88 6172 61 Blackwell Publishing LtdOxford UKBIJBiological Journal - PDF document

1 88 , 61Ð72 61 Blackwell Publishing LtdOx
88 , 61Ð72 61 Blackwell Publishing LtdOxford, UKBIJBiological Journal of the Linnean Society0024-4066The Linnean Society of London, 2006? 20068816172Original ArticleBAIKAL AND CASPIAN SEALSJ. U. PALO AND R. V€IN…L€ The enigma of the landlocked Baikal and Caspian seals addressed through phylogeny of phocine mitochondrial sequences innish Museum of Natural History, POB 17, FI-00014 University of Helsinki, Finland 2 Department of Biological and Environmental Sciences, University of Helsinki, Finland species are involved in an apparent polytomy with the boreal harbour seal, glacial relicts Ð Paratethys Ð Phocidae Ð phylogeography Ð Pleistocene glaciations Ð Pusa . INTRODUCTION The origins and phylogenetic relationships of theendemic landlocked seals in Lake Baikal and in theCaspian Sea present a long-lasting biogeographicalenigma (No

2 rdqvist, 1899; Chapskii, 1955b; Davies,1
rdqvist, 1899; Chapskii, 1955b; Davies,1958; McLaren, 1960; Kozhova & Izmesteva, 1998;Koretsky, 2001). The Baikal seal, Phoca sibirica Gme-lin, and the Caspian seal, Phoca caspica Gmelin,belong to the subfamily Phocinae, northern true seals,in the true seal family Phocidae (Table 1). The twolandlocked species inhabit two old Eurasian continen-tal basins that have been isolated from primarymarine seal habitats for millions of years. Lake Baikal *Hooded sealPhociniHistriophocina (subtribe) Pagophilus groenlandicus Harp seal , cytochrome oxidase I, and cytochromeoxidase II: Cyt b and the COI-COIIgenes were ampliÞed separately. Polymerase chainreaction (PCR) primers for these two regions and addi-tional sequencing primers (Table 2) were designedto match conserved regions in aligned sequence of Table 2. Primers used for ampliÞcation

3 and sequencing " -AACTGTGGCATTTCATTAAAGG
and sequencing " -AACTGTGGCATTTCATTAAAGGL150155 " -CATCATTATTCCCACATGGA Cyt b H163255 " -GGGGTTGTTACCTCTTCCTInternal sequencing primersL66735 " -TAGCCCATGCCGGGAGCATCCOIL70835 " -TATTAGGAATAGTTTGAGCACOIH78685 " -ATTGAGAAAGACATAAGGGTTCOIL78205 repetitions with the taxon input order randomized).The whale and hippopotamus were only used as anoutgroup to root the ML-tree, and were pruned prior tothe actual NPRS analysis, performed with theTREEFINDER software (Jobb, 2003).Although the current analysis is based on singlesequences from each taxon, it may be noted that thereciprocal monophyly of mtDNAs in the three Pusagroup taxa focal to the study is well corroborated inassessments of larger data sets (e.g. RFLP data ofSasaki et al., 2003; our unpublished control regiondata).RESULTSThe data set comprised 3369 nucleotide sites, of which719 we

4 re variable among the nine phocine taxa(
re variable among the nine phocine taxa(Table 3). The alignment contained no gaps and all thesequences consisted of full length open readingframes. Most of the substitutions observed were syn-onymous; there was marked variation in the propor-tions of variable synonymous and nonsynonymoussites among the genes (COI 143:1ÐCyt for a basal position for P. hispida within the Phocina.However, none of the six alternative phylogenetichypotheses presented by the constrained trees inFigure 2 could be strictly ruled out by the data.The likelihood ratio test indicated marked hete-rogeneity of substitution rates among lineages(&lnL = 10.21, d.f. a larger carnivore phylogeny. With a ML tree con-structed under a GTR slightly older (age of Phocina root 18Ð21% of E. bar-batus split).DISCUSSIONPHYLOGENETIC RESOLUTIONDespite the amount of sequence data,

5 the relation- lusion from (or assumptio
the relation- lusion from (or assumption in) most previous exami-nations. The preferred MP phylogeny actually showsP. hispida as a basal Phocina lineage, and a trichot-omy of P. sibirica, (P. caspica + H. grypus) and (P.vitulina + P. largha). Nevertheless, apart from the split, P. sibirica andP. caspica actually are the closest taxa in terms of puresequence similarity (Table 4), and the three Pusa taxawould cluster together in a simple UPGMA phenogram.The evaluation of alternative topologies also showsvirtually no difference between the optimal trees(Fig. 3) and those constrained to reßect Pusa mono-phyly or a P. hispidaÐP. sibirica sister relationship(Fig. 2AÐD). The basal position of P. hispida in theoptimal trees may reßect a slower substitution rate inthat branch.The power of mtDNA data in resolving close clado-genetic events (su

6 ch as those within the Phocina)depends o
ch as those within the Phocina)depends on several factors, including the limitedresolution of the molecular genealogy due to Þnite 10tions, or 106 years for a seal species with a 10-yeargeneration interval (equation 1 in Edwards & Beerli, Biological Journal of the Linnean Society, 2006, 88, 61Ð72interspecies coalescences (1.7Ð2.6 Myr, above). Fromthese considerations, also the continental invasions byP. sibirica and P. caspica would appear to have been landlocked seals in connection with the major glacia-tions since c. 900 kya (Arkhipov et al., 1995) would that the Caspian seal would have immigrated througha Mediterranean connection at this relatively cooltransgressive phase, c. 3.4 Mya.Although our data are only marginally compatiblewith the timescale of the earlier suggested Pliocenelinks, invasion through a direct Pliocene ArcticÐ

7 Caspian waterway is supported by another
Caspian waterway is supported by another biogeo-graphical argument, related to the notion of P. caspicaas a member of the broader ÔArctic marineÕ zoogeo-graphical community in the Caspian basin (Pirozhni-kov, 1937). As with the seal, molecular data suggestthat the invasions of the crustaceans Mysis and Gam-maracanthus most likely predated the major MiddlePleistocene glaciations (VŠinšlŠ, 1995; VŠinšlŠ, Vainio& Palo, 2001); in contrast to the hitherto prevailingglacial invasion hypothesis. The dispersal and envi-ronmental requirements of these taxa are more con-strained than of phocids; evidently, their immigrationhas required a proper direct coldwater connection thatalso would have enabled the access for the seal. suggest the Baikal basin as the origin of dispersal(Politov, Bickham & Patton, 2004; Sukhanova Biological Journal of the

8 Linnean Society, 2006, 88, 61Ð72the cont
Linnean Society, 2006, 88, 61Ð72the continental Caspian and Baikal basins. Althoughthe monophyly cannot be refuted, the radiation ofthese seals appears to have coincided with their diver-gence from other, morphologically more dissimilarmarine taxa (Phocina). The time scale suggested bythe molecular data poorly Þts previous hypotheses ofthe continental invasions of the endemic seals. Rela-tionships appear to be too distant to comply with dis-persal through Middle Pleistocene glacial lakes, andtoo close to conform with a Miocene Paratethyan relictancestry of the Caspian seal. More likely, the radiationof the Phocina started in the northern seas of LatePliocene times 2Ð3 Mya, and was accompanied byinvasion of the continental basins. An Arctic ancestryalso plausibly accounts for the emergence of the whitenatal hair shared by the landlocke

9 d taxa and perpet-uation of the ice bree
d taxa and perpet-uation of the ice breeding habit in the Phocini. Never-theless, the actual geographical conditions that wouldhave facilitated the continental invasions in thosetimes still remain undocumented and enigmatic.ACKNOWLEDGEMENTSSamples were kindly provided by G. F. Jarrell from theAlaska Frozen Tissue Collection (University of AlaskaMuseum), N. J. Gemmel (University of Christchurch),D. Yu. Sherbakov and Lev Mamedov (LimnologicalInstitute RAS, Irkutsk), and E. Helle (Finnish Gameand Fisheries Research Institute). We thank ò. çrna-son, H. HyvŠrinen, M. Koskinen, C. Primmer, and ref-erees for comments on earlier manuscript versions ofour paper. The study was supported by grants fromthe Academy of Finland and the Finnish CulturalFoundation.REFERENCESArbogast BS, Edwards SV, Wakeley J, Beerli P, Slowin-ski JB. 2002. Estimating

10 divergence times from moleculardata on p
divergence times from moleculardata on phylogenetic and population genetic time scales.Annual Review of Ecology and Systematics 33: 707Ð740.Arkhipov SA, Ehlers J, Johnson RG, Wright HEJ. 1995.Glacial drainage towards the Mediterranean during the Mid-dle and Late Pleistocene. Boreas 24: 196Ð206.çrnason ò. 1974. Hoberg EP, Adams AM. 1992. Phylogeny, historical biogeog-raphy, and ecology of Anophrocephalus spp. (Eucestoda: Tet-rabothriidae) among pinnipeds of the Holarctic during the Kozhova OM, Izmesteva LR. 1998. Lake Baikal: Evolutionand Biodiversity. Leiden: Backhyus Publishers.Lopez JV, Cevario S, OÕBrien SJ. 1996. Complete nucle-otide sequences of the domestic cat (Felis catus) mitochon-drial genome and a transposed mtDNA tandem repeat(Numt) in the nuclear genome. Genomics 33: McLaren IA. 1960. On the origin of the Caspian and Bai

11 kal osada D, Crandall KA. 1998. MODELTES
kal osada D, Crandall KA. 1998. MODELTEST: testing the modelof DNA substitution. Bioinformatics 14: 817Ð818.Ray CE. 1976. Biological Journal of the Linnean Society, 2006, 88, 61Ð72Slowinski JB. 2001. Molecular polytomies. Molecular Phylo-genetics and Evolution 19: 114Ð120.Sorenson MD. 1999. ersion 2. Boston, MA: BostonUniversity.Springer MS, Murphy WJ, Eizirik E, OÕBrien SJ. 2003.Placental mammal diversiÞcation and the CretaceousÐTertiary boundary. Proceedings of the National Academy ofSciences, USA 100: Steininger FF, Ršgl F. 1984. Palaeogeography and palins-pastic reconstruction of the Neogene of the Mediterraneanand the Paratethys. In: Dixon JE, Robertson AHF, eds. Thegeological evolution of the eastern mediterranean. Oxford:Blackwell ScientiÞc Publications, 659Ð674.Sukhanova LV, Smirnov VV, Smirnova-Zalumi NS,Kirilchik SV, Shimiz

12 u I. 2004. Grouping of Baikal omulCorego
u I. 2004. Grouping of Baikal omulCoregonus autumnalis migratorius Georgi within theC. lavaretus complex conÞrmed by using a nuclear DNAmarker. Annales Zoologici Fennici 41: 41Ð49.Swofford DL. 1998. PAUP*. Phylogenetic Analysis Using Par-simony *and Other Methods, Version 4. Sunderland, MA:Sinauer Associates,.Taimisto J. 1990. Norpan (Phoca hispida Schr.), baikalinhyl-keen (Phoca sibirica Gmelin) ja kaspianhylkeen (Phocacaspica Gmelin) kallon morfometriaa. MSc Thesis, Univer-sity of Helsinki.Timoshenko YK. 1975. Craniometric features of seals of thegenus Pusa. Rapports et Proces-Verbaux des Reunions. Con-seil International Pour lÕExploration de la Mer 169: 161Ð164.Ursing BM, Arnason U. 1998. Analyses of mitochondrialgenomes strongly support a hippopotamus-whale clade. Pro-ceedings of the Royal Society of London, Series B, BiologicalSc