Groundwater amphipods in Iceland: Population structure and phylogenetics Etienne Kornobis Dissertation submitted in partial fulfillment of a Philosophiae Doctor degree in Biology Advisor: Dr. Snæbjörn Pálsson PhD Committee: Pr. Bjarni K. Kristjánsson Pr. Jörundur Svavarsson Opponents: Pr. Christophe Douady Dr. Guðmundur Guðmundsson Faculty of Science School of Engineering and Natural Sciences University of Iceland Reykjavik, october 2011 Groundwater amphipods in Iceland: Population structure and phylogenetics Dissertation submitted in partial fulfillment of a Philosophiae Doctor degree in Biology Copyright c 2011 Etienne Kornobis All rights reserved Faculty of Science School of Engineering and Natural Sciences University of Iceland Sæmundargötu 2 101 Reykjavik Iceland Telephone: 525 4000 Bibliographic information: Etienne Kornobis, 2011, Groundwater amphipods in Iceland: Population struct- ure and phylogenetics, PhD dissertation, Faculty of Science, University of Ice- land, 166 pp. ISBN 978-9935-9064-1-0 Printing: Háskólaprent Reykjavik, Iceland, October 2011 Abstract Crymostygius thingvallensis and Crangonyx islandicus are two endemic species of groundwater amphipods which were recently discovered in Iceland. C. thing- vallensis is uncommon, but represents a new monotypic family Crymostygidae. C. islandicus is widespread in the geologically youngest parts of the island and belongs to a genus which is widely distributed in North America and Eurasia. Both species belong to the superfamily Crangonyctoidea, exclusively composed of freshwater species. Iceland is geographically isolated and was fully covered by an ice sheet during the last glacial maximum, 21 000 years ago, hence the Icelandic biota is characterized by extremely low endemism and low species diversity. Thus, the discovery of these two endemic freshwater species, on an island isolated in the midst of the Atlantic ocean raised questions about their colonization and their survival during the glacial periods of the Ice age. Sub- glacial refugia have been hypothesized to explain the occurrence of these two amphipods in Iceland which might have colonized the country via an ancient land bridge, now submerged, connecting Iceland and Greenland. In this thes- is, molecular and morphological markers were used to test this hypothesis. In addition, cryptic diversity, common in groundwater species, was assessed with mitochondrial and nuclear markers. Finally, the variation in secondary structure of the nuclear ITS regions has been characterized and its influence on the phylogeny reconstruction evaluated. The mitochondrial markers (COI and 16S) support that C. islandicus populations diverged within the island for up to five million years and consequently they survived repeated glaciati- ons in Iceland in subglacial refugia. The nuclear marker ITS1 is supporting, though weakly, the phylogeographic pattern observed with the mitochondrial markers. Though the populations of C. islandicus showed high genetic di- vergence, different species delimitation methods led to different conclusions about the potential species complex status of C. islandicus. The taxonomic status of both Icelandic species is supported by phylogenies based on nuclear markers (18S, 28S, 5.8S, ITS1 and ITS2). These phylogenies show high di- vergence between Crangonyx species from Eurasia and from North America. C. islandicus is more closely related genetically to Crangonyx species from i North America, supporting the hypothesis of a colonization via Greenland. Morphological data led to a different pattern, C. islandicus clustering with species both from North America and Europe which might be due to conver- gent evolution in morphological traits. In addition, nuclear phylogenies show that Crangonyx, Synurella and Stygobromus genera are polyphyletic and might need taxonomic revision to restore their monophyly. The secondary structures of the ITS2 for Crangonyctoidea species are highly divergent from the common structure described for Metazoans and appeared to have little impact on the phylogenetic reconstruction. ii Útdráttur Crymostygius thingvallensis og Crangonyx islandicus eru tvær einlendar teg- undir grunnvatnsmarflóa sem fundust nýlega á Íslandi. C. thingvallensis er sjaldgæf og myndar hún nýja ætt marflóa. C. islandicus er útbreidd á gos- belti Íslands, og tilheyrir ættkvísl sem er útbreidd í N-Ameríku, Evrópu og Asíu. Báðar tegundirnar tilheyra yfirættinni Crangonyctoidea sem finnst ein- göngu í ferskvatni. Lífríki Íslands einkennist af lítilli tegundafjölbreytni og fáum einlendum tegundum. Hefur það verið rakið til landfræðilegrar einangr- unar Íslands og þess að landið var hulið jökli fyrir um 21000 árum. Fundur tveggja einlendra grunnvatnsmarflóa vakti því athygli og spurningar vöknuðu um hvernig þær höfðu borist til landsins. Tilgáta hefur verið sett fram um að þær hafi lifað af á Íslandi á ísöld og að þær hafi í raun fylgt Íslandi allt frá myndun þess eða mögulega numið land um landbrú milli Grænlands og Íslands. Í ritgerðinni er þessi tilgáta prófuð með aðferðum sem byggja á greiningu DNA sameinda. Annað meginviðfangsefni ritgerðarinnar er að greina flokkun teg- undanna út frá DNA sameindum og útlitseiginleikum marflónna. Auk þess var athugað hvort dulinn breytileiki greindist í hvatbera og kjarna DNA, en slíkt er algengt meðal grunnvatnstegunda. Breytileiki í annars stigs byggingu svokallaðra ITS DNA raða var einnig lýst og áhrif þess á flokkunartré tegund- anna athuguð. Breytileiki í hvatberagenunum COI og 16S RNA styðja að C. islandicus lifði af endurtekin kuldaskeið ísaldar undir jökli. Athugun á ITS1 innröðum í kjarna DNA gefur stuðning við þessa ályktun en sýnir mun minni aðgreiningu milli svæða en hvarberagenin. Erfðafræðileg aðgreining milli stofna C. islandicus gefur vísbendingu um ólíkar tegundir sé að ræða innan Íslands. Flokkunarleg staða beggja tegundanna byggð á RNA genum í kjarna DNA (18S, 28S, 5.8S, ITS1 og ITS2) styður fyrri flokkun byggða á útlitseinkennum. Flokkunartréið sýnir hinsvegar skýra aðgreiningu milli Crangnyx tegunda milli Evrasíu og N-Ameríku. C. islandicus er skyldari Crangonyx tegundum í Amer- íku, það styður landnám frá Grænlandi. Flokkunartré byggt á útlitseinkennum var þó öðruvísi, C. islandicus flokkaðist samkvæmt því bæði með tegundum frá N-Ameríku og Evrópu. Slíkt frávik gæti mögulega stafað af samhliða þróunar meðal útlitseiginleikanna. Flokkunartré byggð á kjarna DNA sýndi einnig að iii ættkvíslirnar Crangonyx, Synurella og Stygobromus eru fjölstofna og að þörf sé að athuga frekar flokkun þeirra byggða á útlitseiginleikum. Annars stigs bygg- ing ITS2 meðal Crangonyctoidea tegunda víkur frá hinni almennu byggingu sem hefur verið lýst fyrir vefdýr (metazoa) og hafði hún lítil áhrif á byggingu ættartrésins. iv List of Papers The present thesis is based on three accepted papers and two currently under review. The papers will be referred in the text by their respective numbers as following: • Paper 1: Kornobis E, Pálsson S, Kristjánsson BK, Svavarsson J (2010). Molecular evidence of the survival of subterranean amphipods (Arthro- poda) during Ice Age underneath glaciers in Iceland. Molecular Ecology 19: 2516-2530. • Paper 2: Kornobis E, Pálsson S (2011). Discordance in variation of the ITS region and the mitochondrial COI gene in the subterranean amphi- pod Crangonyx islandicus. Journal of Molecular Evolution (in press), DOI: 10.1007/s00239-011-9455-2. • Paper 3: Kornobis E, Pálsson S, Sidorov DA, Holsinger JR, Kristjáns- son BK (2011). Molecular taxonomy and phylogenetic affinities of two groundwater amphipods, Crangonyx islandicus and Crymostygius thing- vallensis, endemic to Iceland. Molecular Phylogenetics and Evolution 58: 527-539. • Paper 4: Kornobis E, Pálsson S. (under review). Phylogenies of Cran- gonyctoidea species based on the ITS regions (ITS1 and ITS2). • Paper 5: Kornobis E, Pálsson S, Svavarsson J. (under review). Classi- fication of Crangonyx islandicus based on morphological characters and comparison with molecular phylogenies. v Acknowledgments It is my great pleasure to acknowledge all the persons which played a role, directly or indirectly, in the realization of this thesis. I had the opportunity during these four years to work in an outstanding environment and especially to be advised by a remarkable doctoral committee. I would like to greatly thank my advisor, Snæbjörn Pálsson, for his constant care, his bonhomie and his contagious interest in Science. I have been truly lucky to be supervised by such mentor and I wish to be able to keep on working with him. Thanks a lot also to Bjarni K. Kristjánsson for the awesome sampling trips, always pertinent suggestions and for creating the highest latitude beer club that I know. I have been delighted as well to work with Jörundur Svavarsson and I thank him for guiding me through the mysteries of Crustacean morphology. Thanks to Christophe Douady and Guðmundur Guðmundson for accepting to be my opponents for this thesis. I am grateful to the University of Iceland Research Fund and the Icelandic Research Council (Rannís) for their financial support. I am indebted to all the contributors of free softwares (GNU/Linux, Emacs, R, LATEX, Python, Inkscape, Gimp, Phylip ...) which were of great use during the realization of this thesis. Thanks also to all my lab mates which made this stay so cheerful. I would particularly like to thank Ubaldo, Eduardo, Dileepa, Sindri, Hlynur, Ehsan, Laurène, Baldur and many more. Thanks to Oli Patrick to have revived the cake club. I still feel sorry to have missed one of his french speeches.