Conflicting phylogenetic signal of nuclear vs mitochondrial DNA markers in midwife toads (Anura, Discoglossidae, Alytes): Deep coalescence or ancestral …

HISTÓRIA EVOLUTIVA DOS SAPOS-PARTEIROS (Alytes spp.) NA PENÍNSULA IBÉRICA Análise filogenética e filogeográfica, reconstrução de um cenário biogeográfico e implicações taxonómicas

Maria Helena Aguiar Gonçalves

Dissertação apresentada à Faculdade de Ciências da Universidade do Porto para a obtenção do grau de Doutor em Biologia

Porto 2007

À minha família

Nota Prévia Na elaboração desta dissertação, e nos termos do artigo 8º do Decreto Lei nº 388/70, foi efectuado o aproveitamento total dos resultados de trabalhos já publicados ou submetidos, os quais integram alguns capítulos da presente tese. Em todos estes trabalhos, a candidata participou na obtenção, análise e discussão dos resultados, bem como na elaboração da sua forma publicada.

Agradecimentos

«Só o que sonhamos é o que verdadeiramente somos, porque o mais, por estar realizado, pertence ao mundo e a toda a gente.» Bernardo Soares, Livro do Desassossego

Agradecimentos A elaboração e finalização do presente trabalho só foi possível graças ao incentivo, empenho e apoio de inúmeras pessoas e instituições, às quais gostaria de expressar o meu sincero reconhecimento. A todos aqueles que contribuíram para esta tese e que, por lapso de memória, não foram aqui devidamente reconhecidos, as minhas sinceras desculpas e profundo agradecimento. Ao meu orientador, Prof. Dr. Nuno Ferrand de Almeida, a quem devo ter chegado até aqui, agradeço o incentivo dado ao longo de todos estes anos, ter sempre acreditado no meu trabalho e a total independência e liberdade que me concedeu nas mais variadas etapas desta tese. As enriquecedoras discussões científicas, enorme entusiasmo e dedicação na redacção deste manuscrito foram, sem dúvida, cruciais para a finalização deste trabalho com êxito. Gostaria, ainda, de deixar aqui uma palavra de reconhecimento pelo seu empenho na criação do CIBIO e por ter contribuído para que a investigação científica seja hoje uma realidade em Portugal. Ao meu co-orientador, Doutor Mario García-París, agradeço a forma excepcional como me recebeu no Museo Nacional de Ciências Naturales (MNCN) e as deslocações a Portugal para valiosas discussões dos resultados. Nunca me esquecerei da sua boa disposição

durante

uma

fantástica

saída

de

campo,

que

incluiu

um

roteiro

gastronómico e cultural por terras de “nuestros hermanos”, nem da forma apaixonada como, ao fim de tantos anos, encara a biologia e a ciência. To Prof. Trevor Beebee I would like to acknowledge all the laboratory facilities at School of Biological Sciences, University of Sussex (Brighton, UK) during my short visit to develop primers for the amplification of microsatellite loci in A. cisternasii. To Femmie Kraaijeveld-Smit I would like to thank her patience teaching me the “first steps” at the molecular lab in Brighton and the warm and nice way as, jointly with her husband Ken, received me at their home. For almost a month they were my family and even today I remember the nice evenings at the small village of Lewes.

Agradecimentos

To Graham Rowe I thank the helpful advice and assistance during the microsatellite cloning in Brighton. His permanent good mood softened the hard task of counting thousands of bacterial colonies. To Pim Arntzen I would like to acknowledge his help in the writing of the initial PhD project to submit to FCT and the contacts established with Prof. Trevor Beebee. Ao Iñigo Martínez-Solano quero agradecer a ajuda prestada na análise e discussão de resultados e todas as críticas e sugestões apresentadas. A colaboração por nós estabelecida funcionou tão bem, que espero que continue no futuro. Ao Ricardo Pereira, que, no âmbito do seu estágio de licenciatura, partilhou comigo uma importante etapa do trabalho de campo e laboratorial, demonstrando uma excelente capacidade de trabalho e maturidade científica, agradeço toda a ajuda na análise dos dados e as frutíferas discussões e sugestões que resultaram em trabalhos tão interessantes. A todos os meus colegas do Centro de Investigação em Biodiversidade e Recursos Genéticos (CIBIO), Antigoni, Armando Geraldes, Bárbara, Chico dos Lobos, Catarina, Cláudia, Elsa, Gonçalo, Hélder, James, Miguel Carneiro, Miguel Carretero, Pedro Cardia, Pedro Esteves, Paulo Célio, Raquel Vasconcelos, Ricardo, Rui, Sara Rocha, Vasco, Xana, Zé Carlos, Zef e, em particular, aos meus colegas de gabinete Rita, Sequeira e Tex o meu sincero agradecimento pela excelente camaradagem e ambiente de trabalho proporcionados ao longo de todos estes anos. A muitos deles devo o que hoje sei sobre técnicas laboratoriais e de análise genética. Em especial, quero agradecer ao Rui Faria por me ter ensinado a “ler” os géis de acrilamida, ao James Harris por todas as sugestões e correcções dos manuscritos em inglês, e à Alexandra Sá Pinto, Catarina Pinho, Fernando Sequeira e José Teixeira a ajuda prestada com alguns programas de análise e as valiosas sugestões e críticas que em muito valorizaram esta tese. A todos os meus colegas do MNCN, Annie Machordom, Carina Cunha, Cristina Grande, David Buckley, Elena Gonzaléz, Ernesto Recuero, Eva Albert, Iñigo Martínez-Solano, Jesus Manzanilla, Joaquim Reis, Marina Alcobendas, Mário García-París e Regina Cunha, que me proporcionaram uma excelente estadia em Madrid e me fizeram sentir em casa, o meu sincero agradecimento pela enorme disponibilidade que sempre foi

Agradecimentos

demonstrada e pela ajuda prestada nas tarefas laboratoriais “menos recomendadas” a uma grávida. A todos os que, através do envio de amostras ou da ajuda prestada durante o trabalho de campo, tornaram possível a amostragem planeada para este trabalho o meu profundo reconhecimento, nomeadamente a A Antúnez, A Bermejo, A Fonseca, A Gosá, A Loureiro, B Thiesmeier, C Grande, C Soares, D Donaire-Barroso, DJ Harris, E Albert, E Recuero, F Kraaijeveld-Smit, F Sequeira, G Themudo, I Martínez-Solano, JW Arntzen, J Bosch, JC Brito, J Manzanilla, J Teixeira, JP Veja, L Arregui, M Carretero, M. García-París, M Tejedo, M Vences, N Sillero, P Galán, R Jehel, R Márquez, R Pereira, R Ribeiro, R Vasconcelos e S Rocha. Sem a vossa ajuda este trabalho não se teria concretizado. Às gestoras de ciência do CIBIO, Sara Ferreira e Sandra Rodrigues, agradeço a sempre pronta ajuda na resolução dos problemas burocráticos e logísticos que foram surgindo ao longo de todos estes anos e que, tantas vezes, ultrapassaram as suas competências. À Ana Sara Monteiro agradeço o pronto envio dos muitos artigos científicos que lhe fui solicitando ao longo de todos estes anos em que esteve, primeiro, em Reading e, depois, em Oxford e ainda os divertidos fins-de-semana que passamos juntas durante a minha estadia em Inglaterra. Ao Nuno Vargas agradeço o design da capa desta tese. Ao Dr. Manuel Mouta Faria agradeço a cedência do desenho incluído na capa. Ao Miguel Carretero e especialmente ao seu pai agradeço a forma calorosa como me receberam em sua casa durante uma curta estadia em Barcelona para frequentar o curso de pós-graduação em Filogenias y Genealogías de DNA: Inferencia y Aplicaciones. À Fundação para a Ciência e Tecnologia agradeço a atribuição de uma bolsa de doutoramento com a referência SFRH/BD/3375/2000. Ao Instituto de Conservação da Natureza e Biodiversidade agradeço a atribuição das autorizações necessárias para recolha de amostras.

Agradecimentos

A todos os meus amigos e, em particular, às “cuscas”, Ana Sara, Catarina, Rita “mamã”, Rita, Sandra e Sara, agradeço todo o encorajamento, amizade e diversão que pontuaram estes últimos anos. Os nossos jantares das sextas-feiras foram sempre uma excelente fonte de encorajamento e espero que continuem por muitos e bons anos. Aos meus pais, o meu eterno reconhecimento pelo seu empenho incondicional que me permitiu chegar até aqui, por todo o apoio e compreensão. À minha mãe, agradeço ter-me libertado de muitas das difíceis tarefas domésticas, a prontidão com que sempre se disponibilizou para ficar com a Maria nos momentos mais críticos e o exemplo de força e coragem. Ao José Teixeira, meu eterno companheiro, agradeço o amor, carinho e incentivo e, particularmente, toda paciência e compreensão nas fases mais críticas deste trabalho, acreditando sempre que juntos ultrapassaríamos mais esta importante etapa das nossas vidas. Agradeço ainda a sua dedicada e enorme ajuda, quer na redacção quer na parte gráfica deste manuscrito e a forma extraordinária como soube gerir a grande quantidade de tarefas parentais que, nos últimos tempos, sobre ele recaíram. À minha filha Maria, que nasceu durante estes atribulados anos, agradeço o sacrifício de férias à procura de sapinhos e a assistir a congressos de Herpetologia, e as horas e fins-de-semana a ela roubados. Foi o seu lindo sorriso que muitas vezes me deu força para continuar.

Índice

Índice Resumo Summary Résumé Capítulo 1. Introdução Geral 1.1 A emergência da filogeografia 1.1.1 A importância da utilização de múltiplos marcadores moleculares 1.1.1.1 O DNA-mitocondrial 1.1.1.2 Os microssatélites 1.1.1.3 As genealogias nucleares 1.2 A Península Ibérica: um hotspot de biodiversidade 1.3 Filogenia e história evolutiva do género Alytes 1.4 As espécies em estudo 1.5 Objectivos e organização temática 1.6 Lista dos trabalhos que integram a tese 1.7 Referências bibliográficas

1 2 3 3 4 4 6 10 12 14 15

Capítulo 2. Filogenia e evolução dos sapos-parteiros, Alytes spp. Artigo I. Martínez-Solano I, Gonçalves H, Arntzen JW, García-París M (2004) Phylogenetic relationships and biogeography of midwife toads (Discoglossidae: Alytes). Journal of Biogeography, 31, 603-618.

23

Artigo II. Gonçalves H, Martínez-Solano I, Ferrand N, García-París M (2007) Conflicting phylogenetic signal of nuclear vs mitochondrial DNA markers in midwife toads (Anura, Discoglossidae, Alytes): deep coalescence or ancestral hybridization? Molecular Phylogenetics and Evolution, 44, 494-500.

53

Capítulo 3. Filogeografia do sapo-parteiro-comum (Alytes obstetricans) Artigo III. Gonçalves H, García-París M, Ferrand N (em preparação) Inferring ancient patterns of lineage diversification and extinction in the common midwife toad (Alytes obstetricans) based on mtDNA and nuclear genealogies.

69

Capítulo 4. Variabilidade genética e filogeografia do sapo-parteiroibérico (Alytes cisternasii) Artigo IV. Gonçalves H, Pereira R, Rowe G, Beebee T, Ferrand N (2005) Isolation and characterization of two dinucleotide and four tetranucleotide polymorphic microsatellite loci in the iberian midwife toad Alytes cisternasii. Molecular Ecology Notes, 5, 767-769.

99

Artigo V. Gonçalves H, Martínez-Solano I, Pereira R, García-París M, Ferrand N (submetido) Unexpected high levels of population subdivision in the Iberian Midwife Toad (Alytes cisternasii): evidence from mitochondrial and nuclear genomes.

105

História evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica

Capítulo 5. Discussão Geral 5.1 Filogenia dos sapos-parteiros, Alytes spp. 5.1.1 Análise filogenética do género Alytes com base em dados mitocondriais e caracteres morfológicos Evolução morfológica em Alytes 5.1.2 Análise filogenética do género Alytes com base em dados mitocondriais e nucleares 5.1.3 Biogeografia e história evolutiva do género Alytes 5.2 História evolutiva de Alytes obstetricans e de A. cisternasii 5.2.1 Filogeografia e história evolutiva de A. obstetricans 5.2.1.1 História evolutiva de A. obstetricans 5.2.1.2 Implicações taxonómicas e conservação 5.2.2 Variabilidade genética e filogeografia de A. cisternasii 5.2.2.1 História evolutiva de A. cisternasii 5.2.2.2 Evidência de múltiplos e recentes refúgios 5.2.2.3 Filogeografia comparada do sudoeste da Península Ibérica 5.3 Referências bibliográficas Capítulo 6. Considerações finais e perspectivas futuras 6.1 Referências bibliográficas

137 137 140 141 144 151 151 152 158 160 160 164 167 169 177 180

Resumo

Resumo Os sapos-parteiros (Anura, Discoglossidae, Alytes) englobam cinco espécies distribuídas pelo Norte de África, Península Ibérica e parte da Europa ocidental, sendo actualmente reconhecidos três sub-géneros: Alytes sensu stricto (A. obstetricans), Ammoryctys (A. cisternasii) e Baleaphryne (A. dickhilleni, A. maurus e A. muletensis). Ao longo das últimas décadas, vários estudos tentaram clarificar as relações filogenéticas entre as espécies de Alytes e reconstruir a sua biogeografia histórica com base em dados morfológicos, imunológicos e genéticos. Contudo, ainda não foi proposta uma hipótese filogenética consensual para este género e a opção entre os vários cenários biogeográficos continua ambígua. Assim, o primeiro objectivo desta tese consistiu em investigar as relações filogenéticas das várias espécies e subespécies de Alytes, através da análise de genes mitocondriais (mtDNA) e nucleares, e de caracteres morfológicos, bem como reconstruir a história biogeográfica do género. Deste modo, pretendeu-se desenvolver uma

hipótese

filogenética

robusta

através

da

sequenciação

de

três

genes

mitocondriais (16S, cyt-b e ND4) e de um gene nuclear (β-fibint7) e do estudo da variação morfológica de caracteres osteológicos. Os resultados sugerem a existência de uma diferenciação relativamente elevada entre A. cisternasii e as restantes espécies do género, suportam a monofilia do clado Baleaphryne, e confirmam o estatuto específico proposto para as populações marroquinas de sapo-parteiro. No entanto, a relação filogenética de A. muletensis, A. dickhilleni e A. maurus constitui uma clara tricotomia que continua por resolver. Por outro lado, não foi inicialmente possível esclarecer a posição filogenética de A. o. almogavarii, uma vez que os resultados obtidos mostraram a existência de discrepâncias profundas entre os dois tipos de marcadores moleculares no que diz respeito à posição deste taxon. A combinação dos dados morfológicos, nucleares e mitocondriais sugeriram que as incongruências observadas deverão resultar da existência de polimorfismo ancestral no locus β-fibint7. Contudo, estas discordâncias são também compatíveis com a ocorrência de fluxo génico histórico entre A. o. almogavarii e um proto-Baleaphryne, bem como com fenómenos mais recentes de hibridação e introgressão após contacto secundário com outras linhagens de A. obstetricans. As estimativas da divergência entre os vários grupos foram congruentes com as inferidas em estudos anteriores a partir da análise aloenzimática, com excepção da

História evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica

diferenciação de A. o. almogavarii e da separação entre A. dickhilleni e A. muletensis. Os eventos vicariantes associados nesta fase do trabalho com a radiação do género foram: i) a formação dos grandes lagos salinos no interior da Península Ibérica há cerca de 16 Ma (milhões de anos) ou, alternativamente, as bruscas alterações climáticas registadas na transição entre o período Badeniense médio e tardio há cerca de 14-13,5 Ma (A. cisternasii vs. o ancestral das restantes espécies de Alytes), ii) a estruturação final dos Neo-Pirinéus e a reabertura do Estreito Bético, há cerca de 10-8 Ma (A. o. almogavarii vs. A. obstetricans e vs. o sub-género Baleaphryne), iii) a abertura do estreito de Gibraltar no final da crise Messiniense, há cerca de 5,3 Ma (A. maurus vs. o ancestral do clado A. dickhilleni-A. muletensis), iv) um processo de colonização trans-oceânica, há cerca de 3 Ma (o ancestral de A. muletensis vs. A. dickhilleni); e v) os períodos glaciares do Pleistoceno (diferenciação dentro de A. obstetricans). O segundo objectivo desta tese consistiu na análise dos padrões geográficos de distribuição da diversidade genética do sapo-parteiro-comum (A. obstetricans) através do estudo do polimorfismo de genes mitocondriais e nucleares ao longo de toda a sua área de distribuição. Em particular, procurou-se investigar i) a extensão da congruência entre os dois tipos de marcadores moleculares e o seu grau de sobreposição

com

os

dados

morfológicos,

ii)

a

dinâmica

de

fragmentação,

diferenciação e expansão pós-glacial das diferentes subespécies de A. obstetricans, e iii) a ocorrência de zonas de contacto secundário entre linhagens divergentes e subsequentes processos de miscigenação. Adicionalmente, procurou-se ainda explorar as propriedades do polimorfismo ancestral exibido pelo locus β-fibint7 e avaliar a sua informatividade sobre os processos evolutivos que terão levado à diversificação do clado Baleaphryne. Em primeiro lugar, observou-se um notável grau de variabilidade genética, de acordo com resultados anteriores obtidos através da análise morfológica e aloenzimática. Em segundo lugar, obteve-se uma quase completa concordância entre a distribuição das linhagens mitocondriais e nucleares nas subespécies A. o. almogavarii e A. o. pertinax, reforçando o carácter distintivo destas formas. Em terceiro lugar, foi possível documentar a captura de duas linhagens mitocondriais muito antigas por populações modernas de A. obstetricans: a primeira parece confinada à região central dos Pirinéus espanhóis, enquanto a segunda ocupa uma extensa área geográfica localizada a sul do rio Douro. Estes resultados sugerem que populações ancestrais de A. obstetricans se terão recorrentemente expandido e contraído a partir de um refúgio glacial possivelmente localizado no Nordeste Ibérico. Este facto torna esta espécie única no contexto ibérico pelo facto de apresentar, simultaneamente, padrões muito contrastantes de diversidade genética que permitem inferir uma rápida velocidade de colonização em direcção à Europa Central e, ao

Resumo

contrário, uma lenta dispersão em direcção ao Noroeste peninsular. No que se refere ao polimorfismo ancestral documentado para o locus β-fibint7 foi possível inferir que a permanente ocupação do Nordeste da Península Ibérica por um ancestral de Alytes terá estado, de forma quase simultânea, na origem da expansão que resultou posteriormente na diferenciação do clado Baleaphryne e de um proto-obstetricans. Este cenário é, por isso, incompatível com a sugestão inicial de que os Neo-Pirinéus teriam determinado a diferenciação de A. o. almogavarii. Por outro lado, tanto os dados nucleares como a concatenação de múltiplos genes mitocondriais indicam a quase simultânea diversificação de Baleaphryne durante a crise Messiniense, contrariando a nossa hipótese inicial de uma colonização trans-oceânica para A. muletensis. Finalmente, a combinação do notável grau de diversificação das linhagens de A. o. almogavarii com a sua distribuição geográfica muito restrita cria uma situação difícil de compreender a não ser que se invoque a ocorrência de um processo de especiação incipiente. A confirmar-se, este resultado é não só especialmente relevante do ponto de vista evolutivo, como poderá ter implicações em termos de conservação. Por último, o terceiro objectivo pretendeu estudar a variabilidade genética, a estruturação populacional e os padrões filogeográficos do sapo-parteiro-ibérico (A. cisternasii) em toda a sua área de ocorrência através da análise de um marcador mitocondrial e de um conjunto de microssatélites especialmente desenvolvidos para o efeito. Os resultados obtidos demonstraram a existência de estruturação genética significativa,

com pelo menos quatro linhagens parapátricas que podem ser

consideradas como unidades evolutivas distintas. Estes dados contrastam claramente com a homogeneidade morfológica descrita para esta espécie, adicionalmente suportada pela reduzida variação inter- e intra-populacional obtida com o estudo de aloenzimas. A análise de microssatélites permitiu a detecção de padrões concordantes de diversidade genética em relação ao mtDNA, sugerindo a fragmentação populacional em vários refúgios durante as glaciações do Pleistoceno, seguida por eventos recentes de expansão demográfica e contacto secundário. Finalmente, este trabalho contribui para o número crescente de espécies de distribuição marcadamente mediterrânica com padrões filogeográficos conhecidos, permitindo a progressiva emergência de análises comparativas numa região geográfica até agora pouco estudada. Em particular, foi possível salientar a peculiaridade das regiões montanhosas do sudoeste peninsular, que terão moldado desde há seis milhões de anos e até ao Pleistoceno superior, os padrões de diversidade genética de organismos tão distintos como peixes, anfíbios e répteis.

Summary

Summary The

midwife

toads

(Anura,

Discoglossidae,

Alytes)

comprise

five

species

distributed in North Africa, the Iberian Peninsula and part of Western Europe which are included in three sub-genera: Alytes sensu stricto (A. obstetricans), Ammoryctys (A. cisternasii) and Baleaphryne (A. dickhilleni, A. maurus and A. muletensis). In the last few decades, multiple studies tried to clarify the phylogenetic relationships between Alytes species and reconstruct their historical biogeography based on morphological, immunological and genetic data. However, a consensual phylogenetic hypothesis still does not exist for this genus and an option between the different available biogeographical scenarios remains ambiguous. In this context, the first goal of this thesis consisted in the investigation of the phylogenetic relationships between the different Alytes species and subspecies through the analysis of mitochondrial (mtDNA) and nuclear genes, as well as of morphological characteristics, together with the reconstruction of the historical biogeography of this genus. Accordingly, we tried to develop a robust phylogenetic hypothesis by sequencing three mtDNA genes (16S, cyt-b e ND4) and one nuclear gene (β-fibint7), as well as performing a morphological analysis of osteological characteristics. Our results suggest the existence of a relatively high divergence between A. cisternasii and the remaining species in the genus, support the monophyly of the Baleaphryne clade, and confirm the specific status proposed for the Moroccan midwife toad populations. However, the phylogenetic relationships of A. muletensis, A. dickhilleni and A. maurus exhibit a clear trichotomy and remain to be resolved. On the other hand, the clarification of the phylogenetic position of A. o. almogavarii was not initially possible because the results showed deep discordances between the two types of molecular markers. The combination of morphological, nuclear and mitochondrial data indicated that the observed discordances likely result from the existence of ancestral polymorphism at the β-fibint7 locus. However, these discordances are also compatible with the occurrence of past gene flow between A. o. almogavarii and a proto-Baleaphryne, as well as with more recent hybridization and introgression phenomena after secondary contact with other A. obstetricans lineages. Divergence estimates between the different groups were found to be congruent with those inferred in previous studies from allozyme analysis, with the exception of the differentiation of A. o. almogavarii and the separation between A. dickhilleni and

História evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica

A. muletensis. The vicariant events initially associated to the radiation of the genus were: i) the formation of the great saline lakes in inland Iberia, approximately 16 Mya (millions of years ago) or, alternatively, the sudden climatic changes observed in the Middle-Late Badenian transition, about 14-13.5 Mya (A. cisternasii vs. the ancestral of the remaining Alytes species, ii) the final structuring of the Neo-Pyrenees and the reopening of the Betic Strait, about 10-8 Mya (A. o. almogavarii vs. A. obstetricans and vs. the sub-genus Baleaphryne), iii) the opening of the Strait of Gibraltar at the end of the Messinian Salinity Crisis, about 5.3 Mya (A. maurus vs. the ancestral of the clade A. dickhilleni-A. muletensis), iv) a process of trans-oceanic colonization, about 3 Mya (the ancestral of A. muletensis vs. A. dickhilleni); and v) the Pleistocene glacial periods (differentiation within A. obstetricans). The second goal of this thesis consisted in the analysis of the geographical distribution of patterns of genetic diversity in the common midwife toad (A. obstetricans) through the study of variation in both mitochondrial and nuclear genes encompassing its complete distribution range. In particular, we assessed i) the extension of concordance between the two types of molecular markers and its degree of overlap with morphological data, ii) the dynamics of fragmentation, differentiation and post-glacial expansion of the different A. obstetricans subspecies, and iii) the occurrence of secondary contact zones between divergent lineages and subsequent processes of admixture. In addition, we also explored the properties of the ancestral polymorphism exhibited at the β-fibint7 locus and evaluated its informativeness for the evolutionary processes that ultimately determined the diversification of the Baleaphryne clade. First, we observed a remarkable degree of genetic variability, in agreement with previous results based on morphological and allozyme analyses. Second, we obtained an almost perfect concordance between the distribution of mtDNA and nuclear lineages in the subspecies A. o. almogavarii and A. o. pertinax, strengthening the distinct character of those forms. Third, we could document the capture of two old mtDNA lineages by modern A. obstetricans populations: the first seems confined to the central Spanish Pyrenees, while the second occupies an extended geographical area to the south of the Douro River. These results suggest that ancient populations of A. obstetricans have been expanding and contracting from a glacial refugium that is probably localized in the Iberian Northeast. This fact makes this species unique in the Iberian context because of simultaneously exhibiting highly contrasting patterns of genetic diversity which indicate a rapid spread into Central Europe and, on the contrary, a slow colonization of the Iberian Northwest. In relation with the documented ancestral polymorphism at the β-fibint7 locus we inferred that the persistence of an ancestral of Alytes in the Iberian Northeast was at the origin, in a very short timeframe, of the expansion that subsequently resulted in the

xviii

Summary

differentiation of the Baleaphryne clade and of a proto-obstetricans. This scenario is thus incompatible with our initial suggestion that the Neo-Pyrenees would have determined the differentiation of A. o. almogavarii. On the other hand, both the nuclear data and the concatenation of multiple mitochondrial genes point to the almost simultaneous diversification of Baleaphryne during the Messinian Salinity Crisis, in contrast with our initial hypothesis of a trans-oceanic colonization for A. muletensis. Finally, the combination of the notable degree of lineage diversification within A. o. almogavarii with its very restricted geographical distribution creates a situation that is difficult to understand unless we invoke the occurrence of a process of incipient speciation. In confirmed, this result is not only important from an evolutionary point of view, but may also have relevant implications for the conservation of biodiversity. The last goal of this thesis was the study of the genetic variability, population structuring and phylogeographical patterns in the Iberian midwife toad (A. cisternasii) across its entire geographical range through the analysis of one mtDNA marker and a set of microsatellites that were especially developed for this work. Results showed the existence of a significantly genetic structuring and at least four parapatric mtDNA lineages that may well correspond to independent evolutionary units. These data are in clear contrast with the previously described morphological homogeneity in this species, which is additionally supported by the reduced inter- and intrapopulation diversity inferred from allozymes. The analysis of microsatellites allowed the detection of concordant patterns of genetic diversity in relation with the mtDNA data, suggesting population fragmentation in several refugia during the Pleistocene glaciations followed by recent events of demographic expansion and secondary contact. Finally, this work also contributes for the growing number of species with a marked mediterranean distribution which phylogeographic patterns are well documented, allowing the progressive emergence of comparative analysis in a geographical region that remains poorly known. In particular, it was possible to highlight the peculiarity of the mountainous areas of the Iberian southwest that may have been sculpting the patterns of genetic diversity in organisms as different as fish, amphibians and reptiles since at least six millions years ago.

Résumé

Résumé Les crapauds accoucheurs (Anura, Discoglossidae, Alytes) se regroupent en cinq espèces présentes au Nord de l’Afrique, sur la Péninsule Ibérique et sur une partie de l’Europe Occidentale et constituent trois sous-genres: Alytes sensu stricto (A. obstetricans), Ammoryctys (A. cisternasii) et Baleaphryne (A. dickhilleni, A. maurus et A. muletensis). Ces dernières décennies, plusieurs études ont essayé de clarifier les relations phylogénétiques entre les différentes espèces d’Alytes et de reconstruire sa biogéographie historique fondée sur des données morphologiques, immunologiques et génétiques. Cependant, une hypothèse phylogénétique consensuelle reste à définir et le choix pour un des différents scénarios biogéographiques reste ambiguë. Dans ce cadre, le premier objectif de cette thèse concernait l’investigation des relations phylogénétiques des différentes espèces et sous-espèces d’Alytes à travers l’analyse de gènes mitochondriaux (mtDNA) et nucléaires, et de caractéristiques morphologiques, aussi bien que la reconstruction biogéographique de ce genre. De cette façon, nous avons essayé de développer une hypothèse phylogénétique robuste à travers le séquençage de trois gènes mitochondriaux (16S, cyt-b et ND4) et d’un gène nucléaire (β-fibint7), et aussi à travers l’étude de la variation morphologique de caractères ostéologiques. Les résultats suggèrent l’existence d’une différenciation relativement forte entre A. cisternasii et les autres espèces du genre, supportent la monophylie du clade Baleaphryne, et confirment le statut spécifique suggéré pour les populations

marocaines

de

crapaud

accoucheur.

Pourtant,

les

relations

phylogénétiques de A. muletensis, A. dickhilleni et A. maurus forment une trichotomie évidente et persistent sans résolution. D’un autre côté, la position phylogénétique de A. o. almogavarii n’a pas été clarifiée parce que les résultats obtenus ont montré l’existence

de

moléculaires.

très La

fortes

discordances

combinaison

des

entre

données

les

deux

types

morphologiques,

de

marqueurs

nucléaires

et

mitochondriales suggère que les discordances observées résultent probablement de l’existence d’un polymorphisme ancestral dans le locus β-fibint7. Cependant, ces discordances sont aussi compatibles avec l’existence de flux géniques historiques entre A. o. almogavarii et un proto-Baleaphryne ou avec des phénomènes plus récents d’hybridation et d’introgression après le contact secondaire avec d’autres lignées de A. obstetricans.

História evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica

Les estimations de divergence entre les différents groupes sont cohérents avec celles antérieurement publiées à partir de l’analyse alloenzymatique, à l’exception de la différenciation de A. o. almogavarii et de la séparation entre A. dickhilleni et A. muletensis. Les événements de vicariance initialement associés à la radiation du genre sont: i) la formation des grands lacs salins à l’intérieur de la Péninsule Ibérique, il y a environ

16

Ma

(millions

d’années)

ou,

de

façon

alternative,

les

drastiques

changements climatiques documentés pour la transition entre la période Badeniense moyen et tardif, il y a environ 14-13,5 Ma (A. cisternasii vs. l’ancêtre des autres espèces d’Alytes), ii) la structuration finale des Neo-Pyrénées et l’ouverture du Détroit Bétique, il y a environ 10-8 Ma (A. o. almogavarii vs. A. obstetricans et vs. le sousgenre Baleaphryne), iii) l’ouverture du Détroit de Gibraltar à la fin de la crise Messinique, il y a environ 5,3 Ma (A. maurus vs. l’ancêtre du clade A. dickhilleni-A. muletensis), iv) un processus de colonisation trans-océanique, il y a environ 3 Ma (l’ancêtre de A. muletensis vs. A. dickhilleni); et finalement v) les périodes glaciaires du Pléistocène (différenciation au sein de A. obstetricans). Le

deuxième

objectif

de

cette

thèse

concernait

l’analyse

des

patrons

géographiques de distribution de la diversité génétique du crapaud accoucheur commun

(A.

obstetricans)

à

travers

l’étude

du

polymorphisme

de

gènes

mitochondriaux et nucléaires tout au long de son aire de répartition. Plus précisément, nous avons recherché i) l’extension de congruence entre les deux types de marqueurs moléculaires et son degré de concordance avec les données morphologiques, ii) la dynamique de fragmentation, de différenciation et d’expansion post-glaciale des différentes sous-espèces de A. obstetricans, et iii) l’occurrence de zones de contact secondaire entre les lignées divergentes et les processus de mélange subséquentes. De plus, nous avons aussi exploré les propriétés du polymorphisme ancestral montré pour le locus β-fibint7 et évalué son informativité sur les processus évolutifs qui ont déterminé la diversification du clade Baleaphryne. D’abord, nous avons observé un remarquable degré de variabilité génétique, en accord avec résultats antérieurs obtenus avec l’analyse morphologique et alloenzymatique. Deuxièmement, nous avons obtenu

une

concordance

presque

complète

entre

la

distribution

des

lignées

mitochondriales et nucléaires au sein des sous-espèces A. o. almogavarii et A. o. pertinax, supportant le caractère distinctif de ces formes. Troisièmement, il nous a été possible de documenter la capture de deux lignées mitochondriales très anciennes par des populations modernes de A. obstetricans: la première semble être confinée à la région centrale des Pyrénées espagnoles, alors que la deuxième est distribuée sur une grande aire géographique localisée au sud du fleuve Douro. Ces résultats suggèrent que les populations ancestrales de A. obstetricans ont subi plusieurs expansions et contractions à partir d’un refuge glacial éventuellement localisé au nord-est de la

Résumé

Péninsule Ibérique. Cette observation confère un caractère unique à A. obstetricans dans le contexte ibérique par le fait de présenter, simultanément, des patrons très contrastés de diversité génétique qui permettent d’envisager une vitesse de colonisation très rapide en direction de l’Europe Central et, au contraire, une dispersion bien plus lente en direction du Nord-ouest péninsulaire. En ce qui concerne le polymorphisme ancestral documenté pour le locus β-fibint7 il a été possible d’estimer que l’occupation permanente du Nord-est de la Péninsule Ibérique par un ancêtre d’Alytes a été à l’origine, de façon presque simultanée, de l’expansion qui a abouti plus tard à différenciation du clade Baleaphryne, d’un côté, et d’un protoobstetricans, d’un autre côté. Ce scénario est, pour cette raison, incompatible avec notre suggestion initiale que les Neo-Pyrénées auraient déterminé la différenciation de A. o. almogavarii. D’autre part, les données nucléaires comme la concaténation de plusieurs gènes mitochondriaux indiquent une diversification presque simultanée de Baleaphryne pendant la crise Messinienne, en contradiction avec notre hypothèse initiale d’une colonisation trans-océanique pour A. muletensis. Finalement, la combinaison du degré notable de diversification des lignées de A. o. almogavarii avec sa distribution géographique très restreinte conduit à une situation difficile à comprendre à moins que l’on suggère l’existence d’un processus de spéciation incomplete. Si nous arrivons à confirmer ce résultat, il sera non seulement d’une grande importance d’un point de vue évolutif, mais il pourrait aussi avoir des conséquences sur le plan de la conservation de la biodiversité. Finalement, le dernier objectif concernait l’étude de la variabilité génétique, de la structuration

populationnelle

et

des

patrons

phylogéographiques

du

crapaud

accoucheur ibérique (A. cisternasii) dans toute son aire de répartition à travers l’analyse d’un marqueur mitochondrial et d’un ensemble de marqueurs microsatellites spécialement développés à cet effet. Les résultats obtenus ont montré l’existence d’une structuration génétique significative, avec au moins quatre lignées parapatriques que l’on peut considérer comme des unités évolutives distinctes. Ces donnés contrastent nettement avec l’homogénéité morphologique décrite dans cette espèce, et supportée par la très faible variation intra- et inter-populationnelle détectée avec les allozymes. L’analyse des marqueurs microsatellites a permis la description de patrons

de

variation

concordants

avec

l’ADN

mitochondrial,

suggérant

la

fragmentation populationnelle de l’espèce en plusieurs refuges pendant les glaciations du Quaternaire, suivi par des événements récents d’expansion démographique et de contact secondaire. Finalement, ce travail est aussi une contribution pour le nombre croissant d’espèces de distribution typiquement méditerranéenne avec des patrons phylogéographiques

connus,

permettant

l’émergence

progressive

d’analyses

comparatives dans une région jusqu’a présent très peu étudiée. Plus particulièrement,

História evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica

il a été possible de montrer la singularité des régions de montagnes moyennes du sud-ouest péninsulaire, qui ont sculpté depuis plus de six millions d’années jusqu’au présent les patrons de diversité génétique d’organismes aussi différents que les poissons, les amphibiens et les reptiles.

CAPÍTULO 1 Introdução Geral

Introdução Geral

1. Introdução Geral O conhecimento da variabilidade genética é essencial para a reconstrução da história

evolutiva

das

espécies,

tendo

o

desenvolvimento

de

novas

técnicas

moleculares permitido a generalização deste tipo de estudos a um vasto número de organismos em todo o planeta durante a última década (Avise, 2000). Quando considerada num contexto geográfico, a estruturação desta variabilidade é o produto de fenómenos contemporâneos de dinâmica populacional e de relações históricas entre as populações, sendo o seu conhecimento essencial para a manutenção de espécies ou populações ecológica e geneticamente viáveis (Frankham, 2003; Avise, 2004). Em organismos com uma baixa capacidade de dispersão, tais como os anfíbios (por exemplo, Duellman & Trueb, 1986; mas ver Smith & Green, 2005), o estudo dos efeitos de fenómenos vicariantes e dos padrões de diversidade genética que exibem são muito informativos sobre os processos que determinaram as suas histórias evolutivas.

1.1 A emergência da filogeografia No sentido de analisar e interpretar os factores históricos que conduziram à distribuição geográfica actual das unidades evolutivas, a nível intra e interespecífico, surgiu, em finais da década de 80, uma nova disciplina – a filogeografia (Avise et al., 1987). Esta disciplina funciona como elo de ligação entre a genética populacional e a sistemática, integrando vários domínios científicos, como a genética molecular e populacional, a filogenia, a ecologia, a etologia, a paleogeografia e a paleontologia (Avise, 2000). Desde finais da década de 90 que o número de trabalhos publicados sobre a filogeografia

de

várias

espécies

de

fauna

e

flora

tem

vindo

a

aumentar

consideravelmente (Avise, 2000; Hewitt, 2000), existindo já compilados inúmeros estudos para vários taxa de todas as regiões do planeta (ver revisão em Hewitt, 2004; Emerson & Hewitt, 2005). A denominada filogeografia comparada, que consiste na procura e identificação de padrões filogeográficos comuns entre os taxa (Taberlet et al., 1998; Arbogast & Kenagy, 2001), é uma ferramenta fundamental para identificar áreas importantes para a conservação por albergarem uma elevada diversidade

1

Capítulo 1

genética, um critério que complementa a tradicional selecção baseada na diversidade específica (Moritz & Faith, 1998). A concordância dos padrões filogeográficos entre diferentes espécies permite, frequentemente, revelar o papel dos fenómenos climatéricos históricos na subestruturação dos organismos, bem como identificar refúgios glaciares, rotas de colonização pós-glacial e zonas de contacto secundário comuns (por exemplo, Bernatchez & Wilson, 1998; Schneider et al., 1998; Taberlet et al., 1998; Sullivan et al., 2000; Arbogast & Kenagy, 2001; Carstens et al., 2005; Feldman & Spicer, 2006; Gómez & Lunt, 2007). A distribuição actual das espécies é determinada por requisitos ecológicos e factores históricos. Entre estes, os períodos glaciares do Pleistoceno (entre 1,8 milhões de anos (Ma) e 10.000 anos atrás) constituem um excelente exemplo dos efeitos de flutuações climáticas drásticas na distribuição da flora e fauna do Hemisfério Norte (Hewitt, 1996, 1999, 2000; Taberlet et al., 1998). As oscilações climáticas do Pleistoceno afectaram a maior parte das espécies das regiões temperadas, causando sucessivas flutuações demográficas e alterações nas suas áreas de distribuição, que conduziram a numerosos fenómenos de extinção local, retracção e fragmentação em regiões meridionais, durante os períodos glaciares, e à sua posterior expansão para norte e formação de múltiplas zonas de contacto secundário durante os períodos inter-glaciares (Avise, 2000; Hewitt, 2000, 2004). Os eventos recorrentes de fragmentação das populações promoveram a diferenciação dos grupos populacionais isolados pelo aumento do efeito de deriva genética, induzindo diversas pressões selectivas, que conduziram, por sua vez, a adaptações locais e, em alguns casos, ao isolamento reprodutivo entre populações. Contudo, para a maior parte dos organismos, a especiação completa parece não ter resultado dos fenómenos vicariantes ocorridos durante o Pleistoceno (Avise et al., 1998; Hewitt 1996, 2004).

1.1.1 A importância da utilização de múltiplos marcadores moleculares A

análise

combinada

de

diferentes

marcadores

moleculares

revela-se

de

primordial importância no estudo da história evolutiva das espécies. Vários trabalhos (Pamilo & Nei, 1988; Edwards & Beerli, 2000; Ballard et al., 2002; Shaw, 2002; Machado & Hey, 2003; Monsen & Blouin, 2003) têm vindo a demonstrar que a utilização de apenas um locus, ou de um só tipo de marcador, pode conduzir a inferências incompletas e eventualmente discordantes sobre os padrões evolutivos das espécies, devido ao limite de resolução que cada um apresenta e à possibilidade de ocorrência de fenómenos selectivos e estocásticos.

2

Introdução Geral

1.1.1.1 O DNA-mitocondrial Durante as últimas duas décadas, a análise da variação do DNA mitocondrial (mtDNA) foi a principal fonte de informação em estudos filogeográficos e filogenéticos (Avise 2000, 2004). As características do mtDNA que contribuíram para esta situação incluem a facilidade de extracção e obtenção de dados, elevada taxa de mutação quando comparada com os genes nucleares, a ausência de recombinação para a generalidade das espécies (mas ver Tsaousis et al., 2005) e o facto de ser haplóide e de hereditariedade uniparental (Avise, 2000). O modo de transmissão uniparental do mtDNA reflecte-se no tamanho do efectivo populacional que é, em média, quatro vezes inferior ao dos genes nucleares, sendo um marcador molecular mais sensível a fenómenos estocásticos e de deriva genética (Moore, 1995). No entanto, é também actualmente reconhecido que o mtDNA apresenta várias desvantagens para estudos filogenéticos e filogeográficos (ver revisão em Zhang & Hewitt, 2003; Ballard & Whitlock, 2004), incluindo a eliminação de polimorfismo ancestral, introgressão e interpretações evolutivas enviesadas no caso de fêmeas e machos das espécies em estudo apresentarem selecção sexual ou diferirem em determinadas características biológicas importantes, como a dispersão (FitzSimmons et al., 1997; Alves et al., 2003; Melo-Ferreira et al., 2005). Além disso, a análise do mtDNA corresponde ao estudo de um único locus e a informação daí retirada traduz a história evolutiva dessa molécula e não necessariamente da espécie em estudo (Bazin et al., 2006). Assim, e associado ao recente desenvolvimento de técnicas e métodos de análise mais sofisticados, tem-se verificado um aumento dos estudos que utilizam diferentes marcadores moleculares, tais como microssatélites e genealogias nucleares. Além disso, o crescente desenvolvimento de métodos baseados na teoria da coalescência (Knowles & Maddison, 2002; Hey & Machado, 2003; Hey & Nielsen, 2004) tem permitido

traçar

cenários

filogeográficos

mais

complexos,

bem

como

testar

explicitamente diferentes hipóteses explicativas dos dados obtidos. 1.1.1.2 Os microssatélites Os microssatélites consistem na repetição em tandem de dois até seis pares de bases que, devido à sua elevada taxa de mutação, elevado polimorfismo por locus e modo de transmissão mendeliano, permitem o estudo de aspectos relacionados com a história evolutiva mais recente das populações (Jarne & Lagoda, 1996; Angers & Bernatchez, 1998), assim como inferir a ocorrência de bottlenecks ou de expansões populacionais (Goldstein & Schlötterer, 1999; Beebee & Rowe, 2001; Jehle et al., 2001; Jehle & Arntzen, 2002). Inicialmente, os microssatélites foram considerados

3

Capítulo 1

marcadores pouco informativos para estudos filogeográficos (Beaumont & Bruford, 1999) devido à ocorrência de homoplasia (Paetkau et al., 1997; Estoup et al., 2002). Contudo, vários trabalhos têm vindo a demonstrar a sua potencialidade para resolver filogenias a escalas temporais substancialmente profundas (Angers & Bernatchez, 1998). Desta forma, tem-se assistido ao aumento do emprego de microssatélites em estudos

de

genética

populacional,

bem

como

ao

desenvolvimento

de

novas

metodologias de análise, nomeadamente a aplicação de métodos de agrupamento Bayesianos (Pritchard et al., 2000). 1.1.1.3 As genealogias nucleares Os genes nucleares autossómicos são uma fonte importante no estudo dos processos históricos, demográficos e selectivos que determinam a arquitectura genética das espécies (Hare, 2001). Estes marcadores não evidenciam muitas das desvantagens acima nomeadas para o mtDNA, uma vez que revelam a história de machos e fêmeas e são menos susceptíveis a fenómenos estocásticos e de deriva genética, permitindo a percepção da história evolutiva das espécies a uma escala mais profunda. Além disso, o facto de recombinarem permite-nos inferir sobre o grau de miscigenação de grupos populacionais diferenciados, revelando-se úteis para o estudo de zonas híbridas (Antunes et al., 2002; Baird, 2006; Godinho et al., 2006). No entanto, a sua baixa taxa de mutação e o seu tamanho populacional efectivo elevado, responsáveis pela manutenção frequente de polimorfismo ancestral, bem como a ocorrência de recombinação, podem dificultar ou enviesar as inferências obtidas a partir destes marcadores (Zhang & Hewitt, 2003).

1.2 A Península Ibérica: um hotspot de biodiversidade Na Europa, as penínsulas do sul constituíram importantes refúgios durante as glaciações do Quaternário abrigando a maior parte das espécies que colonizam hoje o Norte e Centro (Hewitt, 1996; Taberlet & Cheddadi, 2002; Weiss & Ferrand, 2007). Neste

contexto,

a

Península

Ibérica

apresenta

diversas

características

que

favoreceram a sobrevivência a longo prazo de diferentes espécies, a ocorrência de uma enorme diversidade biológica e importantes níveis de endemismo, constituindo, como tal, um cenário ideal para o estudo da história evolutiva das espécies e suas populações, e uma área prioritária para a conservação da biodiversidade. Vários factores geo-morfológicos e ambientais contribuíram para a importância desta península como hotspot de biodiversidade, nomeadamente: i) a sua localização

4

Introdução Geral

geográfica no extremo Sudoeste da Europa, e separação do restante continente europeu pelos Pirinéus; ii) a relativa proximidade ao continente africano (o Estreito de Gibraltar tem apenas 14 Km de extensão); iii) a complexidade geográfica desta região, que apresenta várias cadeias montanhosas com orientação Este-Oeste, associada

a

uma

área

relativamente

grande

(580.000

Km2);

iv)

a

elevada

heterogeneidade climatérica, devido à sua diversidade fisiográfica e às influências climáticas do Atlântico e do Mediterrâneo; e v) uma história geológica e climática complexa (Tabela 1). A importância desta região como centro de diversidade genética e origem de endemismos reflecte-se também ao nível herpetofaunístico, tendo os processos de diferenciação em alopatria favorecido o aparecimento de numerosos taxa endémicos, tanto a nível específico como subespecífico (Barbadillo et al., 1997). Tabela 1 Principais eventos paleogeográficos que ocorreram na Península Ibérica. Evento paleogeográfico

Datação (Ma)

Referência

Formação de grandes lagos salinos no interior da península

ca. 16

Anadón et al., 1989

Isolamento do Maciço Bético-Rifenho

16-14

Weijermars, 1991

12-6

Weijermars, 1991

Fragmentação do Maciço Bético-Rifenho Crise Messiniense – Secagem do Mar Mediterrâneo

5,9-5,3

Abertura do Estreito de Gibraltar

5,33

Períodos glaciares do Pleistoceno

1,8-11.500

Duggen et al., 2003 Krijgsman et al., 1999 Krijgsman et al., 1999 Wilson et al., 1999

Vários estudos filogeográficos vieram demonstrar a importância da Península Ibérica como refúgio glaciar para inúmeras espécies animais e vegetais que actualmente se distribuem pelo Centro e Norte da Europa, e que sobreviveram aos ciclos glaciares nesta região (por exemplo, Taberlet et al., 1998; Hewitt 1999; Riberon et al., 2001; Michaux et al., 2003, 2005; Ribera & Vogler, 2004; Brito, 2005; Rowe et al. 2006). Muitos destes trabalhos, juntamente com numerosos estudos filogeográficos de espécies endémicas desta região, ao contrário de indicarem a Península Ibérica como um único refúgio durante os períodos glaciares do Pleistoceno, têm vindo a demonstrar existência de vários “refúgios” dentro do refúgio ibérico e uma elevada concordância nos actuais padrões de distribuição da diversidade genética de inúmeras espécies de fauna e flora (para uma revisão, ver Gómez & Lunt, 2007). Assim, a ocorrência de fenómenos vicariantes nesta região terá conduzido à diferenciação de populações, que embora geograficamente próximas, sobreviveram aos períodos

5

Capítulo 1

glaciares do Pleistoceno em refúgios isolados (por exemplo Branco et al., 2000, 2002; Alexandrino et al., 2000, 2002; Paulo et al., 2001; García-París et al., 2003; Geraldes et al., 2006; Martínez-Solano et al., 2006; Albert et al., 2007; Pinho et al., 2007).

1.3 Filogenia e história evolutiva do género Alytes O género Alytes Wagler 1829 deve o seu nome comum de “sapos-parteiros” ao facto de os machos apresentarem como característica particular o transporte dos ovos nas patas posteriores durante a incubação, libertando-os mais tarde em meio aquático, onde eclodem.

Figura 1 Distribuição do género Alytes e das subespécies descritas para Alytes obstetricans (adaptado de Fonseca, 1999; García-París & Martínez-Solano, 2001). A área com pontos de interrogação representa zonas onde a identificação das subespécies de A. obstetricans é problemática.

Este género engloba cinco espécies que se distribuem essencialmente pela região mediterrânica ocidental (Figura 1): Alytes (Alytes) obstetricans (Laurenti, 1768), amplamente distribuído pela Europa Ocidental, Alytes (Ammoryctis) cisternasii Boscá, 1879, endémico do Centro e Sudoeste da Península Ibérica, Alytes (Baleaphryne) muletensis (Sanchiz & Adrover, 1979), endémico da ilha de Maiorca (Baleares), Alytes (Baleaphryne) dickhilleni Arntzen & García-París, 1995, endémico do Sudeste da

6

Introdução Geral

Península Ibérica, e Alytes maurus Pasteur & Bons, 1962, que ocorre em isolados populacionais no Norte de África, em Marrocos (Bons & Geniez, 1996; Crespo, 1997; Grossenbacher, 1997; García-París & Arntzen 2002; Román, 2002). As diferenças genéticas e morfológicas existentes entre estas espécies deram origem à subdivisão do género em três grupos ou subgéneros: Alytes, Ammoryctis e Baleaphryne. No entanto, o estatuto sub-genérico de Baleaphryne continua em discussão, porque, se do ponto de vista filogenético se justifica designar desta forma o grupo A. maurus - A. muletensis - A. dickhilleni (Fromage et al., 2004), por outro lado, as afinidades morfológicas de A. dickhilleni e A. maurus com A. obstetricans, correspondentes a um morfotipo pouco especializado, muito distinto do morfotipo trepador característico de A. muletensis, já não apoiam a manutenção desta categoria. O género Alytes é um grupo bastante antigo que conserva numerosos caracteres primitivos e que divergiu dos restantes Discoglossídeos durante o Cretáceo, período para o qual existem registos fósseis na Ásia Central atribuídos a este grupo (Sanchiz, 1998). Os subgéneros Alytes e Ammoryctis, monofiléticos do ponto de vista morfológico e genético (Arntzen & García-París, 1995; Fromage et al., 2004) divergiram no Mioceno, há cerca de 10 a 16 Ma. De acordo com Arntzen & GarcíaParís (1995), o isolamento do Maciço Bético-Rifenho, há cerca de 14-16 Ma, teria sido o evento vicariante que conduziu à especiação em alopatria de A. cisternasii e um proto-A. obstetricans. A população continental ter-se-á diferenciado em A. cisternasii, que se adaptou aos solos arenosos do Sudoeste da Península Ibérica, enquanto que a população do Maciço Bético-Rifenho terá dado origem a um proto-A. obstetricans. Mais tarde, a fragmentação deste Maciço terá isolado A. obstetricans do ancestral de A. dickhilleni e A. muletensis. Ainda segundo Arntzen & García-París (1995), A. obstetricans terá colonizado a Península Ibérica durante a crise Messiniense, há cerca de 5,9–5,3 Ma, ficando o ancestral de A. dickhilleni e A. muletensis isolado nas montanhas Béticas. No final deste período, o enchimento da bacia do Mediterrâneo terá isolado as ilhas Baleares e conduzido à especiação de A. dickhilleni e A. muletensis. Pelo contrário, Altaba (1997) sugere que a divergência entre A. cisternasii e A. obstetricans terá ocorrido em consequência do isolamento causado pela formação de grandes lagos salinos no Centro da Península Ibérica. Posteriormente, o Maciço Bético-Rifenho teria sido ocupado pelo proto-A. obstetricans e a fragmentação deste Maciço causado o isolamento de A. muletensis. A divergência de A. obstetricans e A. dickhilleni estaria relacionada com a reabertura do estreito Bético. Um terceiro cenário biogeográfico, semelhante ao de Arntzen & García-París (1995), foi recentemente proposto por Fromage e colaboradores (2004). Contudo, ainda não foi proposta uma hipótese filogenética consensual para este género e a opção entre os vários cenários biogeográficos continua ambígua.

7

Capítulo 1

A

B

ca. 15-16 Ma

A. cisternasii

ca. 8-6 Ma

A. obstetricans (i)

Arco Bético

C

Ancestral dos restantes Alytes

(ii)

D

ca. 5,33 Ma

Ancestral de A. dickhilleni/ A. muletensis e A. maurus

ca. 4 Ma

A. cisternasii A. obstetricans

A. obstetricans (iii)

A. dickhilleni

A. dickhilleni

(iii)

A. muletensis

A. muletensis

A. maurus

A. maurus

Figura 2 Cenários paleogeográficos para a origem e colonização das diferentes espécies do género Alytes. As setas indicam as rotas de dispersão assumidas e as linhas a tracejado identificam os seguintes eventos vicariantes: (i) a abertura do estreito do mar Bético; (ii) fragmentação da região Bética; (iii) barreiras marinhas após o final da crise salina Messiniense (adaptado de Fromhage et al., 2004).

Vários estudos tentaram clarificar as relações filogenéticas entre as espécies do género Alytes e conduziram a diversas alterações taxonómicas, nomeadamente a modificação do estatuto de algumas formas e a descrição de novas espécies e subespécies. Clarke (1984) descreveu a osteologia de A. muletensis, inicialmente considerado um fóssil – Alytes (Baleaphryne) talaioticus (Sanchiz & Adrover, 1979; Alcover & Mayol, 1980; Sanchiz & Alcover, 1982) - e comparou-a com a das restantes espécies do género Alytes conhecidas na altura (A. cisternasii e A. obstetricans). Sanchiz (1984) e Maxson & Szymura (1984) propuseram hipóteses filogenéticas baseadas em caracteres morfológicos e dados imunológicos, respectivamente. Ambos os estudos mostraram a existência de uma diferenciação relativamente elevada entre A. cisternasii e as restantes espécies do género. O trabalho de Arntzen & García-París (1995), baseado no estudo electroforético de proteínas, revelou a existência de uma quarta espécie - A. dickhilleni - e de uma elevada subestruturação genética dentro de A. obstetricans, resultando no reconhecimento da subespécie A. o. almogavarii para a região

da

Catalunha.

Recentemente,

García-París

&

Martínez-Solano

(2001)

descreveram a subespécie A. o. pertinax com base no estudo morfológico, na electroforese de proteínas e sequenciação de mtDNA. Llorente e colaboradores (1995), baseados na análise biogeográfica do género Alytes na Península Ibérica, sugerem que a forma A. obstericans maurus poderia constituir uma nova espécie, critério este que foi de facto adoptado por Donaire-Barroso & Bogaerts (2003) e Fromhage e

8

Introdução Geral

colaboradores (2004). Estes resultados contrariam os obtidos por Arntzen & Szymura (1984) e Maxson & Szymura (1984), que tinham inicialmente descrito uma reduzida diferenciação genética em populações de Alytes de ambos os lados do estreito de Gibraltar. Do ponto de vista das relações filogenéticas entre as diferentes espécies do género, a análise aloenzimática realizada por Arntzen & García-París (1995) colocou A. dickhilleni como o grupo irmão de A. muletensis, e A. obstetricans como irmão do clado A. muletensis + A. dickhilleni (Figura 3A). Por outro lado, Altaba (1997) reanalisou os dados de Arntzen & García-París (1995) e encontrou suporte para uma relação irmã entre A. obstetricans e A. dickhilleni (ver, também, Arntzen & GarcíaParís, 1997). Recentemente, Fromhage e colaboradores (2004), com base na análise filogenética de sequências de mtDNA, consideraram A. maurus o grupo irmão de A. muletensis ou de A. muletensis + A. dickhilleni (Figura 3B).

A

B

Figura 3 Filogenia do género Alytes. A – Com base na análise filogenética de aloenzimas (adaptado de Arntzen & García-París, 1995). B – Com base na análise filogenética de sequências de mtDNA (adaptado de Fromage et al., 2004).

9

Capítulo 1

1.4 As espécies de estudo O sapo-parteiro-comum (A. obstetricans) é o representante do género com a distribuição

mais

ampla,

estando

presente

em

Portugal,

Espanha,

França,

Luxemburgo, Holanda, Bélgica, Alemanha e Suíça (Grossenbacher, 1997). Na Península

Ibérica

encontra-se

em

grande

parte

da

metade

setentrional,

nomeadamente nos sistemas montanhosos do Centro e no litoral oriental, ocorrendo desde do nível do mar até cerca dos 2400 m nos Pirinéus (Márquez & Rosa, 1997). Esta espécie ocupa uma grande diversidade de habitats, que vão desde zonas rochosas de montanha até campos agrícolas e, inclusivamente, áreas urbanas, encontrando-se associada a massas de água permanente que lhe possibilitam o prolongado desenvolvimento larvar (Arntzen & García-París, 1995). Numerosos estudos revelaram a existência de uma elevada diferenciação morfológica e genética das populações de A. obstetricans (Crespo, 1979; Crespo, 1982; Arntzen & Szymura, 1984; Viegas & Crespo, 1985; Rosa et al., 1990; Arntzen & García-París, 1995; García-París, 1995; García-París & Martínez-Solano, 2001; Fonseca et al., 2003; Fromage et al., 2004), o que resultou no reconhecimento de quatro subespécies (Figura 1): i) A. o. obstetricans (Laurenti, 1768), que inclui as populações da Europa Central, penetrando em Espanha pelo extremo Oeste dos Pirinéus e ao longo dos Montes Cantábricos; ii) A. o. boscai Lataste, 1879, presente no Norte e Centro de Portugal, Galiza, Oeste de Castilla-León e Sistema Central espanhol; iii) A. o. almogavarii Arntzen & García-París, 1995, distribuído desde os Pirinéus orientais, na vertente francesa (Geniez & Crochet, 2003), passando pelo vale do Rio Ebro e alcançando a Serra de Guadarrama, perto de Madrid, Espanha (García-París, 1995); e iv) A. o. pertinax García-París & Martínez-Solano, 2001, presente nas regiões do Centro e Este de Espanha. No entanto, os limites da distribuição geográfica destas subespécies

e

respectivas

relações

filogenéticas

ainda

não

se

encontram

completamente esclarecidos. De facto, Fonseca e colaboradores (2003), com base na análise do DNA mitocondrial, propõem que os limites de A. o. obstetricans se estendam até ao rio Douro. Os dados aloenzimáticos mostram a ocorrência de hibridação entre A. o. almogavarii e A. o. obstetricans, nos Pirinéus, e provavelmente em extensas áreas do Norte e Centro de Espanha (Arntzen & García-París, 1995; García-París, 1995).

10

Alytes obstetricans boscai

© Steven Busack

Alytes obstetricans almogavarii © Luis Garcia

Alytes obstetricans pertinax

© Jordi Baucells

© Víctor Císcar

Alytes obstetricans obstetricans

© José Teixeira

© Jan Van Der Voort

Introdução Geral

Alytes dickhilleni

Alytes muletensis

© Jan Van Der Voort

© Jeroen Speybroeck

Alytes maurus

Alytes cisternasii

Figura 4 Imagens das espécies e subsespécies de Alytes spp. estudadas.

11

Capítulo 1

O sapo-parteiro-meridional (A. cisternasii) distribui-se por grande parte das bacias hidrográficas do Guadiana, Tejo e Douro, desde o nível do mar até aos 1210 m de altitude, na vertente Norte do Sistema Central espanhol (García-París et al., 1990). O limite setentrional da sua distribuição abrange o Nordeste de Portugal e as províncias espanholas de Zamora, Salamanca, Ávila, Valladolid e Segóvia. O limite oriental é definido, de Norte a Sul, pelas províncias de Segóvia, Guadalajara, Madrid, Toledo, Ciudad Real, Jaén, Córdoba, Sevilha e Huelva (Márquez & Crespo, 2002). Esta espécie ocorre,

essencialmente,

em

regiões

pertencentes

ao

piso

bioclimático

mesomediterrânico, por norma em zonas com solos arenosos. A análise aloenzimática desta espécie indicou uma reduzida variação inter e intrapopulacional (DNei ≤0,02) e não foi detectada evidência para uma subestruturação geográfica (Arntzen & GarcíaParís, 1995). Este facto poderá ter resultado da persistência das populações de A. cisternasii num único refúgio mediterrânico durante o Pleistoceno seguida pela expansão recente e ocupação da área de distribuição actual, ou, alternativamente, a resolução dos marcadores utilizados não foi suficiente para revelar uma estrutura genética críptica sem reflexo morfológico. O clado Baleaphryne inclui as restantes três espécies de sapo-parteiro, todas descritas

muito

recentemente.

O

sapo-parteiro-balear

(A.

muletensis)

ocorre

unicamente na Serra Tramuntana, na região noroeste da ilha de Maiorca, e foi descoberto em 1980 depois de, anteriormente, se pensar que estava definitivamente extinto (Sanchiz & Adrover, 1979; Alcover & Mayol, 1980; Sanchiz & Alcover, 1982). Distingue-se claramente das suas congéneres pelo facto de mostrar um conjunto de adaptações que lhe permitem trepar facilmente paredes rochosas associadas às regiões de torrentes que habitam (Sanchiz, 1984). O sapo-parteiro-bético (A. dickhilleni) tem uma distribuição geográfica fragmentada e restrita às montanhas do sudeste ibérico, tendo sido descrita como uma espécie distinta apenas em 1995 (Arntzen & García-París, 1995). Finalmente, o sapo-parteiro-marroquino (A. maurus) ocorre em populações muito fragmentadas nas montanhas do Rif e ainda no Médio Atlas (Donaire-Barroso & Bogaerts, 2003). Durante bastante tempo, foi considerada uma subespécie de A. obstetricans, tendo-se mesmo sugerido a possibilidade de ter sido introduzida pelo Homem em tempos históricos. As suas características biológicas são pouco conhecidas.

1.5 Objectivos e organização temática O presente trabalho tem como objectivo reconstruir a história evolutiva dos sapos-parteiros (Alytes spp.) na Península Ibérica, procurando explicar os processos

12

Introdução Geral

micro e macro evolutivos que determinaram a diversidade específica e os padrões de diferenciação

actualmente

encontrados.

Pretende-se

investigar

os

fenómenos

históricos e contemporâneos que modelam a diversidade genética intra e interespecífica deste género na Península através da utilização de diferentes ferramentas moleculares, assim como discutir as suas implicações taxonómicas, e comparar os nossos resultados com os padrões observados em espécies co-distribuídas. Tendo por base toda a informação prévia existente sobre este género, foram definidos os seguintes objectivos específicos: i) Investigar as relações filogenéticas entre as várias espécies e subespécies de Alytes através da análise de genes mitocondriais e nucleares e reconstruir a história biogeográfica do género (capítulo 2); ii) Analisar os padrões geográficos de distribuição da diversidade genética do sapo-parteiro-comum (A. obstetricans) na Península Ibérica através da análise do polimorfismo de genes mitocondriais e nucleares (capítulo 3); iii) Estudar

a

variabilidade

genética,

estruturação

populacional

e

padrões

filogeográficos do sapo-parteiro-ibérico (A. cisternasii) em toda a sua área de distribuição através da sequenciação e análise de um gene mitocondrial e de microssatélites desenvolvidos especificamente para este efeito (capítulo 4). Os temas abordados neste trabalho são apresentados sob a forma de cinco artigos científicos, já publicados, submetidos ou em fase final de preparação, que pretendem dar resposta aos objectivos específicos anteriormente propostos. Assim, esta tese está organizada em seis capítulos, sendo o capítulo 1 uma introdução geral ao tema, que destaca a informação já existente e os objectivos específicos desta dissertação. Após a introdução, surgem os capítulos 2, 3 e 4, constituídos pelos artigos científicos produzidos e onde se expõem os resultados obtidos neste trabalho. O capítulo 5 consiste na discussão integrada dos principais resultados obtidos, nomeadamente a análise filogenética de Alytes spp., a definição de um novo cenário biogeográfico, a descrição e interpretação dos padrões filogeográficos de A. obstetricans e A. cisternasii, e ainda as suas implicações taxonómicas. Por último, no capítulo 6, tecemse algumas considerações finais sobre o trabalho realizado e sugerem-se perspectivas futuras de trabalho. O primeiro artigo (Artigo I, capítulo 2), publicado no Journal of Biogeography, consiste no desenvolvimento de uma hipótese filogenética robusta para o género

13

Capítulo 1

Alytes (Anura: Discoglossidae), com base na análise de sequências de DNA mitocondrial (citocromo b e 16S RNA) e de caracteres osteológicos, e na discussão das suas implicações para a reconstrução da história biogeográfica deste grupo. No segundo artigo (Artigo II, capítulo 2), publicado na revista Molecular Phylogenetics and Evolution, são apresentados novos dados genéticos obtidos para amostras de todas as espécies e subespécies reconhecidas de Alytes através da sequenciação de um gene nuclear (β-fibrinogénio), que já tinha demonstrado ser útil na inferência das relações filogenéticas noutras espécies de anfíbios. Os resultados obtidos são comparados com as hipóteses filogenéticas desenvolvidas através da análise de caracteres morfológicos e genes mitocondriais e mostra-se a ocorrência de um notável polimorfismo ancestral naquele gene nuclear. No terceiro artigo desta tese (Artigo III, em fase final de preparação, capítulo 3) pretendeu-se

estudar

os

padrões

filogeográficos

do

sapo-parteiro-comum

(A.

obstetricans) na Península Ibérica, com base em sequências de DNA mitocondrial e genealogias nucleares. A análise e interpretação destes dados permitiram reformular as hipóteses biogeográficas anteriormente colocadas e revelaram, ainda, o carácter único desta espécie por apresentar processos de dispersão pós-glacial muito contrastantes. Por último, os quarto e quinto artigos desta tese (capítulo 4) fazem uma análise da variabilidade genética e estruturação populacional do sapo-parteiro-ibérico (A. cisternasii) com base na análise do polimorfismo do DNA mitocondrial e de microssatélites. Neste sentido, foram desenvolvidos microssatélites específicos para A. cisternasii, tendo daqui resultado a publicação do Artigo IV na revista Molecular Ecology Notes. A utilização destes marcadores nucleares, juntamente com marcadores citoplasmáticos (DNA mitocondrial – ND4), permitiu analisar os padrões geográficos de diversidade e estruturação genética e formular as primeiras hipóteses sobre a história evolutiva de A. cisternasii. Estes resultados encontram-se incluídos no Artigo V, que se encontra submetido.

1.6 Lista dos trabalhos que integram a tese Devido à publicação de alguns dos trabalhos em revistas científicas internacionais com apresentações gráficas distintas, optou-se por realizar nesta dissertação uma uniformização da formatação do texto e respectivas imagens, sem qualquer alteração dos conteúdos. Os artigos publicados, submetidos, ou em fase final de preparação, em revistas internacionais da especialidade que integram esta tese são:

14

Introdução Geral

Artigo I. Martínez-Solano I, Gonçalves H, Arntzen JW, García-París M (2004) Phylogenetic relationships and biogeography of midwife toads (Discoglossidae: Alytes). Journal of Biogeography, 31, 603-618. Artigo II. Gonçalves H, Martínez-Solano I, Ferrand N, García-París M (2007) Conflicting phylogenetic signal of nuclear vs mitochondrial DNA markers in midwife toads (Anura, Discoglossidae, Alytes): deep coalescence or ancestral hybridization? Molecular Phylogenetics and Evolution, 44, 494-500. Artigo III. Gonçalves H, García-París M, Ferrand N (em preparação) Inferring ancient patterns of lineage diversification and extinction in the common midwife toad (Alytes obstetricans) based on mtDNA and nuclear genealogies. Artigo IV. Gonçalves H, Pereira R, Rowe G, Beebee T, Ferrand N (2005) Isolation and characterization of two dinucleotide and four tetranucleotide polymorphic microsatellite loci in the Iberian midwife toad Alytes cisternasii. Molecular Ecology Notes, 5, 767-769. Artigo V. Gonçalves H, Martínez-Solano I, Pereira R, García-París M, Ferrand N (submetido). Unexpected high levels of population subdivision in the Iberian Midwife Toad (Alytes cisternasii): evidence from mitochondrial and nuclear genomes.

1.7 Referências bibliográficas Albert E, Zardoya R, García-París M (2007) Phylogeographical and speciation patterns in subterranean worm lizards of the genus Blanus (Amphisbaenia: Blanidae). Molecular Ecology, 16, 1519-1531. Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic subdivision, glacial refugia and postglacial recolonization in the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela). Molecular Ecology, 9, 771-781. Alexandrino J, Arntzen JW, Ferrand N (2002) Nested clade analysis and the genetic evidence for population expansion in the phylogeography of the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela). Heredity, 88, 66–74. Alves PC, Ferrand N, Suchentrunk F, Harris DJ (2003) Ancient introgression of Lepus timidus mtDNA into L. granatensis and L. europaeus in the Iberian Peninsula. Molecular Phylogenetics and Evolution, 27, 70–80. Anadón P, Cabrera L, Roca E (1989) Contexto structural y paleogeográfico de los sistemas lacustres cenozoicos de España. Acta Geologica Hispanica, 24, 167-184.

15

Capítulo 1

Angers B, Bernatchez L (1998) Combined use of SMM and non-SMM methods to infer fine structure and evolutionary history of closely related brook charr (Salvelinus fontinalis, Salmonidae) populations from microsatellites. Molecular Biology and Evolution, 15, 143– 159. Arbogast BS, Kenagy GJ (2001) Comparative phylogeography as an integrative approach to historical biogeography. Journal of Biogeography, 28, 819-825. Alcover

JA, Mayol J (1980) Noticia del hallazgo de Baleaphryne (Amphibia: Discoglossidae) viviente en Mallorca. Doñana, Acta Vertebrata, 7, 266-269.

Anura:

Altaba C (1997) Phylogeny and biogeography of midwife toads: a reappraisal. Contributions to Zoology, 66, 257–262. Antunes A, Templeton A, Guyomard R, Alexandrino P (2002) The Role of Nuclear Genes in Intraspecific Evolutionary Inference: Genealogy of the transferrin Gene in the Brown Trout. Molecular Biology and Evolution, 19, 1272-1287. Arntzen JW, García-París M (1995) Morphological and allozyme studies of midwife toads (Genus Alytes), including the description of two new taxa from Spain. Contributions to Zoology, 65, 5–34. Arntzen JW, García-París M (1997) Phylogeny and biogeography of midwife toads (Alytes, Discoglossidae): a rebuttal. Contributions to Zoology, 66, 263–268. Arntzen JW, Szymura JM (1984) Genetic differentiation between African and European midwife toads (Alytes, Discoglossidae). Bijdragen tot de Dierkunde, 54, 157–162. Avise JC (2000) Phylogeography. The History and Formation of Species. Harvard University Press, Cambridge, Massachusetts. Avise JC (2004) Molecular Markers, Natural History and Evolution. 2nd edn. Sinauer Associates, Sunderland, Massachusetts. Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA, Saunders NC (1987) Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics, 18, 489–522. Avise JC, Walker D, Johns GC (1998) Speciation durations and Pleistocene effects on vertebrate phylogeography. Proceedings of the Royal Society of London B, Biological Sciences, 265, 1707-1712. Baird SJE (2006) Phylogenetics: Fisher’s markers of admixture. Heredity, 97, 81-83. Ballard JWO, Whitlock MC (2004) The incomplete natural history of the mitochondria. Molecular Ecology, 13, 729–744. Ballard JWO, Chernoff B, James AC (2002) Divergence of mitochondrial DNA is not corroborated by nuclear DNA, morphology, or behavior in Drosophila simulans. Evolution, 56, 527– 545. Barbadillo LJ, García-París M, Sanchiz B (1997) Orígenes y relaciones evolutivas de la herpetofauna iberica. Distribución y biogeografía de los anfibios y reptiles en España y Portugal (ed. by JM Pleguezuelos), pp. 47–100. Universidad de Granada, Asociación Herpetológica Española, Granada. Bazin E, Glémin S, Galtier N (2006) Population size does not influence mitochondrial genetic diversity in animals. Science, 312, 570-572. Beaumont MA, Bruford MW (1999) Microsatellites in conservation genetics. Microsatellites: Evolution and applications. (eds. DB Goldstein, C Schlötterer), pp. 165-182. Oxford University Press, New York. Beebee TJC, Rowe G (2001) Application of genetic bottleneck testing to the investigation of amphibian declines: a case study with natterjack toads. Conservation Biology, 15, 266270. Bernatchez L, Wilson CC (1998) Comparative phylogeography of Nearctic and Palearctic fishes. Molecular Ecology, 7, 431-452. Bons J, Geniez P (1996) Anfibios y Reptiles de Marruecos. Asociación Herpetológica Española, Barcelona.

16

Introdução Geral

Branco M, Ferrand N, Monnerot M (2000) Phylogeography of the European rabbit (Oryctolagus cuniculus) in the Iberian Peninsula inferred from RFLP analysis of the cytochrome b gene. Heredity, 85, 307–317. Branco M, Monnerot M, Ferrand N, Templeton AR (2002) Postglacial dispersal of the European rabbit (Oryctolagus cuniculus) on the Iberian Peninsula reconstructed from nested clade and mismatch analyses of mitochondrial DNA genetic variation. Evolution, 56, 792–803. Brito PH (2005) The influence of Pleistocene glacial refugia on tawny owl genetic diversity and phylogeography in western Europe. Molecular Ecology, 14, 3077–3094. Carstens BC, Brunsfeld SJ, Demboski JR, Good JM, Sullivan J (2005) Investigating the evolutionary history of the Pacific Northwest mesic forest ecosystem: hypothesis testing within a comparative phylogeographic framework. Evolution, 59, 1639-1652. Clarke BT (1984). General skeletal morphology. Historia Biológica del Ferreret (eds. H Hemmer, JA Alcover), pp. 45–59. Moll, Mallorca. Crespo EG (1979) Contribuição para o conhecimento da biologia dos Alytes ibéricos, Alytes obstetricans boscai Lataste, 1879 e Alytes cisternasii Boscá, 1879 (Amphibia, Salientia): A problemática da especiação de Alytes cisternasii. Tese de Doutoramento. Universidade de Lisboa, Portugal. Crespo EG (1982) Contribuição para o conhecimento da biologia das espécies ibéricas de Alytes, A. obstetricans boscai, Lataste 1879 e A. cisternasii, Boscá 1879 (Amphibia, Discoglossidae): morfologia dos adultos e dos girinos. Arquivos do Museu Bocage (C), 1, 255–316. Crespo EG (1997) Alytes cisternasii Boscá, 1879. Distribución y Biogeografía de los Anfibios y Reptiles en España y Portugal (ed. JM Pleguezuelos), pp. 126–128. Universidad de Granada, Asociación Herpetológica Española, Granada. Donaire-Barroso D, Bogaerts S (2003) Datos sobre taxonomia, ecologia y biológia de Alytes maurus (Pasteur & Bons, 1962) (Anura; Discoglossidae). Butlletí de la Societat Catalana d'Herpetologia, 16, 25-42. Duellman WE, Trueb L (1986) Biology of Amphibians. McGraw-Hill, New York. Duggen S, Hoernle K, den Bogaard P, Rüpke L, Morgan JP (2003) Deep roots of the Messinian salinity crisis. Nature, 422, 602–606. Edwards SV, Beerli P (2000) Perspective: gene divergence, population divergence, and the variance in coalescence time in phylogeographic studies. Evolution, 54, 1839–1854. Emerson BC, Hewitt GM (2005) Phylogeography. Current Biology, 15, 367-371. Estoup A, Jarne P, Cornuet JM (2002) Homoplasy and mutation model at microsatellite loci and their consequences for population genetics analysis. Molecular Ecology, 11, 1591-1604. Feldman CR, Spicer GS (2006) Comparative phylogeography of woodland reptiles in California: repeated patterns of cladogenesis and population expansion. Molecular Ecology, 15, 2201-2222. FitzSimmons N, Moritz C, Limpus CJ, Pope L, Prince R (1997) Geographic structure of mitochondrial and nuclear gene polymorphisms in Australian green turtle populations and male-biased gene flow. Genetics, 147, 1843–1854. Fonseca A (1999) Filogeografia do sapo parteiro comum Alytes obstetricans através da análise do polimorfismo do citocromo b (DNA mitocondrial). Tese de Mestrado. Universidade do Porto, Portugal. Fonseca A, Arntzen JW, Crespo EG, Ferrand N (2003) Regional differentiation in the common midwife toad (Alytes obstetricans) in Portugal: a picture from mitochondrial DNA. Zeitschrift für Feldherpetologie, 10, 83–89. Frankham R (2003) Genetics and conservation biology. C.R. Biologies, 326, S22–S29. Fromhage L, Vences M, Veith M (2004) Testing alternative vicariance scenarios in western Mediterranean discoglossid frogs. Molecular Phylogenetics and Evolution, 31, 308–322. García-París M (1995) Variabilidad genética y distribución geográfica de Alytes obstetricans almogavarii en España. Revista Española de Herpetología, 9, 133–138.

17

Capítulo 1

García-París M, Martínez-Solano I (2001) Nuevo estatus taxonómico para las poblaciones iberomediterráneas de Alytes obstetricans (Anura: Discoglossidae). Revista Española de Herpetología, 15, 99–113. García-París M, Arntzen JW (2002) Alytes dickhilleni Arntzen & García-París 1995. Atlas y Libro Rojo de los Anfibios y Reptiles de España (eds. JM Pleguezuelos, R Márquez, M Lizana), pp. 76–78. Dirección General de la Conservación de la Naturaleza, Asociación Herpetológica Española, Madrid. García-París M, Astudillo G, Prieto J, Márquez R (1990) Distribución de Alytes cisternasii Boscá, 1879, en el centro de la Península Ibérica. Revista Española de Herpetología, 4, 87-91. García-París M, Alcobendas M, Buckley D, Wake DB (2003) Dispersal of viviparity across contact zones in Iberian populations of fire salamanders (Salamandra) inferred from discordance of genetic and morphological traits. Evolution, 57, 129-143. Geniez P, Crochet, P-A (2003) Confirmation de l’existence, en France, de trois taxons méconnus: Alytes obstetricans almogavarii Arntzen & García-París, 1995 (Amphibia, Discoglossidae), Podarcis hispanica sebastiani (Klemmer, 1964) (Reptilia, Lacertidae) et Natrix natrix astreptophora (Seoane, 1884) (Reptilia, Colubridae). Bulletin de la Société Herpétologique de France, 105, 41-53. Geraldes A, Ferrand N, Nachman MW (2006) Contrasting patterns of introgression at X-linked loci across the hybrid zone between subspecies of the European rabbit (Oryctolagus cuniculus). Genetics, 173, 919–933. Godinho R, Mendonca B, Crespo EG, Ferrand N (2006) Genealogy of the nuclear β-fibrinogen locus in a highly structured lizard species: comparison with mtDNA and evidence for intragenic recombination in the hybrid zone. Heredity, 96, 1-10. Goldstein DB, Schlötterer C (1999) Microsatellites: Evolution and Applications. Oxford University Press, Oxford. Gómez A, Lunt DH (2007) Refugia within refugia: patterns of phylogeographic concordance in the Iberian Peninsula. Phylogeography in Southern European Refugia: Evolutionary Perspectives on the Origin and Conservation of European Biodiversity (eds. S Weiss, N Ferrand), pp. 155-188. Springer, Dordrecht, The Netherlands. Grossenbacher K (1997) Alytes obstetricans (Laurenti, 1768). Atlas of amphibians and reptiles in Europe (eds. J-P Gasc et al.), pp. 94–95. Societas Europaea Herpetologica, Paris. Hare MP (2001) Prospects for nuclear gene phylogeography. Trends in Ecology and Evolution, 16, 700-706. Hewitt GM (1996) Some genetic consequences of the Ice Ages and their role in divergence and speciation. Biological Journal of the Linnean Society, 58, 247–266. Hewitt GM (1999) Postglacial recolonization of the European biota. Biological Journal of the Linnean Society, 68, 87–612. Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature, 405, 907–913. Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary. Philosophical Transactions of the Royal Society of London B, Biological Sciences, 359, 183–195. Hey J, Machado CA (2003) The study of structured populations – new hope for a difficult and divided science. Nature Reviews Genetics, 4, 535–543. Hey J, Nielsen R (2004) Multilocus methods for estimating population sizes, migration rates and divergence time, with applications to the divergence of Drosophila pseudoobscura and D. persimilis. Genetics, 167, 747-760. Knowles LL, Maddison WP (2002) Statistical phylogeography. Molecular Ecology, 11, 2623– 2635. Krijgsman W, Hilgen FJ, Raffi I, Sierro FJ, Wilson DS (1999) Chronology, causes and progression of the Messinian salinity crisis. Nature, 400, 652-655. Jarne P, Lagoda PJL (1996) Microsatellites, from molecules to populations and back. Trends in Ecology and Evolution, 11, 424-429. Jehle R, Arntzen JW (2002). Review: microsatellite markers in amphibian conservation genetics. Herpetological Journal, 12, 1-9. 18

Introdução Geral

Jehle R, Arntzen JW, Burke T, Krupa AP, Hödl W (2001). The annual number of breeding adults and the effective population size of syntopic newts (Triturus cristatus, T. marmoratus). Molecular Ecology, 10, 839-850. Llorente GA, Montori A, Santos X, Carretero MA (1995) Atlas dels amfibis i rèptils de Catalunya i Andorra. El Brau, Figueres. Machado CA, Hey J (2003) The causes of phylogenetic conflict in a classic Drosophila species group. Proceedings of the Royal Society of London B, 270, 1193–1202. Márquez R, Rosa HD (1997) Alytes obstetricans (Laurenti, 1768). Distribución y Biogeografía de los Anfibios y Reptiles en España y Portugal (ed. JM Pleguezuelos), pp. 131–133. Universidad de Granada, Asociación Herpetológica Española, Granada. Márquez R, Crespo EG (2002) Alytes cisternasii Boscá, 1879. Atlas y Libro Rojo de los Anfibios y Reptiles de España (eds. JM Pleguezuelos, R Márquez, M Lizana), pp. 73-75. Dirección General de Conservación de la Naturaleza y Asociación Herpetológica Española, Madrid. Martínez-Solano I, Teixeira J, Buckley D, Garcia-París M (2006) Mitochondrial DNA phylogeography of Lissotriton boscai (Caudata, Salamandridae): evidence for old, multiple refugia in an Iberian endemic. Molecular Ecology, 13, 3375–3388. Maxson LR, Szymura JM (1984) Relationships among Discoglossid frogs: an albumin perspective. Amphibia–Reptilia, 5, 245–252. Melo-Ferreira J, Boursot P, Suchentrunk F, Ferrand N, Alves PC (2005) Invasion from the cold past: extensive introgression of mountain hare (Lepus timidus) mitochondrial DNA into three other hare species in northern Iberia. Molecular Ecology, 14, 2459-2464. Michaux JR, Magnanou E, Paradis E, Nieberding C, Libois R (2003) Mitochondrial phylogeography of the Woodmouse (Apodemus sylvaticus) in the Western Palearctic region. Molecular Ecology, 12, 685–697. Michaux JR, Libois R, Filippucci M-G (2005) So close and so different: comparative phylogeography of two small mammal species, the Yellow-necked fieldmouse (Apodemus flavicollis) and the Woodmouse (Apodemus sylvaticus) in the Western Palearctic region. Heredity, 94, 52–63. Monsen KJ, Blouin MS (2003) Genetic structure in a montane ranid frog: restricted gene-flow and nuclear-mitochondrial discordance. Molecular Ecology, 12, 3275–3286. Moore WS (1995) Inferring phylogenies from mtDNA variation: mitochondrial-gene trees versus nuclear-gene trees. Evolution, 49, 718–726. Moritz C, Faith DP (1998) Comparative phylogeography and the identification of genetically divergent areas for conservation. Molecular Ecology, 7, 419-429. Paetkau D, Waits LP, Clarkson PL, Craighead L, Strobeck C (1997) An empirical evaluation of genetic distance statistics using microsatellite data from bear (Ursidae) populations. Genetics, 147, 1943–1957. Pamilo P, Nei M (1988) Relationships between gene trees and species trees. Molecular Biology and Evolution, 5, 568-583. Paulo OS, Dias C, Bruford MW, Jordan WC, Nichols RA (2001) The persistence of Pliocene populations through the Pleistocene climatic cycles: evidence from the phylogeography of an Iberian lizard. Proceedings of the Royal Society of London B, Biological Sciences, 268, 1625–1630. Pinho C, Harris DJ, Ferrand N (2007) Contrasting patterns of population subdivision and historical demography in three western Mediterranean lizard species inferred from mitochondrial DNA variation. Molecular Ecology, 16, 1191-1205. Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics, 155, 945-959. Ribera I, Vogler AP (2004) Speciation of Iberian diving beetles in Pleistocene refugia (Coleoptera, Dysticidae). Molecular Ecology, 13, 179–193. Riberon A, Miaud C, Grossenbacher K, Taberlet P (2001) Phylogeography of the Alpine salamander, Salamandra atra (Salamandridae) and the influence of the Pleistocene climatic oscillations on population divergence. Molecular Ecology, 10, 2555-2560.

19

Capítulo 1

Román A (2002) Alytes muletensis (Sanchiz y Adrover, 1979). Atlas y Libro Rojo de los Anfibios y Reptiles de España (eds. JM Pleguezuelos, R Márquez, M Lizana), pp. 79–81. Dirección General de la Conservación de la Naturaleza, Asociación Herpetológica Española, Madrid. Rosa HD, Viegas AM, Crespo EG (1990) Genetic structure of Portuguese populations of midwife toads, with special reference to an isolate of Alytes obstetricans. Portugaliae Zoologica, 1, 15-25. Rowe G, Harris DJ, Beebee TJC (2006) Lusitania revisited: a phylogeographic analysis of the natterjack toad Bufo calamita across its entire biogeographical range. Molecular Phylogenetics and Evolution, 39, 335–346. Sanchiz B (1984) Análisis filogenético de la tribu Alytini (Anura, Discoglossidae) mediante el estudio de su morfoestructura ósea. Historia Biológica del Ferreret (eds. H Hemmer, JA Alcover), pp. 61–108. Moll, Mallorca. Sanchiz B (1998). Encyclopedia of Paleoherpetology, Part IV (Salientia). Friedrich Pfeil, München. Sanchiz B, Adrover R (1979) Anfibios fósiles del Pleistoceno de Mallorca. Doñana, Acta Vertebrata, 4, 5–25. Sanchiz B, Alcover JA (1982) Un nou discoglossid (Amphibia: Anura) de l’Holocé de Menorca. Bulleti de la Institució Catalana d’Historia Natural, 48, 99-105. Schneider CJS, Cunningham M, Moritz C (1998) Comparative phylogeography and the history of vertebrates in the wet tropics rainforests of Australia. Molecular Ecology, 7, 487-498. Shaw K (2002) Conflict between nuclear and mitochondrial DNA phylogeny of recent species radiation: what mtDNA reveals and conceals about modes of speciation in Hawaiian crickets. Proceedings of the National Academy of Sciences, USA, 99, 16122–16127. Smith MA, Green DM (2005) Dispersal and the metapopulation paradigm in amphibian ecology and conservation: are all amphibian populations metapopulations? Ecography, 28, 110– 128. Sullivan J, Arellano D, Rogers DS (2000) Comparative phylogeography of Mesoamerican Highlands Rodents: concerted versus independent response to past climatic fluctuations. American Naturalist, 155, 755-768. Taberlet P, Cheddadi R (2002) Quaternary refugia and persistence of biodiversity. Science, 297, 2009-2010. Taberlet P, Fumagalli L, Wust-Saucy AG, Cosson JF (1998) Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology, 7, 453-464. Tsaousis AD, Martin DP, Ladoukakis ED, Posada D, Zouros E (2005) Widespread recombination in published animal mtDNA sequences. Molecular Biology and Evolution, 22, 925–933. Viegas AM, Crespo EG (1985) Sur la structure génétique de deux “populations” allopatriques d’Alytes obstetricans boscai et d’Alytes cisternasii (Amphibia, Discoglossidae) du Portugal. Alytes, 4, 1-11. Weijermars R (1991) Geology and tectonics of the Betic zone, SE Spain. Earth-Science Reviews, 31, 153-236. Weiss S, Ferrand N (2007) Phylogeography in Southern European Refugia: Evolutionary Perspectives on the Origin and Conservation of European Biodiversity. Springer, Dordrecht, The Netherlands. Wilson RCL, Drury SA, Chapman JL (1999) The Great Ice Age. The Open University, London, New York. Zhang D-X, Hewitt GM (2003) Nuclear DNA analyses in genetic studies of populations: practice, problems and prospects. Molecular Ecology, 12, 563-584.

20

CAPÍTULO 2 Filogenia e evolução dos sapos-parteiros, Alytes spp. Artigo I. Martínez-Solano I, Gonçalves H, Arntzen JW, García-París M (2004) Phylogenetic relationships and biogeography of midwife toads (Discoglossidae: Alytes). Journal of Biogeography, 31, 603-618. Artigo II. Gonçalves H, Martínez-Solano I, Ferrand N, García-París M (2007) Conflicting phylogenetic signal of nuclear vs mitochondrial DNA markers in midwife toads (Anura, Discoglossidae, Alytes): deep coalescence or ancestral hybridization? Molecular Phylogenetics and Evolution, 44, 494-500.

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Journal of Biogeography, 31, 603-618

Phylogenetic relationships and biogeography of midwife toads (Discoglossidae: Alytes) I. MARTÍNEZ-SOLANO1, H. A. GONÇALVES1,2,3, J. W. ARNTZEN2,4 & M. GARCÍA-PARÍS1 1

Museo Nacional de Ciencias Naturales, CSIC, Madrid, Spain Centro de Investigação em Biodiversidade e Recursos Genéticos (CIBIO/UP), Campus Agrário de Vairão, Vairão, Portugal 3 Departamento de Zoologia e Antropologia da Faculdade de Ciências da Universidade do Porto, Praça Gomes Teixeira, Portugal 4 Department of Vertebrates, National Museum of Natural History, RA Leiden, The Netherlands

2

ABSTRACT Aim: To devise a robust phylogenetic hypothesis for midwife toads (Alytes: Anura: Discoglossidae) and to discuss its implications for the reconstruction of the biogeographical history of the group. Location: Western Palearctic. Methods: Analysis of sequences of mitochondrial DNA (cytochrome b and 16S RNA, 861 bp) and 29 characters of cranial osteology of all species and subspecies within Alytes. Results: Phylogenetic analyses support a sister group relationship between Alytes dickhilleni and A. muletensis, and of this clade with A. maurus. The monophyly of A. obstetricans is controversial; in particular, the phylogenetic position of A. obstetricans almogavarii is uncertain. The estimated dates for the cladogenetic events within Alytes are congruent with those derived from independent analyses (allozymes), except for the differentiation of A. o. almogavarii and the split between A. dickhilleni and A. muletensis. Main conclusions: The phylogeny based on the analysis of morphological and mtDNA data differs from previous hypotheses in the positions of A. o. almogavarii and A. maurus. Events associated with the radiation of Alytes are the formation of large inland saline lakes in Iberia c. 16 Mya and the 11 ºC dramatic decrease in average annual temperature during the Middle–Late Badenian transition c. 14–13.5 Mya (A. cisternasii vs. the ancestor of the remaining clades), the structuring of the NeoPyrenees and the reopening of the Betic Strait c. 10–8 Mya (A. o. almogavarii vs. A. obstetricans and vs. the subgenus Baleaphryne), the opening of the Gibraltar Strait at the end of the Messinian Salinity Crisis at c. 5.3 Mya (A. maurus vs. the ancestor of the A. dickhilleni A. muletensis clade) and a transmarine colonization event at c. 3 Mya (the ancestor of A. muletensis vs. A. dickhilleni). Following the new hypothesis, A. maurus, previously considered a subspecies of A. obstetricans, deserves species status. Second, A. o. almogavarii is a well-differentiated lineage that was isolated from other A. obstetricans more than c. 5 Mya, but later lost its genetic and specific identity following secondary contact, hybridization and introgression with the main stock. The presence of a marked morphological and genetic diversity within A. obstetricans renders reconstruction of the evolutionary history of the genus more complicated than previously appreciated. Keywords: Alytes, Amphibia, Anura, biogeography, evolution, mitochondrial DNA, osteology, phylogeny.

23

Capítulo 2

INTRODUCTION The genus Alytes Wagler 1829 (midwife toads) consists of five extant species distributed in three subgenera: Alytes (Alytes) obstetricans (Laurenti, 1768), widely distributed in western Europe, Alytes (Ammoryctis) cisternasii Boscá, 1879, endemic to the centre and southwest of the Iberian Peninsula, Alytes (Baleaphryne) muletensis (Sanchiz & Adrover, 1979) from the island of Mallorca in the Balearic archipelago, Alytes (Baleaphryne) dickhilleni Arntzen & García-París 1995, endemic to the southeastern Iberian Peninsula (Crespo, 1997; Grossenbacher, 1997; García-París & Arntzen 2002; Román, 2002) and Alytes maurus Pasteur & Bons, 1962, a taxon inhabiting the mountains in Morocco (Bons & Geniez, 1996). A. maurus, which was previously considered a subspecies of A. obstetricans, is now considered a different species. Alytes (Baleaphryne) talaioticus (Sanchiz & Alcover, 1982), described from the Holocene of Menorca (Balearic Islands), is now considered a synonym of A. muletensis (Sanchiz, 1998). Several studies have tried to clarify the phylogenetic relationships among the species of the genus, particularly in the 1980s after the discovery of A. muletensis. Clarke (1984) described the osteology of A. muletensis and compared it with the other species of Alytes known at that time (A. cisternasii and A. obstetricans). Sanchiz (1984) and Maxson & Szymura (1984) proposed phylogenetic hypotheses based on morphological characters of the skeleton and immunological data, respectively. Both studies agreed on proposing a sister-group relationship between A. cisternasii and the remaining taxa. Molecular techniques (allozyme electrophoresis) helped to reveal the existence of a fourth species, the Betic midwife toad A. dickhilleni (Arntzen & García-París, 1995). This study also revealed a high level of genetic substructuring within A. obstetricans, which resulted in the recognition of A. o. almogavarii Arntzen & García-París, 1995, distributed from southern France to north of the Ebro river (García-París, 1995). At present, other three subspecies are recognized: the nominotypical subspecies A. o. obstetricans (Laurenti, 1768), distributed across western Europe and the northern Iberian Peninsula (Navarra, País Vasco and Cantabrian Mountains); A. o. boscai Lataste, 1879, present in northern and central Portugal, Galicia, western Castilla-León as well as along the Sistema Central; and the recently described A. o. pertinax GarcíaParís & Martínez-Solano, 2001, present in the central and eastern regions of the Iberian Peninsula (García-París & Martínez-Solano, 2001). The protein electrophoretic analyses of Arntzen & García-París (1995) placed A. dickhilleni as the sister group of A. muletensis; Alytes obstetricans as sister of the clade A. muletensis + A. dickhilleni; and, finally, A. cisternasii as the sister taxon to all

24

Artigo I. Phylogenetic relationships and biogeography of midwife toads

other species of Alytes (Arntzen & García-París, 1995). The molecular clock applied to these data linked the divergence among lineages with several well-documented vicariant events and biogeographic barriers, such as the opening of the Gibraltar Strait (see, for example, Busack, 1977, 1986). Accordingly, Arntzen & García-París (1995) proposed a biogeographic scenario accounting for the successive cladogenetic events and presented a biogeographical hypothesis of Alytes in the Western Mediterranean. They proposed the formation of the Betic Strait c. 16 Mya as the vicariant event accounting for speciation in allopatry of A. cisternasii and proto-A. obstetricans. Later, the fragmentation of the BeticRiffean Massif isolated A. obstetricans from the ancestor of A. dickhilleni and A. muletensis in different islands. Alytes obstetricans would have then colonized the Iberian Peninsula through terrestrial connections during the Messinian Salinity Crisis. At the end of this period, the refilling of the Mediterranean Basin isolated the Balearic Islands and thus initiated the speciation of A. muletensis and A. dickhilleni. Altaba (1997) re-analysed the same dataset using a different coding scheme and found support for a sister-group relationship between A. obstetricans and A. dickhilleni (but see Arntzen & García-París, 1997). This finding led him to propose a different biogeographic scenario. According to this author, the speciation of A. cisternasii was triggered by the formation of large inland saline lakes in central Iberia at c. 16 Mya (Anadón et al., 1989). Later, the Sub-Betic Massif (including the Balearic Promontory) was occupied by proto-A. obstetricans. The fragmentation of this Massif caused the isolation of A. muletensis. The divergence of A. obstetricans and A. dickhillleni was placed in the Upper Tortonian at c. 8 Mya, in association with the reopening of the Betic Strait. A third hypothesis has recently been proposed by Fromhage et al. (2004). These authors used mtDNA sequences to construct a phylogenetic hypothesis of Alytes including A. maurus, which was recovered as sister to A. muletensis or to (A. muletensis + A. dickhilleni). Their biogeographic scenario is similar to that of Arntzen & García-París (1995), be it with consideration of A. maurus and without A. o. almogavarii, the most differentiated taxon within A. obstetricans. Hence, no comprehensive phylogenetic hypothesis of the genus has as yet been proposed and choosing between alternative biogeographic scenarios remained ambiguous. Both Altaba (1997) and Arntzen & García-París (1995) suggested ways to test the validity of the biogeographic scenarios that were derived from the same dataset (that is, allozyme data of Arntzen & García-París, 1995). For example, the osteology of the Betic midwife toad has not yet been described and, consequently, this information has not been included in previous phylogenetic analyses based on morphological characters (Sanchiz, 1984; Clarke, 1988). Besides, little is known about the evolutionary history of Alytes populations of the North African taxon A. maurus, which

25

Capítulo 2

was not included in the protein electrophoresis study of Arntzen & García-París (1995) and for which morphological information is scant. The works of Arntzen & Szymura (1984) and Maxson & Szymura (1984) showed a low level of genetic differentiation between Alytes populations from both sides of the Gibraltar Strait. However, if the phylogenetic hypothesis of Arntzen & García-París (1995) is correct, the degree of genetic differentiation of A. maurus should be at least equivalent to that found for A. muletensis and A. dickhilleni, a point supported by Fromhage et al. (2004). The fossil record was also suggested as an independent means of discriminating between both biogeographic hypotheses (Altaba, 1997). We have compiled new morphological and molecular data for all recognized species and subspecies of Alytes. We analysed 861 base pairs (bp) corresponding to partial sequences of the 16S rDNA (16S) and cytochrome b (cyt-b) mitochondrial genes. We also studied morphological variation in the cranial bones of all species and examined intraspecific variability in A. obstetricans in order to produce a robust hypothesis of the evolutionary history of the genus. As Arntzen & García-París (1997: 266) stated, ‘the best reason to reject any biogeographical reconstruction is the lack of support for the phylogeny on which it is based’. Consequently, only a robust phylogeny will provide an adequate background to rigorously test and choose among the possible alternatives.

MATERIAL AND METHODS Molecular data We obtained partial sequences of the mitochondrial genes 16S (c. 530 bp) and cyt-b (349–385 bp) for representatives of all species and subspecies of Alytes (Table 1). Discoglossus jeanneae Busack, 1986 was used as outgroup. Amplification and sequencing Tissue samples were obtained from specimens that were already used in the electrophoretic analyses of Arntzen & García-París (1995) (see Table 1), or from live animals that were released at their place of capture. Whole genomic DNA was extracted from small amounts of frozen or ethanol preserved tissue using NaCl following a protocol modified from Miller et al. (1988). We sequenced 530 bp of the large 16S subunit ribosomal mtDNA gene and 349–385 bp of the cytochrome b gene. Amplifications were done by the polymerase chain reaction (PCR) (Saiki et al., 1988), using the primers ‘MVZ15’ (Moritz et al., 1992) and ‘cyt b2’ (Kocher et al., 1989) for cyt-b, and the primers ‘16Sar’ and ‘16Sbr’ (Palumbi et al., 1991) for 16S. PCR

26

Artigo I. Phylogenetic relationships and biogeography of midwife toads

reactions were performed as indicated in García-París et al. (2003). The amplified fragments were sequenced in an automated DNA sequencer (ABI PRISM 3100) using the PCR primers and following the manufacturer’s instructions. GenBank accession numbers for sequences obtained are AY442019–AY442026 and AY514027–AY514035 for cytochrome b and AY442027–AY442036 for 16S. Sequence alignment and analyses All sequences were compiled using Sequence NavigatorTM version 1.0.1 (Applied Biosystems), and 16S sequences were manually aligned and adjusted by comparison with published models of secondary structure for 16S (Ortí & Meyer, 1997). Pairwise comparisons of observed proportional sequence divergences (p-distances) and transition—transversion ratios were calculated with

PAUP*4.0b10

(Swofford, 2002).

Table 1 Sample localities for the specimens used in the molecular study and GenBank accession numbers. MNCN: Museo Nacional de Ciencias Naturales, CSIC, Madrid, Spain. ----------- indicates sequence not obtained. GenBank Number (Cyt-b; 16S)

Taxa

Locality

Voucher

Alytes cisternasii

Cadalso de los Vidrios, Madrid, Spain

No voucher

AY442019; -----------

Alcaracejos, Córdoba, Spain

No voucher

------------; AY442027

Alytes dickhilleni

Sierra de Gádor, Almería

No voucher

AY442020; AY442028

Alytes maurus

Chefchaouen, Morocco

MNCN 40768

AY442021; AY442029

Ketama, Morocco

MNCN 20917

AY442022; AY442030

Alytes muletensis

Sierra de Tramuntana, Mallorca, Spain

No voucher

AF 128915; AY442031

A. obstetricans almogavarii

Berga, Barcelona, Spain

MNCN 16760

AY442023; AY442032

Ibón de Piedrafita, Huesca, Spain

MNCN 16755

AY514027; -----------

A. obstetricans boscai

A. obstetricans obstetricans

A. obstetricans pertinax

Discoglossus jeanneae

Peñalara, Madrid, Spain

No voucher

AY402024; AY442033

Vinuesa, Soria, Spain

MNCN 19901

AY514028; -----------

A Coruña, Spain

MNCN 20848

AY514029; -----------

Puerto de San Isidro, Asturias, Spain

MNCN 20838

AY402025; AY442034

Puebla de Lillo, León, Spain

MNCN 20919

AY514030; -----------

Bochum, Westfalia, Germany

MNCN 19893

AY514031; -----------

Valdelaguna, Madrid, Spain

No voucher

AY402026; AY442035

Benicassim, Castellón, Spain

MNCN 2003820039

AY514032–33; ----------

Ciruelas, Guadalajara, Spain

MNCN 20942

AY514034; -----------

Villanueva de la Torre, Guadalajara, Spain

No voucher

AY514035; -----------

Riópar, Albacete, Spain

MGP 2241

AF 128906; AY442036

27

Capítulo 2

Osteological data We studied 25 disarticulated skeletons and 13 cleared, doublestained specimens of all described species and subspecies of Alytes (Table 2). Most of these specimens were already used in the electrophoretic studies of Arntzen & García-París (1995), see material listed op. cit. Specimens of Discoglossus galganoi Capula, Nascetti, Lanza, Bulllini & Crespo, 1985 and D. jeanneae were used as outgroup. Variation in the following cranial bones was examined: premaxillar, nasal, frontoparietal, prooticexoccipital, maxillar, squamosal, pterygoid, sphenethmoid, vomer and parasphenoid (Fig. 1). The anatomical nomenclature is that of Roček (1980), Clarke (1984, 1988) and Sanchiz (1998). The material examined is deposited at the Museo Nacional de Ciencias Naturales (MNCN) (Spain). The definition of characters is presented in Appendix 1. Table 2 Sample localities for the specimens used in the morphological study, sample size (N) and voucher numbers. MNCN: Museo Nacional de Ciencias Naturales, CSIC, Madrid, Spain. Taxa

Locality

N

Voucher

Alytes cisternasii

San Agustín de Guadalix, Madrid, Spain

2

MNCN 15505-15506

Alytes dickhilleni

Alytes maurus

Valdemorillo, Madrid, Spain

1

MNCN 15515

Piñuécar, Madrid, Spain

1

MNCN 15517

Pelahustán, Toledo, Spain

1

MNCN 40396

Ventas de Zafarraya, Granada, Spain

2

MNCN 16757, 20950

Paterna del Madera, Albacete, Spain

4

MNCN 16782, 19881-19882, 19886

Chefchaouen, Morocco

1

MNCN 40768

Ketama, Morocco

1

MNCN 20917

Alytes muletensis

Sierra de Tramuntana, Mallorca, Spain

5

MNCN 15373, 15410-15413

A. obstetricans almogavarii

Berga, Barcelona, Spain

5

MNCN 16750, 16774-16777

Ibón de Piedrafita, Huesca, Spain

1

MNCN 20886

Tuy, Pontevedra, Spain

2

MNCN 19896, 19899

Venta del Obispo, Ávila, Spain

1

MNCN 20871

A Coruña, Spain

1

MNCN 20852

Rhür Basin, Germany

2

MNCN 19890-19891

Puerto de San Isidro, Asturias

2

MNCN 19857, 19863

Benicassim, Castellón, Spain

3

MNCN 16767, 16773, 20039

Caudiel, Castellón, Spain

1

MNCN 20865

Alfara de Algimia, Valencia Spain

2

MNCN 20882, 20885

Los Corrales, Huelva, Spain

1

MNCN 13877

El Berrueco, Madrid, Spain

1

MNCN 15130

A. obstetricans boscai

A. obstetricans obstetricans A. obstetricans pertinax

Discoglossus galganoi

Discoglossus jeanneae

28

Porto Covo, Baixo Alentejo, Portugal

1

MNCN 15136

San Agustín de Guadalix, Madrid, Spain

1

MNCN 15145

Artigo I. Phylogenetic relationships and biogeography of midwife toads

The geographic origin of the material used for the comparative analysis within A. obstetricans was selected as to confidently assign specimens to subspecies. This is relevant in view of the extended reticulation between subspecies along contact zones (García-París, 1995; Fonseca et al., 2003). We selected specimens of A. o. almogavarii from Berga (Barcelona), type locality of the subspecies, A. o. boscai from Tuy (Pontevedra), type locality of the subspecies (García-París & Martínez-Solano 2001; but see also Fonseca et al., 2003), and A. o. obstetricans from the Rühr Basin, in Germany. We added the only available specimens of A. o. pertinax from Caudiel and Benicassim, in eastern Spain. Specimens of A. dickhilleni were selected from the geographically most distant populations: Paterna del Madera (Albacete) and Ventas de Zafarraya (Granada), as preliminary data had suggested the presence of geographic variation.

Figure 1 Dorsal view of the skull of an adult specimen of Alytes dickhilleni, showing the localization of the bones mentioned in this study. Cartilage is shown in black and bone in grey (dark grey for bones located on the ventral plane). The right tympanic ring has been removed to better visualize the bones in that region.

Phylogenetic analyses Phylogenetic inference was based on maximum parsimony (MP), maximum likelihood (ML) and Bayesian analyses. MP phylogenies were estimated with PAUP*4.0b10

using the exhaustive search algorithm. We used non-parametric

bootstrapping (1000 pseudoreplicates) to assess the stability of internal branches in the cladograms (Felsenstein, 1985). Each base position was treated as an unordered character with four alternative states and gaps were treated as missing data. We used

MODELTEST

3.06 (Posada & Crandall, 1998) to find the best model of

molecular evolution that fit the data for subsequent ML analyses. The model of

29

Capítulo 2

evolution obtained when the outgroup is excluded is TrN + I. The plotting of corrected vs. uncorrected genetic distances suggested the existence of saturation when the outgroup was included in the analysis (Fig. 2). We also conducted a phylogenetic analysis on the basis of 29 morphological characters (see Appendices 2 and 3) with

PAUP*4.0b10,

using the exhaustive search

algorithm. All characters were considered unordered. Branch support was assessed with non-parametric boostrapping (1000 replicates) and decay indices (Bremer, 1994). Bayesian analyses were conducted with

MR.BAYES

3.0 (Huelsenbeck & Ronquist,

2001). Analyses were initiated with random starting trees and run for 100,000 generations. Bayesian analyses were carried out for molecular data (cyt-b + 16S) with and without the addition of morphological characters. We tested for substitution rate constancy between taxa using a molecular clock likelihood ratio test (Felsenstein, 1988) as implemented in

TREE-PUZZLE

5.0 (Strimmer

& von Haeseler, 1996). The null hypothesis of clocklike behaviour was rejected (delta = 32.91, d.f. = 8, P < 0.01). Consequently, in order to estimate the ages of the clades recovered in the phylogenetic analyses, branch lengths of an ML tree were transformed according to the non-parametric rate smoothing method (Sanderson, 1997). We used two different molecular clock calibrations for the estimation of transformed branch lengths: one based on immunological distances (IDU) and another one based on proteins. Maxson (1984) calculated a distance of 12 IDU between A. obstetricans and A. muletensis, which corresponds to 7.5–6.5 Mya of lineage separation (García-París & Jockusch, 1999). On the other hand, Arntzen & García-París (1995), based on allozymes, estimated the genetic distance between A. muletensis and A. dickhilleni to be 0.27–0.31 IDU, which, following the calibration of Beerli et al. (1996) of 0.08–0.10 DNei Mya-1, corresponds to 3.9–2.7 Mya of independent evolutionary history. Phylogeographic analyses Evolutionary relationships among populations within species as inferred from the analysis of gene genealogies are represented better by networks of interconnected haplotypes than by the bifurcating patterns usually recovered by methods of phylogenetic inference (Posada & Crandall, 2001). Consequently, in order to study intraspecific variation in A. obstetricans, we constructed a maximum parsimony haplotype network with the software TCS 1.13 (Clement et al., 2000) with cytochrome b sequences (385 bp) from 13 individuals representing all subspecies (Table 1).

30

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Figure 2 Scatter plot of corrected vs. uncorrected genetic distances in the cyt-b + 16S dataset. Above (a) with outgroup Discoglossus; below (b) without Discoglossus.

RESULTS Phylogenetic relationships Of the 861 nucleotide positions studied, 114 were variable and 79 of these were phylogenetically informative. Sequence divergence (uncorrected p-distances) within the ingroup ranged from 0.2% to 6.9% for 16S and from 0.29% to 15.5% for cyt-b (Table 3). Substitutions in the cyt-b sequences of the ingroup involved two aminoacid replacements. The divergence found between the ingroup taxa and the outgroup

31

Capítulo 2

(Discoglossus) ranged from 12.1% to 14.0% for 16S and from 20.3% to 24.0% for cyt-b. Changes in the cyt-b sequences involved seven aminoacid substitutions between the ingroup and the outgroup. The empirical ratio of transitions to transversions was 3.8:1 in the combined dataset. Table 3 Genetic distances (P-uncorrected) between ingroup taxa (above diagonal: 16S; below diagonal: cytochrome b). Taxon

1

2

3

4

5

6

7

8

9

1. Alytes cisternasii

-

0.0692

0.0672

0.0673

0.0672

0.0573

0.0613

0.0632

0.0593

2. Alytes dickhilleni

0.1547

-

0.0275

0.0275

0.0256

0.0314

0.0315

0.0334

0.0295

3. Alytes maurus (Chefchaouen)

0.1519

0.0774

-

0.0040

0.0197

0.0197

0.0197

0.0217

0.0177

4. Alytes maurus (Ketama)

0.1519

0.0802

0.0029

-

0.0197

0.0197

0.0197

0.0217

0.0178

5. Alytes muletensis

0.1547

0.0860

0.0774

0.0802

-

0.0276

0.0276

0.0296

0.0256

6. A. o. almogavarii

0.1203

0.0917

0.0630

0.0659

0.0831

-

0.0118

0.0138

0.0099

7. A. o. boscai

0.1289

0.0946

0.0659

0.0688

0.0917

0.0201

-

0.0020

0.0020

8. A. o. obstetricans

0.1232

0.0946

0.0659

0.0688

0.0917

0.0201

0.0057

-

0.0040

9. A. o. pertinax

0.1318

0.0974

0.0745

0.0774

0.1003

0.0286

0.0086

0.0143

-

An exhaustive maximum parsimony analysis produced two equally parsimonious trees [286 steps; consistency index (CI) = 0.846, retention index (RI) = 0.679]. The consensus tree is shown in Fig. 3. Alytes cisternasii is the sister taxon to the remaining species of Alytes, rendering a monotypic Ammoryctis. Alytes dickhilleni and A. muletensis form a sister group and A. maurus is sister to A. dickhilleni and A. muletensis, constituting the Baleaphryne clade. Alytes obstetricans is a monophyletic group, conforming the subgenus Alytes, with A. o. almogavarii sister to the other taxa. The topologies resulting from the ML non-parametric bootstrap analysis (best tree ln L = 2465.69) without enforcing a molecular clock and the majority rule consensus tree from the Bayesian analysis differ in the position of A. o. obstetricans relative to the other subspecies of A. obstetricans. The topology obtained in the MP consensus tree differs from these in the position of A. o. almogavarii only (Fig. 4). ML and Bayesian analyses recover three clades forming a polytomy (A. o. almogavarii, the clade formed by the other subspecies of A. obstetricans and the clade formed by A. maurus, A. dickhilleni and A. muletensis) to which A. cisternasii is the sister group; while MP results in A. o. almogavarii sister to the other subspecies of A. obstetricans.

32

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Figure 3 Phylogenetic relationships within Alytes. Strict consensus of two equally parsimonious trees (L = 286 steps; consistency index (CI) = 0.846, retention index (RI) = 0.679), constructed for the combined cyt-b and 16S data set. Seventy-nine characters were parsimony informative. Non-parametric bootstrap values based on 1000 pseudoreplicates using the ‘branch and bound’ algorithm are shown above branches.

Figure 4 Bootstrap consensus tree recovered by ML and Bayesian analyses for the phylogenetic relationships within Alytes. Figures at nodes indicate bootstrap support in the ML analyses (above node) and posterior probabilities in the Bayesian analysis of molecular data (below nodes, left) and combined molecular and morphological dataset (below nodes, right).

33

Capítulo 2

Specific differences in cranial osteology are described in Appendix 2. Figures 5 and 6 show the cleared and doublestained skulls of all Alytes taxa in dorsal view. Numbers in parentheses indicate character number in the matrix of morphological characters (presented in Appendix 3). The phylogenetic analyses of morphological characters produced 12 equally parsimonious trees (38 steps, CI = 0.816, RI = 0.611, Fig. 7). Decay indices (not shown) and bootstrap values indicate weak support for the few branches retained in the strict consensus tree. Alytes cisternasii is recovered as the sister group to the rest of the species of Alytes, among which there is a basal polytomy formed by A. o. boscai, A. o. obstetricans, A. o. pertinax and a clade containing A. o almogavarii, A. dickhilleni, A. muletensis and A. maurus. Within this clade, relationships are unresolved, except that A. muletensis and A. maurus are sister taxa. The topology recovered in the Bayesian analysis of the combined (molecular and morphological) dataset is identical to that obtained with the molecular data set exclusively, with, however, lower posterior clade probability values for some clades, notably for the sister-group relationship of A. muletensis and A. dickhilleni (Fig. 4).

Figure 5 Dorsal views of double-stained, cleared skulls of specimens of all species of Alytes: A. muletensis (MNCN 15373), A. cisternasii, (MNCN 40396), A. maurus (MNCN 20917), A. dickhilleni (MNCN 19886) and A. obstetricans (MNCN 16750). Scale equals 5 mm.

Figure 8 shows an ultrametric tree constructed with corrected ML branch lengths, with divergence times according to both calibrations shown at the nodes. Following these estimates, the earliest split within Alytes took place between A. cisternasii and the other species, between 18.5 and 12.8 Mya. The separation of A. obstetricans and the clade A. maurus + A. dickhilleni + A. muletensis took place 9.0–6.2 Mya. Alytes

34

Artigo I. Phylogenetic relationships and biogeography of midwife toads

obstetricans almogavarii diverged from the other A. obstetricans subspecies 7.5–5.2 Mya, while divergence among A. o. pertinax, A. o. boscai and A. o. obstetricans took place between 3.1 and 2.1 Mya. The split between the North African A. maurus and the clade A. muletensis + A. dickhilleni took place 5.6–3.8 Mya, before the divergence between the Balearic A. muletensis and the Betic A. dickhilleni which occurred 3.9–2.7 Mya.

Figure 6 Dorsal views of double-stained, cleared skulls of specimens of all subspecies of Alytes obstetricans: A. o. boscai (MNCN 20871), A. o. obstetricans (19863); A. o. pertinax (MNCN 20885) and A. o. almogavarii (MNCN 16750). Scale equals 5 mm.

Figure 7 Strict consensus of 12 equally parsimonious trees based on 29 osteological characters in Alytes. Solid and open boxes indicate non-ambiguous and ambiguous change state changes, respectively (numbered as in Appendix 1).

35

Capítulo 2

Figure 8 Ultrametric tree constructed with corrected ML branch lengths following the method of Sanderson (1997). Divergence time estimates shown at the nodes were calculated from two different calibrations: one based on immunological distances (Maxson, 1984, below), and other one based on proteins (Beerli et al., 1996, above). Postulated vicariant events are shown at nodes.

Phylogeographic analysis We observed 10 distinct haplotypes among 13 sequences across all four subspecies of A. obstetricans. This produced two independent networks: one comprising the one sample of A. o. almogavarii from the province of Huesca; and the other connecting the other haplotypes with no more than eight mutational steps. Within the latter, four groups were established separated by one non-observed haplotype encompassing (Fig. 9): (1) Germany and the Cantabrian mountains; (2) Galicia, Sierra de Guadarrama (Sistema Central, Madrid) and Soria (Sistema Ibérico); (3) Castellón, Guadalajara and southeastern Madrid; and (5) Barcelona.

36

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Figure 9 Haplotype network of cyt-b sequences of subspecies within A. obstetricans. The circle represents a non-observed haplotype that connects haplotypes observed in A. o. obstetricans (A. o. o.), A. o. boscai (A. o. b.), A. o. pertinax (A. o. p.) and A. o. almogavarii from Barcelona (A. o. a.). Transverse bars indicate mutational steps. Dashed lines indicate the geographic location of an independent haplotype of A. o. almogavarii from Huesca.

DISCUSSION Phylogenetic relationships within Alytes Our results support the sister group relationship between A. muletensis and A. dickhilleni proposed by Arntzen & García-París (1995) and Fromhage et al. (2004). A sister group relationship between A. obstetricans and A. dickhilleni (Altaba, 1997) is not supported. According to the combined morphological and molecular dataset, A. maurus is sister to A. dickhilleni and A. muletensis and not to A. obstetricans (see Arntzen & Szymura, 1984; Maxson & Szymura, 1984). This clarifies some of the problems of the biogeographic scenario identified by Arntzen & García-París (1995) (see below). A third, independent type of evidence, the fossil record, might also support this hypothesis, if the presence of Baleaphryne in the Early Pleistocene (‘Villafranchian’) of Morocco was to be confirmed (Hossini, 2001).

37

Capítulo 2

The phylogenetic position of A. o. almogavarii remains uncertain. MP analyses include this taxon within A. obstetricans. ML and Bayesian analyses, on the other hand, recover a polytomy of three clades: (1) Baleaphryne, including A. maurus + A. muletensis + A. dickhilleni; (2) A. o. almogavarii; and (3) the remaining subspecies of A. obstetricans (A. o. boscai, A. o. obstetricans and A. o. pertinax). Both analyses suggest that A. o. almogavarii might have had a long independent evolutionary history. In fact, this taxon is well differentiated from other subspecies at the allozyme level, presenting several private alleles (Arntzen & García-París, 1995). However, large areas of hybridization and a high degree of gene flow between A. o. almogavarii and both A. o. obstetricans and A. o. pertinax have also been reported (García-París, 1995). The combination of allozyme, mtDNA and morphological data (with some character states resembling those observed in A. muletensis) strongly suggest that A. o. almogavarii might have been a well-differentiated lineage that progressively lost its genetic and specific identity through extensive hybridization and introgression following secondary contact among previously isolated lineages. From a taxonomic and phylogenetic perspective, our data confirm the species status for the Moroccan populations, a position already taken by Llorente et al. (1995) and Fromhage et al. (2004). Therefore, this taxon should be referred to as: Alytes (Baleaphryne) maurus Pasteur & Bons, 1962 stat. nov. Our data are congruent with the recognition of three subgenera: Alytes, only including A. (A.) obstetricans, Ammoryctis, composed of A. (A.) cisternasii, and Baleaphryne comprising three species A. (B.) dickhilleni, A. (B.) maurus, and A. (B.) muletensis.

Morphological evolution in Alytes Alytes cisternasii and A. muletensis are so distinctive morphologically that their discovery led to the erection of new genera for each of them: Ammoryctis Lataste, 1879 and Baleaphryne Sanchiz & Adrover, 1979, respectively. Their morphological differentiation has been related to functional adaptations (Crespo, 1982; Sanchiz, 1984). Accordingly, A. cisternasii is a fossorial species, specialized in using its forearms for burrowing, whereas A. muletensis has a locomotive morphotype associated with climbing. These adaptations are reflected in the robust and ossified skeleton of A. cisternasii, including markedly ossified frontoparietals with reduced dorsal fontanelles and well-developed lateral rami of the prootic (Clarke, 1984, 1988). In A. muletensis, the skeleton is less ossified than in the other species of Alytes, and the skull is the least ossified within the genus. The frontoparietals have a wide, not subdivided dorsal fontanelle, although the medial processes of the frontoparietals

38

Artigo I. Phylogenetic relationships and biogeography of midwife toads

suggest an anterior–posterior division. Besides, the paired ossification nuclei that form the sphenethmoid do not fuse completely in adult specimens, a unique feature in Alytes (although the condition found in A. maurus is very similar to that of A. muletensis). This has been related to heterochronic processes in the A. muletensis lineage, by which development would have been truncated in its initial stages (Clarke, 1984, 1988; Sanchiz, 1984). However, the distribution of some character states is not in line with this interpretation, including the high counts of maxillary and premaxillary teeth in some specimens of A. o. almogavarii and A. muletensis. The number of maxillary teeth generally increases in the course of development of Discoglossidae, if not in most Anura (Clarke, 1988). The uncertain phylogenetic position of A. o. almogavarii has implications for hypotheses on the ancestral morphology of Alytes. Sanchiz (1984) considered the ‘generalist’ morphotype of A. obstetricans as the ancestral condition within the genus. Alternatively, the ancestral condition in Alytes may have been a mixture of ‘generalized’ and ‘specialized’, such as exhibited by extant A. o. almogavarii. Such polymorphisms might have represented an important source for morphological specialization that was later exploited by some taxa within the Baleaphryne clade, A. muletensis in particular.

Biogeography Our estimates for divergence times between lineages are largely congruent with those proposed by Arntzen & García-París (1995, see their Fig. 5), with two exceptions: A. o. almogavarii A. obstetricans (c. 3 vs. c. 6 Mya) and A. muletensis A. dickhilleni (c. 4.5 vs. 3 Mya). These discordances might be caused by the absence of A. maurus in the protein electrophoresis study, by problems with the calibration of the molecular

clock,

or,

in

the

case

of

A.

o.

almogavarii,

by

differential

cytoplasmic/nuclear introgression among lineages, which would render the use of a molecular clock unreliable. At any rate, some of the geological events proposed by these authors do not fit into the time frame associated with the new phylogeny. This, and the differences between our and previous phylogenetic hypotheses (Arntzen & García-París, 1995; Altaba, 1997; Fromhage et al., 2004) requires the proposal of a new biogeographic scenario. In the Lower Miocene, the faunistic exchange between Asia, Europe and Africa may have been widespread (Barbadillo et al., 1997) and the ancestral stock of Alytes may, in this epoch, have settled in Iberia (Fig. 10A), although it might also have been endemic to this area. A settlement of the proto-Alytes in the Betic–Riffean Massif, as postulated by all previous hypotheses, is not possible since the massif only just

39

Capítulo 2

started to emerge in this period (Weijermars, 1991). The vicariant event that promoted the split between the A. cisternasii clade and the remaining stock of Alytes might have been the formation of large inland saline lakes in central Iberia c. 16 Mya (Anadón et al., 1989), as suggested by Altaba (1997). Alternatively, it could be correlated with dramatic climatic changes in the Middle–Late Badenian transition (Böhme, 2003). The ancestor of Alytes cisternasii would have remained restricted to the arid sandy soils of the southwestern Iberian Peninsula, whereas the ancestor of all other Alytes extended over the rest of Iberia (Fig. 10B). Later, this stock expanded and entered the BeticRiffean Massif that emerged c. 14 Mya (Weijermars, 1991) (Fig. 10C). Altaba (1997) suggested that at this time, the ancestor of A. muletensis could have reached the Balearic Promontory through a postulated terrestrial connection, however, none of the proposed phylogenies support this hypothesis. The final structuring of the Neo-Pyrenees (Oosterbroek & Arntzen, 1992) and the reopening of the Betic Strait after the marine transgression of the Upper Tortonian that isolated the Sub-Betic region (López Martínez, 1989) fragmented Alytes populations into three lineages: the ancestor of A. o. almogavarii north of the Pyrenees, proto-A. obstetricans in Iberia and the ancestor of Baleaphryne (A. dickhilleni, A. maurus and A. muletensis) in the Sub-Betic region (Fig. 10D). The divergence of the lineage leading to A. o. almogavarii from the A. obstetricans lineage is thus older than previously estimated and it is almost contemporary to the separation of A. obstetricans from the Riffean–Betic–Balearic clade. This, together with the unresolved position of A. o. almogavarii in our phylogenetic hypothesis, allows rejection of the hypothesis of an early differentiation of A. obstetricans in the Betic–Riffean Massif (Altaba, 1997). This Massif remained occupied by the ancestors to the Baleaphryne clade: A. maurus, A. muletensis and A. dickhilleni exclusively. Although the fragmentation of the BeticRiffean Massif at 8–6 Mya might have initiated the separation of the lineages leading to A. maurus and the ancestor of A. muletensis and A. dickhilleni, the split of these two lineages is also consistent with the opening of the Strait of Gibraltar at the end of the Messinian Salinity Crisis 5.3 Mya (Krijgsman et al., 1999) (Fig. 10E). Alytes muletensis and A. dickhilleni diverged more recently, at c. 3 Mya, long after the formation of the Balearic Islands. This suggests that the ancestor of A. muletensis has reached the Balearic Islands through a transmarine colonization process (Fig. 10F). Events of long-distance colonization, though uncommon in amphibians, have been invoked to explain other biogeographic patterns (see, for example, Feller & Hedges, 1998; Vences et al., 2003). The differentiation within A. obstetricans may be related to the formation of the main fluvial drainages in the Iberian Peninsula (Arntzen & García-París, 1995) or, alternatively, be associated with the glacial refugia of Pleistocene (Fig. 9).

40

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Figure 10 Historical biogeography of the genus Alytes. Note that for lineages subject to cladogenesis a grey shading is used instead of the black one. (A) The ancestor of extant species of Alytes occupied Iberia. (B) The formation of large inland saline lakes in central Iberia or drastic climatic changes promoted the differentiation of the A. cisternasii (Ac) clade vs. the remaining Alytes stock. (C) The latter taxon expanded its range, including the Sub-Betic Massif. (D) The structuring of the NeoPyrenees and the reopening of the Betic Strait after the marine transgression of the Upper Tortonian promoted the divergence of the A. o. almogavarii (Aoa) clade and Baleaphryne, respectively, vs. the ancestor of A. obstetricans (Ao). (E) The opening of the Gibraltar Strait isolated the ancestors of A. maurus (Ama) and [A. dickhilleni (Ad) plus A. muletensis (Amu)] on opposite shores of the incipient Mediterranean Sea. The ancestor of A. muletensis reached the Balearic Islands. (F) Subspecies of A. obstetricans differentiate and expand their ranges. Alytes o. obstetricans contacts and deeply introgress A. o. almogavarii which loses its nuclear identity.

41

Capítulo 2

The singular evolutionary history of Alytes is characterized by the existence of high levels of morphological and genetic diversity within some lineages (A. obstetricans) but not in others (A. cisternasii, A. muletensis). The reconstructed spatio-temporal associations render the reconstruction of the evolutionary history of the genus more complicated than previously appreciated. As for A. obstetricans, discordances between nuclear (García-París, 1995) and mitochondrial markers (this study) in the populations of northern Guadalajara and Sistema Ibérico suggest the existence of a past, wide intermixing area in the northern Iberian subplateau involving populations of A. o. almogavarii, A. o. boscai and A. o. pertinax. Further studies focusing on the geographic distribution of genetic variability within A. obstetricans and the phylogenetic position of A. o. almogavarii are required for testing hypotheses of fragmentation and subsequent reticulation within these lineages (García-París, 1995; Fonseca et al., 2003).

ACKNOWLEDGMENTS We thank AD Buscalioni and B Sanchiz for constructive comments on earlier drafts of the manuscript, A Antúnez, J Bosch, P Galán, A Gosá, C Grande, R Márquez, B Thiesmeier and M Vences for help in collecting or sending material, B Sanchiz and the Department of Photography of the MNCN for help with the figures of osteological variation in the skulls of Alytes taxa, and JE González for curatorial assistance. We especially thank DB Wake for support and availability of the DNA lab of the Museum of Vertebrate Zoology at Berkeley. We acknowledge the ‘Consejerías de Medio Ambiente’ of the Baleares (special thanks to Álvaro Román) and Madrid, Spain, which provided the permits to collect the biological material used in this work. The DNA research has been partially funded by projects 07M/0109/2000 and 07M/0900/2002 of the Comunidad de Madrid, Spain. The work of IMS is funded by a MNCN-CSIC-CAM predoctoral fellowship and HAG is supported by a PhD grant (SFRH/BD/3375/2000) from FCT (Fundação para a Ciência e a Tecnologia, Portugal).

REFERENCES Altaba C (1997) Phylogeny and biogeography of midwife toads: a reappraisal. Contributions to Zoology, 66, 257–262. Anadón P, Cabrera L, Roca E (1989) Contexto structural y paleogeográfico de los sistemas lacustres cenozoicos de España. Acta Geologica Hispanica, 24, 167–184.

42

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Arntzen JW, García-París M (1995) Morphological and allozyme studies of midwife toads (Genus Alytes), including the description of two new taxa from Spain. Contributions to Zoology, 65, 5–34. Arntzen JW, García-París M (1997) Phylogeny and biogeography of midwife toads (Alytes, Discoglossidae): a rebuttal. Contributions to Zoology, 66, 263–268. Arntzen JW, Szymura JM (1984) Genetic differentiation between African and European midwife toads (Alytes, Discoglossidae). Bijdragen tot de Dierkunde, 54, 157–162. Barbadillo LJ, García-París M, Sanchiz B (1997) Orígenes y relaciones evolutivas de la herpetofauna iberica. Distribución y biogeografía de los anfibios y reptiles en España y Portugal (ed. by JM Pleguezuelos), pp. 47-100. Universidad de Granada, Asociación Herpetológica Española, Granada. Beerli P, Hotz H, Uzzell T (1996) Geologically dated sea barriers calibrate a protein clock for Aegean water frogs. Evolution, 50, 1676–1687. Böhme M (2003) The Miocene Climatic Optimum: evidence from ectothermic vertebrates of Central Europe. Palaeogeography, Palaeoclimatology, Palaeoecology, 195, 389–401. Bons J, Geniez P (1996) Anfibios y Reptiles de Marruecos. Asociación Herpetológica Española, Barcelona. Boulenger GA (1897) The tailless Batrachians of Europe: Part I. Ray Society, London. Bremer K (1994) Branch support and tree stability. Cladistics, 10, 295–304. Busack SD (1977) Zoogeography of amphibians and reptiles in Cádiz province, Spain. Annals of the Carnegie Museum, 46, 285–316. Busack SD (1986) Biogeographic analysis of the herpetofauna separated by the formation of the Strait of Gibraltar. National Geographic Research, 2, 17–36. Clarke BT (1984) General skeletal morphology. Historia Biológica del Ferreret (ed. by H Hemmer and JA Alcover), pp. 45–59. Moll, Mallorca. Clarke BT (1988) Evolutionary relationships of the Discoglossoid frogs: osteological evidence. PhD dissertation, London Polytechnic, London. Clement MD, Posada D, Crandall KA (2000) TCS: a computer program to estimate gene genealogies. Molecular Ecology, 9, 1657-1660. Crespo EG (1982) Contribuição para o conhecimento da biologia das espécies ibéricas de Alytes, A. obstetricans boscai, Lataste 1879 e A. cisternasii, Boscá 1879 (Amphibia, Discoglossidae): morfologia dos adultos e dos girinos. Arquivos do Museu Bocage (C), 1, 255–316. Crespo EG (1997) Alytes cisternasii Boscá, 1879. Distribución y Biogeografía de los Anfibios y Reptiles en España y Portugal (ed. by JM Pleguezuelos), pp. 126–128. Universidad de Granada, Asociación Herpetológica Española, Granada. Feller AE, Hedges SB (1998) Molecular evidence for the early history of living amphibians. Molecular Phylogenetics and Evolution, 9, 509–516. Felsenstein J (1985) Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal of Molecular Evolution, 17, 368–376. Felsenstein J (1988) Phylogenies from molecular sequences: inference and reliability. Annual Review of Genetics, 22, 521–565. Fonseca A, Arntzen JW, Crespo EG, Ferrand N (2003) Regional differentiation in the common midwife toad (Alytes obstetricans) in Portugal: a picture from mitochondrial DNA. Zeitschrift für Feldherpetologie, 10, 83–89. Fromhage L, Vences M, Veith M (2004) Testing alternative vicariance scenarios in western Mediterranean discoglossid frogs. Molecular Phylogenetics and Evolution, 31, 308–322. García-París M (1995) Variabilidad genética y distribución geográfica de Alytes obstetricans almogavarii en España. Revista Española de Herpetología, 9, 133–138. García-París M, Arntzen JW (2002) Alytes dickhilleni Arntzen & García-París 1995. Atlas y Libro Rojo de los Anfibios y Reptiles de España (ed. by JM Pleguezuelos, R Márquez, M

43

Capítulo 2

Lizana), pp. 76–78. Dirección General de la Conservación de la Naturaleza, Asociación Herpetológica Española, Madrid. García-París M, Jockusch EL (1999) A mitochondrial DNA perspective on the evolution of Iberian Discoglossus (Amphibia: Anura). Journal of Zoology (London), 248, 209–218. García-París M, Martínez-Solano I (2001) Nuevo estatus taxonómico para las poblaciones iberomediterráneas de Alytes obstetricans (Anura: Discoglossidae). Revista Española de Herpetología, 15, 99–113. García-París M, Alcobendas M, Buckley D, Wake DB (2003) Dispersal of viviparity across contact zones in Iberian populations of Firer Salamanders (Salamandra) inferred from discordance of genetic and morphological traits. Evolution, 57, 129–143. Grossenbacher K (1997) Alytes obstetricans (Laurenti, 1768). Atlas of amphibians and reptiles in Europe (ed. by J-P Gasc et al.), pp. 94–95. Societas Europaea Herpetologica, Paris. Hossini S (2001) Les anoures (Amphibiens) du Pléistocène inférieur (‘‘Villafranchien’’) du Jebel Irhoud (carrière ‘‘Ocre’’), Maroc. Annales de Paléontologie, 87, 79–97. Huelsenbeck JP, Ronquist F (2001) 17, 754–755.

MRBAYES:

Bayesian inference of phylogeny. Bioinformatics,

Kocher TD, Thomas WK, Meyer A, Edwards SV, Paabo S, Villablanca FX, Wilson AC (1989) Dynamics of mitochondrial DNA evolution in animals. Proceedings of the National Academy of Sciences USA, 86, 6196–6200. Krijgsman W, Hilgen FJ, Raffi I, Sierro FJ, Wilson DS (1999) Chronology, causes and progression of the Messinian salinity crisis. Nature, 400, 652–655. Lataste F (1879) Sur un nouveau genre de Batracien anoure d’Europe. Compte Rendus Hebdomadaire des Séances de l’Academie des Sciences, Paris, 88, 983–985. Llorente GA, Montori A, Santos X, Carretero MA (1995) Atlas dels Amfibis i Reptils de Catalunya i Andorra. Edicions el Brau, Barcelona. López Martínez N (1989). Tendencias en Paleobiogeografía. El futuro de la biogeografía del pasado. Paleontología (ed. by E Aguirre), pp. 271–296. CSIC, Madrid. Maxson LR (1984) Relationships of Baleaphryne to the discoglossid genera Alytes, Bombina and Discoglossus. Historia Biológica del Ferreret (ed. by H Hemmer, JA Alcover), pp. 193– 198. Moll, Mallorca. Maxson LR, Szymura JM (1984) Relationships among Discoglossid frogs: an albumin perspective. Amphibia–Reptilia, 5, 245–252. Miller SA, Dykes DD, Polesky HF (1988) A simple salting procedure for extracting DNA from human nucleated cells. Nucleic Acids Research, 16, 215. Moritz C, Schneider CJ, Wake DB (1992) Evolutionary relationships within the Ensatina eschscholtzii complex confirm the ring species interpretation. Systematic Biology, 41, 273–291. Oosterbroek P, Arntzen JW (1992) Area-cladograms of Circum-Mediterranean taxa in relation to Mediterranean palaeography. Journal of Biogeography, 19, 3–20. Ortí G, Meyer A (1997) The radiation of characiform fishes and the limits of resolution of mitochondrial ribosomal DNA sequences. Systematic Biology, 46, 75–100. Palumbi SR, Martin AP, Romano S, McMillan WO, Stice L, Grabowski G (1991) The simple fool’s guide to PCR. Special Publications, Department of Zoology, University of Hawaii, Honolulu. Posada D, Crandall KA (1998) 14, 817–818.

MODELTEST:

testing the model of DNA substitution. Bioinformatics,

Posada D, Crandall KA (2001) Intraspecific gene genealogies: trees grafting into networks. Trends in Ecology and Evolution, 16, 37–45. Roček Z (1980) Cranial anatomy of frogs of the family Pelobatidae Stannius, 1856, with outlines of their phylogeny and systematics. Acta Universitatis Carolinensis, 1–164.

44

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Román A (2002) Alytes muletensis (Sanchiz y Adrover, 1979). Atlas y Libro Rojo de los Anfibios y Reptiles de España (ed. by JM Pleguezuelos, R Márquez, M Lizana), pp. 79–81. Dirección General de la Conservación de la Naturaleza, Asociación Herpetológica Española, Madrid. Saiki RK, Delfand DH, Stooffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science, 239, 487–491. Sanchiz B. (1984) Análisis filogenético de la tribu Alytini (Anura, Discoglossidae) mediante el estudio de su morfoestructura ósea. Historia Biológica del Ferreret (ed. by H Hemmer, JA Alcover), pp. 61–108. Moll, Mallorca. Sanchiz B (1998) Encyclopedia of Paleoherpetology, Part IV (Salientia). Friedrich Pfeil, München. Sanchiz B, Adrover R (1979) Anfibios fósiles del Pleistoceno de Mallorca. Doñana, Acta Vertebrata, 4, 5–25. Sanchiz B, Alcover JA (1982) Un nou discoglossid (Amphibia: Anura) de l’Holocé de Menorca. Butlleti de la Institució Catalana d’Historia Natural, 48, 99–105. Sanderson MJ (1997) A non parametric approach to estimating divergence times in the absence of rate constancy. Molecular Biology and Evolution, 14, 1218–1231. Strimmer K, von Haeseler A (1996) Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Molecular Biology and Evolution, 13, 964–969. Swofford D. (2002) PAUP*: Phylogenetic analysis using parsimony (* and other methods), Version 4. Sinauer Associates, Sunderland, MA. Vences M, Vieites DR, Glaw F, Brinkmann H, Kosuch J, Veith M, Meyer A (2003) Multiple overseas dispersal in amphibians. Proceedings of the Royal Society of London, B, 270, 2435–2442. Weijermars R (1991) Geology and tectonics of the Betic zone, SE Spain. Earth-Science Reviews, 31, 153–236.

45

Capítulo 2

APPENDIX 1 MORPHOLOGICAL CHARACTERS: DEFINITIONS AND CHARACTER STATES Some of the studied characters have been used or modified from Clarke (1984, 1988). 1. Frontoparietal. Medial process present, well developed at the medial border, extending to the sagital axis (0); absent or poorly developed, not reaching that level (1). 2. Frontoparietal. Paraoccipital process present (0) or absent (1). 3. Frontoparietal. Dorsal fontanelles. A single anterior fontanelle with triangular shape (0); a double fontanelle formed by an anterior and a posterior fontanelle and separated by the medial processes of the frontoparietal, with a slight constriction at that point (1); two fontanelles, anterior and posterior, separated by the medial processes of the frontoparietal, that contact at the sagital axis (2). 4. Frontoparietal. Lateral margin in the orbitary area clearly protruding with respect to the anterior portion of the lateral margin (0) or narrower and not protruding (1). 5. Maxillar. Posterior process present (0) or absent (1). 6. Maxillar. Zygomaxillar process present (0) or absent (1). 7. Maxillar. Palatine process present (0) or absent (1). 8. Maxillar. Posterior process straight (0) or slightly convex distally (1). 9. Maxillar. Longitudinal groove in the base of the teeth row present (0) or absent (1). 10. Maxillar. Pterygoid process present (0) or absent (1). 11. >Nasals. Maxillar process elongated, extending to the nasal process of the maxillar (0) or short, not reaching the maxillar (1). 12. Nasals. Length of the rostral process roughly equivalent to one-third of the total length of the nasal (0); length of the rostral process roughly equivalent to one-fourth of the total length of the nasal (1). 13. Nasals. Ratio total length : maximum width (measured at the level of the maxillar process) is wider than long (0) or of subequal length or slightly longer than wide (1). 14. Paraoccipital. Paraoccipital crests present (0) or absent (1). 15. Parasphenoid. Posterior process long, clearly surpassing the posterior margin of the parasphenoid allae (0) or short, not surpassing that margin (1). 16. Parasphenoid. The cultriform process is narrow at the level of the allae, general shape biconvex (0) or cultriform process uniformly wide, general shape pointed (1).

46

Artigo I. Phylogenetic relationships and biogeography of midwife toads

17. Parasphenoid. Well-marked transverse keels in the allae (0) or transverse keels absent (1). 18. Parasphenoid. The allae are narrower in their proximal end and wider distally (0) or allae uniformly wide or narrowing distally (1). 19. Premaxillar. Number of teeth:greater than or equal to14 (0), or between 13 and 10 (1) or less than10 (2). 20. Prootic + exoccipital. Prootic process short and wide, does not extend over the external border of the orbital fossa (0) or elongated and narrow, extending over the external border of the orbital fossa and reaching the internal border of the pterygoid fossa (1). 21. Pterygoid. Ventral expansion absent (0) or present (1). 22. Sphenethmoid. Unpaired (0) or two paired pieces incompletely fused (1). 23. Sphenethmoid. Anterior process short, not surpassing the anterior margins of the lateral processes of the sphenethmoid (0) or elongated, clearly surpassing the anterior margins of the lateral processes (1). 24. Sphenethmoid. Lateral processes uniformly wide (0) or narrowing distally (1). 25. Sphenethmoid. In frontal view, the ventral border is clearly wider than the dorsal border (0) or the ventral and dorsal borders are approximately equally wide (1). 26. Squamosals. In a dorsal view, the zygomatic rami of the squamosals diverge caudally

in

anteromedial-posterolateral

orientation

(0)

or

diverge

rostrally

in

posteromedial-anterolateral orientation (1). 27. Squamosals. Otic and interior rami well developed (one third to half of the total length of the squamosal, measured from the apical end of the zygomatic ramus to an axis connecting the distal ends of the otic and interior rami) (0); poorly developed, less than one-third of the total length of the squamosal (1). 28. Squamosal. Zygomatic ramus elongated, more than two times the length of the otic ramus (0) or short and blunt (1). 29. Vomer. Posterior choanal process uniformly wide and bifurcated (0) or pointed (spine shape) (1).

47

Capítulo 2

APPENDIX 2 DESCRIPTION OF CHARACTERS AND CHARACTER STATES

Frontoparietal: The extension of the median process of the frontoparietal does not reach the sagital axis in A. muletensis, A. dickhilleni, A. maurus, A. o. almogavarii and A. o. obstetricans, whereas A. o. boscai and A. cisternasii possess a well-developed medial process of the frontoparietal (1). In A. o. pertinax, both character states are observed. Alytes cisternasii is characterized by the absence of a paraoccipital process, which is present in the other taxa (2). Besides, the dorsal fontanelle is in A. cisternasii clearly divided by the medial processes of the frontoparietals (3). The posterior fontanelles are especially small. The other species have a dorsal fontanelle with the typical ‘shoe sole’ shape described by Boulenger (1897). The posterior part of the frontoparietal expands laterally in all Alytes species, but to a lesser extent in A. muletensis and some specimens of A. o. almogavarii (4). Maxillar: The posterior process is curved to some degree in all Alytes except A. muletensis, which have posterior processes typically straight or slightly curved distally (8). Nasal: The nasals of A. cisternasii are wider than long, while the other species present nasals of subequal length, or that clearly are longer than wide (13). Parasphenoid: The posterior process is less developed in A. cisternasii than in the other species of Alytes (15). In the former, the posterior process does not surpass the posterior margin of the parasphenoid allae, in contrast to the other taxa. In most Alytes species, the cultriform process has a constriction at its base (16), which results in a general biconvex appearance. The basal constriction is absent in A. muletensis, A. maurus and some specimens of A. o. almogavarii, whose cultriform processes have a general pointed shape. More or less developed transverse keels are present in the parasphenoid allae of all Alytes taxa, except A. muletensis, A. dickhilleni, A. maurus and A. o. almogavarii (17). The width of the allae is also variable (18). In A. muletensis and A. cisternasii they are narrow in their proximal portion and wider distally. In A. dickhilleni, A. maurus and the subspecies of A. obstetricans (with the exception of some specimens of A. o. pertinax), the allae are uniform in width throughout their total length or narrower distally. 48

Artigo I. Phylogenetic relationships and biogeography of midwife toads

Premaxillar: The number of premaxillary teeth shows variation in Alytes (19). Alytes muletensis and several specimens of A. o. almogavarii have high counts of premaxillary teeth (14 or more), while other species have counts between 10 and 12, except for A. cisternasii that has generally eight or nine premaxillary teeth. Prootic + exoccipital: In all species of Alytes except A. cisternasii, the prootic process is relatively short, not reaching the lateral border of the orbital fossa (20). In A. cisternasii, the prootic process reaches or surpasses the external border of the orbital fossa. Pterygoid: Specimens of A. muletensis have a ventral expansion in the pterygoid, which is absent in the other species of Alytes (21). Sphenethmoid (Fig. A1): In A. muletensis, the paired ossification nuclei that later in the ontogeny confirm the sphenethmoid do not fuse completely as in the other species of Alytes (22). An intermediate condition is found in A. maurus: the fusion of both ossification nuclei is incomplete (Fig. A1). There is also variation in the relation between the length of the anterior process of the sphenethmoid and that of the anterior border of the anterolateral processes (23). In A. muletensis and A. maurus, the anterior process is relatively short, never clearly surpassing the anterior border of the anterolateral processes as in the other species of Alytes. Alytes o. pertinax and A. o. almogavarii are polymorphic with respect to this character. The relative length of the dorsal and ventral borders of the sphenethmoid in frontal view is less than 1 in A. dickhilleni and greater than 1 in A. cisternasii and A. obstetricans (25). This character is inapplicable to A. muletensis and A. maurus because of the incomplete fusion of the ossification centres of the sphenethmoid in this species (see above). Squamosal: The degree of development of the otic and interior rami is lesser in A. cisternasii when compared to other Alytes species (27). Also, the zygomatic ramus is shorter and blunter in A. cisternasii than in the other taxa (28). Vomer (Fig. A2): The posterior choanal process is distally bifurcated, not pointed and of uniform width in A. dickhilleni, A. maurus and some specimens of A. o. almogavarii, and pointed distally and not bifurcated in other Alytes species (29). Characters 5–7, 9–12, 14, 24 and 26 show no marked variation at the level of the ingroup.

49

Capítulo 2

Figure A1 Variation in the sphenethmoid across species of Alytes. Scale equals 3 mm. Figure A2 Anatomic terminology and variation in the vomer across species of Alytes. Scale equals 3 mm.

50

1

0

0

0-1

1

1

1

0

1

1

Taxon

Discoglossus

A. o. boscai

A. o. pertinax

A. o. almogavarii

A. o. obstetricans

A. dickhilleni

A. cisternasii

A. muletensis

A. maurus

0

0

1

0

0

0

0

0

0

2

1

1

2

1

1

1

1

1

0

3

Characters

0

1

0

0

0

0-1

0

0

0

4

1

1

1

1

1

1

1

1

0

5

1

1

1

1

1

1

1

1

0

6

1

1

1

1

1

1

1

1

0

7

1

0-1

1

1

1

1

1

1

0

8

1

1

1

1

1

1

1

1

0

9

1

1

1

1

1

1

1

1

0

10

1

1

1

1

1

1

1

1

0

11

1

1

1

1

1

1

1

1

0

12

1

1

0

1

1

1

1

1

0

13

1

1

1

1

1

1

1

1

0

14

0

0

1

0

0

0

0

0

0

15

1

1

0

0

0

0-1

0

0

0

16

1

1

0

1

0

1

0

0

0

17

1

0

0

1

1

1

0-1

1

0

18

1

0

2

1

1

0-1

1

1

0

19

1

1

0

1

1

1

1

1

0

20

1

0

1

1

1

1

1

1

0

21

1

1

0

0

0

0

0

0

0

22

0

0

1

1

1

0-1

0-1

1

0

23

1

1

1

1

1

1

1

1

0

24

?

?

1

0

1

1

1

1

0

25

1

1

1

1

1

1

1

1

0

26

0

0

1

0

0

0

0

0

0

27

0

0

1

0

0

0

0

0

0

28

0

1

1

0

1

0-1

1

1

0

29

Artigo I. Phylogenetic relationships and biogeography of midwife toads

APPENDIX 3 CHARACTER MATRIX

51

Artigo II. Phylogenetic relationships of midwife toads

Molecular Phylogenetics and Evolution, 44, 494-500.

Conflicting phylogenetic signal of nuclear vs mitochondrial DNA markers in midwife toads (Anura, Discoglossidae, Alytes): deep coalescence or ancestral hybridization? 1,2

H. GONÇALVES

3

1

2

, I. MARTÍNEZ-SOLANO , N. FERRAND & M. GARCÍA-PARÍS

1

CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos, Campus Agrário de Vairão, 4485-661 Vairão, Portugal and Departamento de Zoologia e Antropologia – Faculdade de Ciências da Universidade do Porto, Praça Gomes Teixeira, 4099-002 Porto, Portugal 2 Museo Nacional de Ciencias Naturales, CSIC, José Gutiérrez Abascal, 2, 28006 Madrid, Spain 3 Museum of Vertebrate Zoology. 3101 Valley Life Sciences Bldg., Berkeley, California 94720-3160, U.S.A. ABSTRACT Midwife toads (Anura, Discoglossidae, Alytes) comprise five species distributed in northern Africa, the Iberian Peninsula and parts of Western Europe. Current taxonomic consensus recognizes three subgenera: Alytes sensu stricto (A. obstetricans), Ammoryctys (A. cisternasii) and Baleaphryne (A. dickhilleni, A. maurus and A. muletensis). Previous attempts to resolve their phylogenetic relationships studied variation in nuclear (allozymes) and mitochondrial DNA (mt-DNA) markers and in osteological characters. However, a well-resolved, robust phylogenetic hypothesis is still lacking because of conflicting signals and limited resolution of the different markers used. In particular, the phylogenetic position of A. o. almogavarii is controversial. This taxon is well differentiated with respect to other Alytes taxa according to both nuclear and mt-DNA data, although the former suggest high levels of gene flow with A. obstetricans, the most widespread species in the genus. In this paper we provide new data from mitochondrial (nicotin adenin dehydrogenase subunit 4, nad4, ~800 base pairs) and nuclear (β-fibrinogen intron 7, ~600 base pairs) markers in all described Alytes taxa to reassess the phylogeny of the genus. Analyses of both datasets show conflicting phylogenetic signals related to the placement of A. o. almogavarii. Specimens from the type locality do not cluster with other samples of A. o almogavarii either with respect to their mt-DNA (very divergent mt-DNA haplotypes that are sequentially sister to all other representatives of the A. obstetricans clade) or to nuclear DNA (where individuals from the type locality present very divergent haplotypes that cluster with the Balephryne clade (including A. muletensis + A. maurus + A. dickhilleni) whereas other individuals cluster with A. obstetricans). The combination of morphology, nuclear and mt-DNA data suggests an old origin for A. o. almogavarii, which, coupled with putative hybridization events with the incipient lineages of A. obstericans and Baleaphryne has resulted in a reticulated pattern of evolution within the genus Alytes. Keywords: Amphibia, Anura, Alytes, phylogeny, β-fibrinogen intron 7, mitochondrial DNA, hybridization.

53

Capítulo 2

INTRODUCTION Amphibians, as well as other organisms with limited dispersal abilities and a prevalent vicariant mode of speciation, are particularly prone to reticulate when two previously isolated lineages establish secondary contact (e.g. García-París et al., 2003; Weisrock et al., 2005; Mulcahy et al., 2006; Weisrock & Larson, 2006; but see Hoskin et al., 2005). One case that exemplifies this problem is the long-standing debate on the phylogeny of midwife toads (Alytes Wagler, 1829), resulting from the partial

lack

of

congruence

among

mtDNA,

morphology,

and

allozyme

based

hypotheses (Arntzen & García-París, 1995; Fromhage et al., 2004; Martínez-Solano et al., 2004). Midwife toads present an exclusive reproductive mode including an elaborated parental care, in which the male carries a string of eggs on its hindlimbs until eclosion. The genus Alytes comprises five species distributed in three subgenera, all of them restricted to the extreme western Palearctic region (Figure 1): Alytes (Alytes) obstetricans (Laurenti, 1768), widely distributed in western Europe, with four currently recognized subspecies (Figure 1), Alytes (Ammoryctis) cisternasii Boscá, 1879, endemic to the center and southwest of the Iberian Peninsula, Alytes (Baleaphryne) muletensis (Sanchiz & Adrover, 1979) from the island of Mallorca in the Balearic archipelago, Alytes (Baleaphryne) dickhilleni Arntzen & García-París, 1995, endemic to the southeastern Iberian Peninsula, and Alytes (Baleaphryne) maurus Pasteur and Bons, 1962, which is patchily distributed in the mountains of northern Morocco (Bons & Geniez, 1996; Grossenbacher, 1997; Crespo, 1997; García-París & Arntzen, 2002; Román, 2002). The discovery of “living-fossils” (Sanchiz & Adrover, 1979; Alcover & Mayol, 1980) and the application of molecular techniques (Arntzen & García-París, 1995; GarcíaParís & Martínez-Solano, 2001; Martínez-Solano et al., 2004) resulted in taxonomic changes in the group, including description of new species and subspecies and changes in the taxonomic status of some of them. Several studies have tried to clarify the phylogenetic relationships among the species of the genus. Sanchiz (1984) and Maxson & Szymura (1984) proposed phylogenetic hypotheses based on morphological characters of the skeleton and on immunological data, respectively. Based on protein electrophoretic data, Arntzen & García-París (1995) placed Alytes dickhilleni as the sister group of Alytes muletensis and Alytes obstetricans as the sister taxon of the clade A. muletensis + A. dickhilleni. Altaba (1997) reanalyzed the dataset of Arntzen & García-París (1995) using a different coding scheme and found support for a sistergroup relationship between A. obstetricans and A. dickhilleni (but see Arntzen & García-París, 1997). Recently, Fromhage et al. (2004) and Martínez-Solano et al.

54

Artigo II. Phylogenetic relationships of midwife toads

(2004) used mtDNA sequences to construct a phylogenetic hypothesis of Alytes including A. maurus, which was recovered as the sister taxon to A. muletensis or to (A. muletensis + A. dickhilleni). Martínez-Solano et al. (2004) included sequences of A. o. almogavarii, the most genetically differentiated taxon within A. obstetricans (Arntzen & García-París, 1995; García-París, 1995), and also proposed a phylogenetic hypothesis based on osteological data. Morphological and molecular studies agreed in proposing a sister-group relationship between A. cisternasii and the remaining taxa. However, there are still unresolved issues. For example, based on mtDNA data, support for the monophyly of the Baleaphryne clade (including A. muletensis, A. maurus and A. dickhilleni) is weak, and the phylogenetic placement of A. o. almogavarii is uncertain (Fromhage et al., 2004; Martínez-Solano et al., 2004). Some nuclear introns are relatively rapidly evolving molecular markers that can provide additional information on a species phylogeny (but see Hare & Palumbi, 2003). Recent studies on mammals (Carroll & Bradley, 2005; Yu & Zhang, 2005; Matocq et al., 2007), birds (Feldman & Omland, 2005; Johnson & Clayton, 2000; Moyle, 2004; Prychitko & Moore, 1997, 2000; Weibel & Moore, 2002), reptiles (Creer et al., 2003; Giannasi et al., 2001; Godinho et al., 2005) and amphibians (Sequeira et al., 2006) have shown the utility of the β-fibrinogen intron 7 (β-fibint7) for phylogenetic analysis of closely related taxa. We present here new nuclear (β-fibint7) and mitochondrial (ND4-tRNALEU) data for all recognized species and subspecies of Alytes. Our goal is to produce independent hypotheses (nuclear vs mitochondrial) to infer phylogenetic relationships in Alytes and to assess the possible role of recent reticulation between deeply divergent lineages in the evolutionary history of the genus. We have increased the geographical sampling within A. obstetricans, including two populations of A. o. almogavarii, in an attempt to provide additional resolution with respect to previous studies using mtDNA to infer phylogenetic relationships in the genus (Fromhage et al., 2004; Martínez-Solano et al., 2004). Our results show deep discordances between the two molecular markers regarding the phylogenetic placement of A. o. almogavarii. We discuss the alternative scenarios in which this discrepancy might have arisen and reinterpret all previous morphological and molecular evidence (nuclear and mitochondrial) in the light of the new results.

55

Capítulo 2

MATERIAL AND METHODS Sampling Twenty-four specimens, including all the species and subspecies of the genus Alytes, were analyzed (see Fig. 1 and Table 1). Samples were obtained from toe tips of adult specimens or tail tips of larvae, which were collected and immediately released in the field.

Figure 1 Distribution map of species and subspecies of Alytes with indication of the geographical origin of samples analysed in this study. Areas with question marks represent zones where assignment of populations to subspecies of A. obstetricans is problematical. Sample codes as in Table 1.

DNA extraction, amplification and sequencing Total genomic DNA was extracted from frozen or ethanol-preserved tissue following the standard high-salt protocol of Sambrook et al. (1989) or using a Dneasy Tissue Kit (QIAGEN) following the manufacturers’ protocol. Two gene regions including one mitochondrial gene and adjacent tRNAs (ND4-tRNALEU) and one nuclear intron (β-

56

Artigo II. Phylogenetic relationships of midwife toads

fibint7) were amplified using the following combination of primers and PCR thermal cycling conditions. For ND4-tRNALEU we used the primers ND4 and Leu, previously described by Arévalo et al. (1994), with an initial denaturation 94ºC for 5 min; followed by 35 cycles of denaturation at 94ºC, 1 min, annealing at 56ºC, 45 s, extension at 72ºC; 1 min, and final extension of 72ºC, 5 min. For β-fibint7 a two-step amplification procedure was used, with a combination of two primer pairs (PCR1: FIBX7 and FIBX8; PCR2: BFXF and BFXR) as described by Sequeira et al. (2006). Reactions were performed in the following conditions: initial denaturation at 94ºC for 5 min, followed by 40 cycles of denaturation at 94ºC, 40 s, annealing at 50ºC (PCR1) and 56ºC (PCR2), 1 min, extension at 72ºC, 1.30 min, and a final extension of 72ºC, 7 min. Amplified fragments of the PCR2 reaction were sequenced in both directions using the primers BFXF and BFXR. PCR products were sequenced using the ABI Prism Big Dye Terminator Cycle sequencing protocol in an ABI Prism 310 automated sequencer (Applied Biosystems). Sequences were checked by eye, edited, and then aligned using the program

BIOEDIT

version 7.0.1 (Hall, 1999). All sequences generated

for this study are deposited in GenBank under Accession Numbers EF441291EF441343 (Table 1). Table 1 Sample localities and GenBank accession numbers. Taxa

Sample code

Locality

GenBank Accession Nos. ND4

β-fibint7

Alytes cisternasii

HG112 CER17/CER01

Río Adaja, Ávila, Spain Cercal, Portugal

EF441313 EF441314

EF441342 EF441343

Alytes dickhilleni

HG103

Puerto de las Crucetillas, Sierra de Alcaraz, Albacete, Spain

EF441309

EF441338

Alytes maurus

MAR03 MAR04

Bab Bou Idir, Taza, Morocco

EF441311 EF441312

EF441340 EF441341

Alytes muletensis

MAI06

Mallorca, Spain

EF441310

EF441339

A. o. almogavarii

BR03 BR13 CAT10 CAT12 HS16

Rasos de Peguera, Berga, Spain

EF441305 EF441306 EF441307 EF441308 EF441304

EF441332/EF441333 EF441334/EF441335 EF441336 EF441337 EF441331

PO01 OUR03 ZA04 HG076 VAL22 MTM01 SMA25 DO02 JUB3 NA37 MGP115 HG120 HG123

Tuy, Pontevedra, Spain Penalva, Ourense, Spain Valdefinjas, Zamora, Spain Puerto Nuevo, S. Gata, Spain Alfena, Valongo, Portugal Cinfães, S. Montemuro, Portugal Rib. S. Bento, S. S. Mamede, Portugal Rhür Basin, Germany Jublains, France Irati, Navarra, Spain Tolivia, Asturias, Spain Belmonte de Tajo, Madrid, Spain Buenache de la Sierra, Cuenca, Spain

EF441296 EF441295 EF441297 EF441301 EF441300 EF441302 EF441303 EF441291 EF441292 EF441293 EF441294 EF441298 EF441299

EF441320 EF441319 EF441321 EF441327 EF441325/EF441326 EF441328/EF441329 EF441330 EF441315 EF441316 EF441317 EF441318 EF441322/EF441323 EF441324

A. obstetricans

Llinars, Lleida, Spain Ibón de Piedrafita, Huesca, Spain

57

Capítulo 2

Phylogenetic analyses Phylogenetic analyses were performed with

PAUP*version

4.0b10 (Swofford,

2002), using the criteria of maximum parsimony (MP, heuristic search with TBR branch swapping and random additions of taxa) and maximum likelihood (ML, same searching strategy). Branch support was evaluated via 1000 bootstrap replicates (100 in ML analyses), with 10 random addition replicates to minimize the influence of sequence order in the resulting topologies. Bayesian analyses were performed with MrBayes version 3.1.1 (Ronquist & Huelsenbeck, 2003). We ran five Metropolis coupled Monte Carlo Markov chains (MC3) for five million generations, sampling every 1,000 generations. We discarded 500 trees as burn in, after checking for stationarity and convergence of the chains with the software

TRACER

version 1.3 (Rambaut &

Drummond, 2004). For ML and Bayesian analyses, the optimal substitution model was selected using the Akaike information criterion as implemented in ModelTest version 3.7 (Posada & Crandall, 1998). In all analyses, A. cisternasii was used as the outgroup because both morphological and molecular data support its sister-group relationship with all other species of Alytes (see refs. above), and also because previous studies have suggested saturation problems when more distant outgroups (e.g. Discoglossus) are included in the mtDNA analyses (Martínez-Solano et al., 2004). Prior to combining the datasets from both markers for subsequent phylogenetic analyses, we tested the null hypothesis of homogeneity of both partitions (mtDNA vs nuclear) by performing the incongruence length difference test (ILD, Farris et al., 1994) as implemented in

PAUP*version

4.0b10 (10,000 replicates, invariable sites

excluded). Finally, Shimodaira–Hasegawa tests (Shimodaira & Hasegawa, 1999) and SOWH

tests (Swofford et al., 1996) were used to test for differences between

competing phylogenetic hypotheses. For these topology tests, the phylogenetic hypothesis based on mtDNA data, with a monophyletic A. obstetricans including samples from Berga and Huesca (A. o. almogavarii) (see Fig. 2A) was used as the reference, and used as the “constraint” topology against which the alternative hypothesis based on the nuclear tree was tested. SH tests were performed in

PAUP

under the same ML model as in the original tree searches. For SOWH tests, 100 datasets were simulated in SeqGen version 1.3.2 (Rambaut & Grassly, 1997), with the same ML model and enforcing the monophyly of A. obstetricans including the samples from Berga and Huesca. A null distribution of likelihood scores was then generated analyzing the simulated datasets in

PAUP

with heuristic searches. Finally, the 95%

confidence intervals of the null distribution were calculated and compared with the observed likelihood value in the nuclear dataset (see Goldman et al., 2000).

58

Artigo II. Phylogenetic relationships of midwife toads

Since previous studies have hypothesized the existence of recombinant alleles in the β-fibint7 locus (Godinho et al., 2006), we estimated the minimum number of recombination events in the history of the β-fibint7 using the four-gamete test (Hudson & Kaplan, 1985) as implemented in

DNASP

version 3.51 (Rozas & Rozas,

1999).

RESULTS AND DISCUSSION The mtDNA alignment consisted of 24 sequences of 814 nucleotides except for the sample of A. muletensis, which was 761 base pairs long. Of these, 535 were constant in the alignment and 187 were parsimony informative. The nuclear dataset included 48 sequences, two per individual sequenced. Most individuals analyzed were homozygous (19 out of 24). Individuals VAL22, MTM01, HG120, BR03 and BR13 were heterozygous and alleles were inferred using

PHASE

software (Stephens et al., 2001), using a minimum probability of 0.90 as a cut-off value. The final alignment included 634 base pairs, including 23 gaps, mostly of 1 or 2 base pairs in length except for a 12 base pair insertion observed in all ingroup samples and missing in A. cisternasii. Of the remaining 611 characters, 537 were constant and 74 were parsimony informative. An A+T biased content of 70% was observed in βfibint7 sequences, which is concordant with values reported for this intron in other species (Creer et al., 2003; Godinho et al., 2005; Johnson & Clayton, 2000; Prychitko & Moore, 1997, 2000; Sequeira et al., 2006; Yu & Zhang, 2005). According to Prychitko & Moore (1997) this feature of A-T rich sequences is a peculiarity of noncoding regions that are not under functional constraint. The ILD test rejected the null hypothesis of congruence between the partitions (nuclear vs. mitochondrial, p=0.001), which were thus analyzed separately. The topologies obtained for both markers are shown in Figure 2. There are obvious discrepancies between the nuclear and mitochondrial gene trees, and these concern the phylogenetic position of A. o. almogavarii. The topologies of the trees recovered for ND4 include two main clades: one weakly supported including species of Baleaphryne and the other comprising all samples of A. obstetricans. The two populations of A. o. almogavarii studied do not cluster together in the analyses, but they are both recovered as part of the A. obstetricans clade (Figure 2A). As noted above, the Baleaphryne clade is weakly supported, except in the ML analysis (Figure 2A). These results are similar to those obtained in previous studies (Fromhage et al., 2004; Martínez-Solano et al., 2004) and provide additional support for the monophyly of A. obstetricans including samples of A. o. almogavarii.

59

Capítulo 2

Figure 2 Maximum likelihood phylograms of ND4 (A) and β-fibint7 (B) sequence data in Alytes. Sample codes as in Table 1. Bootstrap values (MP/ML) and posterior probabilities of relevant nodes are indicated above branches. Suffixes a and b in codes in the nuclear DNA phylogram refer to different alleles present in heterozygous individuals. The conflicting position of the samples of A. o. almogavarii is highlighted (arrows and rectangles). Note differences in scale between A and B.

In contrast, the topology obtained based on our nuclear marker is significantly different (Shimodaira–Hasegawa test, dif. likelihood=192.83668, p.0.0001, SOWH test, 95% cutoff for differences in likelihood as calculated from the simulated dataset=82.77