Anomalous variation in mitochondrial genomes of White-crowned (Zomotrichia leucophrys) and Golden-crowned (Z. atricapilla) Sparrows: Pseudogenes, hybridization, or incomplete lineage sorting?

Auk, The, Jan 2001 by Weckstein, Jason D, Zink, Robert M, Blackwell-Rago, Rachelle C, Nelson, Douglas A

The White-crowned Sparrow (Zonotrichia leucophrys) is a common breeding bird in scrubby habitat of the Pacific coast, montane, and boreal regions of North America; five subspecies are recognized (Fig. 1A). The Golden-crowned Sparrow (Z. atricapilla) has a more restricted breeding range and is not subdivided into subspecies (Fig. 1B). The full species status of these two sparrows is undisputed by ornithologists. Despite considerable breeding season sympatry, there are few hybrids known. Because a previous restriction site analysis (Zink et al. 1991) of mitochondrial DNA (mtDNA) suggested that these two species are very closely related, we compared mtDNA sequences and allozyme data from those species and the other North American congeners (Z. querula, Z. albicollis).

Partial sequences from two mitochondrial genes in Golden-crowned and White-crowned sparrows were virtually identical. Neither species is reciprocally monophyletic in a haplotype tree, and two haplotypes are shared between several White-crowned and Golden-crowned sparrows. Considered together, Z. leucophrys and Z. atricapilla mtDNA sequences possess less variation than that found in single populations of passeriform bird species. In contrast, one fixed allozyme difference and several frequency differences (Zink 1982) indicated that Z. leucophrys and Z. atricapilla have been evolving independently for a considerable period of time. In this paper, we evaluate four possible explanations that could account for that anomalous lack of mtDNA differentiation between Z. leucophrys and Z. atricapilla: (1) accidental amplification of a nuclear pseudogene, (2) hybrid origin of either Z. leucophrys or Z. atricapilla, (3) incomplete lineage sorting, or (4) past introgressive hybridization.

Materials and Methods.-We sequenced 985 base pairs (bp) from the mtDNA genome in each of 13 Golden-crowned and 22 White-crowned sparrows that encompassed all described subspecies. Of these 985 bp, 433 bp are from a coding gene, cytochrome-- b, and 552 bp are from the non-coding mtDNA control region. DNA was extracted from muscle using standard methods (Ellegren 1992, Lansman et al. 1981). We used the polymerase chain reaction (Kocher et al. 1989) and standard thermocycling regimes to amplify a 433 bp segment of the mtDNA cytochrome-b gene with primers L14841 (Kocher et al. 1989), LCBKLICKA (5'-CCTTTACTATGGCTCATACC, designed by the authors), and H15299 (Hackett 1992), and a 1,000 bp segment of the mtDNA control region with primers LCR4 and HPHE-1 (Tarr 1993). We used the above-listed primers and standard dideoxy DNA sequencing methods (Hillis et al. 1990) to obtain 433 bp of sequence from the cytochrome-b gene and 552 bp of the mtDNA control region. The extreme similarity of the sequences allowed us to align them visually. Each unique sequence was considered a haplotype, and only one representative of each haplotype was used for phylogenetic analysis. Harris' Sparrow (Z. querula) and White-throated Sparrow (Z. albicollis), the closest relatives to White-crowned and Golden-crowned sparrows (Zink and Blackwell 1996), were used as outgroups to root the crowned sparrow haplotype tree. We used the computer programs PAUP (Swofford 1990) for phylogenetic analysis, MEGA (Kumar et al. 1993) to compute pairwise Kimura (1980) two-parameter distances, and Arlequin (Schneider et al. 1997) to compute nucleotide diversity (7r). Collecting localities, collection dates, specimen voucher numbers, and sequences are associated with the GenBank accession numbers (U40173, U40175, U40185, U40186, AF305744-AF305776, AF308245-AF308281, AY008087-AY008123). We used allozyme data from Zink (1982), who analyzed 39 presumptive loci using standard starch gel electrophoresis.

Results and Discussion.-Four salient findings emerged: (1) we found extraordinarily low levels of sequence divergence between Z. leucophrys and Z. atricapilla, despite distinct plumage, song, and allozymes (Zink 1982); (2) some individuals of Z. leucophrys and Z. atricapilla share mtDNA haplotypes; (3) nucleotide diversity within the pool of Z. leucophrys and Z. atricapilla sequences is extremely low; and (4) Z. leucophrys and Z. atricapilla are not reciprocally monophyletic.

First, allozyme distance (Rogers 1972) between these two undisputed species was 0.0448 (Zink 1982), whereas Kimura (1980) two-parameter distances computed from mtDNA sequences ranged from 0 to 0.62% and averaged 0.24% over all individuals. Comparison of relative mtDNA and allozyme distances places the discrepancy in perspective (Fig. 2). MtDNA distance between White-crowned and Golden-crowned sparrows is very short relative to their congeners, whereas the allozyme distance is relatively longer. MtDNA evolves as rapidly or more rapidly than allozymes (Brown et al. 1979); therefore, given the observed allozyme distance (0.0448), divergence between mtDNA haplotypes of the two sparrow species should be an order of magnitude greater than that observed. In a comparison (not shown) of 122 pairs of passeriform mtDNA and allozyme distances, we found no pairs of species that showed a similar pattern.