BEHAVIORAL AND GENETIC IDENTIFICATION OF A HYBRID VIDUA: MATERNAL ORIGIN AND MATE CHOICE IN A BROOD-PARASITIC FINCH

Transcription

1 The Auk 121(1): , 2004 BEHAVIORAL AND GENETIC IDENTIFICATION OF A HYBRID VIDUA: MATERNAL ORIGIN AND MATE CHOICE IN A BROOD-PARASITIC FINCH R B. P 1,3 M D. S 2 1 Museum of Zoology and Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan 48109, USA; and 2 Department of Biology, Boston University, Boston, Massachuse s 02215, USA A. Hybrid male Vidua were observed in the field and recorded to document host song mimicry. The mtdna of one male was sequenced to identify the maternal parent. The hybrid males mimicked songs of Melba Finch (Pytilia melba), the usual host of Long-tailed Paradise Whydah (V. paradisaea), but the mtdna matched that of indigobirds (V. chalybeata or another species), which parasitize and mimic other estrildid finches. This combination of song behavior and genetics is consistent with a two-generation history that began with a female indigobird (e.g. V. chalybeata) laying in a Melba Finch nest rather than in a nest of her usual host (e.g. Red-billed Firefinch [Lagonosticta senegala]). Her daughter, genetically an indigobird, imprinted on her Melba Finch foster parents and then mated with a male paradise whydah mimicking Melba Finch song. She also laid eggs in Melba Finch nests. Her son, the male hybrid carrying his grandmother s indigobird mtdna, learned and later mimicked Melba Finch song. Genetic identification of the maternal species origin of this hybrid supports a model of mate choice based on mimetic song in the Vidua finches. Received 15 February 2003, accepted 5 October R. Se observaron machos híbridos del género Vidua en el campo y se grabaron sus vocalizaciones para documentar la imitación del canto de su hospedero. También se secuenció el ADN mitocondrial (ADNmt) de uno de los machos para identificar a la especie madre. Los machos híbridos imitaron el canto de la especie Pytilia melba, el hospedero usual de Vidua paradisaea, pero el ADNmt coincidió con el de una viuda (V. chalybeata u otra especie) que parasita e imita el canto de otros pinzones estríldidos. Esta combinación del comportamiento de canto y la genética es consistente con una historia de dos generaciones que comenzó con una viuda hembra (e.g. V. chalybeata) que puso un huevo en un nido de P. melba y no en un nido de su hospedero habitual, Lagonosticta senegala. Su hij a, genéticamente una viuda, se improntó en sus padres putativos (P. melba) y luego se apareó con un macho de la especie V. paradisaea, que imitaba el canto de P. melba. Ella también puso huevos en nidos de P. melba. Su hij o, el macho híbrido que llevaba el ADNmt de viuda de su abuela, aprendió y más tarde imitó el canto de P. melba. La identificación genética de la especie materna que originó a este híbrido apoya un modelo de selección de pareja basado en cantos miméticos en el género Vidua. B brood-parasitic finches of the genus Vidua provides a mechanism for reproductive isolation of populations associated with different host species. Males learn and mimic the songs of their host (Nicolai 1973, Payne et al. 1998) and females use male song mimicry to discriminate among potential mates, preferring males reared by the same host species (Nicolai 1964, Payne et al. 2000). Occasional hybridization may occur when females do not actively choose their mate on the basis of song. Females may reproduce with males mimicking the songs 3 rbpayne@umich.edu of a different host (1) if other male traits (e.g. plumage, flight displays) are more important than song in female choice, (2) if female choice is constrained by the availability of males singing the appropriate host song, or (3) if males engage in unsolicited copulation with females of other parasitic species. Unsolicited copulation leads to extrapair fertilizations in other songbirds (Payne and Payne 1989, Westneat 1990, Westneat et al. 1990). Alternatively, behavioral imprinting could lead to hybridization (Grant and Grant 1997) if female Vidua imprint on an alternative host during the period of parental care and as a result actively choose to mate with males of another species. 156

2 January 2004] Identification of a Hybrid Vidua 157 Two hybrid male Vidua were observed and tape-recorded at Lochinvar National Park in Zambia in 1973 (Payne 1980). Judging from morphology, those males were the product of a cross between a Long-tailed Paradise Whydah (V. paradisaea), the specialist broodparasite of Melba Finch (Pytilia melba; Nicolai 1969), and one of three indigobird species in the area: Purple Indigobird (V. purpurascens), Village Indigobird (V. chalybeata), or Twinspot Indigobird (V. codringtoni). The male hybrids mimicked Melba Finch song and not the songs of any normal indigobird host species, and therefore must have been reared in Melba Finch nests (Payne 1980). Because male indigobirds were seen to engage in unsolicited matings with female whydahs, Payne (1980) reasoned that the mother of the hybrids was a whydah that was fertilized by an indigobird and then laid eggs in nests of her usual host, the Melba Finch. Employing a method common in studies of songbird hybridization (Gelter et al. 1992, Joseph and Moritz 1993, Jung et al. 1994, Rohwer 1994), we used mitochondrial DNA (mtdna) to identify the maternal origin of a hybrid male Vidua, testing the expectation that it would have paradise whydah mtdna. The unexpected result of the genetic analysis, in combination with data on the hybrid s song, supports a two-generation model of hybridization in which female mate choice based on male song leads to hybridization in the generation a er a female Vidua lays in the nest of an alternative host. We discuss the implications of this result for understanding mate choice and egglaying behavior in parasitic finches and pa erns of genetic variation among Vidua species. M One of the hybrid males was collected and prepared as a specimen in 1973 (UMMZ ). A single primary feather was plucked from that specimen and DNA was extracted from the calamus using a QIAamp Tissue Kit (Qiagen, Valencia, California) supplemented with 30 µl of 100 mg ml 1 dithiothreitol (DTT) added to the tissue digestion buffer (Cooper 1994). The 5 half of the control region was amplified and sequenced using primers IndigoC1F1 and FinchC1R1 (Sorenson and Payne 2001). The resulting sequence was compared to those for all Vidua species. The unexpected result was confirmed by repeating the entire process with a second feather, the extract from which was used to amplify the 3 half of NADH dehydrogenase subunit 2 (ND2) using primers L5758 and H6313 (Sorenson et al. 1999). Sequences reported here have been deposited in GenBank (accession numbers: AF407022, AF407108, AY322693, AY323559, AY323565, AY324263). R The hybrid males were morphologically intermediate between V. paradisaea and indigobirds (see Payne 1980 for photographs). The hybrid male that was collected was all black, with a long tail that consisted of inner rectrices 170 mm in length and rectrices number 2 that were 194 mm in length and enclosed the inner pair. Rectrices number 2 were slender and shaped like those of V. paradisaea rather than those of Broad-tailed Paradise Whydah (V. obtusa), the other species of paradise whydah at Lochinvar. Wing length of the hybrid was 75 mm, intermediate between wing lengths of V. paradisaea and indigobirds (Payne 1980). The Long-tailed Paradise Whydah mimics the songs of Melba Finch, its estrildid finch host species, and those songs are unlike those of firefinches (Lagonosticta senegala and L. rhodopareia) or Peters Twinspot (Hypargos niveogu atus), the hosts of the three indigobird species that breed at Lochinvar (Payne et al. 1993). The songs of the hybrid Vidua clearly matched those of Melba Finch, the usual host species of V. paradisaea (see Payne 1980 for audiospectrograms). Contrary to expectation, mtdna sequences from the hybrid specimen were clearly indigobird sequences and not paradise whydah sequences for both the control region and ND2 (Fig. 1). Because mtdna is maternally inherited, the mother of the hybrid was therefore an indigobird rather than a paradise whydah. The identity of the mother s species within the indigobird species complex is uncertain, because the indigobirds breeding at Lochinvar and elsewhere in southern Africa are only marginally differentiated in mtdna haplotype frequencies and a number of closely related haplotypes are shared among species (Klein and Payne 1998, Sorenson et al. 2003). Mitochondrial DNA nucleotide sequences, however, clearly distinguish indigobirds from paradise whydahs (Fig. 1) and all other Vidua species (Sorenson et al. 2003). D The combined data on morphology, mimetic song behavior, and mtdna allow a

3 158 P S [Auk, Vol. 121 F. 1. Comparison of control region and ND2 sequences for the hybrid Vidua (UMMZ ), V. purpurascens (UMMZ ), V. paradisaea (RBP.A81, specimen in National Museum of Malawi), and V. obtusa (UMMZ ). Only variable sites are shown. Nucleotides matching the hybrid are indicated by dots. In the control region and ND2 fragments, respectively, the hybrid differs from the V. purpurascens individual at 1 of 490 alignment positions (0.2%) and 3 of 549 positions (0.5%). By comparison, genetic divergence in the control region ranges 0.8% among southern indigobird haplotypes. The hybrid differs from the whydahs at positions in the control region ( %) and positions in ND2 ( %). reconstruction of the events producing the hybrid male Vidua. The hybrid mimicked the song of Melba Finch, the host of Long-tailed Paradise Whydah, whereas its mtdna indicates that its mother was an indigobird. The intermediate male breeding plumage between that of an indigobird and a paradise whydah therefore indicates that the father was a paradise whydah. Figure 2 presents two alternative models for the production of this hybrid. In the first (Fig. 2A), a female paradise whydah mates with a male indigobird (whether by choice or coercion), then lays an egg in a nest of her usual host, Melba Finch. Her son, the male hybrid, learns Melba Finch song from his foster parents and carries his mother s paradise whydah mtdna. That model, however, is inconsistent with the observation that the hybrid male has indigobird mtdna. In a second model consistent with all of the empirical observations (Fig. 2B), a female indigobird mates with a male indigobird as usual but lays an egg in the nest of a Melba F. 2. Two models for the origin of a male paradise whydah indigobird hybrid. See text for a description of the events involved. Note that two reciprocal models produce male hybrids with the song of an indigobird host, but those are not relevant to the present example in which the hybrid was already known to have Melba Finch song. In addition, a model similar to that shown in (B) posits a male indigobird offspring in the first generation. If that male imprints on Melba Finches and sings Melba Finch song as an adult, it might attract a female Long-tailed Paradise Whydah that would lay in a Melba Finch nest, producing a hybrid son in the second generation that has Melba Finch song and paradise whydah mtdna, the same combination as in (A). Finch rather than in the nest of her usual host, a firefinch (if the indigobird was V. purpurascens or V. chalybeata) or a Peters Twinspot (if the indigobird was V. codringtoni). In the next generation, her daughter (a bird that was not knowingly observed in the field) is genetically an indigobird but imprints on her Melba Finch foster parents and learns their songs. As a result, she chooses to mate with a male paradise whydah mimicking Melba Finch song and lays

4 January 2004] Identification of a Hybrid Vidua 159 her eggs in nests of Melba Finch, the host species on which she is imprinted (see Payne et al. 2000). Her son, the male hybrid, learns and later mimics Melba Finch song and also carries his grandmother s indigobird mtdna. Captive cross-fostering experiments support the essential elements of the second model. Young male indigobirds imprint on songs of their foster species and sing the foster species songs as adults, even when the foster species is not their normal host (Nicolai 1973, Payne et al. 1998). Female Vidua also imprint on the songs of a novel foster species and as adults are sexually a racted to male Vidua that mimic those songs (Payne et al. 2000). In addition, those females parasitize nests of the species that reared them, even when active nests of their normal host species are available in the same aviary (Payne et al. 2000). Thus, infrequent hybridization among Vidua species may result from a two-generation process that begins with an egg-laying mistake rather than mating between males and females that are imprinted on different hosts. Throughout this process, females actively choose to mate with males based on their hostmimetic songs (Fig. 2B). Alternatively, female Vidua might choose males on the basis of plumage (Barnard 1990, Oakes and Barnard 1994) rather than song. In particular, female indigobirds might prefer males with long tails like those of male paradise whydahs, perhaps because they retain an ancestral sensory bias (Ryan and Rand 1993). Phylogenetic analyses suggest that short-tailed indigobirds may be derived from long-tailed ancestors (Klein and Payne 1998, Sorenson and Payne 2001). A hybrid with indigobird mtdna could be produced by a female indigobird a racted to the long tail of a male paradise whydah. That hybrid, however, would have Melba Finch song only if his firefinch-reared mother also laid in a Melba Finch nest, a second unexpected event in the same generation. Furthermore, there is no evidence that female indigobirds prefer males with long tails; female indigobirds implanted with estradiol did not approach or solicit male indigobirds with experimentally elongated tails (R. B. Payne et al. unpubl. data). The two-generation model presented above (Fig. 2B) requires a single unexpected event in the first generation (an indigobird laying in a Melba Finch nest) and provides a clear explanation for both female mate choice (based on song) and egg-laying behavior in the second generation. This model is consistent with experimental evidence that female Vidua choose males that mimic the songs of their own foster species (Payne 1973a, b; Payne et al. 2000). Given that different Vidua species mimic the nestling mouth pa erns of their respective hosts, our model requires that estrildid finch hosts at least occasionally rear brood-parasitic Vidua with mouth colors and pa erns unlike those of their own young. Red-billed Firefinches (L. senegala), which are normally parasitized by a mimetic indigobird, will rear other species in captivity, as will other estrildid finches (Immelmann et al. 1965, Goodwin 1982). In our aviaries, L. senegala have successfully reared to fledging and independence nestlings of Bluecapped Cordon-bleu (Uraeginthus cyanocephalus) and Goldbreast (Amandava subflava), two species that are likewise successful in rearing other estrildids as well as Vidua with mouth patterns different from their own young (Payne et al. 2001). In addition, various combinations of estrildids, including both closely related species and species in different genera, produce hybrid offspring in captivity (Lynch 1989, Fehrer 1993). Nestlings of nearly all estrildid species have different palate marks and colors (Immelmann et al. 1965, Goodwin 1982) and hybrid nestlings have mouth pa erns unlike those of either parent species, yet hybrids are reared by mixedspecies pairs to fledging and independence. The occurrence of hybrids between different estrildid species and successful cross-fostering demonstrate that parent finches can rear young with mouth pa erns and colors unlike those of their own species. Perhaps the most interesting implication of the Lochinvar hybrid is that it serves as a model of the usual mode of hybridization among indigobirds, a species complex in which mtdna haplotypes are shared among morphologically distinct species (Sorenson et al. 2003). Hybrids between two indigobird species are likely produced by essentially the same chain of events but go unnoticed because the species are similar in size and color and the females of some species are not morphologically distinguishable (Payne 1996). In contrast, the indigobird whydah hybrid could be recognized by its unusual morphology. As in Figure 2B, the two-generation process begins with a female indigobird laying

5 160 P S [Auk, Vol. 121 in the nest of an estrildid host that is already associated with another indigobird species. The resulting offspring imprint on the alternate host, leading to mixed mating and the production of genetic hybrids in the second generation. If the mislaid egg produces a female, as in the example presented here, a second-generation hybrid male will carry mtdna of the maternal parasitic species but his mimicry songs will indicate that he was reared by the alternate host. When a mislaid egg produces a male indigobird, he learns the song of his new foster species and thereby a racts female indigobirds of the species normally associated with the alternate host. In such a case, a second-generation male hybrid would carry mtdna consistent with his song mimicry (i.e. mtdna of the indigobird species normally associated with the host mimicked by the male). In principle, such hybrids might be detected with nuclear genetic markers or by their morphology, but limited differentiation among indigobird species makes identification of hybrids difficult at best. Song mimicry provides direct evidence of a male indigobird s host, and field observations are consistent with occasional mistakes in laying. In southern Africa, ~1% of males mimic songs of an estrildid finch other than their normal host species, and those males a ract female indigobirds of the species that normally parasitizes the alternate host (Payne 1973a, Payne et al. 1993). Female Vidua can be reared by an alternative foster species, imprint on that species, mate with males that mimic the song of that species, and lay selectively in nests of that species (Payne et al. 1998, 2000, 2001). Those behavioral mechanisms appear to be responsible for host colonization and speciation in the group but also may lead to hybridization among established species. The genetic similarity of indigobird species likely reflects both retained ancestral polymorphism following recent speciation and ongoing hybridization and introgression (Takahata and Slatkin 1984, Hoelzer and Melnick 1994, Sorenson et al. 2003). A Field work in Zambia was encouraged and aided by B. Chipeta, J. Colebrook-Robjent, R. J. Dowse, K. Hustler, K. Klitz, D. M. Lewis, and R. Stjernstedt. Fieldwork and collecting was permi ed by the Department of Wildlife and National Parks, Lusaka. D. Mindell provided space in the molecular genetics laboratory at the University of Michigan Museum of Zoology. R. Diener helped with tests of female response to tail length in captive indigobirds, and W. A. Searcy advised on experimental technique. Field and genetic research was supported by the University of Michigan Faculty Research Fund, the National Geographic Society, and the National Science Foundation. L C B, P Male ornament size, sexual display intensity and female sexual response in a parasitic African finch. Animal Behaviour 39: C, A DNA from museum specimens. Pages in Ancient DNA: Recovery and Analysis of Genetic Material from Paleontological, Archaeological, Museum, Medical, and Forensic Specimens (B. Herrmann and S. Herrmann, Eds.). Springer- Verlag, New York. F, J Interspecies-Kreuzungen bei cardueliden Finken und Prachtfinken. Pages in Typen des Lebens (S. Scherer, Ed.). Studium Integrale, Pascal Verlag, Berlin. G, H. P., H. T, L. G Evidence from hatching success and DNA fingerprinting for the fertility of hybrid Pied Collared flycatchers Ficedula hypoleuca albicollis. Ibis 134: G, D Estrildid Finches of the World. British Museum (Natural History), London. G, P. R., B. R. G Hybridization, sexual imprinting and mate choice. American Naturalist 149:1 28. H, G. A., D. J. M Pa erns of speciation and limits to phylogenetic resolution. Trends in Ecology and Evolution 9: I, K., J. S, H. E. W Prachtfinken, 2nd ed. vol. 1: Astrilde. Verlag Hans Limberg, Aachen, Germany. J, L., C. M Hybridisation between the White-browed and Atherton scrubwrens: Detection with mitochondrial DNA. Emu 93: J, R. E., E. S. M, R. C. F Behavior and parentage of a Whitethroated Sparrow Dark-eyed Junco hybrid. Wilson Bulletin 106: K, N. K., R. B. P Evolutionary associations of brood parasitic finches (Vidua) and their host species: Analyses of mitochondrial restriction sites. Evolution 52: L, J. D The gauge of speciation: On the frequencies of modes of speciation. Pages in Speciation and Its Consequences (D.

6 January 2004] Identification of a Hybrid Vidua 161 O e and J. A. Endler, Eds.). Sinauer Associates, Sunderland, Massachuse s. N, J Der Brutparasitismus der Viduinae als ethologisches Problem. Zeitschri für Tierpsychologie 21: N, J Beobachtungen an Paradieswitwen (Steganura paradisaea L., Steganura obtusa Chapin) und der Strohwitwe (Tetraenura fischeri Reichenow) in Ostafrika. Journal für Ornithologie 110: N, J Das Lernprogramm in der Gesangsausbildung der Strohwitwe Tetraenura fischeri Reichenow. Zeitschri für Tierpsychologie 32: O, E. J., P. B Fluctuating asymmetry and mate choice in Paradise Whydahs, Vidua paradisaea: An experimental manipulation. Animal Behaviour 48: P, R. B. 1973a. Behavior, mimetic songs and song dialects, and relationships of the parasitic indigobirds (Vidua) of Africa. Ornithological Monographs, no. 11. P, R. B. 1973b. Vocal mimicry of the Paradise Whydahs (Vidua) and response of female whydahs to the songs of their hosts (Pytilia) and their mimics. Animal Behaviour 21: P, R. B Behavior and songs in hybrid parasitic finches. Auk 97: P, R. B Field identification of the indigobirds. Bulletin of the African Bird Club 3: P, R. B., L. L. P Heritability estimates and behaviour observations: Extra-pair matings in Indigo Buntings. Animal Behaviour 38: P, R. B., L. L. P, M. E. D. N, K. H Species status and distribution of the parasitic indigobirds Vidua in east and southern Africa. Pages in Proceedings of the Eighth Pan-African Ornithological Congress (R. T. Wilson, Ed.). Royal Museum for Central Africa, Tervuren, Belgium. P, R. B., L. L. P, J. L. W Song learning in brood parasitic indigobirds Vidua chalybeata: Song mimicry of the host species. Animal Behaviour 55: P, R. B., L. L. P, J. L. W, M. D. S Imprinting and the origin of parasite host species associations in brood parasitic indigobirds Vidua chalybeata. Animal Behaviour 59: P, R. B., J. L. W, L. L. P Parental care in estrildid finches: Experimental tests of a model of Vidua brood parasitism. Animal Behaviour 62: R, S Two new hybrid Dendroica warblers and a new methodology for inferring parental species. Auk 111: R, M. J., A. S. R Sexual selection and signal evolution: The ghosts of biases past. Philosophical Transactions of the Royal Society of London, Series B 340: S, M. D., J. C. A, D. E., D, T. Y, D. P. M Primers for a PCR-based approach to mitochondrial genome sequencing in birds and other vertebrates. Molecular Phylogenetics and Evolution 12: S, M. D., R. B. P A single ancient origin of brood parasitism in African finches: Implications for host parasite coevolution. Evolution 55: S, M. D., K. M. S, R. B. P Speciation by host switch in brood parasitic indigobirds. Nature 424: T, N., M. S Mitochondrial gene flow. Proceedings of the National Academy of Sciences USA 81: W, D. F Genetic parentage in the Indigo Bunting: A study using DNA fingerprinting. Behavioral Ecology and Sociobiology 27: W, D. F., P. W. S, M. L. M The ecology and evolution of extra-pair copulations in birds. Current Ornithology 7: Associate Editor: M. T. Murphy