História evolutiva dos lagartos anões

(Lygodactylus, Gekkonidae) no continente Sul

Americano

Flávia Mól Lanna

UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE

CENTRO DE BIOCIÊNCIAS

DEPARTAMENTO DE ECOLOGIA

PROGRAMA DE PÓS-GRADUAÇÃO EM ECOLOGIA

Adrian Antonio Garda (Orientador)

Fernanda de Pinho Werneck (INPA) (Coorientadora)

Marcelo Coelho Miguel Gehara (AMNH – USA) (Coorientador)

Natal/2017

FLÁVIA MÓL LANNA

HISTÓRIA EVOLUTIVA DOS LAGARTOS ANÕES

(Lygodactylus, Gekkonidae) NO CONTINENTE SUL

AMERICANO

Dissertação apresentada ao Programa

de Pós-Graduação em Ecologia da

Universidade Federal do Rio Grande

do Norte como parte das exigências

para obtenção do Grau de Mestre

ORIENTADOR: Adrian Antonio Garda

COORIENTADORES: Fernanda de Pinho Werneck

Marcelo Coelho Miguel Gehara

Natal/2017

Universidade Federal do Rio Grande do Norte - UFRN

Sistema de Bibliotecas - SISBI

Catalogação de Publicação na Fonte. UFRN - Biblioteca Setorial Prof. Leopoldo Nelson - ­Centro de Biociências - CB

Lanna, Flávia Mól.

História evolutiva dos lagartos anões (Lygodactylus,

Gekkonidae) no continente Sul Americano / Flávia Mól Lanna. - Natal, 2017.

79 f.: il.

Dissertação (Mestrado) - Universidade Federal do Rio Grande do

Norte. Centro de Biociências. Programa de Pós-Graduação em Ecologia.

Orientador: Prof. Dr. Adrian Antonio Garda.

Coorientadora: Profa. Dra. Fernanda de Pinho Werneck.

Coorientador: Dr. Marcelo Coelho Miguel Gehara.

1. Caatinga - Dissertação. 2. Chaco - Dissertação. 3. Complexo

de espécies - Dissertação. 4. Florestas Tropicais Sazonalmente

Secas - Dissertação. 5. Hipótese do Arco Pleistocênico -

Dissertação. 6. Rio São Francisco - Dissertação. I. Garda, Adrian Antonio. II. Werneck, Fernanda de Pinho. III. Gehara, Marcelo

Coelho Miguel. IV. Universidade Federal Do Rio Grande do Norte.

V. Título.

RN/UF/BSE-CB CDU 574

FLÁVIA MÓL LANNA

HISTÓRIA EVOLUTIVA DOS LAGARTOS ANÕES

(Lygodactylus, Gekkonidae) NO CONTINENTE SUL

AMERICANO

Dissertação apresentada ao Programa de

Pós-Graduação em Ecologia da

Universidade Federal do Rio Grande do

Norte como parte das exigências para

obtenção do Grau de Mestre

Data da defesa: 21 de fevereiro de 2017

BANCA EXAMINADORA

___________________________

Dr. Adrian Antonio Garda

Presidente/Orientador | UFRN

___________________________ ___________________________

Dr. Fabrícius Maia Chaves Bicalho Domingos Dra. Simone Nunes Brandão

Membro externo | UnB Membro externo | UFRN

“Na vida, não vale tanto o que temos,

nem tanto importa o que somos.

Vale o que realizamos com aquilo que possuímos e,

acima de tudo, importa o que fazemos de nós”

– Chico Xavier –

AGRADECIMENTOS

Mais uma etapa está sendo concluída, outra página virada deste livro de histórias em

que cada capítulo é um desafio encerrado com um final feliz, cheio de personagens que vou levar

pra sempre comigo. Foram tantas pessoas importantes nesses dois anos de mestrado que fica até

difícil agradecer o ombro amigo e os conselhos de todo mundo da forma como merecem.

Agradeço primeiramente ao meu orientador Adrian Garda pela confiança depositada

em mim, pela oportunidade de realizar esse projeto, paciência nos momentos de aperto, apoio e

ensinamentos do início ao fim desses processo.

Ao meu coorientador Marcelo Gehara pela prontidão em responder minhas dúvidas,

pelas horas e horas discutindo as análises e pela crucial direção nos momentos de dúvida.

À minha coorientadora Fernanda Werneck por confiar à mim este projeto, por todo

ensinamento durante este período, pela recepção em Manaus e todo o suporte laboratorial no

INPA e na BYU.

À vocês três em conjunto pela especial oportunidade de realizar a coleta de dados nos

Estados Unidos, uma experiência incrivelmente enriquecedora profissionalmente, culturalmente

e pessoalmente.

Ao Guarino Colli por ter viabilizado toda a viagem para os Estados Unidos e idas a

campo, além das doações dos tecidos.

Ao Miguel Rodrigues pelo auxílio às idas a campo e doações dos tecidos.

À Eliana Oliveira e ao Gabriel Costa pelas sugestões na qualificação.

Ao Jack Sites Jr. por ter me recebido com muito carinho em seu laboratório na

Brigham Young University.

Ao Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) pelos

dois anos de bolsa e ao programa de Pós-Graduação em Ecologia/UFRN pelo suporte.

Ao amigos do LAR-UFRN (David Lucas Rohr, Eliana Oliveira, Emanuel Fonseca,

Felipe Camurugi, Felipe Magalhães, Marilia Lion, Ricardo Marques, Ricardo Rodrigues, Sarah

Mângia, Vinícius São-Pedro e Willianilson Pessoa) pelas discussões de trabalho, pelos

momentos de diversão, idas a campo, papos cabeça e algumas poucas piadas boas. Obrigada

pessoal!

Aos amigos do Sites Lab (César Aguilar, Derek Tucker, Juan Santos, Luciano J. Ávila,

Mariana Morando, Perry L. Wood e Randy Klabacka) pela recepção agradável, toda a ajuda no

laboratório, piqueniques e passeios por lugares maravilhosos de Provo e região.

Aos amigos do INPA que me receberam de braços abertos, me auxiliaram no

laboratório e me apresentaram a região (Alan Filipe, Gabriela Farias, Erik Choueri, Jéssica dos

Anjos e, em especial, Lídia Martins).

Ao pessoal da república Tanquetão (Marina Vergara, Carol Vergara, Jessé Ramos,

Édina Vergara, Cíntia Pinheiro (Tida), Clarinha, Natália Pires (Pocas), Nicolas Penna, Maysa

Gomes, Gustavo Paterno, Alexsander Hada, Felipe Camurugi e Morena), por todas as festas,

noites de músicas, brincadeiras, comilanças, descontrações nos momentos de tensão, amizade,

carinho e noites passadas em claro em batalhas de outra realidade.

Aos amigos de Natal e Minas Gerais ainda não citados acima, com os quais

compartilhei bons momentos e uma gostosa amizade. Um especial obrigado aos grandes amigos

Vinícius São-Pedro e Eliana Oliveira pelo carinho e companheirismo.

À toda minha família, que apesar da saudade, respeitou minha ausência e distância.

Agradeço principalmente ao meu pai (Eduardo) e minha mãe (Teresa), que sempre me apoiaram

e me incentivaram a seguir meus sonhos, e também ao meu irmão (Arthur), que sempre esteve

presente apesar de distante durante esse tempo. Vocês são o meu alicerce! O apoio de vocês foi

fundamental pro meu crescimento.

E por último, mas de forma alguma menos importante, agradeço ao El (Emanuel

Fonseca), meu amor, meu companheiro, parceiro em todas as aventuras. Te agradeço por estar

sempre ao meu lado e por superarmos juntos os desafios que encontramos e que nos dispomos a

enfrentar. Sem você tudo isso não teria sido possível. Obrigada de coração!

SUMÁRIO

LISTA DE TABELAS ....................................................................................................................... I

LISTA DE FIGURAS ........................................................................................................................ II

RESUMO ........................................................................................................................................ V

ABSTRACT .................................................................................................................................... VII

INTRODUÇÃO GERAL.................................................................................................................... 1

CHAPTER I: Out of Africa: a cryptic speciation history of a small traveler gecko in South

America .................................................................................................................... 11

ABSTRACT ..................................................................................................................................... 13

INTRODUCTION ............................................................................................................................. 14

MATERIAL AND METHODS ........................................................................................................... 17

Taxon sampling ....................................................................................................................... 17

Sequencing .............................................................................................................................. 17

Phylogenetic relationships ...................................................................................................... 18

Divergence time estimates ...................................................................................................... 18

Species delimitation ................................................................................................................ 19

RESULTS ....................................................................................................................................... 20

Genetic data and phylogenetic analyses ................................................................................. 20

Divergence times and species delimitation ............................................................................. 20

DISCUSSION ................................................................................................................................... 21

Monophyly of Lygodactylus in South America ...................................................................... 21

Pleistocenic Arc Hypothesis ................................................................................................... 22

Cryptic diversity...................................................................................................................... 23

CONCLUSIONS ............................................................................................................................... 25

ACKNOWLEDGEMENTS ................................................................................................................. 25

REFERENCES ................................................................................................................................. 26

TABLES ......................................................................................................................................... 35

FIGURES ........................................................................................................................................ 38

SUPPORTING INFORMATION ................................................................................................. 41

Tables ................................................................................................................................ 41

Figures ............................................................................................................................... 42

References ......................................................................................................................... 43

CHAPTER II: The role of the São Francisco River on the diversification of a dwarf gecko

endemic to the semiarid Caatinga, Northeastern Brazil........................................... 44

ABSTRACT ..................................................................................................................................... 46

INTRODUCTION ............................................................................................................................. 47

MATERIAL AND METHODS ........................................................................................................... 49

Sample collection and sequencing .......................................................................................... 49

Gene tree ................................................................................................................................. 50

Assignment of genetic lineages .............................................................................................. 51

Haplotype network and DNA polymorphism ......................................................................... 52

Species tree estimation ............................................................................................................ 52

Phylogeographic reconstruction .............................................................................................. 53

Testing the SFR barrier ........................................................................................................... 54

RESULTS ....................................................................................................................................... 55

Assignment of lineages ........................................................................................................... 55

Gene trees, species tree, haplotype network and DNA polymorphism .................................. 56

Phylogeographic reconstruction .............................................................................................. 56

Testing the SFR as a barrier .................................................................................................... 57

DISCUSSION ................................................................................................................................... 57

CONCLUSIONS ............................................................................................................................... 59

ACKNOWLEDGEMENTS ................................................................................................................. 59

REFERENCES ................................................................................................................................. 60

TABLES ......................................................................................................................................... 66

FIGURES ........................................................................................................................................ 68

SUPPORTING INFORMATION ................................................................................................. 72

Tables ................................................................................................................................ 72

Figures ............................................................................................................................... 75

References ......................................................................................................................... 77

I

LISTA DE TABELAS

CAPÍTULO I:

Table 1: Samples used in this study with respective voucher, localities number (associated with

Figure 1) and GenBank number (if available) .............................................................................. 35

Table 2: Genetic distance (uncorrelated p-distance) among South American Lygodactylus

species suggested by SpedeSTEM. Upper values are from ND2 (mtDNA) and lower values are

from RAG1 (nuDNA) marker ....................................................................................................... 37

Table S1: Information about markers and PCR protocols used in this study .............................. 41

CAPÍTULO II:

Table 1: Genetic statistics of northern and southern lineages of Lygodactylus klugei (according

to GMYC results) for each locus .................................................................................................. 66

Table 2: Genetic distance (p-distance) between and within northern and southern lineages

(according to GMYC results) for the four markers ...................................................................... 67

Table S1: Information about Lygodacylus samples used in this study, with locality numbers and

coordinates .................................................................................................................................... 72

Table S2: Information about gene, primers, and PCR protocols used in this study .................... 74

II

LISTA DE FIGURAS

CAPÍTULO I:

Figure 1: Distribution map of South American Lygodactylus samples. The biomes of the open

diagonal are in a gray scale: Chaco (CH) in dark gray, Cerrado (CE) in gray and Caatinga (CA)

in black, highlighting for Seasonally Dry Tropical Forests enclaves in Cerrado also in black.

Orange dots represents L. klugei samples, green dot Lygodactylus sp. 1 (Santo Inácio), blue dot

Lygodactylus sp. 2 (Condeúba), light red Lygodactylus sp. 3 (São Domingos), and dark red L.

wetzeli ........................................................................................................................................... 38

Figure 2: Maximum Likelihood concatenated tree with the individuals gathered in five highly

supported groups (pp = 100). Node numbers correspond to 1000 ML bootstrap values. Different

colors represent distinct groups .................................................................................................... 39

Figure 3: Bayesian species tree and divergence times for South American Lygodactylus. The

African Lygodactylus were collapsed only on the representation for better visualization of the

tree. For the whole species tree, see Supporting Information, Figure S1. The node numbers

correspond to posterior probability. Nodes with posterior probability higher than 99% are

marked with an asterisk (*). Outgroup 1 correspond to L. angularis, outgroup 2 to L. chobiensis

and L. kimhowelli, and outgroup 3 to the other 15 species of African Lygodactylus used here.

Abbreviations: Pli, Pliocene; P, Pleistocene; BA, Bahia State; GO, Goiás State ......................... 40

Figure S1: Bayesian species tree and divergence time for Lygodactylus. Complete tree

encompassing the 18 African species. Abbreviations: Pli, Pliocene; P, Pleistocene .................... 42

III

CAPÍTULO II:

Figure 1: Map of the sample localities of Lygodactylus klugei and its related lineage according

GMYC results. Pink circles correspond to northern lineage. Green circles correspond to southern

lineage. São Francisco River is represented in blue and its paleocurse is represented in grey in a

dotted line...................................................................................................................................... 68

Figure 2: Lineage assignment based on the Generalized Mixed Yule Coalescent (GMYC)

method. (A) gene tree generated by GMYC. The two most probable groups are showed in red,

with one individual not allocated for any of them. (B) heat map represents Bayesian

implementations of the GMYC (bGMYC). Darker colors indicate lower probability of grouping

while lighter colors indicate higher probability. Probabilities higher than 50% were used to

assign the lineages......................................................................................................................... 69

Figure 3: Haplotype network for (A) ND4, (B) DMXL1, (C) DNAH3, and (D) PRLR markers

according to their respective Bayesian gene tree. The size of each circle is proportional to the

haplotype frequency. The small blue correspond to the number of mutational steps. Pink circles

represent northern lineage and green circles represent southern lineage (according to GMYC

results) ........................................................................................................................................... 70

Figure 4: Bayesian spatiotemporal diffusion of mtDNA for Lygdactylus klugei in six time

frames. Lighter shades represent older diffusion events and darker shades represent younger

diffusion events ............................................................................................................................. 71

IV

Figure S1: Estimated gene trees for ND4 (A), DMXL1 (B), DNAH3 (C), and PRLR (D).

Individuals are highlighted according to GMYC results. Individuals in pink correspond to

northern lineage and green correspond to southern lineage. Nodes with posterior probability

higher than 95% are marked with an asterisk (*) ......................................................................... 75

Figure S2: Genetic structure of Lygodactylus klugei based on nuclear markers performed in

Structure. (A) probability of individuals' assignment to each population (green and red). (B) map

with localities colored according Structure population assignment; Green triangles correspond to

population 1 and red circles correspond to population 2; Blue line correspond to the São

Francisco River and A correspond to São Francisco River paleocourse ...................................... 76

V

RESUMO

Quais os processos e mecanismos responsáveis pela diversificação das espécies? Essa é uma

questão antiga que tem sido revolucionada com o avanço tecnológico, computacional e

metodológico, e tem sido agora compreendida de uma forma que antes não era possível. A

filogeografia é uma multidisciplina que utiliza ferramentas derivadas da biogeografia, filogenia

molecular e genética de populações para entender o contexto da distribuição dos genes no tempo

e espaço. O presente estudo utiliza análises filogenéticas e filogeográficas para inferir os

processos determinantes na diversificação de lagartos do gênero Lygodactylus nas Florestas

Tropicais Sazonalmente Secas (FTSS) da América do Sul. No primeiro capítulo nós

investigamos as relações entre os Lygodactylus Sul Americanos, buscando entender a influência

do Arco Pleistocênico em sua diversificação e se essas espécies representam um grupo

monofilético. Através de análises filogenéticas e de delimitação de espécie, nós recuperamos o

monofiletismo do grupo quando comparado com as espécies Africanas e reconhecemos L. klugei

como um complexo de espécies crípticas. Nós sugerimos o aumento de duas para cinco espécies

de Lygodactylus na América do Sul. O tempo de divergência entre L. klugei e as espécies

candidatas endêmicas das FTSSs não foi congruente com a hipótese do Arco Pleistocênico.

Porém, a fragmentação das FTSS pode ter influenciado na divergência de L. wetzeli e uma

espécie candidata endêmica de um enclave de FTSS no Cerrado (São Domingos, região do Vale

do Paranã). No segundo capítulo investigamos a diversificação dentro da Caatinga, testando o

papel do rio São Francisco (RSF) como barreira geográfica nesse bioma. Utilizamos um lagarto

endêmico dessa região (L. klugei) como modelo de estudo. Nós delimitamos as possíveis

linhagens, investigamos as relações filogenéticas entre elas, a história de difusão espaço-

temporal e, para testar a hipótese do rio (barreira para fluxo gênico), nós utilizamos uma análise

VI

de migração. Nós recuperamos duas linhagens estruturadas de acordo com o RSF: uma ao norte e

outra ao sul do rio. A divergência dessas linhagens ocorreu à 295 mil anos atrás, congruente com

a mudança do curso do RSF para seu atual curso. Não encontramos influência do paleocurso do

RSF na estruturação de L. klugei.

Palavras-chave: Caatinga, Chaco, complexo de espécies, Florestas Tropicais Sazonalmente

Secas, hipótese do Arco Pleistocênico, hipótese do rio, rio São Francisco

VII

ABSTRACT

Which processes and mechanisms are responsible for species diversification? This old question

has been revolutionized with technological, computational and methodological advancements,

and is now being understood in a way that was previously not possible. Phylogeography is a

multidiscipline that uses tools derived from biogeography, molecular phylogeny, and population

genetics to understand the context of gene distribution in time and space. The present study uses

phylogenetic and phylogeographic analyses to infer determinant processes in the diversification

of the lizard genus Lygodactylus in Seasonally Dry Tropical Forests (SDTF) in South America.

In the first chapter we investigate the relationships among South American Lygodactylus species,

seeking to understand the influence of the Pleistocenic Arc on its diversification and whether

these species represent a monophyletic group. Through phylogenetics and species delimitation

analyses we recovered the monophyly of the group in relation to African species and recognized

L. klugei as a cryptic species complex. We suggest that Lygodactylus in South America actually

comprises five species instead of two. The divergence time among L. klugei and candidate

species endemic to SDTFs was not congruent with the Pleistocenic Arc Hypothesis. However,

we suggest that the fragmentation of SDTFs likely influenced the divergence of L. wetzeli, and of

a candidate species endemic to a SDTF enclave within the Cerrado biome (São Domingos, Vale

do Paranã region). In the second chapter we investigate the diversification within the Caatinga,

testing the role of the São Francisco River (SFR) as a prominent geographic barrier. We used a

lizard endemic to this region (L. klugei) as study model. We delimited the existent lineages,

investigated the genetic relationships between them, the spatio-temporal diffusion history, and

used a migration analysis to test the riverine hypothesis (barrier to gene flow). We recovered two

lineages structured in respect to the SFR: a northern and a southern one. Lineage divergence

VIII

occurred 295 kya, congruent with the course change of the SFR to its current position. We found

no influence of the paleo-SFR on L. klugei structure.

Keywords: Caatinga, Chaco, Pleistocenic Arc Hypothesis, riverine hypothesis, São Francisco

River, Seasonally Dry Tropical Forests, species complex

1

INTRODUÇÃO GERAL

O continente Sul-Americano apresenta uma enorme diversidade de espécies.

Entretanto, sua história evolutiva ainda é pouco conhecida (Turchetto-Zolet, et al. 2013). A

compreensão dos processos que levaram à diversificação das espécies Sul-Americanas foi

intensificada na última década com o surgimento de ferramentas, como a filogeografia, que

tornaram o teste de hipóteses alternativas mais viáveis (Riddle, et al. 2008). A filogeografia é a

interface da filogenia molecular, a biogeografia e a genética de populações no estudo da

distribuição espacial dos genes no espaço e no tempo entre espécies próximas ou populações de

uma mesma espécie (Avise 2000). Análises filogeográficas vem sendo consideradas importantes

para testar hipóteses biogeográficas e identificar diferentes linhagens evolutivas, como por

exemplo barreiras geográficas separando populações e até mesmo espécies (Carnaval, et al.

2009; Fitzpatrick, et al. 2009; Shepard and Burbrink 2008; Thomé, et al. 2010). Inicialmente,

muita atenção foi dada aos biomas de florestas úmidas da América do Sul (Turchetto-Zolet, et al.

2013), e apenas recentemente os biomas de vegetação aberta começaram a ser investigados e os

processos que levaram à diversificação das espécies nessa região estão sendo melhor

compreendidos (Gamble, et al. 2012; Magalhaes, et al. 2014; Recoder, et al. 2014; Werneck, et

al. 2012; Werneck, et al. 2015).

As áreas de formações abertas estão distribuídas na forma de uma diagonal na América

do Sul, ligando a região Nordeste do Brasil ao Paraguai, Argentina e Bolívia, englobando os

biomas Caatinga, Cerrado e Chaco (Werneck 2011). Apesar de existirem espécies

compartilhadas por esses três biomas, cada um tem sua biota característica com espécies

endêmicas. Inicialmente, pesquisadores sugeriram que a Caatinga era um bioma com uma baixa

2

diversidade de espécies, não possuindo uma fauna característica, onde suas espécies estavam

amplamente distribuídas pela Diagonal de Formações Abertas (Mares, et al. 1981; Pennington, et

al. 2000; Vanzolini 1988, 1974, 1976; Werneck 2011). Atualmente, mesmo ainda sendo pouco

estudada, sabe-se que a diversidade nessa região é muito maior do que afirmada anteriormente, o

que mostra que esses estudos se basearam em coletas pouco representativas e amostragem

geográfica insuficiente do bioma (Rodrigues 2003).

A Caatinga, um bioma exclusivamente brasileiro, está distribuído principalmente no

nordeste do Brasil, se estendendo também ao norte de Minas Gerais, na região que segue o Rio

São Francisco e o médio Rio Jequitinhonha (Prado 2003). A Caatinga apresenta um clima quente

e seco durante a maior parte do ano, com um curto período de chuvas no inverno, que coincide

com o solstício de verão (Prado 2003). Sua vegetação é predominantemente decídua e espinhosa,

podendo variar de florestas altas e secas à afloramentos rochosos com cactos, bromélias e

arbustos baixos (Prado 2003). A produtividade primária, diferentemente das florestas tropicais

úmidas, é maior no inverno por ser a estação chuvosa e, consequentemente, o período de maior

crescimento vegetal (Pennington, et al. 2006).

A Caatinga é a maior área contínua de Florestas Tropicais Sazonalmente Secas (FTSS)

(Werneck, et al. 2012). As FTSSs englobam esse bioma e também formações com características

semelhantes e que possuem: média de temperatura anual de 17º C, livre de geadas, com duas

estações bem definidas (seca – chuvosa) e precipitação entre 200 e 2000 mm anuais (Murphy

and Lugo 1986). Na região Neotropical, essa unidade fitogeográfica ocorre descontinuamente

desde a América do Sul até a América do Norte (Pennington, et al. 2009). Alguns remanescentes

isolados de FTSS ocorrem como enclaves no Cerrado, em áreas com condições edáficas

favoráveis (Silva and Bates 2002; Werneck and Colli 2006).

3

De acordo com a hipótese do Arco Pleistocênico, as FTSSs atualmente representam uma

pequena porção do contínuo que já foram um dia (Pennington, et al. 2004). Esse contínuo é

conhecido como Arco Pleistocênico (Prado and Gibbs 1993), e atualmente está fragmentado e

possui três grandes núcleos: Caatinga (Brasil), Missiones (rio Paraguai-Paraná) e Montes

subandinos (Bolívia e Argentina) (Pennington, et al. 2000; Prado 2000). A suposta formação do

Arco Pleistocênico teria possibilitado a difusão de espécies endêmicas das FTSSs.

Adicionalmente, a ruptura do arco em manchas isoladas teria assim produzido condições

favoráveis à especiação alopátrica, aumentando a diversidade de espécies endêmicas das FTSSs

(Pennington, et al. 2000). Evidências sugerem que o Arco Pleistocênico tenha alcançado

tamanho máximo durante o último máximo glacial no Pleistoceno (Prado and Gibbs 1993).

Entretanto, Werneck, et al. (2011) sugerem que ele possa ter ocorrido anteriormente, entre o final

do Plioceno e o início do Pleistoceno.

Outra hipótese de diversificação para a Caatinga envolve a atuação dos grandes rios.

Os rios são considerados importantes agentes de especiação, atuando, por exemplo, como uma

barreira ao fluxo gênico entre populações (Garda and Cannatella 2007; Ribas, et al. 2012;

Wallace 1854; Werneck, et al. 2015). O Rio São Francisco é o rio perene de maior extensão da

Caatinga e é conhecido por estruturar populações e até mesmo espécies em suas margens opostas

(Faria, et al. 2013; Passoni, et al. 2008; Rodrigues 2003). Evidências geomorfológicas indicam

que o paleocurso desse rio diferia consideravelmente do curso atual, sendo responsável pela

diversificação de algumas espécies (Nascimento, et al. 2013; Werneck, et al. 2015). As paleo-

dunas do Médio do São Francisco, formadas durante o Quaternário, contém um grande número

de espécies de lagartos endêmicos, muitos dos quais restritos a uma ou outra margem do Rio São

Francisco (Rodrigues 2003; Werneck, et al. 2012; Werneck, et al. 2015). Todos esses fatores

4

reforçam o papel do Rio São Francisco como uma das mais importantes barreiras para

diversificação de espécies na Caatinga.

Assim, tanto a formação e a fragmentação do Arco Pleistocênico como a presença do

Rio São Francisco e suas mudanças de curso ao longo do tempo devem ter contribuído para a

formação e a manutenção da biodiversidade da Caatinga. O presente trabalho teve como objetivo

testar essas hipóteses por meio de técnicas filogeográficas. Para tanto, nós utilizamos um gênero

de lagarto da família Gekkonidae, uma família cosmopolita composta por lagartos pequenos e

com pele muito delicada, onde a maioria das espécies possui hábitos noturnos (Vanzolini, et al.

1980). Das 1103 espécies dessa família, apenas oito ocorrem no Brasil, duas delas pertencendo

ao gênero utilizado aqui, Lygodactylus (Uetz 2016).

Lygodactylus contém aproximadamente 64 espécies, das quais apenas duas ocorrem na

América do Sul (Gamble et al., 2011). Lygodactylus wetzeli (Smith, Martin & Swain, 1977)

ocorre no Brasil no estado do Mato Grosso do Sul e também no Paraguai e na Bolívia (Smith, et

al. 1977; Uetz 2016), apresentando distribuição restrita ao Chaco (Vanzolini 1974, 1976).

Enquanto isso, Lygodactylus klugei (Smith, Martin & Swain, 1977) ocorre no Brasil, é endêmico

da Caatinga, mas até então poderia ser encontrado também em enclaves de FTSS no Cerrado, no

estado de Goiás, considerado assim endêmico das FTSS (Rodrigues 2003; Werneck and Colli

2006). Ambas espécies são de pequeno porte (podendo atingir pouco mais de 5cm), diurnas,

arborícolas e insetívoras (Galdino, et al. 2011; Smith, et al. 1977).

Nós utilizamos as espécies Sul-Americanas de Lygodactylus para compreender os

processos responsáveis pela diversificação na diagonal de formações abertas, mais

especificamente testando: (i) a hipótese do Arco Pleistocênico; e (ii) o papel do rio São Francisco

como barreira. A dissertação foi dividida em dois capítulos. No primeiro, buscamos entender

5

melhor a relação do gênero Lygodactylus no continente Sul-Americano e investigar (i) se o

tempo de divergência entre L. klugei da Caatinga e dos enclaves de FTSS no Cerrado

corresponde à época em que teria ocorrido a fragmentação do Arco Pleistocênico, (ii) se os

Lygodactylus na América do Sul representam um complexo de espécies, e (iii) se esse grupo é

monofilético indicando um único evento de colonização a partir das linhagens africanas. No

segundo capítulo focamos na Caatinga e utilizando o L. klugei, restrito à Caatinga, buscamos (i)

compreender os processos que levaram à diversificação dessa espécie, e (ii) o papel do rio São

Francisco para a estruturação populacional e genealógica.

Referências bibliográficas

Avise JC. 2000. The History and Formation of Species. Harverd University press, Cambridge,

Massachusetts.

Carnaval AC, Hickerson MJ, Haddad CFB, Rodrigues MT, Moritz C 2009. Stability predicts

genetic diversity in the Brazilian Atlantic Forest hotspot. Science (Washington D C) 323:

785–789.

Faria MB, Nascimento FF, Oliveira JAd, Bonvicino CR 2013. Biogeographic determinants of

genetic diversification in the mouse opossum Gracilinanus agilis (Didelphimorphia:

Didelphidae). Journal of Heredity 104: 613–626.

Fitzpatrick SW, Brasileiro CA, Haddad CFB, Zamudio KR 2009. Geographical variation in

genetic structure of an Atlantic Coastal Forest frog reveals regional differences in habitat

stability. Molecular Ecology 18: 2877–2896.

6

Galdino CAB, Passos DC, Zanchi D, Bezerra CH 2011. Lygodactylus klugei (NCN). Sexual

dimorphism, habitat, diet. Herpetological Review 42: 275–276.

Gamble, T., Bauer, A.M., Colli, G.R., Greenbaum, E., Jackman, T.R., Vitt, L.J., Simons, A.M.,

2011. Coming to America: multiple origins of New World geckos. Journal of Evolutionary

Biology 24, 231–244.

Gamble T, Colli GR, Rodrigues MT, Werneck FP, Simons AM 2012. Phylogeny and cryptic

diversity in geckos (Phyllopezus; Phyllodactylidae; Gekkota) from South America's open

biomes. Molecular Phylogenetics and Evolution 62: 943-953.

Garda AA, Cannatella DC 2007. Phylogeny and biogeography of paradoxical frogs (Anura,

Hylidae, Pseudae) inferred from 12S and 16S mitochondrial DNA. Molecular

Phylogenetics and Evolution 44: 104–114. doi: 10.1016/j.ympev.2006.11.028

Magalhaes ILF, Oliveira U, Santos FR, Vidigal THDA, Brescovit AD, Santos AJ 2014. Strong

spatial structure, Pliocene diversification and cryptic diversity in the Neotropical dry forest

spider Sicarius cariri. Molecular Ecology 23: 5323-5336. doi: 10.1111/mec.12937

Mares MA, Willig MR, Steilein KE, Lacher-Jr TE 1981. The mammals of northeastern Brazil: a

preliminary assessment. Annals of Carnegie Museum 50: 81–137.

Murphy PG, Lugo AE 1986. Ecology of Tropical Dry Forest. Annual Review of Ecology and

Systematics 17: 67–88.

Nascimento FF, Lazar A, Menezes AN, Durans AdM, Moreira JC, Salazar-Bravo J, D'Andrea

PS, Bonvicino CR 2013. The role of historical barriers in the diversification processes in

7

open vegetation formations during the Miocene/Pliocene using an ancient rodent lineage as

a model. PLoS One 8: 1–13.

Passoni JC, Benozzati ML, Rodrigues MT 2008. Phylogeny, species limits, and biogeography of

the Brazilian lizards of the genus Eurolophosaurus (Squamata: Tropiduridae)

as inferred from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 46:

403–414.

Pennington RT, Lavin M, Oliveira-Filho A 2009. Woody plant diversity, evolution, and ecology

in the tropics: perspectives from Seasonally Dry Tropical Forests. Annual Review of

Ecology, Evolution, and Systematics 40: 437–457. doi:

10.1146/annurev.ecolsys.110308.120327

Pennington RT, Lavin M, Prado DE, Pendry CA, Pell SK, Butterworth CA 2004. Historical

climate change and speciation: neotropical seasonally dry forest plants show patterns of

both Tertiary and Quaternary diversification. Philosophical Transactions of the Royal

Society of London B Biological Sciences 359: 515–537.

Pennington RT, Lewis GP, Ratter JA. 2006. An Overview of the Plant Diversity, Biogeography

and Conservation of Neotropical Savannas and Seasonally Dry Forests. In. Neotropical

Savannas and Seasonally Dry Forests: Plant Diversity, Biogeography, and Conservation. p.

1–30.

Pennington RT, Prado DE, Pendry CA 2000. Neotropical seasonally dry forests and Quaternary

vegetation changes. Journal of Biogeography 27: 261–273.

8

Prado DE. 2003. As Caatingas da América do Sul. In: Leal IR, Tabarelli M, da Silva JMC,

editors. Ecologia e Conservação da Caatinga. Editora Universitária UFPE. p. 3–73.

Prado DE 2000. Seasonally dry forests of tropical South America: from forgotten ecosystems to

a new phytogeographic unit. Edinburgh Journal of Botany 57: 437–461. doi:

10.1017/S096042860000041X

Prado DE, Gibbs PE 1993. Patterns of species distributions in the dry seasonal forests of South

America. Annals of the Missouri Botanical Garden 80: 902–927.

Recoder RS, Werneck FDP, Teixeira M, Jr., Colli GR, Sites JW, Jr., Rodrigues MT 2014.

Geographic variation and systematic review of the lizard genus Vanzosaura (Squamata,

Gymnophthalmidae), with the description of a new species. Zoological Journal of the

Linnean Society 171: 206–225.

Ribas CC, Maldonado-Coelho M, Smith BT, Cabanne GS, d'Horta FM, Naka LN 2012. Towards

an integrated historical biogeography of the neotropical lowland avifauna: combining

diversification analysis and landscape evolution. Ornitologia Neotropical 23: 187–206.

Riddle BR, Dawson MN, Hadly EA, Hafner DJ, Hickerson MJ, Mantooth SJ, Yoder AD 2008.

The role of molecular genetics in sculpting the future of integrative biogeography. Progress

in Physical Geography 32: 173–202. doi: 10.1177/0309133308093822

Rodrigues MT. 2003. Herpetofauna da Caatinga. In: Leal IR, Tabarelli M, Silva JMC, editors.

Ecologia e Conservação da Caatinga. Brasil: Universidade Federal de Pernambuco. p. 181–

236.

9

Shepard DB, Burbrink FT 2008. Lineage diversification and historical demography of a sky

island salamander, Plethodon ouachitae, from the Interior Highlands. Molecular Ecology

17: 5315–5335. doi: 10.1111/j.1365-294X.2008.03998.x

Silva JMC, Bates JM 2002. Biogeographic patterns and conservation in the South American

Cerrado: a tropical savanna hotspot. Bioscience 52: 225–233.

Smith HM, Martin RL, Swain TA 1977. A new genus and two new species of south american

geckos (Reptilia: Lacertilia). Papeis Avulsos de Zoologia 30: 195–213.

Thomé MTC, Zamudio KR, Giovanelli JGR, Haddad CFB, Baldissera Jr. FA, Alexandrino J

2010. Phylogeography of endemic toads and post-Pliocene persistence of the Brazilian

Atlantic Forest. Molecular Phylogenetics and Evolution 55: 1018–1031. doi:

10.1016/j.ympev.2010.02.003

Turchetto-Zolet AC, Pinheiro F, Salgueiro F, Palma-Silva C 2013. Phylogeographical patterns

shed light on evolutionary process in South America. Molecular Ecology 22: 1193–1213.

doi: 10.1111/mec.12164

Uetz P 2016. The Reptile Database. Disponível em http://www.reptile-database.org.

Vanzolini PE. 1988. Distributional Patterns of South American Lizards. In: Heyer WR,

Vanzolini PE, editors. Proceedings of a Workshop on Neotropical Distribution Patterns.

Academia Brasileira de Ciências, Rio de Janeiro, Brasil. p. 317–342.

Vanzolini PE 1974. Ecological and geographical distribution of lizards in Pernambuco,

Northeastern Brasil (Sauria). Papeis Avulsos de Zoologia 18: 61–90.

10

Vanzolini PE 1976. On the lizards of a cerrado-caatinga contact: evolutionary and

zoogeographical implications (Sauria). Papeis Avulsos de Zoologia 29: 111–119.

Vanzolini PE, Ramos-Costa AMM, Vitt LJ. 1980. Repteis da Caatinga. Academia Brasileira de

Ciências, Rio de Janeiro.

Wallace AR 1854. On the Monkeys of the Amazon. Journal of Natural History Series 2 14: 451–

454. doi: 10.1080/037454809494374

Werneck FP 2011. The diversification of eastern South American open vegetation biomes:

Historical biogeography and perspectives. Quaternary Science Reviews 30: 1630–1648.

doi: 10.1016/j.quascirev.2011.03.0 09

Werneck FP, Colli GR 2006. The lizard assemblage from seasonally dry tropical forest enclaves

in the Cerrado biome, Brazil, and its association with the Pleistocenic Arc. Journal of

Biogeography 33: 1983–1992.

Werneck FP, Costa GC, Colli GR, Prado DE, Sites JW, Jr. 2011. Revisiting the historical

distribution of Seasonally Dry Tropical Forests: new insights based on palaeodistribution

modelling and palynological evidence. Global Ecology and Biogeography 20: 272–288.

Werneck FP, Gamble T, Colli GR, Rodrigues MT, Sites JW, Jr. 2012. Deep diversification and

long-term persistence in the South American 'Dry Diagonal': integrating continent-wide

phylogeography and distribution modeling of geckos. Evolution 66: 3014–3034.

11

Werneck FP, Leite RN, Geurgas SR, Rodrigues MT 2015. Biogeographic history and cryptic

diversity of saxicolous Tropiduridae lizards endemic to the semiarid Caatinga. BMC

Evolutionary Biology 15: 1–24.

12

– CHAPTER I –

OUT OF AFRICA: A CRYPTIC SPECIATION

HISTORY OF A SMALL TRAVELER GECKO IN

SOUTH AMERICA

Manuscript to be submitted to Molecular Phylogenetics and Evolution

13

Out of Africa: a cryptic speciation history of a small traveler gecko in South America

Flávia M. Lanna1*, Fernanda P. Werneck2, Marcelo Gehara3, Emanuel M. Fonseca1, Guarino R.

Colli4, Jack W. Sites Jr5, Miguel T. Rodrigues6, Adrian A. Garda7

1 Programa de Pós-Graduação em Ecologia, Universidade Federal do Rio Grande do

Norte, Campus Universitário, Lagoa Nova, 59078-900, Natal, RN, Brazil.

2 Coordenação de Biodiversidade, Programa de Coleções Científicas Biológicas,

Instituto Nacional de Pesquisas da Amazônia (INPA), 69067–375, Manaus, Amazonas, Brazil.

3 American Museum of Natural History, Department of Herpetology, 79th St. Central

Park West, New York, NY 10024.

4 Departamento de Zoologia, Universidade de Brasília, 70910–900 Brasília, DF, Brazil.

5 Department of Biology and Bean Life Science Museum, Brigham Young University,

Provo, UT 84602.

6 Departamento de Zoologia, Instituto de Biociências, Universidade de São Paulo,

05508–090, São Paulo, SP, Brazil.

7 Departamento de Botânica e Zoologia, Centro de Biociências, Universidade Federal do

Rio Grande do Norte, Campus Universitário, Lagoa Nova, 59078-900, Natal, RN, Brazil.

* Corresponding author. E-mail address: flaviamollanna@gmail.com (F.M. Lanna)

14

Abstract

The Pleistocenic Arc Hypothesis (PAH) posits that South American Seasonally Dry Tropical

Forests (SDTF) were interconnected during Pleistocene glacial periods, enabling the expansion

of species ranges that were subsequently fragmented, promoting speciation. The lizard genus

Lygodactylus occurs in Africa, Madagascar, and South America and represents an interesting

case of unbalanced diversity and distribution across these continents. While many African

Lygodactylus species are recognized as complexes of cryptic species, the only two species

described in South America have a disjoint distribution occurring in SDTFs and the Chaco

biomes. We use a phylogenetic approach based on mitochondrial (ND2) and nuclear (RAG-1)

markers and a sampling encompassing the known range of South American Lygodactylus to

investigate if: (i) divergence timing of L. klugei is congruent with the PAH; (ii) the species

currently recognized correspond to species complexes; and (iii) South American Lygodactylus

are monophyletic. Species delimitation analysis suggested the existence of five species, two of

which correspond to the already described taxa, and three representing new candidate species.

Divergence times among L. klugei and the other species endemic to the SDTFs were not

congruent with the PAH. However, fragmentation of the once broader and continuous SDTFs

likely influenced the divergence of L. wetzeli and Lygodactylus sp. 3 (from São Domingos

region). Our molecular results corroborate the monophyly of South American Lygodactylus.

Keywords: Caatinga, Chaco, dwarf gecko, Lygodactylus, monophyly, Pleistocenic Arc

Hypothesis, species complex, STDF.

15

1. Introduction

The Neotropics are considered one of the most diverse regions on Earth (Antonelli and

Sanmartín, 2011; Myers et al., 2000), and different diversification processes were responsible for

this great biodiversity through time (Rull, 2011). The interest to understand diversification

drivers increased during the past decades with the increase of molecular-based biogeographic

approaches (Riddle et al., 2008). Studies aimed to unveil the processes responsible for

Neotropical biodiversity patterns have mostly focused on wet biomes, such as the Atlantic Forest

and the Amazon rainforest (Turchetto-Zolet et al., 2013). Still, a large portion of the region is

covered by the less studied open formations, which have also been subject to some of the same

evolutionary forces that drove speciation on forested areas and other regional determinants

(Turchetto-Zolet et al., 2013; Werneck, 2011). The glaciation periods during the Pleistocene

likely promoted expansion of the open areas, connecting otherwise isolated fragments (Prado and

Gibbs, 1993). The current fragmented distribution of Seasonally Dry Tropical Forests (SDTFs),

for example, has been used as evidence for a previously more widespread distribution of this

dominium (Prado, 1991; Prado and Gibbs, 1993).

The SDTFs are forest formations that occur in frost-free tropical regions, marked by a

highly seasonal rainfall and severe droughts (less then 1800 mm/year) (Murphy and Lugo, 1986).

They usually occur on fertile soils with low levels of aluminum and moderate to high pH

(Pennington et al., 2006). Most of the vegetation is deciduous, loosing more than half of arboreal

cover during the dry season (Murphy and Lugo, 1986; Pennington et al., 2006). SDTFs have a

patched distribution throughout the Neotropical region, and in South America the largest

remaining areas represent three nuclei: Caatinga (northeastern Brazil), Misiones (along

Paraguay-Paraná rivers), and Piedmont (northwestern Argentina and southwestern Bolivia)

16

(Prado and Gibbs, 1993; Pennington et al., 2000). Some small and isolated patches occur in

favorable soil conditions, as enclaves within the Cerrado savanna biome in central Brazil (Silva

and Bates, 2002; Werneck and Colli, 2006).

The Caatinga biome (the largest nucleus of SDTF in South America) was initially

considered to have low diversity, no characteristic fauna and no endemic species, sharing its

biota with the two other biomes of the diagonal of open formations - Cerrado and Chaco (Mares

et al., 1981; Vanzolini, 1974, 1976, 1988). Currently, the idea of the Caatinga as a poor biome

was abandoned and the number of recognized endemic species has increased in the past years

(Carvalho et al., 2016; Leal et al., 2003; Recoder et al., 2014). This erroneous interpretation was

based in unrepresentative collections and insufficient geographic sampling in this biome

(Rodrigues, 2003).

The Pleistocenic Arc Hypothesis (PAH) posits that the disjoint distribution of present-

day SDTFs results from the fragmentation of a previously more extensive and uninterrupted

formation that reached its maximum extension during the dry-cool Last Glacial Maximum

(LGM) period of late Pleistocene (Prado, 1991; Prado and Gibbs, 1993). The SDTFs would have

then retracted during subsequent humid-warm periods causing allopatric speciation by

vicariance, what would explain the presence of endemic species on remaining patches

(Pennington et al., 2000). However, environmental niche models projected to the past did not

recovered an expansion of SDTFs during LGM (Werneck et al., 2011). Thus, a different time of

expansion was proposed, where the Pleistocenic Arc would have occurred during the early

Pliocene/lower Pleistocene, with the fragmentation of SDTFs occurring before the LGM

(Werneck et al., 2011). Indeed, the few papers using molecular data to test the PAH found that

17

plant species isolated in SDTF patches diverged before the LGM (Caetano et al., 2008; Collevatti

et al., 2012; Pennington et al., 2004).

The genus Lygodactylus comprises 64 species, 62 occurring in Africa and Madagascar

and two in South America (Uetz, 2016). These Gekkonidae lizards are small, cryptic, arboreal,

and diurnal (Galdino et al., 2011; Vitt, 1995). Although African Lygodactylus had received much

attention in the past years (Castiglia and Annesi, 2011; Malonza et al., 2016; Mezzasalma et al.,

2017; Puente et al., 2005; Röll et al., 2010; Travers et al., 2014), South American Lygodactylus

have been poorly studied, and no molecular study involving the two species was ever performed.

South American Lygodactylus have a disjoint distribution and similar morphologies.

Lygodactylus wetzeli (Smith, Martin & Swain, 1977) occurs in the State of Mato Grosso do Sul

in Brazil, Paraguay, and Bolivia, and is mostly restricted to the Chaco (Uetz, 2016; Vanzolini,

1974, 1976), a biome with strong seasonality (high temperatures on summer and frosts on

winter), compact soils with poor drainage, rainfall from 1000 mm/year to 500 mm/year and

floods during the summer (Pennington et al., 2000; Prado, 1993). Lygodactylus klugei (Smith,

Martin & Swain, 1977) has a widespread distribution in the Caatinga biome (Rodrigues, 2003;

Smith et al., 1977), also occurring on enclaves of SDTF inserted in the Cerrado biome (the

Brazilian savannas) (Werneck and Colli, 2006). The presence of L. klugei on SDTFs enclaves

and its absence on the Cerrado biome may indicate previous connections between the Caatinga

and SDTFs enclaves. Indeed, this species was considered endemic of SDTFs of the Pleistocenic

Arc (Werneck and Colli, 2006). Many of the African Lygodactylus formed complexes of cryptic

species before being properly investigated (Malonza et al., 2016; Portik et al., 2013; Röll, 2005;

Röll et al., 2010; Travers et al., 2014). Likewise, many lizards from the Caatinga are now

18

considered part of cryptic species complexes (Oliveira et al., 2015; Recoder et al., 2014;

Werneck et al., 2012; Werneck et al., 2015).

Herein we investigate the potential cryptic diversity of South American Lygodactylus

and the possible influence of the Pleistocenic Arc in its diversification. We used a molecular

phylogenetic approach to test the following hypotheses: (i) the diversification between L. klugei

from Caatinga and L. klugei from SDTFs enclaves will agree with the time originally (LGM) or

subsequently (early Pliocene–late Pleistocene) suggested for the PAH; (ii) L. klugei corresponds

to a complex of cryptic species; (iii) South American Lygodactylus are monophyletic.

2. Material and Methods

2.1 Taxon sampling

We sequenced a total of 25 individuals of South American Lygodactylus encompassing

eight localities: one in the Chaco, one in a SDTFs enclave within the Cerrado biome, and six in

the Caatinga (Figure 1). These individuals are deposited in three zoological collections: Coleção

Herpetológica da Universidade Federal do Rio Grande do Norte (UFRN), Coleção Herpetológica

da Universidade de Brasilia (CHUNB), and Coleção Herpetológica do Museu de Zoologia da

Universidade de São Paulo (MZUSP). To improve species tree calibration and test

monophyletism in this group, we used sequences of 18 African Lygodactylus species available in

GenBank (Table 1).

2.2 Sequencing

We extracted genomic DNA from muscle, liver, finger or tail tissues preserved in 95–

100% ethanol, with a DNA Purification Kit (Wizard®, Promega). We amplified ND2

19

(mitochondrial marker - mtDNA) and RAG1 (nuclear marker - nuDNA) using a standard

polymerase chain reaction (PCR) technique. We chose these markers based on published

phylogenetic data for Lygodactylus (Travers et al., 2014). For PCR protocols and markers details

see Supporting Information (Table S1). We purified PCR products using polyethylene glycol

(PEG 8000), prepared sequencing reactions using BigDye terminator kit v.3.1 (Applied

Biosystems), precipitated the products with EDTA/Ethanol, and produced sequences using an

ABI 3130xl sequencer (Applied Biosystems) at INPA Sequencing Center (Laboratório Temático

de Biologia Molecular, Instituto Nacional de Pesquisas da Amazônia, Manaus/AM, Brazil). We

assembled, edited for ambiguous bases, and aligned sequences using Muscle algorithm (Edgar,

2004) in Geneious v8.1.7 (Biomatters). We used PHASE v2.1.1 (Stephens et al., 2001; Stephens

and Wiens, 2003) to determine the pair of alleles with higher probability for nuclear sequences.

2.3 Phylogenetic relationships

To recover phylogenetic relationships of South American Lygodactylus we inferred a

Maximum Likelihood (ML) concatenated tree combining both markers using a Randomized

Axelerated Maximum Likelihood approach in RAxML v7.2.6 (Stamatakis, 2014). The same

model of nucleotide substitution (GTR+Gamma) was assigned for both markers, as this is the

option the program offers. Two hundred independent searches and one thousand bootstrap

replicates were used to assess nodal support.

2.4 Divergence time estimates

To infer divergence times among South American Lygodactylus we considered the

concatenated tree topology that separated samples in five groups (see Results, Figure 2). We

20

expected a divergence time among the sister African Lygodactylus and the South American

Lygodactylus to be of approximately 25 million years (My) (Gamble et al., 2011), and the

diversification timing within the South American Lygodactylus to fall within the Pleistocene if a

Pleistocene Arc fragmentation promoted speciation of this genus. We estimated a time calibrated

Bayesian species tree using *Beast implemented in BEAST 1.8.2 (Drummond et al., 2012) using

the five groups recovered by the gene tree as species assignments (See Results). We calibrated

the *Beast tree using a mitochondrial mutation rate of 1.15% per million years as suggested for

geckos and other lizards (Arnold et al., 2008). We conducted three independent runs of 3x108

generations, sampling every 3x104 steps, with a Yule speciation process prior and an

uncorrelated lognormal relaxed clock. To calibrate the molecular clock, we set up mtDNA

ucld.mean parameter using normal prior (mean: 0.0115; standard deviation: 0.002). In addition,

for the nuDNA ucld.mean we used a default gamma prior and for ucld.stdev we used an

exponential prior (mean: 0.5). We checked convergence among runs and effective sample sizes

above 200 using Tracer v1.6 (Drummond and Rambaut, 2007). We used TreeAnnotator

(Drummond et al., 2012) to calculate a maximum clade credibility tree, excluding the first one

thousand trees as burn-in.

2.5 Species delimitation

In order to test if groups recovered with high support in the concatenated gene tree fit as

candidate species, we used a species delimitation analysis, with SpedeSTEM 2 (Ence and

Carstens, 2010). SpedeSTEM estimates a ML species tree from mitochondrial and nuclear gene

trees to compute AIC (Akaike Information Criterion) scores of different models based on a line

of evidence (e.g. molecular). We generated gene trees required for this analysis in BEAST 1.8.2

21

(Drummond et al., 2012) and used JModelTest 2 (Darriba et al., 2012) to determine the

nucleotide substitution models. We calculated nucleotide diversity for each gene and

subsequently the mean between them as an approximation of θ (Theta). To account for a possible

influence of theta on incomplete lineage sorting, we used a variation of θ/4 and 4θ of theta

values. We tested 52 models, with k (number of species) varying from 1 to 5. For this analysis

we used only South American species and Lygodactylus angularis as an outgroup based on the

phylogenetic relationships with South American group (see Supporting information, Figure S1).

We also estimated genetic distances among groups as the uncorrected p-distance for both

mitochondrial and nuclear markers using Mega 7.0 (Kumar et al., 2016).

3. Results

3.1 Genetic data and phylogenetic analyses

We obtained sequences from 67 specimens (25 for South American and 42 for African

Lygodactylus). Final aligned sequences comprise 540 base pairs (bp) for ND2 and 302 bp for

RAG1.

The ML concatenated tree recovered with high bootstrap support the monophyly of

Lygodactylus from South America (Figure 2). Five major groups were identified within South

American Lygodactylus: L. klugei (encompassing L. klugei from four sampling localities in the

Caatinga), Lygodactylus Santo Inácio (population from north Bahia State, Brazil), Lygodactylus

Condeúba (population from south Bahia State, Brazil), L. wetzeli (encompassing L. wetzeli

individuals from Paraguay) and Lygodactylus São Domingos (population from STDF enclaves

within the Cerrado in São Domingos, Goiás State, Brazil; previously identified as L. klugei).

22

3.2 Divergence times and species delimitation

The monophyly of South American Lygodactylus observed in the concatenated tree

was confirmed by the Bayesian species tree (Figure 3). Divergence between African and South

American species of Lygodactylus occurred at 29 Mya (18–44 My HPD interval) (Figure 3).

Clades within South American Lygodactylus also had high nodal support values (all posterior

probabilities > 0.99). The divergence time among these groups varied from 1.9 Mya to 22.8 Mya

(Figure 3). SpedeSTEM analysis recognized five species as the most likely model under three

values of θ (θ=0.05, 0.2 and 0.8) and the genetic distances among these groups varied from 3.8 –

33.3% for ND2 and 0.3 – 3% for RAG1 (Table 2). Taking into consideration the genetic

distances and divergence times estimated between the closest groups and the results from the

species delimitation analysis, we suggest three new candidate species of Lygodactylus in South

America: Lygodactylus sp. 1 (Santo Inácio), Lygodactylus sp. 2 (Condeúba) and Lygodactylus sp.

3 (São Domingos).

4. Discussion

4.1 Monophyly of Lygodactylus in South America

Our results corroborate the monophyly and existence of cryptic species in South

American Lygodactylus. Previous phylogenies involving this group are scarce and have used

only one individual of L. klugei (Gamble et al., 2011; Pyron et al., 2013). Because no sample of

L. wetzeli or the candidate species identified herein were included in such analyses, it was

impossible to ascertain if South American species were monophyletic in respect to African

species. If not, the region either experienced multiple colonization events or some species would

have arrived before the separation of Africa and South America.

23

Although L. klugei and L. wetzeli are not sister species, the South American clade is

monophyletic, corroborating a single colonization event of South America from African

Lygodactylus. Gamble et al. (2011) suggested that the first lineage of Lygodactylus reached

South America by a trans-Atlantic dispersal around 25 Mya. The divergence time between South

American and African Lygodactylus estimated using all five species identified was around 29

Mya (HPD 95%: 18 – 44 Mya), corroborating the hypothesis of Gamble et al. (2011).

4.2 Pleistocenic Arc Hypothesis

Lygodactylus klugei was considered to occur across the Caatinga biome and SDTFs

enclaves within Cerrado. For such reason, it was considered an appropriate taxon to test the

Pleistocenic Arc Hypothesis (Werneck and Colli, 2006). However, based on our species

delimitation results, we suggest that Lygodactylus from SDTFs enclaves is actually a new species

(Lygodactylus sp. 3), and its divergence time does not support the PAH. Divergence times

among candidate species and L. klugei occurred during the Miocene (at least 8 Mya), before the

hypothesized existence of the Pleistocenic Arc (Prado and Gibbs, 1993; Werneck et al., 2011).

The diagonal of open formations extends from northeastern Brazil to northwestern

Argentina, encompassing three South American biomes: Caatinga, Cerrado, and Chaco (Mayle,

2004; Pennington et al., 2000). An historical connection between Caatinga and Chaco has been

previously suggested (Vanzolini, 1974), but subsequent studies have shown that SDTFs endemic

species are different from Chaco endemics, and hence excluded the Chaco from the Pleistocenic

Arc (Colli, 2005; Pennington et al., 2000; Prado, 1993; Prado and Gibbs, 1993). Also,

palaeodistribution modeling of the SDTFs never predicted stable areas of SDTFs in areas

currently occupied by the Chaco (Werneck et al., 2011). Conversely, phylogenetic relationships

24

recovered in our study suggest a historical connection between lineages/species from the Chaco

and SDTFs enclaves. The divergence time between Lygodactylus sp. 3 (São Domingos) and L.

wetzeli dates to the late Pleistocene. The original hypothesis suggested the occurrence of the

Pleistocene Arc during the Last Glacial Maximum (Prado, 1991; Prado and Gibbs, 1993). Based

on paleomodeling, Werneck et al. (2011) suggested that the Pleistocene Arc might have existed

in the early Pliocene/lower Pleistocene, and that SDTFs were fragmented during the LGM. Our

results match with the timing suggested by Werneck et al. (2011) for the Pleistocenic Arc.

Few studies dated divergence times to test the PAH based on SDTF’s endemic taxa

(Collevatti et al., 2012; Pennington et al., 2004). Divergence times similar to the one recovered

here were found for SDTFs endemic trees (Pennington et al., 2004). Authors concluded that the

fragmentation of SDTFs that once formed the Pleistocene Arc probably had no influence on the

biogeographic pattern detected. Nevertheless, they did not rule out possible effects of Pleistocene

climatic changes on speciation within the different SDTF nuclei (Pennington et al., 2004).

4.3 Cryptic diversity

We propose the existence of three undescribed species of Lygodactylus in South

America. Cryptic species have also been identified for other lizard species within the open

diagonal (Domingos et al., 2014; Guarnizo et al., 2016; Recoder et al., 2014; Rodrigues, 2003;

Werneck et al., 2015). Indeed, this pattern seems to be recurrent, and the underestimation of the

biodiversity of the open diagonal is becoming even more evident, especially for the Caatinga.

Indeed, according to a phylogeography study of L. klugei (FML, in prep.), two

candidate species and L. klugei are endemic or occur mostly in the Caatinga. Once treated as a

SDTF endemic (Werneck and Colli, 2006), L. klugei is mostly restricted to the Caatinga, with

25

some localities in neighbor biomes such as the Atlantic Forest (our sample 1, from Espírito

Santo, Rio Grande do Norte, for example). Lygodactylus sp. 2 is known only from Condeúba in

southwestern Bahia State, northeast Brazil. This species occurs in microhabitats similar to other

Lygodactylus from the Caatinga, albeit at higher elevations (680 m against 15–550 m for L.

klugei across most of the Caatinga). The Caatinga of Southern Bahia is poorly studied, and

further fieldwork is necessary to clarify the distribution of this new species and potential

processes involved in the diversification of Lygodactylus and perhaps other species.

Quaternary sand dunes of São Francisco River (SFR - the largest perennial river in the

Caatinga) are a key center of endemism for Caatinga vertebrates (Barreto et al., 2002; Lencioni-

Neto, 1994; Nascimento et al., 2013; Rocha, 1995; Rodrigues and Juncá, 2002). The sand dunes

are desert-like formations that have a characteristic composition, making it a very different

landscape from the rest of the Caatinga (Rodrigues, 1996). Accordingly, Lygodactylus sp. 1 is

apparently endemic to one of these Quaternary dunes which, together, contain impressive levels

of endemism for squamates in the Caatinga (Passoni et al., 2008; Rodrigues, 1996; Rodrigues,

2003; Werneck et al., 2015). Approximately 37% of lizards and amphisbaenians and 16% of

snakes found in the Caatinga are endemic to these sand dunes (Rodrigues, 2003). Although such

paleodunes were dated to the Quaternary, their size suggest that the semiarid condition in this

area may date back to the Tertiary (Barreto et al., 2002). Despite this uniqueness, the SFR sand

dunes are still unprotected by formal protected areas.

Despite the occurrence of Lygodactylus in SDTFs enclaves, this genus is absent from

the adjacent Cerrado biome (Colli, 2005; Werneck and Colli, 2006). Indeed, the Central

Brazilian Plateau (CBP) geographically separates L. klugei and L. wetzeli. A previous hypothesis

suggested that the uplift of the CBP was responsible for a vicariant speciation in Lygodactylus,

26

followed by differentiation in the Caatinga and Chaco, and subsequent extinction in Cerrado

(Vanzolini, 1963). Our results show that L. klugei and L. wetzeli are not sister species and their

divergence time is older than the uplift of the CBP. Lygodactylus sp. 3 (São Domingos) shares a

more recent common ancestor with L. wetzeli and has diverged more recently than the uplift of

the CBP. Hence, there is no evidence for a role of the CBP on the diversification of

Lygodactylus.

5. Conclusions

South American Lygodactylus is monophyletic and originated from a single

colonization event into the New World from an African ancestor around 29 Mya. South

American Lygodactylus represents a complex of cryptic species, including three endemic or

associated to the Caatinga species (L. klugei, Lygodactylus sp. 1, and Lygodactylus sp. 2), one

related to the Chaco (L. wetzeli), and one species endemic to a SDTFs enclave within Cerrado

(Lygodactylus sp. 3). Although there is no evidence that the PAH accounts for the diversification

of L. klugei, we suggest an influence of SDTFs fragmentation on the split of the ancestor of L.

wetzeli and Lygodactylus sp. 3.

Acknowledgments

We are grateful to researchers that worked in collaboration with us at AAG, GRC, and

MTR laboratories for help with fieldwork and samples donation. We also thank Felipe

Magalhães and Willianilson Pessoa for help with fieldwork. We are grateful to Erik Choueri for

laboratorial assistance. FML and EMF thanks Conselho Nacional de Desenvolvimento Científico

e Tecnológico (CNPq) for their graduate scholarships. FPW thanks National Science Foundation

27

DDIG award (DEB-1210346), Science Without Borders Program from CNPq (#374307/2012-1),

CAPES/Fulbright (#15073722–2697/06–8), and the Partnerships for Enhanced Engagement in

Research (PEER) program for financial support. GRC thanks Coordenação de Apoio à Formação

de Pessoal de Nível Superior (CAPES), CNPq, Fundação de Apoio à Pesquisa do Distrito

Federal (FAPDF), and PEER program for financial support. MTR thanks Fundação de Amparo à

Pesquisa do Estado de São Paulo (FAPESP 2003/10335-8 and 2011/50146-6), CNPq, and

Dimensions of Biodiversity Program [FAPESP (BIOTA, 2013/50297-0), NSF (DOB 1343578),

and NASA]. AAG thanks CNPq (563352/2010-8, 552031/2011-9, 431433/2016-0 and

457463/2012-0) and CAPES (23038.005577/2012-28 and 23038.009565/2013-53) for financial

support.

References

Antonelli, A., Sanmartín, I., 2011. Why are there so many plant species in the Neotropics? Taxon

60, 403–414.

Arnold, E.N., Vasconcelos, R., Harris, D.J., Mateo, J.A., Carranza, S., 2008. Systematics,

biogeography and evolution of the endemic Hemidactylus geckos (Reptilia, Squamata,

Gekkonidae) of the Cape Verde Islands: based on morphology and mitochondrial and

nuclear DNA sequences. Zoologica Scripta 37, 619–636.

Barreto, A.M.F., Suguio, K., Oliveira, P.E., Tatumi, S.H., 2002. Campo de Dunas Inativas do

Medio Rio Sao Francisco, BA. Sítios geológicos e paleontológicos do Brasil,

Departamento Nacional de Producao Mineral-DNPM.

28

Caetano, S., Prado, D., Pennington, R.T., Beck, S., Oliveira Filho, A.T., Spichiger, R., Naciri,

Y., 2008. The history of Seasonally Dry Tropical Forests in eastern South America:

inferences from the genetic structure of the tree Astronium urundeuva (Anacardiaceae).

Molecular Ecology 17, 3147–3159.

Carvalho, A.L.G., Sena, M.A., Peloso, P.L.V., Machado, F.A., Montesinos, R., Silva, H.R.,

Campbell, G., Rodrigues, M.T., 2016. A new Tropidurus (Tropiduridae) from the semiarid

Brazilian Caatinga: evidence for conflicting signal between mitochondrial and nuclear loci

affecting the phylogenetic reconstruction of South American collared lizards. American

Museum Novitates, 1–66.

Castiglia, R., Annesi, F., 2011. The phylogenetic position of Lygodactylus angularis and the

utility of using the 16S rDNA gene for delimiting species in Lygodactylus (Squamata,

Gekkonidae). Acta Herpetol 6, 35–45.

Collevatti, R.G., Terribile, L.C., Lima-Ribeiro, M.S., Nabout, J.C., Oliveira, G., Rangel, T.F.,

Rabelo, S.G., Diniz-Filho, J.A.F., 2012. A coupled phylogeographical and species

distribution modelling approach recovers the demographical history of a Neotropical

seasonally dry forest tree species. Molecular Ecology 21, 5845–5863.

Colli, G.R., 2005. As origens e a diversificação da herpetofauna do Cerrado. In: Scariot, A.,

Souza-Silva, J.C., Felfili, J.M. (Eds.), Cerrado: Ecologia, Biodiversidade e Conservação,

Brasília: Ministério do Meio Ambiente, pp. 247–264.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models, new

heuristics and parallel computing. Nature Methods 9, 772.

29

Domingos, F.M.C.B., Bosque, R.J., Cassimiro, J., Colli, G.R., Rodrigues, M.T., Santos, M.G.,

Beheregaray, L.B., 2014. Out of the deep: Cryptic speciation in a Neotropical gecko

(Squamata, Phyllodactylidae) revealed by species delimitation methods. Mol Phylogenet

Evol 80, 113–124.

Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by sampling trees.

BMC Evolutionary Biology 7, 214.

Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with

BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29, 1969–1973.

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high

throughput. Nucleic Acids Research 32, 1792–1797.

Ence, D.D., Carstens, B.C., 2010. SpedeSTEM: a rapid and accurate method for species

delimitation. Molecular Ecology Resources 11, 473–480.

Galdino, C.A.B., Passos, D.C., Zanchi, D., Bezerra, C.H., 2011. Lygodactylus klugei (NCN).

Sexual dimorphism, habitat, diet. Herpetological Review 42, 275–276.

Gamble, T., Bauer, A.M., Colli, G.R., Greenbaum, E., Jackman, T.R., Vitt, L.J., Simons, A.M.,

2011. Coming to America: multiple origins of New World geckos. Journal of Evolutionary

Biology 24, 231–244.

Guarnizo, C.E., Werneck, F.P., Giugliano, L.G., Santos, M.G., Fenker, J., Sousa, L.,

D'Angiolella, A.B., Dos Santos, A.R., Strussmann, C., Rodrigues, M.T., Dorado-

Rodrigues, T.F., Gamble, T., Colli, G.R., 2016. Cryptic lineages and diversification of an

endemic anole lizard (Squamata, Dactyloidae) of the Cerrado hotspot. Mol Phylogenet

Evol 94, 279–289.

30

Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: Molecular Evolutionary Genetics Analysis

Version 7.0 for Bigger Datasets. Molecular Biology and Evolution 33, 1870–1874.

Leal, I.R., Tabarelli, M., Silva, J.M.C., 2003. Ecologia e Conservação da Caatinga, Recife: Ed.

Universitária da UFPE.

Lencioni-Neto, F., 1994. Une nouvelle espèce de Chordeiles (Aves, Caprimulgidae) de Bahia

(Brésil). Alauda 62, 241–245.

Malonza, P.K., Granthon, C., Williams, D.A., 2016. A new species of dwarf gecko in the genus

Lygodactylus (squamata: Gekkonidae) from central Kenya. Zootaxa 4061, 418–428.

Mares, M.A., Willig, M.R., Steilein, K.E., Lacher-Jr, T.E., 1981. The mammals of northeastern

Brazil: a preliminary assessment. Annals of Carnegie Museum 50, 81–137.

Mayle, F.E., 2004. Assessment of the Neotropical dry forest refugia hypothesis in the light of

palaeoecological data and vegetation model simulations. Journal of Quaternary Science 19,

713–720.

Mezzasalma, M., Andreone, F., Aprea, G., Glaw, F., Odierna, G., Guarino, F.M., 2017.

Molecular phylogeny, biogeography and chromosome evolution of Malagasy dwarf geckos

of the genus Lygodactylus (Squamata, Gekkonidae). Zoologica Scripta 46, 42–54.

Murphy, P.G., Lugo, A.E., 1986. Ecology of Tropical Dry Forest. Annual Review of Ecology

and Systematics 17, 67–88.

Myers, N., Mittermeier, R.A., Mittermeier, C.G., Fonseca, G.A.B., Kent, J., 2000. Biodiversity

hotspots for conservation priorities. Nature 403, 853–858.

Nascimento, F.F., Lazar, A., Menezes, A.N., Durans, A.d.M., Moreira, J.C., Salazar-Bravo, J.,

D'Andrea, P.S., Bonvicino, C.R., 2013. The role of historical barriers in the diversification

31

processes in open vegetation formations during the Miocene/Pliocene using an ancient

rodent lineage as a model. PLoS One 8, 1–13.

Oliveira, E.F., Gehara, M., São-Pedro, V.A., Chen, X., Myers, E.A., Burbrink, F.T., Mesquita,

D.O., Garda, A.A., Colli, G.R., Rodrigues, M.T., Arias, F.J., Zaher, H., Santos, R.M.L.,

Costa, G.C., 2015. Speciation with gene flow in whiptail lizards from a Neotropical xeric

biome. Molecular Ecology 24, 5957–5975.

Passoni, J.C., Benozzati, M.L., Rodrigues, M.T., 2008. Phylogeny, species limits, and

biogeography of the Brazilian lizards of the genus Eurolophosaurus (Squamata:

Tropiduridae) as inferred from mitochondrial DNA sequences. Mol Phylogenet Evol 46,

403–414.

Pennington, R.T., Lavin, M., Prado, D.E., Pendry, C.A., Pell, S.K., Butterworth, C.A., 2004.

Historical climate change and speciation: neotropical seasonally dry forest plants show

patterns of both Tertiary and Quaternary diversification. Philosophical Transactions of the

Royal Society of London B Biological Sciences 359, 515–537.

Pennington, R.T., Lewis, G.P., Ratter, J.A., 2006. An overview of the plant diversity,

biogeography and conservation of Neotropical savannas and Seasonally Dry Forests.

Neotropical Savannas and Seasonally Dry Forests: Plant Diversity, Biogeography, and

Conservation, pp. 1–30.

Pennington, R.T., Prado, D.E., Pendry, C.A., 2000. Neotropical seasonally dry forests and

Quaternary vegetation changes. Journal of Biogeography 27, 261–273.

32

Portik, D.M., Travers, S.L., Bauer, A.M., Branch, W.R., 2013. A new species of Lygodactylus

(Squamata: Gekkonidae) endemic to Mount Namuli, an isolated 'sky island' of northern

Mozambique. Zootaxa 3710, 415–435.

Prado, D.E., 1991. A critical evaluation of the floristic links between Chaco and Caatingas

vegetation in South America. University of St Andrews, St Andrews.

Prado, D.E., 1993. What is the Gran Chaco vegetation in South America? II. A redefinition.

Candollea 48, 615–629.

Prado, D.E., Gibbs, P.E., 1993. Patterns of species distributions in the dry seasonal forests of

South America. Annals of the Missouri Botanical Garden 80, 902–927.

Puente, M., Thomas, M., Vences, M., 2005. Phylogeny and biogeography of Malagasy dwarf

geckos, Lygodactylus Gray, 1864: preliminary data from mitochondrial DNA sequences

(Squamata: Gekkonidae. African Biodiversity. Springer US, pp. 229–235.

Pyron, R.A., Burbrink, F.T., Wiens, J.J., 2013. A phylogeny and revised classification of

Squamata, including 4161 species of lizards and snakes. BMC Evolutionary Biology 13, 1–

53.

Recoder, R.S., Werneck, F.D.P., Teixeira, M., Jr., Colli, G.R., Sites, J.W., Jr., Rodrigues, M.T.,

2014. Geographic variation and systematic review of the lizard genus Vanzosaura

(Squamata, Gymnophthalmidae), with the description of a new species. Zoological Journal

of the Linnean Society 171, 206–225.

Riddle, B.R., Dawson, M.N., Hadly, E.A., Hafner, D.J., Hickerson, M.J., Mantooth, S.J., Yoder,

A.D., 2008. The role of molecular genetics in sculpting the future of integrative

biogeography. Progress in Physical Geography 32, 173–202.

33

Rocha, P.L.B., 1995. Proechimys yonenagae, a new species of spiny rat (Rodentia: Echimyidae)

from fossil sand dunes in the Brazilian Caatinga. Mammalia 59, 537–549.

Rodrigues, M.T., 1996. Lizards, snakes, and amphisbaenians from the Quaternary Sand Dunes of

the Middle Rio São Francisco, Bahia, Brazil. Journal of Herpetology 30, 513–523.

Rodrigues, M.T., 2003. Herpetofauna da Caatinga. In: Leal, I.R., Tabarelli, M., Silva, J.M.C.

(Eds.), Ecologia e Conservação da Caatinga, Brasil: Universidade Federal de Pernambuco,

pp. 181–236.

Rodrigues, M.T., Juncá, F.A., 2002. Herpetofauna of the Quaternary sand dunes of the middle

Rio Sao Francisco: Bahia: Brazil. VII. Typhlops amoipira sp. nov., a possible relative of

Typhlops yonenagae (Serpentes, Typhlopidae). Papeis Avulsos de Zoologia 42, 325–333.

Röll, B., 2005. Variability in a common species: the Lygodactylus capensis complex from

southern and eastern Africa (Reptilia, Gekkonidae). In: Huber, B.A., Lampe, K.H. (Eds.),

African Biodiversity – Molecules, Organisms, Ecosystems, Springer, Berlin, pp. 237–244.

Röll, B., Prohl, H., Hoffmann, K.P., 2010. Multigene phylogenetic analysis of Lygodactylus

dwarf geckos (Squamata: Gekkonidae). Mol Phylogenet Evol 56, 327–335.

Rull, V., 2011. Neotropical biodiversity: timing and potential drivers. Trends Ecol Evol 26, 508–

513.

Silva, J.M.C., Bates, J.M., 2002. Biogeographic patterns and conservation in the South American

Cerrado: a tropical savanna hotspot. Bioscience 52, 225–233.

Smith, H.M., Martin, R.L., Swain, T.A., 1977. A new genus and two new species of south

american geckos (Reptilia: Lacertilia). Papeis Avulsos de Zoologia 30, 195–213.

34

Stamatakis, A., 2014. RAxML Version 8: A tool for Phylogenetic Analysis and Post-Analysis of

Large Phylogenies. Bioinformatics.

Stephens, M., Smith, N.J., Donnelly, P., 2001. A new statistical method for haplotype

reconstruction from population data. The American Journal of Human Genetics 68, 978–

989.

Stephens, P.R., Wiens, J.J., 2003. Explaining species richness from continents to communities:

the time-for-speciation effect in emydid Turtles. The American Naturalist 161, 112–128.

Travers, S.L., Jackman, T.R., Bauer, A.M., 2014. A molecular phylogeny of Afromontane dwarf

geckos (Lygodactylus) reveals a single radiation and increased species diversity in a South

African montane center of endemism. Mol Phylogenet Evol 80, 31–42.

Turchetto-Zolet, A.C., Pinheiro, F., Salgueiro, F., Palma-Silva, C., 2013. Phylogeographical

patterns shed light on evolutionary process in South America. Molecular Ecology 22,

1193–1213.

Uetz, P., 2016. The Reptile Database. Disponível em http://www.reptile-database.org.

Vanzolini, P.E., 1963. Problemas Faunísticos do Cerrado. In: Ferri, M. (Ed.), Simpósio Sobre o

Cerrado, pp. 307–320.

Vanzolini, P.E., 1974. Ecological and geographical distribution of lizards in Pernambuco,

Northeastern Brasil (Sauria). Papeis Avulsos de Zoologia 18, 61–90.

Vanzolini, P.E., 1976. On the lizards of a Cerrado-Caatinga contact: evolutionary and

zoogeographical implications (Sauria). Papeis Avulsos de Zoologia 29, 111–119.

35

Vanzolini, P.E., 1988. Distributional Patterns of South American Lizards. In: Heyer, W.R.,

Vanzolini, P.E. (Eds.), Proceedings of a Workshop on Neotropical Distribution Patterns,

Academia Brasileira de Ciências, Rio de Janeiro, Brasil, pp. 317–342.

Vitt, L.J., 1995. The ecology of tropical lizards in the Caatinga of northeast Brazil. Occasional

Papers of Oklahoma Museum of Natural History 1, 1–29.

Werneck, F.P., 2011. The diversification of eastern South American open vegetation biomes:

Historical biogeography and perspectives. Quaternary Science Reviews 30, 1630–1648.

Werneck, F.P., Colli, G.R., 2006. The lizard assemblage from seasonally dry tropical forest

enclaves in the Cerrado biome, Brazil, and its association with the Pleistocenic Arc.

Journal of Biogeography 33, 1983–1992.

Werneck, F.P., Costa, G.C., Colli, G.R., Prado, D.E., Sites, J.W., Jr., 2011. Revisiting the

historical distribution of Seasonally Dry Tropical Forests: new insights based on

palaeodistribution modelling and palynological evidence. Global Ecology and

Biogeography 20, 272–288.

Werneck, F.P., Gamble, T., Colli, G.R., Rodrigues, M.T., Sites, J.W., Jr., 2012. Deep

diversification and long-term persistence in the South American 'Dry Diagonal': integrating

continent-wide phylogeography and distribution modeling of geckos. Evolution 66, 3014–

3034.

Werneck, F.P., Leite, R.N., Geurgas, S.R., Rodrigues, M.T., 2015. Biogeographic history and

cryptic diversity of saxicolous Tropiduridae lizards endemic to the semiarid Caatinga.

BMC Evolutionary Biology 15, 1–24.

36

Table 1: Samples used in this study with respective voucher number, localities (associated with

Figure 1) and GenBank number (if available).

Species Specimen ID Locality (Numbers) GenBank number

ND2 Rag 1

L. klugei AAGARDA 2967 Espírito Santo/RN (1) - -

L. klugei AAGARDA 7563 Buíque/PE (2) - -

L. klugei AAGARDA 11633 Quixadá/CE (3) - -

L. klugei FSCHUFPB 3023 Aiuaba/CE (4) - -

Lygodactylus sp. 1 MTR 3347 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3348 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3349 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3352 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3353 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3356 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3358 Santo Inácio/BA (5) - -

Lygodactylus sp. 1 MTR 3359 Santo Inácio/BA (5) - -

Lygodactylus sp. 2 AAGARDA 10526 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10527 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10543 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10544 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10545 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10546 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10547 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10551 Condeúba/BA (6) - -

Lygodactylus sp. 2 AAGARDA 10553 Condeúba/BA (6) - -

Lygodactylus sp. 3 CHUNB 56795 São Domingos/GO (7) - -

Lygodactylus sp. 3 CHUNB 56817 São Domingos/GO (7) - -

L. wetzeli MNHNP 11467 Boquerón, Paraguay (8) - -

L. wetzeli MNHNP 11472 Boquerón, Paraguay (8) - -

L. angularis MVZ 266139 Niassa Province,

Mozambique

KF546229 KF546245

L. angularis MVZ 266140 Niassa Province,

Mozambique

KF546230 KF546246

L. angularis PEMR16821 Mbeya Region, Tanzania KM034121 KM034172

L. bernardi NMZB 17837 Manicaland, Zimbabwe KM034124 KM034175

L. bivittis FGMV 2001.A21 Unavailable JX041380 JQ945314

L. bivittis AMB8955 Madagascar KM034119 KM034170

L. bonsi PEMR 16321 Phalombe District, Malawi KF546235 KF546252

L. bonsi PEMR 16834 Phalombe District, Malawi KF546236 KF546251

L. bradfieldi AMB 7628 Kunene Region, Namibia JX041381 HQ426301

L. capensis MVZ 266133 Unavailable KF546237 KF546254

L. capensis MVZ 266131 Unavailable KF546238 KF546253

L. capensis MCZR 192363 Limpopo, South Africa KM034122 KM034173

L. chobiensis MCZR 190467 Unavailable KF546225 KF546241

L. chobiensis MCZR 190518 Matabeleland North,

Zimbabwe

KF546226 KF546242

L. conradti PEMR 16820 Tanga Region, Tanzania KM034120 KM034171

L. graniticolus MCZR 192327 Limpopo Province, South KM034125 KM034176

37

Africa

L. graniticolus MCZR 192328 Limpopo Province, South

Africa

KM034126 KM034177

L. graniticolus MCZR 192329 Limpopo Province, South

Africa

KM034127 KM034178

L. graniticolus MCZR 192330 Limpopo Province, South

Africa

KM034128 KM034179

L. graniticolus MCZR 192331 Limpopo Province, South

Africa

KM034129 KM034180

L. kimhowelli PEMR 16819 Tanga Region, Tanzania KF546228 KF546244

L. methueni MBUR 01678 Limpopo Province, South

Africa

KM034136 KM034188

L. methueni MBUR 01677 Limpopo Province, South

Africa

KM034137 KM034187

L. methueni MBUR 01692 Limpopo Province, South

Africa

KM034138 KM034189

L. mirabilis FGMV 2000.B3 Unavailable JX041382 HQ426300

L. nigropunctatus MBUR00357 Limpopo Province, South

Africa

KM034145 KM034196

L. nigropunctatus MB314 Limpopo Province, South

Africa

KM034146 KM034197

L. nigropunctatus MBUR00280 Limpopo Province, South

Africa

KM034147 KM034198

L. ocellatus MBUR 00179 Mpumalanga Province,

South Africa

KM034152 KM034203

L. ocellatus PEMR 16421 Mpumalanga Province,

South Africa

KM034153 KM034204

L. ocellatus MBUR 00180 Mpumalanga Province,

South Africa

KM034154 KM034205

L. ocellatus AMB 8591 Mpumalanga Province,

South Africa

KM034155 KM034206

L. ocellatus PEMR 16647 Mpumalanga Province,

South Africa

KM034156 KM034207

L. regulus MVZ 266137 Zambézia Province,

Zimbabwe

KF546233 KF546249

L. regulus MVZ 266138 Zambézia Province,

Zimbabwe

KF546234 KF546250

L. rex PEMR 16289 Phalombe District, Malawi KF546231 KF546247

L. rex PEMR 9770 Phalombe District, Malawi KF546232 KF546248

L. stevensoni MCZR 192298 Matabeleland South,

Zimbabwe

KM034123 KM034174

L. waterbergensis CAS 234227 Limpopo Province, South

Africa

KM034166 KM034217

L. waterbergensis MCZR 192343 Limpopo Province, South

Africa

KM034167 KM034219

L. waterbergensis MCZR 192344 Limpopo Province, South

Africa

KM034168 KM034220

Abbreviations: BA, Bahia State; CE, Ceará State; GO, Goiás State; PE, Pernambuco State; RN, Rio

Grande do Norte State;

38

Table 2: Genetic distances (uncorrelated p-distance) among South American Lygodactylus

species recovered by SpedeSTEM. Upper values are from ND2 (mtDNA) and lower values are

from RAG1 (nuDNA) marker.

L. klugei Lygodactylus

sp. 1 (Santo

Inácio)

Lygodactylus

sp. 2

(Condeúba)

Lygodactylus

sp. 3 (São

Domingos)

L. wetzeli

L. klugei -

Lygodactylus

sp. 1 (Santo

Inácio)

0.190/

0.007

-

Lygodactylus

sp. 2

(Condeúba)

0.301/

0.030

0.333/

0.024

-

Lygodactylus

sp. 3 (São

Domingos)

0.298/

0.027

0.308/

0.021

0.151/

0.010

-

L. wetzeli 0.326/

0.030

0.328/

0.024

0.179/

0.013

0.038/

0.003

-

39

Figure 1: Distribution map of South American Lygodactylus samples. The biomes of the open

diagonal are in a gray scale: Chaco (CH) in dark gray, Cerrado (CE) in gray and Caatinga (CA)

in black. Seasonally Dry Tropical Forests enclaves in Cerrado are also in black. Orange circles

represents L. klugei samples, green circle Lygodactylus sp. 1 (Santo Inácio), blue circle

Lygodactylus sp. 2 (Condeúba), light red circle Lygodactylus sp. 3 (São Domingos), and dark red

circle L. wetzeli.

-20

°-1

0°

-70° -60° -50° -40°

CH

CE

CA

0 500 1000

Km

1

2

3

4

5

67

8

40

Figure 2: Maximum Likelihood concatenated tree showing individuals in five highly supported

groups (Bootstrap = 100). Node numbers correspond to 1000 ML bootstrap values. Different

colors represent distinct groups.

41

Figure 3: Bayesian species tree and divergence times for South American Lygodactylus. The

African Lygodactylus were collapsed only on the representation for better visualization of the

tree. For the species tree with uncollapsed nodes, see Supporting Information, Figure S1. The

node numbers correspond to Bayesian posterior probabilityies. Nodes with posterior probability

higher than 99% are marked with an asterisk (*). Outgroup 1 correspond to L. angularis,

outgroup 2 to L. chobiensis and L. kimhowelli, and outgroup 3 to the other 15 species of African

Lygodactylus used here. Abbreviations: Pli, Pliocene; P, Pleistocene; BA, Bahia State; GO,

Goiás State.

Outgroup 3

Outgroup 2

Outgroup 1

L. sp.3 (São Domingos/GO)

L. wetzeli

L. sp.2 (Condeúba/BA)

L. klugei

L. sp.1 (Santo Inácio/BA)

PMiocene PliOligocenePaleocene

60.0 50.0 40.0 30.0 20.0 10.0 0.0

5.0

*

*

*

0.65

0.97

0.60

*

*

*

Million years before present

42

SUPPORTING INFORMATION

Table S1: Information about markers and PCR protocols used in this study.

Gene Length

(bp)

PCR profile Primer Author

ND2 540 ND2-50 METF1: AAGCTTTCGGGCCCATACC

CO1R1: AGRGTGCCAATGTCTTTGTGRTT

Macey et al. (1997)

Arévalo et al. (1994)

RAG-1 302 RAG1-50 RAG1 F700: GGAGACATGGACACAATCCATCCTAC

RAG1 R700: TTTGTACTGAGATGGATCTTTTTGCA

Bauer et al. (2007)

Bauer et al. (2007)

ND2-50: initial denaturation at 95° C (2:00min); 34 cycles consisting of: (1) denaturation at 95° C

(0:35min), (2) annealing at 50° C (0:35min), and (3) extension at 72° C (1:34min + 4s per cycle); final

rest at 20º C.

RAG1-50: initial denaturation at 95° C (2:00min); 34 cycles consisting of: (1) denaturation at 95° C

(0:35min), (2) annealing at 50° C (0:35min), and (3) extension at 72° C (1:35min); final extension at 72°

C (4:00min); final rest at 20º C.

43

Figure S1: Bayesian species tree and divergence time for Lygodactylus. Complete tree

encompassing the 18 African species. Abbreviations: Pli, Pliocene; P, Pleistocene.

L. angularis

L. sp. 3

L. wetzeli

L. sp. 2

L. klugei

L. sp. 1

L. chobiensis

L. kimhowelli

L. rex

L. regulus

L. bonsi

L. waterbergensis

L. nigropunctatus

L. methueni

L. graniticolus

L. ocellatus

L. bradfieldi

L. capensis

L. stevesoni

L. bernardi

L. conradti

L. mirabilis

L. bivittis

PMiocene PliOligocenePaleocene

60.0 50.0 40.0 30.0 20.0 10.0 0.0

Million years before present

5.0

*

*

*

*

*

0.65

*

0.97

*

*

*

*

*

0.92

0.85

0.43

0.96

0.74

*

0.42

0.52

0.60

44

REFERENCES

Arévalo, E., Davis, S.K., Sites Jr., J.W., 1994. Mitochondrial DNA sequence divergence and

phylogenetic relationships among eight chromosome races of the Scoleporus grammicus

complex (Phrynosomatidae) in Central Mexico. Systematic Biology 43, 387–418.

Bauer, A.M., Silva, A.d., Greenbaum, E., Jackman, T., 2007. A new species of day gecko from

high elevation in Sri Lanka, with a preliminary phylogeny of Sri Lankan Cnemaspis

(Reptilia, Squamata, Gekkonidae). Zoosystematics and Evolution 83, 22–32.

Macey, J.R., Larson, A., Ananjeva, N.B., Fang, Z., Papenfuss, T.J., 1997. Two Novel Gene

Orders and the Role of Light-Strand Replication in Rearrangement of the Vertebrate

Mitochondrial Genome. Molecular Biology and Evolution 14, 91–104.

45

– CHAPTER II –

THE ROLE OF THE SÃO FRANCISCO RIVER ON THE

DIVERSIFICATION OF A DWARF GECKO ENDEMIC TO THE

SEMIARID CAATINGA, NORTHEASTERN BRAZIL

Manuscript in preparation

46

The role of the São Francisco River on the diversification of a dwarf gecko endemic to the

semiarid Caatinga, Northeastern Brazil

Flávia M. Lanna1*, Marcelo Gehara2, Fernanda P. Werneck3, Emanuel M. Fonseca1, Guarino R.

Colli4, Jack W. Sites Jr5, Miguel T. Rodrigues6, Adrian A. Garda7

1 Programa de Pós-Graduação em Ecologia, Universidade Federal do Rio Grande do

Norte, Campus Universitário, Lagoa Nova, 59078-900, Natal, RN, Brazil

2 American Museum of Natural History, Department of Herpetology, 79th St. Central

Park West, New York, NY 10024

3 Coordenação de Biodiversidade, Programa de Coleções Científicas Biológicas,

Instituto Nacional de Pesquisas da Amazônia (INPA), 69067–375, Manaus, Amazonas, Brazil

4 Departamento de Zoologia, Universidade de Brasília, 70910–900 Brasília, DF, Brazil

5 Department of Biology and Bean Life Science Museum, Brigham Young University,

Provo, UT 84602

6 Departamento de Zoologia, Instituto de Biociências, Universidade de São Paulo,

05508–090, São Paulo, SP, Brazil

7 Departamento de Botânica e Zoologia, Centro de Biociências, Universidade Federal do

Rio Grande do Norte, Campus Universitário, Lagoa Nova, 59078-900, Natal, RN, Brazil

* Corresponding author. E-mail address: flaviamollanna@gmail.com (F.M. Lanna)

47

Abstract

Species diversification is highly influenced by geomorphological unities, such as mountains,

valleys, and rivers. Rivers may act isolating or reducing gene flow among populations, as hard or

soft barriers, depending on their particular characteristics and on the organism in question. The

São Francisco River (SFR) is the largest perennial river in the Caatinga biome and has been

suggested to act as a barrier for gene flow and dispersion. Herein, we evaluated the role of the

SFR on the diversification of Lygodactylus klugei, a small gecko from the Caatinga. We used a

single locus species delimitation method to define lineages (GMYC), a migration analysis, and

other phylogenetic tests to evaluate the role of the SFR in structuring genetic diversity in this

species. We also evaluate genetic structure based on nuclear markers, testing the number of

population found through an assignment test (Structure) across the species distribution. The

GMYC analysis recovered two mitochondrial lineages structured with respect to the SFR, but

only a single population was identified by nuclear markers. We suggest that the current SFR

course has acted as a geographic barrier for L. klugei since its course changed ~450 kya.

However, we found no evidence for an effect of its paleocourse on the diversification of L.

klugei.

Keywords: Lygodactylus klugei, multilocus, paleocourse, recent barrier, riverine hypothesis,

single locus

48

1. Introduction

Landscape features such as mountains, valleys, and water bodies represent

geomorphological unities that can isolate or reduce gene flow among populations, influencing

diversity patterns at both regional and global scales (Morrone 2009). Rivers can act as a vicariant

event, isolating populations, and as secondary dispersal barriers, decreasing gene flow between

populations. Rivers influenced biogeographic patterns at least in some taxonomic groups, such as

frogs (Garda and Cannatella 2007), monkeys (Wallace 1854), birds (Ribas, et al. 2012), and

lemurs (Pastorini, et al. 2003).

The São Francisco River (SFR) is the largest perennial river of northeastern Brazil. It

runs through a region mostly covered by the Caatinga, the largest continuous block (over

850.000 km2) of Seasonally Dry Tropical Forests (SDTFs). The Caatinga is characterized by high

average temperatures (27–29ºC), highly seasonal climate, and low (300–800 mm/yr) and

unpredictable rainfall (Prado 2003). The SFR course was largely dynamic during the Quaternary,

with geomorphological evidences suggesting that until the Middle Pleistocene its paleocourse

was different from today (Mabesoone 1994). The paleo-SFR ran north towards the equatorial

Atlantic Ocean, where the Piauí and Parnaíba Rivers currently occur (Figure 1). In the middle

Pleistocene, the SFR course changed to its current position, reaching the Atlantic Ocean on the

east coast of Brazil (Mabesoone 1994; Werneck, et al. 2015). This change is supported by the

younger age of the lower part of the São Francisco drainage and by limestone deposits found in

the Remanso-Petrolina area, dated to the Quaternary based on low fossil content, which

evidenced the formation of a shallow lake after the uplift of the Serra Grande and Ibiapaba

mountain ranges (Suguio, et al. 1980). Taken together these evidences point to a change in

49

course of the SFR to Sergipe State (SE) during the Mindel Glaciation (478–424 kya, Mabesoone

1994; Werneck, et al. 2015).

The SFR current course has been shown to act as a barrier to gene flow for lizards and

rodents (Nascimento, et al. 2013; Werneck, et al. 2015). Likewise, the middle-SFR Quaternary

sand dunes, which show high endemism levels for squamates, have different species/populations

restricted to either of its opposite margins (Passoni, et al. 2008; Rodrigues 2003; Werneck, et al.

2012). Such evidences led scientists to propose the São Francisco Riverine hypothesis (SFR-

hypothesis), which suggests the SFR as one of the major barriers influencing endemism and

diversity in the Caatinga (Faria, et al. 2013; Nascimento, et al. 2013; Passoni, et al. 2008;

Rodrigues 2003; Siedchlag, et al. 2010; Werneck, et al. 2015).

However, not all phylogeographic studies have endorsed the SFR-hypothesis (Magalhaes,

et al. 2014; Oliveira, et al. 2015; Recoder, et al. 2014; São-Pedro 2014). The presence of

incomplete lineage sorting (ILS), panmitic populations across the Caatinga, migration between

river margins, and lack of a genetic break matching the current position of the river have been

pointed as inconsistencies with a possible role of the river as a hard barrier. Still, because its

current position is relatively recent, retention of ancestral polymorphism could account for ILS

patterns observed, even if the SFR is a hard barrier to gene flow. Furthermore, uneven sampling

among margins might bias results of population assignment programs such as Structure

(Pritchard, et al. 2000). At last, populations may diversify even under the presence of gene flow

(Nosil 2008), which could prevent the complete sorting of genealogies, while low migration

levels would be consistent with the SFR acting as a soft barrier.

An appropriate test for the SFR-hypothesis would, then, ideally implement models that

explicitly test the existence of panmitic populations versus structured populations with gene

50

flow. Using a multi-locus data set for the gecko Lygodactylus klugei we performed population

and genetic lineages assignment tests to delimit genetic units and test if the divergence time is

congruent with the recent change in the course of the SFR. We further used population genetic

models to test the support for a panmixy model (Pan), which simulates the SFR as totally

permeable to gene flow and not responsible for population differentiation; and a migration

model, or soft barrier model (SB), which represents the hypothesis of the SFR constituting a

permeable barrier driving diversification in the presence of gene flow.

2. Material and Methods

2.1 Sample collection and sequencing

We sequenced 124 samples from 19 localities, focusing in L. klugei, and a few samples of

L. wetzeli and Lygodactylus sp. "São Domingos" (Figure 1, Table S1). We obtained most

samples from tissues deposited in the following zoological collections: Coleção Herpetológica da

Universidade Federal do Rio Grande do Norte (UFRN), Coleção Herpetológica da Universidade

de Brasília (CHUNB) and Museu de Zoologia da Universidade de São Paulo (MZUSP).

We extracted genomic DNA from liver, muscle, finger or tail tissues with DNeasy

Qiagen kit and amplified one mitochondrial (mtDNA) and three nuclear (nuDNA) genes

following standard polymerase chain reaction (PCR) techniques. We chose nuclear genes that

were shown to be sufficiently variable for Squamata (Townsend, et al. 2008; Werneck, et al.

2012) and sequenced all 124 samples for the NADH subunit 4 (ND4, mtDNA). Later, a subset of

one or two individuals from each locality, representing different mtDNA haplotypes were chosen

to be amplified for the following nuclear genes: DMX Like 1 (DMXL1, nuDNA), Dynein

Axonemal Heavy Chain 3 (DNAH3, nuDNA), and Prolactin Receptor (PRLR, nuDNA), totaling

51

59 individuals for each nuclear gene. For PCR protocols see the Supporting Information (Table

S2). PCRs products were cleaned using MANU 30 PCR Milipore plates and resuspended with

ultra-pure water. We prepared sequencing reactions using ABI Big-Dye Terminator v3.1 Cycle

Sequencing Kit in an ABI GeneAmp PCR 9700 thermal cycler, purified products with Sephadex

G-50 Fine (GE Healthcare), and sequenced in an ABI 3730x1 DNA Analyzer at the BYU DNA

Sequencing Center.

We assembled and edited chromatograms in Geneious v8.1.7 (Biomatters),

subsequentially aligning them using the Muscle algorithm (Edgar 2004). We used PHASE v2.1.1

(Stephens, et al. 2001; Stephens and Wiens 2003) for all three nuDNA to determine the most

probable pair of alleles for each gene, using default program options.

2.2 Gene tree

If SFR acts as a barrier, we expect to see gene trees geographically structured with

respect to it, with different clades restricted to either side of the river. We assume that this might

be the case for at least the mitochondrial tree, considering that the current SFR course is recent

(478–424 kya).

To check the genealogical structure of each gene, we estimated Bayesian gene trees for

the mtDNA and for each phased nuDNA in BEAST 1.8.2 (Drummond, et al. 2012). We used L.

wetzeli and Lygodactylus sp. "São Domingos" (see chapter 1) as outgroups to root gene trees for

all genes. Models of nucleotide substitution were determined using Bayesian information

criterion in JModelTest 2 (Darriba, et al. 2012). We conducted three independent BEAST runs to

ensure convergence. Each run had 100 million generations sampled every 10 thousand steps. We

used Tracer v1.6 (Drummond and Rambaut 2007) to visually evaluate chain mixing and effective

52

sample size values (ESS > 200). We removed the first one thousand trees as burn-in and inferred

the maximum clade credibility (MCC) tree using TreeAnnotator (Drummond, et al. 2012).

2.3 Assignment of genetic lineages

We estimated the number of populations for Lygodactylus klugei with a Bayesian

probabilistic genetic clustering implemented in Structure 2.3.4 (Pritchard, et al. 2000). We used

XMFAS2STRUCT (available at http://www.xavierdidelot.xtreemhost.com/clonalframe.htm) to

convert the nuclear alignments into Structure input files. We ran 10 replicates of the analysis for

each k (number of possible clusters) varying from 1 to 5, using a linkage model with independent

allele frequencies approach. We retained 5x105 generations of Markov Chain Monte Carlo

(MCMC) after 1x105 generation of burn-in were discarded. The best value of k was defined on

Structure Harvester (available at http://taylor0.biology.ucla.edu/structureHarvester/), based on

the rate of changes in the log-probability of data between successive k, Δk (Earl and vonHoldt

2012).

The mitochondrial gene tree showed a slight structure with respect to the São Francisco

River (see Results). To account for the possible role of the SFR in structuring L. klugei, we used

a single locus species delimitation' method (GMYC) to determine lineages within this species.

The Generalized Mixed Yule Coalescent (GMYC) method statistically estimates the number of

distinct species or distinct lineages using an ultrametric consensus tree from the mitochondrial

data (Pons, et al. 2006). This analysis is implemented in R (R Core Team 2015) and uses

functions from the package ‘splits’. We ran the GMYC analysis for a MCC mtDNA tree

generated in BEAST 1.8.2 (Drummond, et al. 2012). The GMYC relies on branch lengths and

tree topology, and does not consider phylogenetic uncertainty. To access this uncertainty, we

53

used the Bayesian implementation of GMYC (Reid and Carstens 2012). We used a subsample of

100 trees of the 9000 after burn-in trees constructed in BEAST. We used the package ‘bGMYC’

in R to run the bGMYC analysis. We ran 100,000 generations, with 90,000 generations of burn-

in, and a thinning interval of 100 samples. We set MCMC parameters for the model as follows:

py2=1.2, t1=2, t2=53, start = c(1, 0.5, 5), scale = c(1, 20, 0.5) (Werneck, et al. 2015).

2.4 Haplotype network and DNA polymorphism

To check for haplotype sharing between the lineages delimited by GMYC (see Results),

we constructed haplotype networks. We used Bayesian gene trees (without outgroups) as input

on the program Haploviewer (http://www.cibiv.at/~greg/haploviewer) (Salzburger, et al. 2011).

We calculated DNA polymorphism summary statistics for L. klugei using "ape",

"haplotypes", and "pegas" packages implemented in R (R Core Team 2015), which included:

number of polymorphic sites (S), haplotype number (h), haplotype diversity (Hd), and nucleotide

diversity (π) for each gene and lineage defined by GMYC analysis. We also checked the

uncorrected p-distance of all four genes between and within lineages using Mega 7.0 software

(Kumar, et al. 2016).

2.5 Species tree estimation

Using the GMYC results, we estimated the divergence times between northern and

southern lineages of L. klugei using *BEAST 1.8.2 (Drummond, et al. 2012). We expected a

divergence time between the two lineages less than 478ky if the river is actually a vicariant

barrier for this species. The lack of Lygodactylus fossils prevent precise calibration, so we based

calibration of the L. klugei mtDNA mutation rate on Arnold, et al. (2008) that suggested a

54

mutation rate for gekkonid lizards of 1.15% per million years. We ran three independent species

tree analyses with 2x108 generations each, sampled every 2x104 generation, and using an

uncorrelated lognormal relaxed clock with a Yule speciation process prior. For the first run we

estimated the nuclear rates using default gamma prior for ucld.mean and exponential prior for

ucld.stdev with a mean of 0.5 for the nuDNA, and for the mtDNA we used a normal prior to

calibrate the estimation with mean of 0.0115 substitutions/million of years and standard

deviation of 0.002 for ucld.mean parameter. We implemented the other two runs substituting the

mean and standard deviation by the rates estimated for the three nuclear genes. We assessed

convergence of MCMC runs and effective sample size values (ESS > 200) using Tracer v1.6

(Drummond and Rambaut 2007). We removed the first 103 trees as burn-in and inferred the

MCC tree using TreeAnnotator (Drummond, et al. 2012).

2.6 Phylogeographic reconstruction

We used a diffusion model to reconstruct the phylogeographic history of L. klugei using

the mtDNA in BEAST 1.8.2 (Drummond, et al. 2012). For this, we used a lognormal relaxed

random walk (RRW) model, which is analogous to the relaxed clock model and estimates

dispersal rates among the branches of the phylogeny (Lemey, et al. 2010). To create noise and

separate randomly the identical coordinates, we apply the Jitter option. We used 0.0115 as mean

for the mitochondrial substitution rate, a standard deviation of 0.002 and a Gaussian Markov

random field (GMRF) Bayesian Skyride method to infer demographic dynamic between the

lineages (Minin, et al. 2008). We ran three independent runs with 5 x 108 generations, sampling

every 5 x 104 steps. We checked if the MCMC parameters converged and effective population

size (ESS) reached 200 or more on TRACER v1.6 (Drummond and Rambaut 2007). We

55

discarded the first 10% of trees as burn-in and computed the MCC tree with TreeAnnotator

(Drummond, et al. 2012). We used the MCC tree on the program SPREAD (Bielejec, et al. 2011)

to construct a kml file that contains the spatiotemporal diffusion model of L. klugei lineages and

visualized this kml file on Google Earth.

2.7 Testing the SFR barrier

To test if the SFR present-day course is a barrier for the gene flow, we used Migrate-n

v3.6.11 (Beerli and Palczewski 2010). We separated individuals in two lineages according to the

SFR: lineage 1, encompassing all the individuals located at north of the SFR (localities number:

1 – 13), and lineage 2, encompassing all the individuals south of the SFR (localities number: 14

– 17). We tested four models: (i) a panmictic model, assuming that the Caatinga has only one

panmictic population, (ii) a model with two lineages (1 and 2) and migration in both directions;

(iii) a model with two lineages and unidirectional migration from lineage 1 to lineage 2 (from

north to south); (iv) a model with two lineages and unidirectional migration from lineage 2 to

lineage 1 (from south to north).

The analysis evaluated especially if there was migration or if L. klugei forms a panmitic

population across the entire Caatinga biome. This analysis does not account for incomplete

lineage sorting and, hence, considers only migration as a possible explanation for the observed

data. If the SFR acted as a vicariant barrier in the past, but is not a secondary dispersal barrier

preventing migration, we expected the second model to be selected as the best one. If the SFR is

a barrier, even a soft one, we expected the third or the fourth models to be preferred. If the SFR

is not a barrier for L. klugei the first model should be selected.

56

We estimated model probabilities using the Bezier approximation score following Beerli

and Palczewski (2010). Using a Bayesian approach we estimated mutation-scaled immigration

rates M (m/μ) and mutation-scaled effective population sizes Θ (4Neμ) for each lineage. We used

a thermodynamic static heating with four chains with swap intervals (temperatures: 1.0, 1.5, 3.0,

107). We set the MCMC parameters to run 20,000 steps, sampling every 100, with a burn-in of

50,000 steps.

3. Results

We sequenced a total of 124 samples for mtDNA (ND4) and 59 samples for the

nuDNA (53 for Lygodactylus klugei, two for L. wetzeli and four for Lygodactylus sp. "São

Domingos"). Additional detail on the number of base pairs and sequences for each gene is

available in Table S2 (Supporting Information).

3.1 Assignment of lineages

Structure recovered two populations after Structure Harvester (K=2), but with no

geographic structure and many localities (that have more than one sample) with their individuals

assigned to different populations (Supporting Information, Figure S2). This unstructured

population assignment can mean that L. klugei has only one population, due to migration,

incomplete lineage sorting, or a much recent diversification still not observable in nuDNA.

The GMYC and the bGMYC analyses based on the mtDNA dataset recovered two

lineages structured with respect to the SFR (Figure 1 and 2.a–b), with 50–90% of grouping

probability. Hereafter these are referred to as northern and southern lineages. One individual

located north of the SFR (Jati, Ceará State, locality #5) was assigned to the southern population,

57

while three individuals located south of the SFR (two individuals from Poço Redondo, Sergipe

State, locality #14 and one from Queimadas, Bahia State, locality #17) were assigned to the

northern population.

3.2 Gene trees, species tree, haplotype network and DNA polymorphism

The mitochondrial tree showed a slight structure across both margins of the SFR (Figure

1 and S1). Nuclear trees did not recover the same mitochondrial clades and were not concordant

among each other (Figure S1). The average divergence time between mitochondrial lineages was

295 kya (95% HPD = 110–625 kya). Polymorphism data showed higher haplotype and

nucleotide diversity for the mtDNA than for nuDNA, as expected (Table 1). All nuclear markers

presented low variability, which results from haplotype sharing among localities (Figure 3b–d).

The northern mitochondrial lineage has higher variability for PRLR, while DNAH3 has higher

haplotype and nucleotide diversity for the southern lineage (Table 1). All four markers presented

low genetic distances within lineages (≤ 0.4%), and the same between lineages genetic distances

for the nuclear markers. The mitochondrial marker presented a genetic distance of 3% between

lineages (Table 2).

3.3 Phylogeographic reconstruction

The phylogeographic reconstruction using the mtDNA marker shows that the diffusion

of L. klugei started at the Pernambuco State around 1 Mya, expanding to different directions. The

diffusion reconstruction shows that L. klugei reaches the region south of the SFR for the first

time at 500 kya in northeast Bahia State, and reaches south of the SFR for the second time, now

in northwest Bahia State, 100 kya (Figure 4).

58

3.4 Testing the SFR as a barrier

The best-fit model evaluated in Migrate-n was the (iii), with two lineages (previously

defined) with migration in only one direction, from the lineage located at north of the SFR

(lineage 1) to the lineage located at south of the SFR (lineage 2). The first model (the panmitic

one) was considered the worst among the four models, with zero probability in comparison with

the other models. Effective sample sizes (ESS) were above 1,000 for all parameters (Θ and M).

4. Discussion

Our results corroborate the SFR hypothesis. The São Francisco River is the most

prominent river in Northeastern Brazil and figures as one of the main vicariant barriers to gene

flow along this region. In general, geologic information on rivers from the Neotropical region is

scarce (Potter 1997). Changes on the SFR course have been proposed for distinct periods, until

the river finally reached its current course in the Middle Pleistocene (Mabesoone 1994; Potter

1997). We found a diversification between the two lineages of Lygodactylus klugei at about 295

kya, a timing congruent with the final change of the SFR course to its current position (Werneck,

et al. 2015). The GMYC analyses assigned three individuals, located at the south bank of SFR, to

the northern lineage, and one individual located in northern bank to the southern lineage. This

can be explained either by the recent divergence time between lineages (i.e., incomplete lineage

sorting) or migration across margins. The diffusion model analysis estimated two different

colonization routes to the southern bank of the SFR. The west part of the SFR (northwest Bahia

State) was the most recent one (100 kya) and the fact that this individual was grouped with the

59

northern lineage, combined with the time suggested to this locality colonization, could be

explained by migration and reinforce the role of the SFR as a soft barrier.

Many studies have suggested the SFR as geographic barrier that acted structuring

species/population between riverbanks (Faria, et al. 2013; Nascimento, et al. 2013; Passoni, et al.

2008; Siedchlag, et al. 2010; Werneck, et al. 2015). Those studies used mitochondrial markers

for lineage delimitation, as we did here. Conversely, others species have no evidence for genetic

structure along riverbanks (Magalhaes, et al. 2014; Oliveira, et al. 2015; Recoder, et al. 2014;

São-Pedro 2014). However, studies that fail to recover the SFR as a barrier used multilocus

dataset, including nuclear genes on the population assignment, to infer the evolutionary history.

The final change of the SFR course was a relatively recent event. Nuclear genes have lower rates

of sequence evolution compared with mitochondrial genes and are more suitable to address older

events (Avise 2009). For this reason, nuclear genes may show no resolution regarding the

structuring the role of the SFR.

Unlike for other species (Nascimento, et al. 2013; Werneck, et al. 2015), we found no

evidence for a possible influence of the SFR paleocourse on the diversification of L. klugei. The

individuals of the only locality west of the paleocourse (Figure 1, locality #13) are grouped with

the northern lineage according to GMYC results. The divergence between northern and southern

lineages is older than the divergence of those individuals with the other individuals from the

same lineage. If the paleo-SFR had some influence on the divergence of L. klugei was expected

an older divergence among individuals from the locality west and east of the paleocourse than

from localities north and south of the current SFR course, once the current course is a more

recent barrier.

60

5. Conclusion

Our study recovered two lineages of Lygodactylus klugei in the Caatinga biome. The São

Francisco River current course is the most probable cause of L. klugei diversification. Because

Migrate-N cannot distinguish ILS from migration among opposing river banks, either the SFR

acted as a soft barrier, decreasing but not fully halting gene flow, or that not enough time has

passed for reciprocal monophyly to be achieved. Either way, the role of the SFR as a barrier to

gene flow is corroborated. The paleo-SFR course had no influence on the diversification of this

species, once the oldest diversification of this species is younger than the period of occurrence of

the paleocourse.

Acknowledgements

We are grateful to students, postdocs, and researchers that worked in collaboration with AAG,

GRC, and MTR laboratories for help with fieldwork and samples donation. We are grateful to

César Aguilar, Derek Tucker, Luciano J. Ávila, Mariana Morando, Perry L. Wood, and Randy

Klabacka for laboratorial assistance. We also thank Eliana F. Oliveira and Gabriel C. Costa for

suggestions and comments on the manuscript. FML and EMF thanks Conselho Nacional de

Desenvolvimento Científico e Tecnológico (CNPq) for their graduate scholarships. FPW thanks

National Science Foundation DDIG award (DEB-1210346), Science Without Borders Program

from CNPq (#374307/2012-1), CAPES/Fulbright (#15073722–2697/06–8), and the Partnerships

for Enhanced Engagement in Research (PEER) program for financial support. GRC thanks

Coordenação de Apoio à Formação de Pessoal de Nível Superior (CAPES), CNPq, Fundação de

Apoio à Pesquisa do Distrito Federal (FAPDF), and PEER program for financial support. MTR

thanks Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 2003/10335-8 and

61

2011/50146-6), CNPq, and Dimensions of Biodiversity Program [FAPESP (BIOTA,

2013/50297-0), NSF (DOB 1343578), and NASA]. AAG thanks CNPq (563352/2010-8,

552031/2011-9, 431433/2016-0 and 457463/2012-0) and CAPES (23038.005577/2012-28 and

23038.009565/2013-53) for financial support.

References

Arnold EN, Vasconcelos R, Harris DJ, Mateo JA, Carranza S 2008. Systematics, biogeography

and evolution of the endemic Hemidactylus geckos (Reptilia, Squamata, Gekkonidae) of

the Cape Verde Islands: based on morphology and mitochondrial and nuclear DNA

sequences. Zoologica Scripta 37: 619–636. doi: 10.1111/j.1463-6409.2008.00351.x

Avise JC 2009. Phylogeography: retrospect and prospect. Journal of Biogeography 36: 3–15.

Beerli P, Palczewski M 2010. Unified framework to evaluate panmixia and migration direction

among multiple sampling locations. Genetics 185: 313–326. doi:

10.1534/genetics.109.112532

Bielejec F, Rambaut A, Suchard MA, Lemey P 2011. SPREAD: spatial phylogenetic

reconstruction of evolutionary dynamics. Bioinformatics 27: 2910–2912. doi:

10.1093/bioinformatics/btr481

Darriba D, Taboada GL, Doallo R, Posada D 2012. jModelTest 2: more models, new heuristics

and parallel computing. Nature Methods 9: 772.

Drummond AJ, Rambaut A 2007. BEAST: Bayesian evolutionary analysis by sampling trees.

BMC Evolutionary Biology 7: 214.

62

Drummond AJ, Suchard MA, Xie D, Rambaut A 2012. Bayesian phylogenetics with BEAUti

and the BEAST 1.7. Molecular Biology and Evolution 29: 1969–1973.

Earl DA, vonHoldt BM 2012. STRUCTURE HARVESTER: a website and program for

visualizing STRUCTURE output and implementing the Evanno method. Conservation

Genetics Resources 4: 359–361. doi: 10.1007/s12686-011-9548-7

Edgar RC 2004. MUSCLE: multiple sequence alignment with high accuracy and high

throughput. Nucleic Acids Research 32: 1792–1797. doi: DOI: 10.1093/nar/gkh340

Faria MB, Nascimento FF, Oliveira JAd, Bonvicino CR 2013. Biogeographic determinants of

genetic diversification in the mouse opossum Gracilinanus agilis (Didelphimorphia:

Didelphidae). Journal of Heredity 104: 613–626.

Garda AA, Cannatella DC 2007. Phylogeny and biogeography of paradoxical frogs (Anura,

Hylidae, Pseudae) inferred from 12S and 16S mitochondrial DNA. Molecular

Phylogenetics and Evolution 44: 104–114. doi: 10.1016/j.ympev.2006.11.028

Kumar S, Stecher G, Tamura K 2016. MEGA7: Molecular Evolutionary Genetics Analysis

Version 7.0 for Bigger Datasets. Molecular Biology and Evolution 33: 1870–1874. doi:

10.1093/molbev/msw054

Lemey P, Rambaut A, Welch JJ, Suchard MA 2010. Phylogeography takes a relaxed random

walk in continuous space and time. Molecular Biology and Evolution 27: 1877–1885. doi:

10.1093/molbev/msq067

Mabesoone JM. 1994. Sedimentary Basins of Northeast Brazil: Recife: Federal University of

Pernambuco.

63

Magalhaes ILF, Oliveira U, Santos FR, Vidigal THDA, Brescovit AD, Santos AJ 2014. Strong

spatial structure, Pliocene diversification and cryptic diversity in the Neotropical dry forest

spider Sicarius cariri. Molecular Ecology 23: 5323–5336. doi: 10.1111/mec.12937

Minin VN, Bloomquist EW, Suchard MA 2008. Smooth skyride through a rough skyline:

Bayesian coalescent-based inference of population dynamics. Molecular Biology and

Evolution 25: 1459–1471. doi: 10.1093/molbev/msn090

Morrone JJ. 2009. Evolutionary Biogeography: an Integrative Approach with Case Studies.

Columbia University Press, NewYork.

Nascimento FF, Lazar A, Menezes AN, Durans AdM, Moreira JC, Salazar-Bravo J, D'Andrea

PS, Bonvicino CR 2013. The role of historical barriers in the diversification processes in

open vegetation formations during the Miocene/Pliocene using an ancient rodent lineage as

a model. PLoS One 8: 1–13.

Nosil P 2008. Speciation with gene flow could be common. Molecular Ecology 17: 2103–2106.

Oliveira EF, Gehara M, São-Pedro VA, Chen X, Myers EA, Burbrink FT, Mesquita DO, Garda

AA, Colli GR, Rodrigues MT, Arias FJ, Zaher H, Santos RML, Costa GC 2015. Speciation

with gene flow in whiptail lizards from a Neotropical xeric biome. Molecular Ecology 24:

5957–5975. doi: 10.1111/mec.13433

Passoni JC, Benozzati ML, Rodrigues MT 2008. Phylogeny, species limits, and biogeography of

the Brazilian lizards of the genus Eurolophosaurus (Squamata: Tropiduridae) as inferred

from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 46: 403–414.

64

Pastorini J, Thalmann U, Martin RD 2003. A molecular approach to comparative

phylogeography of extant Malagasy lemurs. Proceedings of the National Academy of

Sciences 100: 5879–5884. doi: 10.1073/pnas.1031673100

Pons J, Barraclough TG, Gomez-Zurita J, Cardoso A, Duran DP, Hazell S, Kamoun S, Sumlin

WD, Vogler AP 2006. Sequence-based species delimitation for the DNA taxonomy of

undescribed insects. Systematic Biology 55: 595–609. doi: 10.1080/10635150600852011

Potter PE 1997. The Mesozoic and Cenozoic paleodrainage of South America: a natural history.

Journal of South American Earth Sciences 10: 331–334.

Prado DE. 2003. As Caatingas da América do Sul. In: Leal IR, Tabarelli M, da Silva JMC,

editors. Ecologia e Conservação da Caatinga. Editora Universitária UFPE. p. 3–73.

Pritchard JK, Stephens M, Donnelly P 2000. Inference of population structure using multilocus

genotype data. Genetics 155: 945–959.

R Core Team 2015. R: A Language and Environment for Statisti- cal Computing. R Foundation

for Statistical Computing, Vienna, Austria. Available from http://www.R-project.org.

Recoder RS, Werneck FDP, Teixeira M, Jr., Colli GR, Sites JW, Jr., Rodrigues MT 2014.

Geographic variation and systematic review of the lizard genus Vanzosaura (Squamata,

Gymnophthalmidae), with the description of a new species. Zoological Journal of the

Linnean Society 171: 206–225.

Reid N, Carstens BC 2012. Phylogenetic estimation error can decrease the accuracy of species

delimitation: a Bayesian implementation of the general mixed Yule-coalescent model.

BMC Evolutionary Biology 12: 196.

65

Ribas CC, Maldonado-Coelho M, Smith BT, Cabanne GS, d'Horta FM, Naka LN 2012. Towards

an integrated historical biogeography of the neotropical lowland avifauna: combining

diversification analysis and landscape evolution. Ornitologia Neotropical 23: 187–206.

Rodrigues MT. 2003. Herpetofauna da Caatinga. In: Leal IR, Tabarelli M, Silva JMC, editors.

Ecologia e Conservação da Caatinga. Brasil: Universidade Federal de Pernambuco. p. 181–

236.

Salzburger W, Ewing GB, von Haeseler A 2011. The performance of phylogenetic algorithms in

estimating haplotype genealogies with migration. Molecular Ecology 20: 1952–1963. doi:

10.1111/j.1365-294X.2011.05066.x

São-Pedro VA 2014. Filogeografia de anfíbios da diagonal de áreas abertas da America do Sul.

[Universidade Federal do Rio Grande do Norte.

Siedchlag AC, Benozzati ML, Passoni JC, Rodrigues MT 2010. Genetic structure, phylogeny,

and biogeography of Brazilian eyelid-less lizards of genera Calyptommatus and

Nothobachia (Squamata, Gymnophthalmidae) as inferred from mitochondrial DNA

sequences. Molecular Biology and Evolution 56: 622–630. doi:

10.1016/j.ympev.2010.04.027

Stephens M, Smith NJ, Donnelly P 2001. A new statistical method for haplotype reconstruction

from population data. The American Journal of Human Genetics 68: 978–989.

Stephens PR, Wiens JJ 2003. Explaining species richness from continents to communities: the

time-for-speciation effect in emydid Turtles. The American Naturalist 161: 112–128.

Suguio K, Barcelos JH, Matsui E editors. Anais XXXI Congresso. 1980 Balneário de Camboriu

(SC).

66

Townsend TM, Alegre RE, Kelley ST, Wiens JJ, Reeder TW 2008. Rapid development of

multiple nuclear loci for phylogenetic analysis using genomic resources: An example from

squamate reptiles. Molecular Phylogenetics and Evolution 47: 129–142. doi:

10.1016/j.ympev.2008.01.008

Wallace AR 1854. On the Monkeys of the Amazon. Journal of Natural History Series 2 14: 451–

454. doi: 10.1080/037454809494374

Werneck FP, Gamble T, Colli GR, Rodrigues MT, Sites JW, Jr. 2012. Deep diversification and

long-term persistence in the South American 'Dry Diagonal': integrating continent-wide

phylogeography and distribution modeling of geckos. Evolution 66: 3014–3034.

Werneck FP, Leite RN, Geurgas SR, Rodrigues MT 2015. Biogeographic history and cryptic

diversity of saxicolous Tropiduridae lizards endemic to the semiarid Caatinga. BMC

Evolutionary Biology 15: 1–24.

67

Table 1: Genetic statistics of northern and southern lineages of Lygodactylus klugei (according

to GMYC results) for each locus.

Locus Population Length (bp) N S H Hd π

ND4 Northern 679 45 32 29 0.904 0.00369

Southern 679 8 9 4 0.643 0.00331

DMXL1 Northern 691 80 3 5 0.379 0.00092

Southern 691 16 0 1 0 0

DNAH3 Northern 698 84 7 5 0.453 0.00081

Southern 698 16 8 5 0.650 0.00252

PRLR Northern 584 90 5 6 0.569 0.00114

Southern 584 14 1 2 0.363 0.00062

N, number of samples; S, number of polymorphic sites; H, number of exclusive haplotypes; Hd, haplotype

diversity; π, nucleotide diversity.

68

Table 2: Genetic distance (p-distance) between and within northern and southern lineages

(according to GMYC results) for the four markers.

Within

Locus Between Northern Southern

ND4 0.037 0.004 0.003

DMXL1 0.001 0.001 0.000

DNAH3 0.002 0.001 0.003

PRLR 0.002 0.001 0.001

69

Figure 1: Map of the sample localities of Lygodactylus klugei and its related lineage according

GMYC results. Pink circles correspond to northern lineage. Green circles correspond to southern

lineage. São Francisco River is represented in blue and its paleocurse is represented in grey in a

dotted line.

-45° -40° -35° -30°

-15

°-1

0°

-5°

0°

70

Figure 2: Lineage assignment based on the Generalized Mixed Yule Coalescent (GMYC)

method. (A) gene tree generated by GMYC. The two most probable groups are showed in red,

with one individual not allocated for any of them. (B) heat map represents Bayesian

implementations of the GMYC (bGMYC). Darker colors indicate lower probability of grouping

while lighter colors indicate higher probability. Probabilities higher than 50% were used to

assign the lineages.

71

Figure 3: Haplotype network for (A) ND4, (B) DMXL1, (C) DNAH3, and (D) PRLR markers

according to their respective Bayesian gene tree. The size of each circle is proportional to the

haplotype frequency. The small blue correspond to the number of mutational steps. Pink circles

represent northern lineage and green circles represent southern lineage (according to GMYC

results).

72

Figure 4: Bayesian spatiotemporal diffusion of mtDNA for Lygdactylus klugei in six time

frames. Lighter shades represent older diffusion events and darker shades represent younger

diffusion events.

73

SUPPORTING INFORMATION

Table S1 Information about Lygodacylus samples used in this study, with locality numbers and

coordinates.

Species Specimen ID Locality (Numbers) Lat Long

L. klugei A 123 Paulo Afonso/BA, Brazil (16) -9.685 -38.666

L. klugei AAGARDA 4046 Paulo Afonso/BA, Brazil (16) -9.685 -38.666

L. klugei AAGARDA 4066 Paulo Afonso/BA, Brazil (16) -9.685 -38.666

L. klugei MTR 916397 Queimadas/BA, Brazil (17) -10.370 -42.360

L. klugei FSCHUFPB 3015 Aiuaba/CE, Brazil (3) -6.648 -40.157

L. klugei FSCHUFPB 3023 Aiuaba/CE, Brazil (3) -6.648 -40.157

L. klugei FSCHUFPB 6427 Aiuaba/CE, Brazil (3) -6.648 -40.157

L. klugei FSCHUFPB 3232 Aiuaba/CE, Brazil (3) -6.648 -40.157

L. klugei FSCHUFPB 3234 Aiuaba/CE, Brazil (3) -6.648 -40.157

L. klugei CHUNB 56575 Jati/CE, Brazil (5) -7.629 -39.044

L. klugei AAGARDA 11890 Milagres/CE, Brazil (4) -7.254 -38.976

L. klugei AAGARDA 11891 Milagres/CE, Brazil (4) -7.254 -38.976

L. klugei AAGARDA 11892 Milagres/CE, Brazil (4) -7.254 -38.976

L. klugei AAGARDA 11893 Milagres/CE, Brazil (4) -7.254 -38.976

L. klugei CHUNB 56574 Milagres/CE, Brazil (4) -7.254 -38.976

L. klugei AAGARDA 11629 Quixadá/CE, Brazil (2) -4.961 -38.974

L. klugei AAGARDA 11633 Quixadá/CE, Brazil (2) -4.961 -38.974

L. klugei AAGARDA 11635 Quixadá/CE, Brazil (2) -4.961 -38.974

L. klugei AAGARDA 11740 Quixadá/CE, Brazil (2) -4.961 -38.974

L. klugei FSCHUFPB 00377 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FSCHUFPB 00927 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FSCHUFPB 00938 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FSCHUFPB 00944 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FSCHUFPB 00948 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FSCHUFPB 00973 Santa Quitéria/CE, Brazil (1) -4.333 -40.151

L. klugei FRD 731 Cabaceiras/PB, Brazil (10) -7.480 -36.280

L. klugei FRD 733 Cabaceiras/PB, Brazil (10) -7.480 -36.280

L. klugei FRD 857 Cabaceiras/PB, Brazil (10) -7.480 -36.280

L. klugei CHUFC L2862 Betânia/PE, Brazil (11) -8.312 -38.196

L. klugei AAGARDA 7563 Buíque/PE, Brazil (12) -8.589 -37.243

L. klugei AAGARDA 8649 Buíque/PE, Brazil (12) -8.589 -37.243

L. klugei AAGARDA 4829 São Raimundo Nonato/PI, Brazil

(13)

-8.745 -42.653

L. klugei AAGARDA 5315 São Raimundo Nonato/PI, Brazil

(13)

-8.745 -42.653

74

L. klugei AAGARDA 2967 Espírito Santo/RN, Brazil (8) -6.332 -35.311

L. klugei AAGARDA 6476 João Câmara/RN, Brazil (6) -5.562 -35.906

L. klugei AAGARDA 5561 João Câmara/RN, Brazil (6) -5.562 -35.906

L. klugei AAGARDA 5645 João Câmara/RN, Brazil (6) -5.562 -35.906

L. klugei AAGARDA 5697 João Câmara/RN, Brazil (6) -5.562 -35.906

L. klugei AAGARDA 5713 João Câmara/RN, Brazil (6) -5.562 -35.906

L. klugei AAGARDA 2890 Macaíba/RN, Brazil (7) -5.885 -35.362

L. klugei AAGARDA 3548 Macaíba/RN, Brazil (7) -5.885 -35.362

L. klugei AAGARDA 5055 Macaíba/RN, Brazil (7) -5.885 -35.362

L. klugei AAGarda 1342 Macaíba/RN, Brazil (7) -5.885 -35.362

L. klugei AAGarda 1505 Macaíba/RN, Brazil (7) -5.885 -35.362

L. klugei FSCHUFPB 5471 Serra Negra do Norte/RN, Brazil (9) -6.596 -37.252

L. klugei FSCHUFPB 5745 Serra Negra do Norte/RN, Brazil (9) -6.596 -37.252

L. klugei FSCHUFPB 5747 Serra Negra do Norte/RN, Brazil (9) -6.596 -37.252

L. klugei FSCHUFPB 00073 Monte Alegre de Sergipe/SE, Brazil

(15)

-10.030 -37.561

L. klugei FSCHUFPB 00104 Monte Alegre de Sergipe/SE, Brazil

(15)

-10.030 -37.561

L. klugei CBC 74 Poço Redondo/SE, Brazil (14) -9.664 -37.683

L. klugei CBC 75 Poço Redondo/SE, Brazil (14) -9.664 -37.683

L. klugei RAS 43 Poço Redondo/SE, Brazil (14) -9.664 -37.683

L. klugei RAS 45 Poço Redondo/SE, Brazil (14) -9.664 -37.683

Lygodactylus sp.

"São Domingos"

CHUNB 33596 São Domingos/GO, Brazil -13.450 -46.450

Lygodactylus sp.

"São Domingos"

CHUNB 35335 São Domingos/GO, Brazil -13.450 -46.450

Lygodactylus sp.

"São Domingos"

CHUNB 56792 São Domingos/GO, Brazil -13.414 -46.370

Lygodactylus sp.

"São Domingos"

CHUNB 56817 São Domingos/GO, Brazil -13.414 -46.370

L. wetzeli MNHNP11467 Depto. Boquerrón, Estancia Amistad,

Paraguay

-24.006

-60.722

L. wetzeli MNHNP11472 Depto. Boquerrón, Estancia Amistad,

Paraguay

-24.006

-60.722

Abbreviations: BA, Bahia State; CE, Ceará State; GO, Goiás State; PE, Pernambuco State; PI, Piauí

State; RN, Rio Grande do Norte State; SE, Sergipe State.

75

Table S2: Information about gene, primers, and PCR protocols used in this study.

Gene Length

(bp)

PCR profile Primer Author

ND4

679 Cytb50 F:

CACCTATGACTACCAAAAGCTCATGTAGAAGC

R: CATTACTTTTACTTGGATTTGCACCA

Arévalo et al. (1994)

DMXL1

691 1Touch-57 F2: GTCTAGGGAGGATGGTTCACATA

R2: GAATGAAGCAAGTGACSAGAAAGA

Werneck et al. (2012)

DNAH3

698 Nu52 F1: GGTAAAATGATAGAAGAYTACTG

R6: CTKGAGTTRGAHACAATKATGCCAT

Townsend et al. (2008)

PRLR

584 ANL63 F1: GACARYGARGACCAGCAACTRATGCC

R3: GACYTTGTGRACTTCYACRTAATCCAT

Townsend et al. (2008)

Cytb50: initial denaturation at 94° C (5:00min); 40 cycles consisting of: (1) denaturation at 94° C

(1:00min), (2) annealing at 50° C (1:00min), and (3) extension at 72° C (1:00min); final extension at 72°

C (7:00min); final rest at 10º C.

1Touch-57: initial denaturation at 94º C (2:45min); 40 cycles consisting of: (1) denaturation at 94º C

(15s), (2) annealing at 57º C (-0.1º C for cycle) (20s), (3) extension at 72º C (1:00min); final extension at

72º C (1:00min); final rest at 10º C.

Nu52: initial denaturation at 94º C (3:00min); 40 cycles consisting of: (1) denaturation at 94º C

(1:00min), (2) annealing at 52º C (1:00min), (3) extension at 72º C (1:00); final extension at 72º C

(5:00min); final rest at 10º C.

ANL63: initial denaturation at 95º C (1:30min); 10 cycles consisting of: (1) denaturation at 95º C (35s),

(2) annealing at 63º C (-0.5º C for cycle) (35s), (3) extension at 72º C (1:00min); 10 cycles consisting of:

(1) denaturation at 95º C (35s), (2) annealing at 58º C (35s), (3) extension at 72º C (1:00min); 15 cycles

consisting of: (1) denaturation at 95º C (35s), (2) annealing at 52º C (35s), (3) extension ate 72º C

(1:00min); final rest at 10º C.

76

Figure S1: Gene trees for ND4 (A), DMXL1 (B), DNAH3 (C), and PRLR (D). Individuals are

highlighted according to GMYC results. Individuals in pink correspond to northern lineage and

green correspond to southern lineage. Nodes with posterior probability higher than 95% are

marked with an asterisk (*).

77

Figure S2: Genetic structure of Lygodactylus klugei based on nuclear markers performed in

Structure. (A) probability of individuals' assignment to each population (green and red). (B) map

with localities colored according Structure population assignment; Green triangles correspond to

population 1 and red circles correspond to population 2; Blue line correspond to the São

Francisco River and A correspond to São Francisco River paleocourse.

78

REFERENCES

Arévalo, E., Davis, S.K., Sites Jr., J.W., 1994. Mitochondrial DNA sequence divergence and

phylogenetic relationships among eight chromosome races of the Scoleporus grammicus

complex (Phrynosomatidae) in Central Mexico. Systematic Biology 43, 387–418.

Townsend, T.M., Alegre, R.E., Kelley, S.T., Wiens, J.J., Reeder, T.W., 2008. Rapid

development of multiple nuclear loci for phylogenetic analysis using genomic resources:

An example from squamate reptiles. Mol Phylogenet Evol 47, 129–142.

Werneck, F.P., Gamble, T., Colli, G.R., Rodrigues, M.T., Sites, J.W., Jr., 2012. Deep

diversification and long-term persistence in the South American 'Dry Diagonal': integrating

continent-wide phylogeography and distribution modeling of geckos. Evolution 66, 3014–

3034.

79