Abstract

Craspedacusta sowerbii named by Lankester in 1880 is by far the most common freshwater medusa jellyfish species in the world. It has a cosmopolitan distribution, being found in every continent but Antarctica, specially in sub-tropical and temperate zones. Even though the presence of jellyfishes in many freshwater environments, it is quite unfamiliar for most people. This animal is a regular inhabitant of those systems during the warm months of the summer (Terry Peard pers. comms.).

The species was described based solely in morphological characteristics. But, the molecular data (16S nuclear marker) generated in this research shows two well-differentiated clades, which might correspond to two different cryptic species under the same name of C. sowerbii. One of these clades groups only specimens from Europe and the other one from all around the world. Within the “Widespread” clade were found at least 5 different haplotypes, which give an account of the overall diversity that might be hidden under the same morphological features.

Also a unique sample from Arizona was identified as a specimen belonging to the Limnocnida sp. genus, which has only been reported to occur in Africa and India. This could constitute evidence of a species introduction

Methods

Samples were taken form NMNH museum specimens and from different external donations.

DNA extractions were performed using a Automated Phenol/Chloroform Autogenprep 965 extractor. PCRs were performed with the primers designed by Cunningham & Buss (1993). Sequencing was completed using a ABI capillary DNA Sequencer. Sequences were edited using Sequencher.

Mitochondrial 16S data from Limnomedusae were aligned using MUSCLE (Edgar 2004). Obvious mistakes in alignment were assessed by eye and corrected manually. Optimal topologies, as measured by two criteria, parsimony and maximum likelihood, were searched for using PAUP* (Swofford 2002). Searches yielded 40 equally most parsimonious trees and three maximally likely topologies. Bootstrap indices assessing node support were calculated under parsimony (1000 replicates) using PAUP* and under maximum likelihood (400 replicates) using GARLI (Zwickl 2006). For maximum likelihood analyses, an assumed model of nucleotide evolution (HKY+G -- two substitution types, transitions and transversions, and a gamma parameter accounting for rate variability across mt16S) was chosen using the Akaike Information Criterion as employed by ModelTest (Posada & Crandall 1998).

Hypothetical Secondary Structures of 16S haplotypes were obtained using Mfold (Zuker 2003), according to the minimum energy criteria. Molecular Morphometrical data were obtained trough helixes and loops measurement. PAUP* was used to perform a Wagner and Fitch parsimony searches, which yielded 2 most parsimonious trees, one for each method.

Discussion

The phylogenetic hypothesis obtained from the molecular data are compatible with the ecology of the species within the Order Limnomedusae, showing a possible transition form the marine environment to the land. The freshwater species are grouped within the same clade.

On the other hand, the phylogenetic hypothesis obtained from the molecular morphometrics data does not present a consistent topology in that sense.
Although there is a incongruence between both hypothesis, the branching pattern within the genus Craspedacusta is mostly the same. There are two principal clades within Craspedacusta sowerbii, the European Group and the Widespread Group. The European Group is composed of specimens only from European countries, whereas the Widespread Group contains individuals from North America, Asia, Europe, and Oceania. The existence of such a diverse group could be the result of non intentional introductions, probably by human means.

Craspedacusta sowerbii has been thought to be a single cosmopolitan species, but these kind of species are becoming questionable due the improbability of a global metapopulation that is continuously interbreeding. These results support the hypothesis of the existence of at least two cryptic species having apparently the same morphology.

The secondary structure hypotheses showed a high variation between taxa. Even small changes between sequences yield a variation in the morphometrics of the 16S. This variation is roughly consistent with the one found with the molecular data.

The diversity of subgroups shown within the Widespread Group could constitute evidence of even more cryptic species, but for addressing this question more sequences from different markers shuold be included in the future.

The confirmation of a specimen belonging to the genus Limnocnida constitutes the first formal report of this organism in America, this genus was reported to occur only in Africa and Asia. This is probably a case of an species introduction.

Conclusions

Craspedacusta sowerbii

There are at least two cryptioc species of Craspedacusta sowerbii, and is likely that there might be even more.

Two main clades under the name Craspedacusta sowerbii were found, one European and another globally Widespread.

One specimen was confirmed as the first report of Limnocnida in America.

More research time is required to unravel the cryptic diversity within Craspedacusta soerbii