ADDITIONAL DATA ON MITOCHONDRIAL DNA OF NORTH AMERICAN LARGE GULL TAXA

Auk, The, Apr 2005 by Gay, Laur�ne, Bell, Douglas A, Crochet, Pierre-Andr�

ABSTRACT

After publication of the Larus phylogeny in Crochet et al. (2002), the taxonomic status of the "Larus thayeri" and the "L. occidentalis" specimens that were used in that study came into question. For each of those species, we sequenced the same mitochondrial DNA regions in new specimens of known identity. In addition, specimens of L. glaucescens were included. Results from using those specimens confirm that L. occidentalis was the first to diverge from the large white-headed gulls. Larus glaucescens, on the contrary, is part of the Arctic species clade, which also includes L. hyperboreus, L. glaucoides, L. thayeri, and L. schistisagus. The three new L. thayeri specimens differ substantially in mitochondrial DNA from the previously used sample of L. thayeri and share the same haplotype with L. glaucescens. The significance of this finding is unclear, because relationships within the "Arctic species" clade are still unresolved; that is attributable to the unusually high incidence of lineage sharing and extremely low divergence of haplotypes in the group. Received 16 April 2004, accepted 13 December 2004.

Key words: control region, cytochrome b, large white-headed gulls, mitochondrial DNA, phylogeny, recent speciation.

Phylog�nie Mitochondriale des Grands Go�lands: De Nouvelles Donn�es en Am�rique du Nord

R�SUM�.-Cette �tude compl�te la phylog�nie des grands go�lands publi�e dans Crochet et al. (2002), o� le statut taxonomique des sp�cimens "thayeri" et "occidentalis" �tait incertain. Les m�mes fragments d'ADN mitochondrial ont �t� s�quenc�s sur de nouveaux sp�cimens de chaque esp�ce, dont l'origine n'est pas ambigu�. Des �chantillons de L. glaucescens ont �galement �t� utilis�s. L'ajout de ces sp�cimens confirme que le taxon L. occidentalis a diverg� en premier au sein des grands go�lands. L. glaucescens au contraire est inclus dans le clade des esp�ces Arctiques, qui regroupe L. hyperboreus, L. glaucoides, L. thayeri et L. schistisagus. Les trois nouveaux sp�cimens de thayeri diff�rent substantiellement de l'�chantillon utilis� pr�c�demment et partagent le m�me haplotype avec les sp�cimens de L. glaucescens. La signification de ce r�sultat ne peut cependant pas �tre �valu�e pour l'instant, les relations au sein du groupe des grands go�lands n'�tant pas encore r�solues du fait de la forte fr�quence de polymorphisme partag� et de la faible divergence entre taxons.

LARGE WHITE-HEADED GULLS constitute a group of species whose systematic relationships are poorly understood. Several recent studies have used genetic markers to address parts of the problem (de Knijff et al. 2001; liebers et al. 2001; Crochet et al. 2002, 2003; liebers and Helbig 2002), but none has yet worked with a large sample of all taxa. Extensive sampling within taxa, in particular, is crucial in this group, because of widespread sharing of polymorphisms among species (Crochet et al. 2003). The study by Crochet et al. (2002) included most of the taxa, but sampling was often limited to one or a few specimens per taxon. In addition, after the study's publication, the validity of some of its results came into question because the identity of some specimens was challenged (K. G. Smith pers. comm., R. C. Banks pers. comm.). The "Larus thayeri" specimen (from the Louisiana State University Museum of Natural Science [LSUMZ]; tissue LSUMZ B-21816, voucher LSUMZ 160609) was collected in an unusual location-unusual in that Louisiana is not part of the normal winter range for L. thayeri. Furthermore, after publication of the study, correspondence with staff at the LSUMZ revealed that the specimen is actually labeled "L. thayeri?" and should not have been used as a representative of the taxon (S. W. Cardiff pers. com.). The L. occidentalis specimen (tissue LSUMZ B-20480) was collected in Grays Harbor, Washington, which is the northern-range limit for that species and lies in the hybrid zone of L. occidentalis and L. glaucescens (Scott 1971; Bell 1996, 1997; Good et al. 2000). It is thus possible that the specimen had an L. glaucescens haplotype, even if the morphology is typical of L. occidentalis. Here, we analyzed additional specimens of L. thayeri and L. occidentalis of known identity, together with several L. glaucescens specimens, to resolve ambiguities in the phylogenetic relationships among those taxa of large white-headed gulls.

METHODS

Specimens of L. occidentalis and L. glaucescens and all available samples of L. thayeri with no identification ambiguity were loaned by the Museum of Vertebrate Zoology, University of California, Berkeley (see Table 1). Localities of the L. occidentalis and L. glaucescens samples were chosen as far from the hybrid zone as possible. Details about the other specimens used here are described elsewhere (see table 2 in Crochet et al. 2002).

DNA was extracted from tissue samples by complete digestion in 10% Chelex 100 (Biorad, Hercules, California) with 5 �L of proteinase K, followed by two periods of 15-min boiling. A 674-base-pair (bp) fragment of the mitochondrial DNA (mtDNA) control region was amplified by polymerase chain reaction (PCR). We used the primers L438: 5'-TCA-CGT-GAA-ATCAGC-AAC-CC-3' and H1561: 5'-CGG-TTAATT-AGG-GTC-TCT-TG-3'. A 307-bp fragment from the cytochrome-b gene was amplified from the same specimen, using the following primers: L15008, 5'-AAC-TTC-GGA-TCT-CTACTA-GG-3' and H15326, 5'-GAA-TAA-GTTGGT-GAT-GAC-TG-3'. The PCR amplification was performed in a 50-�L reaction volume containing 2 �L of DNA solution (variable concentration), 5 �L of 10� buffer (Tris-HCl 100 mmol KCl 500 mmol), 6.25 �mol of MgCl^sub 2^, 8.32 �mol of dNTP, 1.67 �m of each primer, and 0.5 units of Taq DNA polymerase (Eurogentec, Seraing, Belgium). The annealing temperature was 58�C for the control region and 560C for the cytochrome-i? gene. Annealing duration was 45 s. Sequencing reactions were conducted with the ABI Prism BigDye terminator kit (Applied BioSystems, Foster City, California), following the standard ABI cycle sequencing protocol, and were electrophoresed on an ABI Prism 310 genetic analyzer following the manufacturer's recommended procedures. Sequencing primers were L438, HHTR (5'-ATC-GCT-GTT-GTTGAC-ATG-TA-3'), and DLIlF (5'-AAA CCC TTC CAG TGC ACC GGG-3') for the control region, and L15008 for the cytochrome-b gene. Sequences have been deposited in GenBank (accession numbers: AY615681-AY615706).