetically distinctive host species (Hill numbers). We performed ordinary least-squares and phylogenetic generalized least-squares regression to test whether phylogenetic host range is related to eusociality index (which takes higher values with more non-reproducing workers per colony).


Sponge phylogeny and Synalpheus phylogeny showed significant cophylogeny (ParaFit, P=0.002), meaning that more closely related shrimps use more closely related sponges. Accordingly, most Synalpheus species had reduced host range after incorporating the phylogenetic relatedness between sponge hosts. Moreover, phylogenetic host ranges were not higher in eusocial species and did not increase with eusociality index (P=0.52). This suggests that although eusocial species were found in more sponge species, they tend to utilize phylogenetically similar sponge. Therefore, whether eusociality broadens ecological range in shrimps, and whether these patterns reflects differences in competitive ability, remain elusive.