Andreia Miraldo - University of Helsinki

Transcrição

Phylogeography and population dynamics of
secondary contact zones of Lacerta lepida in the
Iberian Peninsula
Andreia Miraldo
A thesis submitted for the degree of Doctor of Philosophy
Norwich, June 2009
© This copy of the thesis has been supplied on condition that anyone who consults it is understood to
recognise that its copyright rests with the author and that no quotation from the thesis, nor any
information derived there from, may be published without the author's prior, written consent.
Phylogeography and population dynamics of secondary contact
zones of Lacerta lepida in the Iberian Peninsula
Ângela Andreia Firmino Miraldo
June 2009
Lacerta lepida is a lizard species that occurs throughout the Iberian Peninsula.
Detailed phylogeographic analysis of the species using mitochondrial DNA and
nuclear sequence data revealed a history of population fragmentation and
diversification in allopatry. Diversification within the species was estimated to have
started in the Miocene probably related to geological events of the region,
nevertheless a strong influence of Pleistocene climatic oscillations were also
detected. Several glacial refugia and demographic range expansions after
diversification in allopatry were detected leading to the establishment of several
secondary contact zones. Detailed analysis of two secondary contact zones within
the species was carried out.
One of the secondary contact zones was characterized by the existence of intraindividual mitochondrial polymorphism. The origin of the polymorphism was
identified to be the result of introgression of mitochondrial DNA fragments from one
lineage into the nuclear genome of the other (Numts), suggesting that hybridization
between the lineages occurred. Detailed phylogeographic analysis of the identified
Numts allowed the inference of lineages recent demographic events. Additionally,
further analysis of the polymorphic samples detected within this contact zone
revealed the existence of low levels of heteroplasmy and mitochondrial DNA
recombination, which until now was rarely reported for natural populations in the
literature.
Gene flow dynamics was assessed in another zone of secondary contact
between two very divergent mitochondrial lineages, located in south-eastern Spain.
The use of mitochondrial DNA and microsatellite data allowed the detection of
restricted gene flow amongst the lineages. It was postulated that the two lineages are
on independent evolutionary paths, and therefore should be considered as two
different species.
The molecular tools used throughout this study revealed that geological events,
climatic changes, hybridization, and speciation have shaped the evolutionary history
of Lacerta lepida.
i
Acknowledgments
I would like to thank Brent Emerson for the supervision and guidance given
throughout the development of this thesis. His constant support, encouragement and
friendship were an invaluable help for the successful completion of this PhD.
I am extremely thankful to Godfrey Hewitt with whom I had the pleasure to share so
many exciting conversations over these four years and during which I was able to
absorb his contagious enthusiasm about Science and History.
To Octavio Paulo I am very thankful for introducing me to the amazing world of
reptiles and for giving me the opportunity to develop this project.
I would also like to thank Sara Goodacre for all the help given by introducing me to
the lab and guiding me through my first months at UEA. Thank you for your time
and for being always so happy to help.
To Paul Dear I am grateful for sparing his time and knowledge with me and
accepting me to join his lab every now and then.
This project would have not been possible without the help from all the people I had
the pleasure to do fieldwork with. Their continuous enthusiasm despite the hard task
they faced every day while chasing lizards is definitely impressive: Pedro
Silveirinha, Rui Osório, Rita Jacinto, Nuno Valente, Juan Pablo de la Vega, Luís
Garcia, Gabriel Marin, Eugenia Zarza-Franco and Brent Emerson.
Special thanks go to my father, Mário Miraldo, and grandfather, António Firmino,
who have spent many weekends developing the best lizard traps ever, which lead to
the successful capture of hundreds of clever lizards!
Aos meus pais, irmã e avós um muito obrigado por todo o apoio, carinho e
inspiração. Obrigado pelos exemplos de preserverança e espírito de aventura que
sempre incutiram em mim. É a voçês que dedico esta tese!
E finalmente para ti Matthew um muito obrigado pela companhia ao longo destes
anos!
This work was funded by Fundacao para a Ciencia e Tecnologia through a PhD scholarship
(SFRH/BD/1696/2004) and a research fellowship (POCTI/BSE/48365/2002)
ii
Crise de vocação
Parte I
Num frágil batel vou navegando
Empoleirado nas vagas alterosas
Olhando para a lua e os astros namorando
O temeroso mar da vida vou sulcando
À procura de enseadas radiosas.
Três anos porém já são passados
Sem ter conseguido o meu intento
Três anos de ilusão amargurados
Três anos a negro já passados
Três anos levados pelo vento.
Impelido pel’ ansiedade que me vence
‘Inda pensando num sonho que se afasta
Olho a lonjura e, com gáudio fremente
Mui longe praia diviso atraente.
Mas a meus olhos fenece num repente
Essa miragem fatídica e nefasta.
Se num dia parecer estar ciente,
Olhando por acaso para o Norte
Que aquela que julgava inexistente
Eu vejo, enfim, com sorriso que se sente
Ser meu na vida e não me deixar na morte,
Esquecerei para sempre este lamento
E a vida não será mais um tormento,
Mas um céu de luz, de paz e boa sorte.
Parte II
E depois de tantos anos que passaram
Muitas praias radiosas encontrei
Umas delas as dores me afagaram
Outras tantas a naufrágios me levaram
Por não serem, afinal, o que sonhei
Ao vencer cada borrasca com bravura
E crer que a vida só vitórias não gerou
Se vejo o sol para além da nuvém escura
Seja qual for a dôr ou a tortura
Desistir da luta eu já não vou.
Como não desistiu a micro-criatura
Que ao vencer milhões em luta dura
No ovo materno me gerou.
A poem by Mário Miraldo
iii
Table of Contents
General abstract
i
Acknowledgments
ii
Poem by Mário Miraldo
iii
Table of contents
iv
1. General introduction ............................................................................................. 1
1.1. The Iberian Peninsula ................................................................................................ 2
1.1.1. Geological history and geographic aspects of the Iberian Peninsula ............. 3
1.1.2. The role of the Iberian Peninsula during the Quaternary climatic oscillations6
1.2. Lacerta lepida ............................................................................................................ 7
1.3. Molecular tools ......................................................................................................... 9
1.3.1 The pitfalls of mitochondrial DNA: Numts, heteroplasmy and recombination11
1.4 Thesis structure ........................................................................................................ 13
1.5. References............................................................................................................... 15
2. Intra-individual mitochondrial DNA polymorphism in a reptilian secondary
contact zone .............................................................................................................. 24
2.1. Abstract ................................................................................................................... 24
2.2. Introduction ............................................................................................................ 25
2.3. Materials and Methods ........................................................................................... 28
2.3.1. Sampling ....................................................................................................... 28
2.3.2. DNA extraction, amplification and sequencing ............................................ 28
2.3.3. Identification of polymorphic individuals and quantification of intra-individual
variation.................................................................................................................. 29
2.3.4. Amplification of the entire cytochrome b gene ........................................... 30
2.3.5. Haplotype network construction ................................................................. 30
2.4. Results ..................................................................................................................... 32
2.4.1 Characterization of polymorphism ................................................................ 32
2.4.2. Characterization of Numts and intra-individual variation............................ 33
2.4.3. Phylogeographic analysis ............................................................................. 35
2.5. Discussion........................................................................................................... 38
2.5.1. Phylogeographic history of lineages L3 and L5............................................. 38
2.5.2. Origin of the polymorphism ......................................................................... 40
2.5.3. Phylogeographic utility of Numts ................................................................. 42
2.6. Conclusion ............................................................................................................... 42
2.7. References ............................................................................................................... 55
3. Phylogeography of Lacerta lepida in the Iberian Peninsula ............................. 62
3.1 Abstract .................................................................................................................... 62
3.2. Introduction............................................................................................................. 63
3.3. Materials and methods ........................................................................................... 67
3.3.1. Sampling strategy collection......................................................................... 67
3.3.2. Laboratory procedures ................................................................................. 68
3.3.3. Phylogeographic and historical demographic analysis ................................. 69
3.3.4. Estimation of divergence times .................................................................... 72
3.4. Results ..................................................................................................................... 74
3.4.1. Mitochondrial DNA data ............................................................................... 74
3.4.2. Nuclear DNA data ......................................................................................... 77
3.4.3. Divergence times .......................................................................................... 78
3.5. Discussion ................................................................................................................ 78
3.5.1. Mitochondrial DNA data ............................................................................... 79
3.5.2. Nuclear DNA data ......................................................................................... 80
3.5.3. Historical biogeography of Lacerta lepida .................................................... 82
3.6. Conclusion ............................................................................................................... 85
3.7. References ............................................................................................................... 99
4. Genetic analysis of a secondary contact zone between Lacerta lepida lepida and
Lacerta lepida nevadensis ....................................................................................... 106
4.1 Abstract .................................................................................................................. 106
4.2. Introduction........................................................................................................... 107
4.3. Materials and methods ......................................................................................... 110
4.3.1. Sampling strategy collection....................................................................... 110
4.3.2. Laboratory procedures ................................................................................ 111
4.3.3. Data analyses .............................................................................................. 112
4.4. Results ................................................................................................................... 116
4.4.1. Mitochondrial DNA data ............................................................................. 116
4.4.2. Nuclear DNA data ....................................................................................... 117
4.4.3. Cline analysis............................................................................................... 119
4.5. Discussion .............................................................................................................. 120
4.5.1. Genetic structure of the contact zone: tension zone vs neutral diffusion . 121
4.5.2. The historical dynamics of lineages contact and introgression.................. 124
4.5.3. Taxonomic and conservation implications ................................................. 125
4.6. References ............................................................................................................. 141
5. Testing for the presence of heteroplasmy in Lacerta lepida through single
molecule PCR ......................................................................................................... 148
5.1. Abstract ................................................................................................................. 148
5.2. Introduction........................................................................................................... 149
5.3. Material and methods ........................................................................................... 152
5.3.1. Sample selection and DNA extraction ........................................................ 152
5.3.2. Estimation of the number of template copies ........................................... 152
5.3.3. Selection of loci and design of PCR primers ............................................... 153
5.3.4. PCR amplifications, scoring and sequencing .............................................. 155
5.4. Results and discussion ........................................................................................... 156
5.4.1. Ruling out Numts ........................................................................................ 156
5.4.2. Ruling out contamination ........................................................................... 157
5.4.3. Heteroplasmy and mtDNA recombination ................................................. 158
5.4.4. Origin of heteroplasmy and recombination in Lacerta lepida ................... 159
5.5. Conclusion ............................................................................................................. 160
5.6. References ............................................................................................................. 165
6. General discussion and conclusions ................................................................. 172
Chapter 1
General introduction
Photo by Nuno Valente
Cerro del Mencal, Pedro Martínez, Andalucia, Spain*
*Sampling site 5 in chapter 4
1. General introduction
One hundred and fifty one years ago Charles Darwin (Darwin, 1958) and
Alfred Wallace (Wallace, 1958) presented to the world their theory about the
evolution of species which has changed forever the way the natural world is
perceived. The revolutionary, yet very simple theory of evolution later compiled in
Darwin’s publication “On the Origin of Species” (Darwin, 1859) was based on two
basic concepts: the concept of a “tree-of-life”, the idea that all species diverged from
a common ancestor along separate pathways; and the concept of natural selection, a
process by which species change and adapt to different environments and thereby
identified as the key mechanism responsible for the branching in the “tree-of-life”.
Although recent research has disclosed that the verticality relation between ancestor
and descent in the Darwinian “tree-of-life” does not always adequately describe how
evolution works (e.g. it does not describe the lateral transfer of genes which is now
known to have play a very important role in the evolution of archaea and some
bacteria), the general concept of the evolution of species holds true making the
theory of evolution one of the most important and comprehensive scientific theories
ever published.
The publication of Darwin’s theory of evolution marked the
emergence of evolutionary biology as a new and prosperous field of research
concerned with the understanding of the mechanisms responsible for the isolation of
populations, their differentiation and ultimately speciation.
However, it is only relatively recently that genetic tools have been developed
that make it possible to describe and quantify the genetic diversity found within
species and gain a detailed understanding of the evolutionary process. An important
contribution of genetic studies is the recognition that almost all species show some
1
level of genetic structure (Avise et al. 2000). Following from this it is now clear that
studying geographic patterns of the distribution of genetic variation can provide
understanding of species evolution. Throughout this thesis questions related to the
origin and distribution of intraspecific genetic diversity and the historical and
contemporary processes involved in creating and maintaining it will be addressed.
These questions will be specifically addressed in Lacerta lepida, a lizard species that
is mainly distributed in the Iberian Peninsula, a region that is known to have played a
key role in species survival and diversification within Europe through several glacial
cycles.
1.1. The Iberian Peninsula
The Iberian Peninsula is extremely rich in species diversity and endemism.
Although occupying only a small part of Western Europe (<6%), as much as 50% of
the European plant and terrestrial vertebrate species occur in this Peninsula.
Additionally, of the approximately 900 European endemic species, 31% occur in the
Iberian Peninsula (Williams et al., 2000). Its importance for biodiversity
conservation has been acknowledged by conservation policies: the Mediterranean
biodiversity hotspot from Conservation International included nearly 80% of the
Iberian Peninsula area (Myers et al., 2000), whereas the European network of
important sites for conservation (Natura 2000 network) included more than 20%.
Furthermore, the high biodiversity and endemism levels found in this peninsula are
accompanied by high levels of intraspecific genetic diversity, as disclosed by several
phylogeographic studies in the region. The reasons behind such diversity richness
are varied, and are thought to be related to long-term species persistence in the
region. Nevertheless, it is known that species distribution and the distribution of
genetic variation are extremely variable in space and time and they are influenced by
many different factors. Amongst them are major geological events, climatic
oscillations and environmental changes, all of which are known to have promoted
vicariant events and have therefore been invoked to explain species diversification
(Avise 2000).
2
1.1.1. Geological history and geographic aspects of the Iberian
Peninsula
The geographic position of the Iberian Peninsula within Europe is extremely
peculiar. The peninsula connects to Europe by a relatively narrow and mountainous
region (the Pyrenees) which is known to constitute an important physical barrier for
species dispersal. At the same time, the Iberian Peninsula is the region within Europe
with the closest proximity to the African continent with separation from northern
Africa by only 14Km across the Strait of Gibraltar. These factors have important
implications in the evolutionary histories and distributions of species within this
peninsula. Indeed, the biogeographic history of the Iberian Peninsula is intimately
associated with North Africa, and this is reflected on the existence of a great number
of closely related species in both sides of the Mediterranean Sea, as detected by early
biogeographic studies (e.g. Busack, 1986).
The biogeography of Iberia and North Africa was greatly affected by
common geological events associated with the Alpine orogeny in the middle
Tertiary. The Alpine orogeny was mainly a mountain-building event which is known
to have had an extreme effect in southern Europe and the Mediterranean basin
(Rosenbaum et al., 2002a; Rosenbaum et al., 2002b). During the Oligocene (± 30
Mya) the area between the Iberian Peninsula and southern France suffered
fragmentation and several land masses became disjunct from the main European
continental block (Fig. 1.1.a). These land masses currently correspond to the Betic
region in Spain, the Rif region in Morocco, the Kabylies in Algeria, and the Balearic
Islands, Sardinia and Corsica throughout the Mediterranean Sea. Since the time of
disjunction of the Betic and Rif land masses from the European continental block to
their final collision with the Iberian Peninsula and North Africa respectively, several
important events are registered which allowed the occurrence of dispersal and
vicariance between these two regions. After disjunction and during the drifting
process, the Betic and Rif regions remained as part of the same land mass for most of
the time. The first land corridor between Iberia and North Africa was established
±15 Mya (Weijermars 1991) (Fig. 1.1.b) with the collision of this block with these
3
two regions simultaneously. Later, ±10-8 Mya, due to the opening of the Betic,
Guadalhorce and Rif sea corridors this important land bridge between the two
continents suffered fragmentation and the connection was broken (Fig. 1.1.c). The
land connection between the Iberian Peninsula and North Africa became reestablished (Fig. 1.1.d) with the sequential closure of the three sea corridors, starting
with the Betic 7.8-7.6 Mya (Krijgsman et al., 2000; Krijgsman et al., 1999),
followed by the Guadalhorce 6.8-6.7 Mya (Martín et al., 2001) and finally the Rif
sea corridor 6.7-6.0 Mya (Krijgsman and Langereis, 2000).
a)
30 Mya
b)
15 Mya
B+R
B+R
c)
10-8 Mya
d)
8-6 Mya
Betic corridor
B
Guadalhorce
R
Rif corridor
Fig. 1.1. Reconstruction of the main geological events of the western Mediterranean from 30 Mya
until 6 Mya. a) 30 Mya parts of the Iberian Peninsula and southern France suffer fragmentation and
disjunction from main land. b) 15 Mya the Betic-Rif land masses establish a transient connection
between Iberia and North Africa. c) 10-8 Mya the Betic massif suffers fragmentation and three sea
corridors emerge: the Betic, the Guadalhorce and the Rif sea corridors. d) 8-6 Mya the sea corridors
close in a sequential order, reconnecting Iberia and North Africa again. Adapted from Rosenbaum et
al. (2002a).
4
More recently ±5.96 Mya, another important event in the geological history
of the Mediterranean basin started with the closure of the connection between the
Atlantic Ocean and the Mediterranean Sea. This resulted in the almost complete
desiccation of the Mediterranean Sea, a period known as the Messinian salinity crisis
(MSC) (Hsu et al., 1977). During the MSC (5.9 to 5.3 Mya) the European and the
North African Mediterranean coasts became connected through extensive land
bridges, again allowing the dispersal of terrestrial species throughout the basin. With
the refilling of the Mediterranean by the opening of the modern Strait of Gibraltar,
terrestrial species distributed on both sides of the Mediterranean became isolated,
although for certain groups, not strongly restricted in their dispersal by sea, exchange
of individuals through the Strait is known to have occurred relatively recently (e.g.
Late Pleistocene) (Alvarez et al., 2000; Carranza et al., 2004; Lenk et al., 1999;
Schmitt et al., 2005).
The series of vicariant and dispersal events that occurred during the
geological history of the Mediterranean basin have left characteristic signatures in
the distributions of some species within this region. The geological events mentioned
above have been invoked to explain biogeographic patterns for a number of species
that exhibit genetic divergences concordant with those events. For example, Buthus
spp. scorpions (Gantenbein and Largiadèr, 2003) show evidence for the occurrence
of dispersal between the Iberian Peninsula and North Africa 15-14 Mya, when these
regions are first known to have been in contact (see above). Events for allopatric
speciation related to the fragmentation of the Betic region between 10-8 Mya have
been invoked to explain why species of midwife toad (Alytes spp.) inhabiting
southern Iberia are more closely related to North African species than to other
Iberian species (Martínez-Solano et al., 2004). The sequential closure of the 3 sea
corridors between 8-6 Mya have also recently been invoked to explain putative
speciation events in the ocellated lizards (Lacerta spp.) (Paulo et al., 2008). Finally,
for many species, divergences between Iberian and North African groups have been
shown to have an origin coinciding with the opening of the Strait of Gibraltar, which
marked the end of the MSC 5.3 Mya (e.g Blanus worm lizards (Vasconcelos et al.,
2006) and Discoglossus frogs (Zangari et al., 2006)).
5
1.1.2. The role of the Iberian Peninsula during the Quaternary
climatic oscillations
In Europe, the effect of Quaternary climatic oscillations in the distribution of
genetic variation has been well studied over recent years and its important role in the
evolution of species has long been recognized (Hewitt, 1996; Hewitt, 2000; Hewitt,
2004). The accumulation of phylogeographic studies within this continent suggests
a general pattern of species contraction into southern refugia during glacial periods
followed by expansions to northern latitudes during warm interglacials (Hewitt,
1996; Hewitt, 1999; Hewitt, 2004; Taberlet et al., 1998), a pattern strongly supported
by pollen and fossil data (Zagwijn, 1992a; Zagwijn, 1992b). These dramatic changes
in the geographic distribution of species have left signatures in the geographic
distribution of genetic variation, with southern populations typically exhibiting
higher genetic diversity than northern ones (Hewitt, 1996; Hewitt, 2000; Hewitt,
2004).
Paleontological and palynological studies from the southern European
peninsulas of Iberia, Italy and the Balkans suggest that these regions have remained
relatively stable throughout the Quaternary, allowing for the persistence and survival
of species during adverse climatic conditions which were more pronounced at
northern latitudes (Roucoux et al., 2005; Tzedakis et al., 2002).
Amongst the
southern European peninsulas, Iberia is the best studied in terms of phylogeography,
and several studies show that species have in fact persisted there through several ice
ages (e.g. Paulo et al., 2001). This pattern of long-term persistence is also supported
by the relatively high biological diversity and endemicity found within this peninsula
(Blondel and Aronson, 1999; García-Barros et al., 2002; Williams et al., 2000). The
long-term persistence of populations in the Iberian Peninsula is intimately associated
with its southern geographic position within Europe, which allowed it to remain
almost entirely ice free during the glaciations, and with the high topographic and
climatic heterogeneity that characterizes it. The latter seems to be intimately
associated with the Northern Atlantic and Mediterranean climatic influences which
maintain distinct microclimates that change in a northwest-southeast direction in this
Peninsula. This climatic gradient is reflected in strong genetic differentiation within
Iberia between western and eastern groups as reported for several species (Batista et
6
al., 2004; Carranza et al., 2006; Paulo et al., 2001; Schmitt and Seitz, 2004).
However the majority of studied species exhibit more complex phylogeographic
structures beyond a simple east-west split. This complexity is associated with
isolation in several distinct refugia, a pattern inferred to be related to the high
physiographic complexity and multiplicity of climates that the Iberian peninsula
offers (for a review on the subject see Gomez and Lunt, 2007 and references
therein). Signatures of post-glacial demographic range expansions and the
establishment of secondary contact zones, with differing degrees of admixture, as a
result of range expansions from distinct glacial refugia are also reported (e.g.
Godinho et al., 2008; Godinho et al., 2006; Martínez-Solano et al., 2006; Sequeira et
al., 2005).
1.2. Lacerta lepida
Lacerta lepida (Daudin 1802) together with Lacerta pater (Lataste 1880),
Lacerta tangitana (Boulenger 1888) and Lacerta princeps (Balnford 1874) form the
ocellated lizards, a subset of approximately 50 species within the genus Lacerta,
which have a continuous distribution in south-western Europe (L. lepida) and North
Africa (L. pater and L. tangitana) and with one species occurring disjunctly (L.
princeps) in parts of the Middle East (Fig. 1.2.). Some authors (e.g. Arnold et al.,
2007) consider that the ocellated lizards should belong to a different genus, Timon,
and therefore it is common to find these four species referred to as Timon spp. As
there is still some controversy around the phylogenetic relationships within the genus
Lacerta (Fu, 1998; Fu, 2000; Harris et al., 1998), the classic nomenclature, Lacerta
spp., will be used throughout this thesis.
The ocellated lizards belong to the family Lacertidae, which in turn belongs
to the order Squamata. Lacerta lepida, the largest European lacertid lizard, is the
only species from the ocellated lizards to occur in Europe, where it occupies almost
all of the Iberian Peninsula, southern France and north-western Italy (Castroviejo
and Mateo, 1998; Mateo and Castroviejo, 1990; Mateo et al., 1996). Four subspecies
within Lacerta lepida are recognized: Lacerta lepida iberica, which occurs in the
7
north-western corner of the Iberia peninsula; Lacerta lepida nevadensis, which
occurs in the south-western parts of Spain, mainly associated with the Betic
mountain ranges; Lacerta lepida lepida, the nominal species and which has the
widest distribution within the group and occurs in all remaining parts of the Iberian
Peninsula, southern France and north-western Italy; and finally Lacerta lepida
oteroi, which is restricted to the island of Salvora in northern Spain. The subspecies
designations are based on morphological patterns, allozymes and in one case (L. l.
oteroi) chromosomes (Castroviejo and Mateo, 1998; Mateo, 1988; Mateo and
Castroviejo, 1990; Mateo and López-Jurado, 1994; Mateo et al., 1996).
Fig. 1.2. Distribution of ocellated lizards (grey shadded area). Lacerta lepida occurs in Europe;
Lacerta tangitana and Lacerta pater occur in Africa and Lacerta princeps occurs in the Middle East.
The majority of research involving L. lepida has been concerned mainly with
its geographic distribution (e.g. Castroviejo and Mateo, 1998; Cheylan and Grillet,
2005; Mateo, 1988; Mateo and López-Jurado, 1994), ecology (e.g. Castilla and
Bauwens, 1992; Castilla et al., 1991; Mateo, 1988), behaviour (e.g. Castilla and
Bauwens, 1989; Castilla and Bauwens, 1990; Mateo and Castanet, 1994; Paulo,
1988) and morphological and cariological differences within the species (e.g. Mateo,
1988). The first study concerned with the distribution of genetic variation within
Lacerta lepida was carried out in 1988 by Mateo, using allozymes as molecular
markers (Mateo, 1988; Mateo et al., 1996). Although the three continental Lacerta
lepida subspecies analysed in those studies formed a homogenous group, relatively
8
higher Nei’s genetic distances were detected between the subspecies L. l. nevadensis
and the other two subspecies. Subsequent to this Paulo (2001) carried out a
phylogeographic study of the species using a more geographically representative
sampling strategy and mitochondrial genealogies. The results suggested a west-east
gradient of genetic diversity and several mitochondrial lineages with nonoverlapping geographic ranges were described. The results suggested a history of
allopatric differentiation in multiple refugia during the Plio-Pleistocene, in
concordance with the patterns described for other species within the Iberian
Peninsula. The geographic subdivision and population divergence detected by Paulo
(2001) do not correspond exactly to the patterns previously obtained by Mateo et al.
(1996) with morphology and allozyme data. More recently, data from a phylogenetic
study regarding Lacerta spp. involving mitochondrial and nuclear genealogies
suggested elevating the subspecies L. l. nevadensis to a new species, due to the high
levels of mitochondrial genetic differentiation detected (Paulo et al., 2008). Despite
this apparently deep historical subdivision within Lacerta lepida, as revealed by the
mitochondrial genealogy (Paulo et al., 2008) and some level of differentiation at
allozyme markers (Mateo et al., 1996), it still remains unclear whether these
divergent subspecies are evolving independently, or if they are in fact coalescing due
to high levels of gene flow at zones of secondary contact.
1.3. Molecular tools
Our understanding of the genetic structure of populations and species is
greatly informed by the field of phylogeography (Avise 2000). Phylogeography
emerged as an integrative discipline bridging the traditional independent fields of
population genetics and molecular phylogenetics, thus providing a valuable
contribution to the understanding of the evolutionary pattern and process.
Phylogeography, as initially formulated, has developed by integrating information
from other fields including demography, historical geography, ethology and
molecular genetics (Avise, 1998), which has considerably enhanced our ability to
9
analyze and interpret the genetic structure of populations and species over space and
time.
The discovery of a fast pace of molecular evolution for mitochondrial DNA
thirty years ago (Brown et al., 1979) strongly shaped the field of phylogeography,
which recognized the great utility of this fast evolving molecule as a source of
information in an evolutionary context (Avise et al., 1987; Harrison, 1989). In fact
until now phylogeography has relied mainly on data derived from this molecule
(Avise, 2004; Avise, 2009). Apart from the high mutation rate generating enough
signal to make inferences about population history over short time frames, other
important characteristics have made this molecule the tool of choice in
phylogeographic studies. In animals, mitochondrial DNA has an almost exclusively
maternal, non-recombining mode of inheritance that enables evolutionary histories to
be reconstructed without the complexities introduced by biparental recombination.
Additionally, the uniparental inheritance of mtDNA and its haploid state reduce its
effective population size to one quarter of that of nuclear genes which facilitates a
rapid lineage sorting. Thus, the power for detecting splitting events when using this
molecule when compared to nuclear genes is highly increased (Moore, 1995).
Nevertheless it has become widely recognized that studies relying solely on
mitochondrial DNA have important limitations (reviewed in Zhang and Hewitt,
2003). As a non-recombining molecule mitochondrial DNA will give us information
of a single locus only and the evolutionary history of that molecule might not be
consistent with the evolutionary history of the organism being studied due to
stochastic processes (Harrison, 1989). Furthermore, retention of ancestral
polymorphism, introgressive hybridization (e.g. Alves et al., 2003), natural selection
(Ballard and Whitlock, 2004; Bazin et al., 2006), sex-biased dispersal (e.g. Piertney
et al., 2000) amongst other factors, can dramatically influence phylogeographic
results based entirely on mitochondrial DNA. These problems can be addressed by
the combined use of nuclear and mitochondrial genes, which have different modes of
inheritance and different mutation rates. Such combined approaches are likely to
provide insights into complex evolutionary histories as the ones that most likely
characterize species that have persisted in the Iberia Peninsula for several ice ages
(e.g. Godinho et al., 2008).
10
1.3.1 The pitfalls of mitochondrial DNA: Numts, heteroplasmy
and recombination
When using mitochondrial DNA as a genetic marker several issues have to
be taken into account in order to avoid erroneous conclusions revealed by the data,
i.e. one has to be aware of the pitfalls of mitochondrial DNA. One of those pitfalls is
the existence of Numts - mitochondrial fragments that are incorporated in the nuclear
genome. It is known that parts of the mitochondrial genome can be transferred to the
nucleus, a phenomenon which has been demonstrated in various taxa (for a review
on the subject see Bensasson et al., 2001; Zhang and Hewitt, 1996). The
mitochondrial fragments that are incorporated in the nucleus have different
evolutionary histories from their mitochondrial counterparts and therefore can
seriously compromise phylogenetic and phylogeographic studies if wrongly
incorporated in mitochondrial datasets (e.g. Arctander and Fjeldsa, 1994).
Nevertheless, when correctly identified Numts can be useful as they represent relicts
of mitochondrial DNA providing therefore information regarding the ancestral state
of mitochondrial genealogies (Bensasson et al., 2001). The correct identification of
mitochondrial DNA sequences, and distinguishing them from Numts, are essential
steps in any study involving mtDNA as a genetic marker.
Other important issues concerning mitochondrial DNA relate to the standard
paradigms about its mode of inheritance in animals which, if proven to be wrong,
could seriously compromise the suitability of mtDNA for many of its uses. One of
the paradigms concerns the lack of recombination in animal mitochondrial DNA.
Apart from some bivalve families, which show a unique type of mtDNA inheritance
and where recombination of mtDNA has been detected (Burzynski et al., 2003;
Ladoukakis and Zouros, 2001a), direct evidence that recombination of mtDNA
occurs is scarce and has only been reported in humans (Kraytsberg et al., 2004) and
in the nematode Meloidogyne javanica (Lunt and Hyman, 1997).
In natural
populations, recent strong evidence suggests that it occurred in one individual of the
Australian frillneck lizard (Chlamydosaurus kingii) (Ujvari et al., 2007) and in one
hybrid of Salmo salar and Salmo trutta (Ciborowski et al., 2007). Nevertheless,
surveys of published data sets suggest that the phenomenon can be more widespread
and frequent (Awadalla et al., 1999; Ladoukakis and Zouros, 2001b; Piganeau et al.,
11
2004; Tsaousis et al., 2005) and recent studies report indirect evidence for its
occurrence in a variety of systems: butterflies (Andolfatto et al., 2003), scorpions
(Gantenbein et al., 2005), silkmoths (Arunkumar et al., 2006) and fish (Guo et al.,
2006). Furthermore, in the last years an increasing number of studies also reported
the occurrence of bi-parental inheritance of mitochondrial DNA, through paternal
leakage. The consequences of this are that more than one type of mitochondrial
DNA (heteroplasmy) are reported to co-occur in the same individual across several
taxa: in birds (Kvist et al., 2003), Drosophila (Kondo et al., 1990; Sherengul et al.,
2006), mice (Gyllensten et al., 1991; Kaneda et al., 1995; Shitara et al., 1998), cows
(Steinborn et al., 1998; Sutovsky et al., 2000), cicadas (Fontaine et al., 2007), mites
(Van Leeuwen et al., 2008), fish (Hoarau et al., 2002; Magoulas and Zouros, 1993),
bees (Meusel and Moritz, 1993), sheep (Zhao et al., 2004) and humans (Kraytsberg
et al., 2004; Schwartz and Vissing, 2002). The discovery of heteroplasmy in such a
wide range of taxa and the recently reported cases of clear recombinants between
divergent mitochondrial genomes in natural populations (Ciborowski et al., 2007;
Ujvari et al., 2007) suggest that these phenomena might be more common than
previously suspected and researchers working with mitochondrial DNA should bear
this in mind.
Taking in account the many pitfalls and problems associated with analyses
that rely solely on data obtained from mitochondrial DNA, this study will employ
the use of both cytoplasmic (maternally inherited) and nuclear (biparentaly inherited)
markers to infer the historical and contemporary processes that have shaped the
evolutionary history of Lacerta lepida within the Iberian Peninsula. The
mitochondrial marker used throughout the thesis is the cytochrome b gene. This gene
is commonly used for phylogenetic and phylogeographic studies in vertebrates,
including studies involving Lacerta lepida. The nuclear marker used is the intron 7
of the β-fibrinogen gene that has also been successfully used as a marker in several
vertebrate phylogeographic and phylogenetic studies (Dolman and Phillips, 2004;
Pinho et al., 2008; Sequeira et al., 2006). In a recent phylogenetic study of the genus
Lacerta (Paulo et al., 2008) it was revealed to have sufficient variation within
Lacerta lepida for phylogeographical inference. In addition, microsatellite DNA
markers are employed to address issues related to a more recent time scale, as the
assessment of recent population history.
12
1.4 Thesis structure
This thesis is composed of four data chapters, each dealing with different
aspects of the evolutionary biology of Lacerta lepida within the Iberia Peninsula.
The first data chapter (chapter 2) assesses the origin of intra-individual
polymorphism at the mitochondrial level detected in individuals distributed through
a putative zone of contact between two divergent mitochondrial lineages in northwestern Iberia. The analysis of mitochondrial DNA sequence data (cytochrome b
gene) from individuals representing the entire distribution area of the two lineages,
and including a detailed sampling at the putative contact zone, provided a detailed
understanding of the recent demographic history of the species in the north-western
region of the Iberian Peninsula and allowed the identification of several Numts as the
source of intra-individual polymorphism.
In chapter 3 the phylogeographic analysis is extended to assess the broader
phylogeographic patterns within Lacerta lepida across its distribution. This was
achieved by using both mtDNA and nDNA derived genealogies. Their contrasting
molecular and population genetic properties facilitated the description of the
phylogeography of this species. Using a coalescence approach exploring the
geographic distribution of ancestral and derived alleles, the probable refugia areas
for each lineage that facilitated their allopatric differentiation were indentified. The
influence of the Quaternary climatic oscillations in generating the phylogeographic
patterns observed was assessed through the estimation of the diversification timings.
The detailed phylogeographic study performed in chapter 3 disclosed the
location of several secondary contact zones between divergent mitochondrial
lineages and the demographic patterns associated with their establishment.
Particularly interesting was the detection of a contact zone between two very
divergent lineages that represent the oldest split within Lacerta lepida, estimated to
have occurred during the Miocene. In chapter 4, gene flow across the putative
contact zone between these two lineages was assessed. This was achieved by using 8
microsatellite loci and mitochondrial DNA data. Although hybridization between the
13
lineages was detected by the presence of F1 hybrids, results suggest that they are on
independent evolutionary trajectories.
In the final data chapter (chapter 5) the polymorphic individuals detected in
chapter 2 are re-analyzed using a single molecule PCR protocol. Results lead to the
detection of low levels of heteroplasmy and evidence for mitochondrial DNA
recombination in Lacerta lepida.
The thesis finishes with a general discussion (chapter 6) where the
implications of the findings of this thesis are discussed in detail and topics for future
research are addressed.
14
1.5. References
Alvarez Y, Mateo JA, Andreu AC, Diaz-Paniagua C, Diez A, Bautista JM (2000)
Brief communication. Mitochondrial DNA haplotyping of Testudo graeca on
both continental sides of the Straits of Gibraltar. Journal of Heredity 91, 3941.
Alves PC, Ferrand N, Suchentrunk F, Harris DJ (2003) Ancient introgression of
Lepus timidus mtDNA into L. granatensis and L. europaeus in the Iberian
Peninsula. Molecular Phylogenetics and Evolution 27, 70-80.
Andolfatto P, Scriber JM, Charlesworth B (2003) No association between
mitochondrial DNA haplotypes and a female-limited mimicry phenotype in
Papilio glaucus. Evolution 57, 305-316.
Arctander P, Fjeldsa J (1994) Andean tapaculos of the genus Scytalopus (Aves,
Rhinocryptidae): a study of speciation patterns using mtDNA sequence data.
In: Conservation genetics (eds. Loeschcke V, Tomiuk J, Jain SK).
Birkhauser, Basel.
Arnold EN, Arribas O, Carranza S (2007) Systematics of the Palaearctic and
Oriental lizard tribe Lacertini (Squamata: Lacertidae: Lacertinae), with
descriptions of eight new genera. Zootoxa 1430.
Arunkumar KP, Metta M, Nagaraju J (2006) Molecular phylogeny of silkmoths
reveals the origin of domesticated silkmoth, Bombyx mori from Chinese
Bombyx mandarina and paternal inheritance of Antheraea proylei
mitochondrial DNA. Molecular Phylogenetics and Evolution 40, 419-427.
Avise JC (1998) The history and purview of phylogeography: a personal reflection.
Molecular Ecology 7, 371-379.
Avise JC (2004) Molecular Markers, Natural History and Evolution, 2nd edn.
Sinauer Associates, Sunderland, Massachusetts.
Avise JC (2009) Phylogeography: retrospect and prospect. Journal of Biogeography
36, 3-15.
Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA,
Saunders NC (1987) Intraspecific phylogeography: the mitochondrial DNA
bridge between population genetics and systematics. Annual Review of
Ecology and Systematics 18, 489-522.
Awadalla P, Eyre-Walker A, Smith JM (1999) Linkage disequilibrium and
recombination in Hominid mitochondrial DNA. Science 286, 2524-2525.
Ballard JWO, Whitlock MC (2004) The incomplete natural history of mitochondria.
15
Batista V, Harris DJ, Carretero MA (2004) Genetic variation in Pleurodeles waltl
Michaelles, 1830 (Amphibia: Salamandridae) across the Strait of Gibraltar
derived from mitochondrial DNA sequences. Herpetozoa 16, 166-168.
Bazin E, Glemin S, Galtier N (2006) Population size does not influence
mitochondrial genetic diversity in animals. Science 312, 570-572.
Bensasson D, Zhang D-X, Hartl DL, Hewitt GM (2001) Mitochondrial pseudogenes:
evolution's misplaced witnesses. Trends in Ecology & Evolution 16, 314-321.
Blondel J, Aronson J (1999) Biology and wildlife of the mediterranean region
Oxford University Press, New York.
Brown WM, George M, Wilson AC (1979) Rapid evolution of animal mitochondrial
DNA. Proceedings of the National Academy of Sciences of the United States
of America 76, 1967-1971.
Burzynski A, Zbawicka M, Skibinski DOF, Wenne R (2003) Evidence for
recombination of mtDNA in the marine mussel Mytilus trossulus from the
Baltic. Molecular Biology and Evolution 20, 388-392.
Busack SD (1986) Biogeographic analysis of the herpetofauna separated by the
formation of the Strait of Gibraltar. National Geographic Research 2, 17-36.
Carranza S, Arnold EN, Wade E, Fahd S (2004) Phylogeography of the false smooth
snakes, Macroprotodon (Serpentes, Colubridae): mitochondrial DNA
sequences show European populations arrived recently from Northwest
Africa. Molecular Phylogenetics and Evolution 33, 523-532.
Carranza S, Harris DJ, Arnold EN, Batista V, Gonzalez de la Vega JP (2006)
Phylogeography of the lacertid lizard, Psammodromus algirus, in Iberia and
across the Strait of Gibraltar. Journal of Biogeography 33, 1279-1288.
Castilla AM, Bauwens D (1989) Reproductive characteristics of the lacertid lizard
Lacerta lepida. Amphibia-Reptilia 10, 445-452.
Castilla AM, Bauwens D (1990) Reproductive and fat body cycles of the lizard,
Lacerta lepida, in central Spain. Journal of Herpetology 24, 261-266.
Castilla AM, Bauwens D (1992) Habitat selection by the lizard Lacerta lepida in a
Mediterranean oak forest Herpetological Journal 2, 27-30.
Castilla AM, Bauwens D, Llorente GA (1991) Diet composition of the lizard
Lacerta lepida in Central Spain. Journal of Herpetology 25, 30-36.
Castroviejo J, Mateo JA (1998) Una nueva subespecie de Lacerta lepida Daudin
1802 (Sauria, Lacertidae) para la Isla de Salvora (España ). Publicaciones de
la Asociacion de Amigos de Doñana 12, 1–21.
16
Cheylan M, Grillet P (2005) Statut passe et actuel du lezard ocelle (Lacerta lepida,
Sauriens, lacertides) en France. Implication en termes de conservation. Vie et
Milleu 55, 15-30.
Ciborowski KL, Consuegra S, Garcia de Leijniz C, Beaumont MA, Wang J, Jordan
WC (2007) Rare and fleeting: an example of interspecific recombination in
animal mitochondrial DNA. Biology Letters 3, 554-557.
Darwin C (1859) On the origin of species by means of Natural Selection, or the
preservation of favoured races in the struggle for Life J. Murray, London.
Darwin C (1958) On the tendency of species to form varieties; and on the
perpetuation of varieties and species by natural means of selection. I. Extract
from an unpublished work on species, II. Abstract of a letter from C. Darwin,
Esq., to Prof. Asa Gray. Journal of the Proceedings of the Linnean Society of
London 3, 45-53.
Dolman G, Phillips B (2004) Single copy nuclear DNA markers characterized for
comparative phylogeography in Australian wet tropics rainforest skinks.
Molecular Ecology Notes 4, 185-187.
Fontaine KM, Cooley JR, Simon C (2007) Evidence for paternal leakage in hybrid
periodical cicadas (Hemiptera: Magicicada spp.). PLoS ONE 2, e892.
Fu J (1998) Toward the phylogeny of the family Lacertidae: implications from
mitochondrial DNA 12S and 16S gene sequences (Reptilia: Squamata).
Molecular Phylogenetics and Evolution 9, 118-130.
Fu J (2000) Towards the phylogeny of the family Lacertidae: Why 4708 base pairs
of mtDNA sequences cannot draw the picture. Biological Journal of the
Linnean Society 71, 203-217.
Gantenbein B, Fet V, Gantenbein-Ritter IA, Balloux Fo (2005) Evidence for
recombination in scorpion mitochondrial DNA (Scorpiones: Buthidae).
Proceedings of the Royal Society B: Biological Sciences 272, 697-704.
Gantenbein B, Largiadèr CR (2003) The phylogeographic importance of the Strait of
Gibraltar as a gene flow barrier in terrestrial arthropods: a case study with the
scorpion Buthus occitanus as model organism. Molecular Phylogenetics and
Evolution 28, 119-130.
García-Barros E, Gurrea P, Luciáñez MJ, Cano JM, Munguira ML, Moreno JC,
Sainz H, Sanz MJ, Simón JC (2002) Parsimony analysis of endemicity and
its application to animal and plant geographical distributions in the IberoBalearic region (western Mediterranean). Journal of Biogeography 29, 109124.
Godinho R, Crespo EG, Ferrand N (2008) The limits of mtDNA phylogeography:
complex patterns of population history in a highly structured Iberian lizard
17
are only revealed by the use of nuclear markers. Molecular Ecology 17,
4670-4683.
Godinho R, Mendonca B, Crespo EG, Ferrand N (2006) Genealogy of the nuclear
beta-fibrinogen locus in a highly structured lizard species: comparison with
mtDNA and evidence for intragenic recombination in the hybrid zone.
Heredity 96, 454-463.
Gomez A, Lunt DH (2007) Refugia within refugia: patterns of phylogeographic
concordance in the Iberian Peninsula. In: Phylogeography of Southern
European Refugia (eds. Weiss S, Ferrand N). Springer, Dordrecht.
Guo X, Liu S, Liu Y (2006) Evidence for recombination of mitochondrial DNA in
triploid Crucian Carp. Genetics 172, 1745-1749.
Gyllensten U, Wharton D, Josefsson A, Wilson AC (1991) Paternal inheritance of
mitochondrial DNA in mice. 352, 255-257.
Harris DJ, Arnold EN, Thomas RH (1998) Relationships of lacertid lizards (Reptilia:
Lacertidae) estimated from mitochondrial DNA sequences and morphology.
Harrison RG (1989) Animal mitochondrial DNA as a genetic marker in population
and evolutionary biology. Trends in Ecology & Evolution 4, 6-11.
Hewitt GM (1996) Some genetic consequences of ice ages, and their role, in
divergence and speciation. Biological Journal of the Linnean Society 58,
247-276.
Hewitt GM (1999) Post-glacial re-colonization of European biota. Biological
Journal of the Linnean Society 68, 87-112.
Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405, 907913.
Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary.
Philosophical Transactions of the Royal Society B: Biological Sciences 359,
183-195.
Hoarau G, Holla S, Lescasse R, Stam WT, Olsen JL (2002) Heteroplasmy and
Evidence for Recombination in the Mitochondrial Control Region of the
Flatfish Platichthys flesus. Molecular Biology and Evolution 19, 2261-2264.
Hsu KJ, Montadert L, Bernoulli D, Cita MB, Erickson A, Garrison RE, Kidd RB,
Melieres F, Muller C, Wright R (1977) History of the Mediterranean salinity
crisis. Nature 267, 399-403.
Kaneda H, Hayashi J, Takahama S, Taya C, Lindahl K, Yonekawa H (1995)
Elimination of paternal mitochondrial DNA in intraspecific crosses during
18
early mouse embryogenesis. Proceedings of the National Academy of
Sciences 92, 4542-4546.
Kondo R, Satta Y, Matsuura ET, Ishiwa H, Takahata N, Chigusa SI (1990)
Incomplete Maternal Transmission of Mitochondrial-DNA in Drosophila.
Genetics 126, 657-663.
Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS, Clayton DA,
Vissing J, Khrapko K (2004) Recombination of human mitochondrial DNA.
Science 304, 981-981.
Krijgsman W, Garcés M, Agustí J, Raffi I, Taberner C, Zachariasse WJ (2000) The
"Tortonian" salinity crisis' of the eastern Betics (Spain). Earth and Planetary
Science Letters 181, 497-511.
Krijgsman W, Hilgen FJ, Raffi I, Sierro FJ, Wilson DS (1999) Chronology, causes
and progression of the Messinian salinity crisis. Nature 400, 652-655.
Krijgsman W, Langereis CG (2000) Magnetostratigraphy of the Zobzit and Koudiat
Zarga sections (Taza-Guercif basin, Morocco): implications for the evolution
of the Rifian Corridor. Marine and Petroleum Geology 17, 359-371.
Kvist L, Martens J, Nazarenko AA, Orell M (2003) Paternal leakage of
mitochondrial DNA in the great tit (Parus major). Molecular Biology and
Ladoukakis ED, Zouros E (2001a) Direct evidence for homologous recombination in
mussel (Mytilus galloprovincialis) mitochondrial DNA. Molecular Biology
and Evolution 18, 1168-1175.
Ladoukakis ED, Zouros E (2001b) Recombination in animal mitochondrial DNA:
Evidence from published sequences. Molecular Biology and Evolution 18,
2127-2131.
Lenk P, Fritz U, Joger U, Wink M (1999) Mitochondrial phylogeography of the
European pond turtle, Emys orbicularis (Linnaeus 1758). Molecular Ecology
8, 1911-1922.
Lunt DH, Hyman BC (1997) Animal mitochondrial DNA recombination. Nature
Genetics, 247.
Magoulas A, Zouros E (1993) Restriction-site heteroplasmy in Anchovy (Engraulis
encrasicolus) indicates incidental biparental inheritance of mitochondrial
DNA. Molecular Biology and Evolution 10, 319-325.
Martín JM, Braga JC, Betzler C (2001) The Messinian Guadalhorce corridor: the last
northern, Atlantic-Mediterranean gateway. Terra Nova 13, 418-424.
19
Martínez-Solano I, Gonçalves HA, Arntzen JW, García-París M (2004) Phylogenetic
relationships and biogeography of midwife toads (Discoglossidae: Alytes).
Journal of Biogeography 31, 603-618.
Martínez-Solano I, Teixeira J, Buckley D, Garcia-Paris M (2006) Mitochondrial
DNA phylogeography of Lissotriton boscai (Caudata, Salamandridae):
evidence for old, multiple refugia in an Iberian endemic. Molecular Ecology
15, 3375-3388.
Mateo JA (1988) Estudio sistematico y zoogeografico de los Lagartos Ocelados,
Lacerta lepida Daudin, 1802, y Lacerta pater (Lataste, 1880), (Sauria:
Lacertidae), Universidad de Sevilla.
Mateo JA, Castanet J (1994) Reproductive strategies in three Spanish populations of
the ocellated lizard, Lacerta lepida (Sauria, Lacertidae). Acta oecologica 15,
215-229.
Mateo JA, Castroviejo J (1990) Variation morphologique et revision taxonomique de
l’espece Lacerta lepida Daudin, 1802 (Sauria, Lacertidae). Bulletin du Museé
de Histoire Naturele de Paris 12, 691–706.
Mateo JA, López-Jurado LF (1994) Variaciones en el color de los lagartos ocelados;
aproximacion a la distribuicion de Lacerta lepida nevadensis Buchholz 1963.
Revista Espanola de Herpetologia 8, 29-35.
Mateo JA, López-Jurado LF, Guillaume CP (1996) Variabilité électrophorétique et
morphologique des lézards ocellés (Lacertidae): un complexe d’espèces de
part et d’autre du détroit de Gibraltar. Comptes Rendus de L’Academie des
Sciences Serie iii-Sciences de la Vie-Life Sciences 319, 737–746.
Meusel MS, Moritz RFA (1993) Transfer of paternal mitochondrial DNA during
fertilization of honeybee (Apis mellifera L.) eggs. Current Genetics 24, 539543.
Moore WS (1995) Inferring phylogenies from mtDNA variation: mitochondrial-gene
trees versus nuclear-gene trees. Evolution 49, 718-726.
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J (2000)
Biodiversity hotspots for conservation priorities. Nature 403, 853-858.
Paulo OS (1988) Estudo eco-etologico da populacao de Lacerta lepida (Daudin
1802) (Sauria, LAcertidae) da ilha da Berlenga, Universidade de Lisboa.
Paulo OS (2001) The phylogeography of reptiles of the Iberian Peninsula,
University of London.
Paulo OS, Dias C, Bruford MW, Jordan WC, Nichols RA (2001) The persistence of
Pliocene populations through the Pleistocene climatic cycles: evidence from
the phylogeography of an Iberian lizard. Proceedings of the Royal Society B:
Biological Sciences 268, 1625-1630.
20
Paulo OS, Pinheiro J, Miraldo A, Bruford MW, Jordan WC, Nichols RA (2008) The
role of vicariance vs. dispersal in shaping genetic patterns in ocellated lizard
species in the western Mediterranean. Molecular Ecology 17, 1535-1551.
Piertney SB, MacColl ADC, Bacon PJ, Racey PA, Lambin X, Dallas JF (2000)
Matrilineal genetic structure and female-mediated gene flow in Red Grouse
(Lagopus lagopus scoticus): and anlysis using mitochondrial DNA. Evolution
54, 279-289.
Piganeau G, Gardner M, Eyre-Walker A (2004) A broad survey of recombination in
animal mitochondria. Molecular Biology and Evolution 21, 2319-2325.
Pinho C, Harris DJ, Ferrand N (2008) Non-equilibrium estimates of gene flow
inferred from nuclear genealogies suggest that Iberian and North African
wall lizards (Podarcis spp.) are an assemblage of incipient species. BMC
Evolutionary Biology 8, 63.
Rosenbaum G, Lister GS, Duboz C (2002a) Reconstruction of the tectonic evolution
of the western Mediterranean since the Oligocene. Journal of the Virtual
Explorer 8, 107-130.
Rosenbaum G, Lister GS, Duboz C (2002b) Relative motions of Africa, Iberia and
Europe during Alpine orogeny. Tectonophysics 359, 117-129.
Roucoux KH, de Abreu L, Shackleton NJ, Tzedakis PC (2005) The response of NW
Iberian vegetation to North Atlantic climatic oscillations during the last 65
kyr. Quaternary Science Reviews 24, 1637-1653.
Schmitt T, Rober S, Seitz A (2005) Is the last glaciation the only relevant event for
the present genetic population structure of the Meadow Brown butterfly
Maniola jurtina (Lepidoptera: Nymphalidae)? Biological Journal of the
Schmitt T, Seitz A (2004) Low diversity but high differentiation: the population
genetics of Aglaope infausta (Zygaenidae: Lepidoptera). Journal of
Biogeography 31, 137-144.
Schwartz M, Vissing J (2002) Paternal inheritance of mtDNA in a patient with
mitochondrial myopathy. European Journal of Human Genetics 10, 239-239.
Sequeira F, Alexandrino J, Rocha S, Arntzen JW, Ferrand N (2005) Genetic
exchange across a hybrid zone within the Iberian endemic golden-striped
salamander, Chioglossa lusitanica. Molecular Ecology 14, 245-254.
Sequeira F, Ferrand N, Harris DJ (2006) Assessing the phylogenetic signal of the
nuclear β-Fibrinogen intron 7 in salamandrids (Amphibia: Salamandridae).
Amphibia-Reptilia 27, 409-418.
21
Sherengul W, Kondo R, Matsuura ET (2006) Analysis of paternal transmission of
mitochondrial DNA in Drosophila. Genes and Genetic Systems 81, 399-404.
Shitara H, Hayashi J, Takahama S, Kaneda H, Yonekawa H (1998) Maternal
inheritance of mouse mtDNA in interspecific hybrids: Segregation of the
leaked paternal mtDNA followed by the prevention of subsequent paternal
leakage. Genetics 148, 851-857.
Steinborn R, Zakhartchenko V, Jelyazkov J, Klein D, Wolf E, Müller M, Brem G
(1998) Composition of parental mitochondrial DNA in cloned bovine
embryos. FEBS Letters 426, 352-356.
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G
(2000) Ubiquitinated sperm mitochondria, selective proteolysis, and the
regulation of mitochondrial inheritance in mammalian embryos. Biology of
Reproduction 63, 582-590.
Taberlet P, Fumagalli L, Wust-Saucy A-G, Cosson J-F (1998) Comparative
phylogeography and postglacial colonization routes in Europe. Molecular
Ecology 7, 453-464.
Tsaousis AD, Martin DP, Ladoukakis ED, Posada D, Zouros E (2005) Widespread
recombination in published animal mtDNA sequences. Molecular Biology
Tzedakis PC, Lawson IT, Frogley MR, Hewitt GM, Preece RC (2002) Buffered tree
population changes in a Quaternary refugium: evolutionary implications.
Science 297, 2044-2047.
Ujvari B, Dowton M, Madsen T (2007) Mitochondrial DNA recombination in a freeranging Australian lizard. Biology Letters 3, 189-192.
Van Leeuwen T, Vanholme B, Van Pottelberge S, Van Nieuwenhuyse P, Nauen R,
Tirry L, Denholm I (2008) Mitochondrial heteroplasmy and the evolution of
insecticide resistance: Non-Mendelian inheritance in action. Proceedings of
the National Academy of Sciences 105, 5980-5985.
Vasconcelos R, Carretero MA, Harris DJ (2006) Phylogeography of the genus
Blanus (worm lizards) in Iberia and Morocco based on mitochondrial and
nuclear markers - preliminary analysis. Amphibia-Reptilia 27.
Wallace AR (1958) On the tendency of species to form varieties; and on the
perpetuation of varieties and species by natural means of selection. III. On
the tendency of varieties to depart indefinitely from the original type. Journal
of the Proceedings of the Linnean Society of London 3, 53-62.
Williams PH, Humphries C, Araujo MB, Lampinen R, Hagemeijer W, Gasc J-P,
Mitchell-Jones T (2000) Endemism and important areas for representing
European biodiversity: a preliminary exploration of atlas data for plants and
terrestrial vertebrates. Belgian Journal of Entomology 2, 21-46.
22
Zagwijn WH (1992a) The beginning of the Ice Age in Europe and its major
subdivisions. Quaternary Science Reviews 11, 583-591.
Zagwijn WH (1992b) Migration of vegetation during the Quaternary in Europe.
Courier Forschungsinstitut Senckenberg 153, 9-20.
Zangari F, Cimmaruta R, Nascetti G (2006) Genetic relationships of the western
Mediterranean painted frogs based on allozymes and mitochondrial markers:
evolutionary and taxonomic inferences (Amphibia, Anura, Discoglossidae).
Biological Journal of the Linnean Society 87, 515-536.
Zhang D-X, Hewitt GM (1996) Nuclear integrations: challenges for mitochondrial
DNA markers. Trends in Ecology & Evolution 11, 247-251.
Zhang D-X, Hewitt GM (2003) Nuclear DNA analyses in genetic studies of
populations: practice, problems and prospects. Molecular Ecology 12, 563584.
Zhao X, Li N, Guo W, Hu X, Liu Z, Gong G, Wang A, Feng J, Wu C (2004) Further
evidence for paternal inheritance of mitochondrial DNA in the sheep (Ovis
aries). Heredity 93, 399-403.
23
Chapter 2
Intra-individual mitochondrial DNA polymorphism in
a reptilian secondary contact zone
Photos by Rita Jacinto
Sampling Lacerta lepida in Sierra Mágina, Andalucia, Spain*
2. Intra-individual mitochondrial DNA
polymorphism in a reptilian secondary
contact zone
2.1. Abstract
In the north-western region of the Iberian Peninsula two divergent mitochondrial
lineages of Lacerta lepida, L3 and L5, occur. Sequence traces with polymorphic nucleotide
sites diagnostic for the two lineages were found in individuals distributed across the
geographic range of L3 but far away from a probable contact zone between the lineages.
This suggests that genetic exchange involving either biparental inheritance or introgression
of mitochondrial fragments of one lineage into the nuclear genome of the other (Numts), has
occurred. To identify the origin of this intra-individual polymorphism a detailed
phylogeographic study of the putative contact zone between the lineages was carried out.
Data suggests that the two mtDNA lineages have diverged in allopatry in two different
refugia. A secondary contact zone located to the south of Douro River was formed between
both lineages as a consequence of a range expansion, predominantly in L5. The intraindividual polymorphism in L3 is concluded to be due to the presence of Numts. The
phylogenetic relatedness of Numts to the mtDNA sequences of both lineages was assessed
and the geographic patterns of association are discussed in detail.
Key words: contact zone, Numts, heteroplasmy, mitochondrial lineages
24
2.2. Introduction
Mitochondrial DNA (mtDNA) is the most widely employed molecular
marker in animal phylogenetic, phylogeographic and population genetic studies. This
preferential use of mtDNA in evolutionary analysis is primarily due to its higher
mutation rate compared to the nuclear genome, and maternal inheritance that
precludes sexual recombination. A variety of mechanisms acting at many stages of
the reproductive process (for a review see Birky, 1995) are responsible for
controlling the strict maternal inheritance of mtDNA in animals. In most animals it
seems that maternal inheritance of mtDNA is due to a combination of several
factors: much lower numbers of mitochondria in the sperm, random replication of
mtDNA within cells and active elimination of paternally derived mitochondria
(Birky, 1995; Birky, 2001).
Under these conditions mtDNA genes within an
individual typically exhibit homoplasmy - sequences and genes are identical copies
of each other.
However, there are two interesting exceptions to this general
observation that can lead to the detection of polymorphism upon sequencing
mitochondrial genes, which can be more apparent within individuals that are the
product of a hybrid history.
The first of these exceptions is heteroplasmy resulting from the inheritance of
both maternal and paternal mtDNA genomes. We refer to this as Biparental
Inheritance (BI) heteroplasmy (Fig. 2.1.). Although not all the mechanisms
responsible for preventing paternal mtDNA transmission are as yet fully understood,
in recent years processes controlling the active elimination of paternal mitochondria
and their mtDNA genomes when inside the egg have been studied in some detail in
several mammal species (Shitara et al., 1998; Sutovsky et al., 1999; Sutovsky et al.,
2000; Sutovsky et al., 2004). After fertilization, sperm mitochondria are tagged
within the egg by ubiquitin (a proteolytic marker) and are later selectively destroyed.
Nevertheless, failure of the successful elimination of all paternal mitochondria in the
25
egg has been shown in hybrid crosses, supporting the idea that the process of active
elimination of sperm mitochondria is species-specific (Kaneda et al., 1995; Sutovsky
et al., 2000). When the mechanism of elimination breaks down, paternal leakage can
occur with the generation of heteroplasmic individuals. With the exception of some
bivalve species, where paternal transmission of mtDNA occurs due to doubly
uniparental inheritance (Skibinski et al., 1994; Zouros et al., 1994), most cases of BI
heteroplasmy reported in the literature are incidental and typically associated with
hybridization events. Such cases of BI mtDNA heteroplasmy have been reported in
birds (Kvist et al., 2003), Drosophila (Kondo et al., 1990; Sherengul et al., 2006),
mice (Gyllensten et al., 1991; Kaneda et al., 1995; Shitara et al., 1998), cows
(Steinborn et al., 1998; Sutovsky et al., 2000), cicadas (Fontaine et al., 2007), mites
(Van Leeuwen et al., 2008), fish (Hoarau et al., 2002; Magoulas and Zouros, 1993),
bees (Meusel and Moritz, 1993), sheep (Zhao et al., 2004) and humans (Kraytsberg
et al., 2004; Schwartz and Vissing, 2002). Further indirect evidence for BI
heteroplasmy comes from recent reports of mtDNA genomes that are the products of
recombination between different species or evolutionary lineages. Ciborowski et al.
(2007) analysed the mtDNA of 717 Atlantic salmons (Salmo salar) and identified a
single recombinant genome between Atlantic salmon and brown trout (Salmo trutta).
Ujvari et al. (2007) also identified a single recombinant genome from an analysis of
79 individuals across a contact zone of two divergent mtDNA haplotypes in the
Australian frillneck lizard Chlamydosaurus kingii. While neither of these latter
studies reported BI heteroplasmy, in both cases BI heteroplasmy had to be a
precursor stage for such recombination events to occur.
The second exception to mtDNA homoplasmy that can be more apparent
within individuals of hybrid ancestry, is polymorphism arising from mitochondrial
pseudogenes incorporated in the nucleus (Numts, as first abbreviated by Lopez et al.,
1994). Numts are common and have been recorded in many taxa (for a review in
Numts see Bensasson et al., 2001; Zhang and Hewitt, 1996). They can arise through
single or several independent translocations to the nucleus and vary in number and
size, from very small translocations comprising only partial fragments of
mitochondrial genes to the incorporation of almost the entire mitochondrial genome
(Richly and Leister, 2004). In vertebrates it has been shown that the mutation rate of
Numts slows down relative to the mtDNA gene regions from which Numts are
derived (Arctander, 1995; Collura and Stewart, 1995; Fukuda et al., 1985; Lu et al.,
26
2002; Smith et al., 1992; Zischler et al., 1995) in line with the slower rate of
sequence evolution in this nuclear genome (Brown et al., 1982). Thus the detection
of Numts that have recently arisen may be difficult, as the divergence between the
Numt and its functional mtDNA copy could be very small or even absent depending
on the size of the fragment under study (Fukuda et al., 1985; Zischler et al., 1995).
Recently Podnar et al. (2007) have demonstrated in Podarcis lizards the existence of
a Numt in P. sicula that is genetically more similar to the mtDNA genome of the
related species P. muralis. Although their data did not allow for definitive
conclusions regarding the origin of the Numt in P. sicula, their study does present an
interesting model whereby Numts of recent origin may be more readily detectable
within individuals of hybrid ancestry (Fig. 2.1.). In essence, a recent Numt that is
genetically identical to the parental genome will remain undetected by PCR when it
co-occurs with the parental genome in an individual. However, hybridisation
involving individuals with divergent mtDNA genomes can result in offspring where
the recent Numt and divergent mtDNA genome reside in the same cells, and thus are
detectable by PCR.
A recent phylogenetic analysis of the genus Lacerta (Paulo et al., 2008)
revealed phylogeographic structuring within the species Lacerta lepida in the Iberian
Peninsula. Clade L of Paulo et al. (2008) is composed of 4 mitochondrial lineages
(L1-L4), with only low levels of divergence among them (Fig. 2.2.). These 4
mtDNA lineages were found to have non-overlapping geographic ranges, supporting
the idea of allopatric differentiation in multiple Iberian refugia during the Pleistocene
(Paulo et al., 2008). Of particular interest for the present study is the north-western
region of Iberia, where two mitochondrial lineages (L3 and L4) occur. Lineage L3 is
distributed to the north of Douro River in Portugal and in the regions of Galicia and
Asturias in Spain. Lineage L4 has the widest distribution of all lineages occurring
from the Atlantic coast of Portugal to southern France and north-western Italy.
Although the study of Paulo et al. (2008) did not go so far as to identify a contact
zone between lineages L3 and L4, several polymorphic sequence traces at nucleotide
sites diagnostic for the lineages (unpublished) were detected within the range of L3
but far away from the probable zone of contact between the two lineages. These
polymorphic sequence traces suggest that (1) genetic exchange has occurred between
these lineages, and (2) the genetic exchange has involved some level of either BI
heteroplasmy or Numts.
27
The central aim of this study is to determine the origin of the polymorphic
sequences previously detected in lineage L3 by increased sampling across the
geographic ranges of L3 and L4 in north-western Iberia. To achieve this aim we
have three specific objectives. The first objective is to define more precisely the
geographic ranges of lineages L3 and L4 and the incidence of intra-individual
mtDNA sequence polymorphism within these. The second objective is to identify a
contact zone between lineages L3 and L4. The third objective is to identify the
refugial areas for each lineage and to infer the probable demographic events
resulting in a contact zone.
2.3. Materials and Methods
2.3.1. Sampling
Lizards were captured under licence during the years 2005 and 2006. In 2005
the north-western corner of Iberia was sampled broadly to focus on identifying the
geographic limits of lineages L3 and L4 previously described by Paulo et al. (2008),
particularly where they come into geographic contact. In 2006, five sites (A-E)
spanning an area of contact (identified by the sequence data from samples collected
in 2005) were sampled. Sites were 20km apart and sampling was targeted at 20
individuals per site. Lizards were captured using tomahawk traps or by hand, and
tissue samples were taken by clipping 1cm of the tail tip that was subsequently
preserved in 100% ethanol. After tissue sampling, animals were immediately
released back into the wild in the place of capture. Geographic coordinates of
sampling sites were recorded with a GPS.
2.3.2. DNA extraction, amplification and sequencing
28
Total genomic DNA was extracted from ethanol-preserved muscle tissue
using a salt extraction protocol (Aljanabi and Martinez, 1997; Sunnucks and Hales,
1996). A fragment of 627 base pairs (bp) of the mitochondrial DNA (mtDNA)
cytochrome b (cytb) gene was amplified using the truncated version of primer
L14841 (Kocher et al., 1989) (CYTBF, 5’-CCA TCC AAC ATC TCA GCA TGA
TGA AA-3’) and the modified version of primer MV16 (Moritz et al., 1992)
(CYTBR, 5’- AAA TAG GAA GTA TCA CTC TGG TTT-3’). Polymerase chain
reactions (PCRs) were performed in a total volume of 25µl, containing 0.5U of Taq
polymerase (BIOTAQTM), 4mM of MgCl2, 0.4mM of each nucleotide (Bioline),
0.4µM of each primer, 2µl of 10x NH4 reaction buffer (Bioline) and approximately
50ng of DNA. PCR amplifications were conducted as follows: DNA was initially
denaturated at 94ºC for 3 min followed by 35 cycles of denaturation at 94ºC for 45s,
annealing at 51ºC for 45s and extension at 72ºC for 45s, plus a final extension step at
72ºC for 5 min. Negative controls (no DNA) were included for all amplifications.
PCR products were visualized on a 2% agarose gel and purified by filtration through
QIAquick® columns (Qiagen) following the manufacturer’s recommendations.
Purified PCR products were sequenced in both directions using the above primers
with reaction mixes consisting of 6.35µl of ddH2O, 1.5µl of primer at 3.5µM, 1µl of
BigDye Terminator v3.1TM (Applied Biosystems) and 1µl of PCR product. Sequence
reactions were performed as follows: initial incubation at 96ºC for 1min; 25 cycles
of incubation at 90ºC for 10s, 50ºC for 5s and 60ºC for 4min. All PCRs and
sequencing reactions were performed in a DNA engine tetrad 2, Peltier
thermocycler, and sequences were obtained using an ABI 3700 capillary sequencer.
DNA sequences were aligned by eye using BioEdit Sequence Alignment Editor 7.01
(Hall, 1999).
2.3.3.
Identification
of
polymorphic
individuals
and
quantification of intra-individual variation
All chromatograms were visually checked to assess sequence quality and the
presence of double peaks. Sequences were classified as polymorphic if at least one
double peak was detected. In order to control for contamination as the source of
29
polymorphism, DNA from 8 polymorphic samples and 2 samples with no signs of
polymorphism were re-extracted and the cytb fragment was amplified and sequenced
using the same conditions as described before. A negative control was used in every
step of the experiment.
Eighteen polymorphic samples from the contact zone were used to quantify
intra-individual variation. For those samples the cytb fragment was amplified and the
fragments were cloned using the StrataCloneTM PCR Cloning Kit (Stratagene),
following the manufacturer’s recommendations. Three samples with no signs of
polymorphism were also cloned for control purposes. For each cloned sample, 6 to
10 positive clones were purified using the QIAprep® Spin Miniprep Kit (Qiagen) and
directly sequenced using the conditions described above.
2.3.4. Amplification of the entire cytochrome b gene
For all polymorphic samples the entire cytb gene was amplified, using
modified versions of primers L14919 (TRNAGLU, 5’- AAC CAC CGT TGT ATT
TCA ACT -3’) and L16064 (TRNATHR, 5’- CTT TGG TTT ACA AGA ACA
ATG CTT TA - 3’) (Burbrink et al., 2000). Primers anneal at tRNA Glu and tRNA
Thr genes respectively. PCR reagent conditions were the same as described for the
cytb fragment and amplifications were conducted as follows: DNA was initially
denatured at 94ºC for 3 min followed by 35 cycles of denaturation at 94ºC for 30s,
annealing at 52ºC for 30s and extension at 72ºC for 90s, plus a final extension step at
72ºC for 3 min. Negative controls (no DNA) were included for all amplifications.
PCR products were visualized on a 2% agarose gel and purified by filtration through
QIAquick® columns (Qiagen) following the manufacturer’s recommendations.
Purified PCR products were sequenced with internal primers specifically designed
for this study: CBF (5’ - AAC CTC CTC TCA GCA ATA CC - 3’) and CBR (5’ –
CCT GTG GGG TTG TTT GAA - 3’). Sequencing conditions were the same as
above.
2.3.5. Haplotype network construction
30
Network approaches have been recommended as more appropriate than
phylogenetic trees as a tool for the representation of intraspecific genetic variation as
they incorporate population processes important at the intraspecific level; persistence of
ancestral haplotypes in the population, lower levels of divergence, multifurcations and
reticulations (Cassens et al., 2005; Posada and Crandall, 2001). Networks also allow
the representation of alternative genealogical pathways revealing ambiguities due to
homoplasies and/or recombination which would not be revealed by a strict consensus
tree (Cassens et al., 2005). Intraspecific gene genealogies were inferred using two
different network construction approaches: median-joining (MJ) (Bandelt et al., 1999)
and statistical parsimony (SP) (Templeton et al., 1992). Both methods have been shown
to perform well and give similar outcomes but MJ seems to generate fewer errors than
the SP approach when missing node haplotypes are identified (Cassens et al., 2005).
The first approach of MJ is to connect all haplotypes in a tree without inferring
ancestral nodes and keeping the tree length to a minimum. The result is a “minimum
spanning network” where all possible minimum length trees are represented in a single
network. As minimum length trees are not always the most parsimonious ones,
consensus sequences (ancestral nodes) are then added to the minimum spanning trees in
order to increase parsimony. The consensus sequences can be biologically interpreted
as extant unsampled sequences or extinct ancestral haplotypes.
The MJ network was computed with the program NETWORK 4.5.0
(www.fluxus-engineering.com) keeping the parameter ε = 0, which does not allow less
parsimonious pathways to be included in the analysis. A SP network was inferred using
the program TCS 1.21 (Clement et al., 2000) with a parsimony confidence limit of
95%. SP networks include less parsimonious alternatives whenever those alternatives
cannot be excluded at the confidence limit chosen.
Ambiguities within networks were resolved following the criteria of Crandall &
Templeton (1993): rare haplotypes occur more often at the tips of cladograms, while
common ones are more likely to be interior, and unique haplotypes are more likely to
be connected to haplotypes from the same population than to haplotypes from different
populations. Ancestral haplotypes were identified using predictions from coalescent
theory that ancestral haplotypes will occur at high frequency, be represented in the
greatest number of populations, have multiple connections to low frequency
haplotypes, and be located at the interior of the network (Crandall & Templeton, 1993;
Posada & Crandall, 2001).
31
2.4. Results
A total of 148 samples were collected: 63 from across the geographic range
of lineages L3 and L4 (sample codes BS1 to BS63), and 85 from 5 populations
spanning the putative contact zone between the two mitochondrial lineages
(Population A, n=7; Population B, n=19; Population C, n=20; Population D, n=14
and Population E, n=24). Sampling sites and number of samples per site are detailed
in Fig. 2.2. and Table 2.1. respectively. Forty samples generated polymorphic
sequences involving: 2 from population A (n=7), 13 from population B (n=19), 18
from population C (n=20) and 7 from outside the transect (samples BS38 to BS44, in
sampling sites 3, 6, 7, 8, 10 and 14). In all cases polymorphic sites included
nucleotide positions and nucleotide states that are diagnostic between lineages L3
and L4 (Fig. 2.3.), strongly suggesting some form of introgression.
2.4.1 Characterization of polymorphism
To distinguish between BI heteroplasmy and Numts a strategy of increasing
amplicon size was adopted.
In the case of BI heteroplasmy a signature of
heteroplasmy in sequence traces irrespective of amplicon size should be expected.
Nevertheless, unless the complete mtDNA genome has been incorporated as a Numt,
there should be an upper limit beyond which a Numt will not be amplified, and
therefore no polymorphism should be detected. For the first step of this strategy the
entire cytb gene (1143 bp) was amplified, which represents only a minor increase in
amplicon size. For all samples which were previously shown to be polymorphic,
homoplasmic sequences were now obtained. Chromatograms of the complete cytb
sequences of all samples had no polymorphic nucleotide sites, indicating that the
polymorphic signal arose from the incorporation of a mitochondrial pseudogene into
the nucleus. The complete cytb sequences thus represent the authentic mtDNA and
revealed that the mtDNA of all polymorphic samples belong to lineage L3.
32
The possibility of co-amplification of Numts during PCR is certainly
determined by their abundance in the genome, but primer specificity also plays an
important role (e.g. Arctander, 1995; Bensasson et al., 2001; Collura and Stewart,
1995). From the analysis of the complete L3 cytb sequences from polymorphic
individuals it was found that both CYTBF and CYTBR primers have four nucleotide
mismatches (Table 2.2.). To compare the specificity of these primers within the L4
lineage we amplified the entire cytb gene for a selection of individuals from this
lineage. While the same single mismatches for CYTBF were present within the L4
lineage, the CYTBR primer was a substantially better match with only a single
mismatch (Table 2.2.). The lower specificity of the CYTBR primer within the L3
lineage could facilitate preferential amplification of Numts, as has been noted in
other studies (e.g. Sorenson and Quinn, 1998). To test this possibility we designed a
new CYTBR primer to have no mismatches to the L3 lineage. Amplification of
individuals that were polymorphic with the original primer pair resulted in the
amplification of L3 mtDNA sequences with no evidence of polymorphism. Indeed
we can identify a single nucleotide polymorphism in the CYTBR priming site within
L3 mtDNA genome that explains why some L3 samples co-amplified Numts and
some did not (Table 2.2.).
2.4.2. Characterization of Numts and intra-individual variation
One hundred and sixty six clones were sequenced from 21 individuals. All
clones sequenced within the 3 homoplasmic individuals (26 clones, samples B3, B5
and C1) represent either the authentic mtDNA sequences or sequences that differ
from the mtDNA of each sample by 1 to 4 unique point mutations (Table 2.3.).
These mutations can be attributed to Taq error, the rate of which has been estimated
to vary between 1.1x10-4 (Tindall and Kunkel, 1988) and 2.0x10-4 (Saiki et al., 1988)
errors per nucleotide per cycle, depending on PCR reaction conditions. According to
these error rates we would expect to have between 2.4 and 4.4 errors in each
amplified fragment.
33
Amongst the 140 clones sequenced from 18 polymorphic individuals, 9
represent sequences from the mitochondrial genome of the cloned samples and 7
represent sequences that differ from the mtDNA sequence by less than 4 mutations.
Those mutations were attributed to Taq and/or cloning errors. Among the remaining
124 clones, it was possible to detect 4 sequences (18 clones) that occur in more than
one sample and thus are considered to represent four different Numts (Numts I to
IV). The remaining 106 sequenced clones are either very similar to one of the Numt
sequences, differing from those by 1 to 4 mutations (56 clones), or are consistent
with recombinants (50 clones) between the different types of sequences present
within each sample (Table 2.3.). As a conservative approach, and taking into account
the above mentioned Taq error rates, all mutations present in sequences that differ by
less than 5 mutations from one of the described Numts will be considered as being
potential Taq and/or cloning errors and thus will be ignored, unless the sequence
occurs in two or more individuals.
As in vitro recombination cannot be excluded as the origin of the
recombinant sequences, and all but one of the recombinant sequences are found in
single individuals, these were eliminated from further analysis. Furthermore, as it
was possible to reproduce in vitro the single recombinant sequence occurring in
more than one sample (3 clones in three different samples) by mixing the parental
DNA sequences in a PCR (data not shown) this sequence was also eliminated from
further analysis. Thus the sequences retained for further analysis are those that are of
known origin from the mitochondrial genome, and those that can be definitively
classified as Numts by their occurrence in one or more individuals with an origin
that cannot be attributable to jumping PCR.
All Numts apart from Numt IV have open reading frames, suggesting a
recent translocation to the nucleus. Pairwise comparisons of uncorrected sequence
divergences (p-distance) within each sample were estimated using PAUP* version 4.0
b10 (Swofford, 2002). The distances (uncorrected p-values) among the pseudogenes
present in each sample and the corresponding mtDNA varies between 1.6% (Numt
III) and 17.0% (Numt IV) (Table 2.3.). Numt I is the most common (54 out of 74
clones) occurring in 14 out of 18 cloned samples and Numt II is the second most
represented (9 clones in 7 samples). Numt III and IV occur in 3 (6 clones) and 5 (5
clones) samples respectively (Table 2.3.).
34
2.4.3. Phylogeographic analysis
For the construction of networks 182 sequences were used (146 generated in
this study and 36 sequences (GB1 to GB36, Table 2.1.) from lineages L3 and L4
from a previous study (Paulo et al., 2008)) that yielded 68 unique haplotypes. Of a
total of 627 sites, 82 (13%) were variable, amongst which 48 (59%) were parsimony
informative. Both methods used for network estimation (MS and SP) resulted in a
single network with the same topology and with two loops (Fig. 2.4.), which were
easily resolved. Haplotype 1 can be identified as the most probable ancestral
haplotype within the network based on the criteria of Crandall & Templeton (1993)
and Posada & Crandall (2001). Haplotype 1 is centrally located within the network
and is the most geographically widespread haplotype, occurring in eight sampling
localities throughout Spain, Portugal and France. This is corroborated by a more
extensive survey of genetic variation within L. lepida (chapter 3), which reveals that
across the range of L. lepida haplotype 1 is the most widespread and frequently
sampled haplotype, connected to numerous low frequency haplotypes each of which
typically have restricted geographic distributions within the distribution of haplotype
1.
The two mitochondrial lineages described by Paulo et al. (2008) (L3 and L4)
are separated in the network by eight mutations. The more detailed sampling in this
study has revealed a further lineage of haplotypes possessing a geographically
distinct distribution from other haplotypes within lineage L4. This subgroup exhibits
a geographically well-defined distribution within the region between the Tagus and
Douro Rivers in central Portugal. This subgroup of phylogeographically distinct
haplotypes will be referred to as lineage L5 (Fig. 2.5.).
2.4.3.1. L5 ancestral haplotypes and ancestral area
Within mtDNA lineage L5 it is possible to identify haplotype 13 as the most
recent common ancestor (MRCA) of all sampled haplotypes. The MRCA and the
closely related descendant haplotypes 14, 15 and 16 occur only in the southern limit
35
of the distribution of lineage L5, near the Tagus River valley. In contrast the most
derived haplotypes within lineage L5 (31, 32, 33, 34, 35 and 36) are nearly all (19
out of 20 samples) distributed in the northern limit of the lineage distribution, just
south of the Douro River. The geographic distribution of haplotypes within lineage
L5 suggests that the region around the Tagus River has most probably functioned as
a refugial area for the lineage, followed by a range expansion northwards towards
the Douro River valley.
2.4.3.2. L3 ancestral haplotypes and ancestral area
Lineage L3 occupies the entire region north of the Douro River in Portugal
and the regions of Galicia and Asturias in Spain (Fig. 2.5.). This lineage is also
found up to 20-30 km to the south of Douro River. Within this lineage haplotype 52
is the MRCA of all sampled haplotypes. Both haplotype 52 and its immediate
descendant, haplotype 46, are most frequent in the southern limit of the geographic
range of L3: of the 35 samples that possess either haplotype 46 or 52, 29 (83%) are
from southern sampling sites (south of Douro River) with the remaining 6 (17%)
from sampling sites of the centre of the lineage distribution (Geres (site 3), Parque
Natural de Montesinho (site 4), Miranda do Douro (site 6) and Macedo de
Cavaleiros (site 7); Fig. 2.2.). In contrast, haplotype 40 and its descendants (41, 42,
43, 44, 45, 50, 51, 57 and 58), all of which are derived from haplotype 46, are nearly
all represented in more northerly sampling sites: of the 30 samples that possess one
of these haplotypes, 9 (30%) are from southern populations, 9 (30%) are from
populations at the centre of the distribution (Geres (site 3) and P. N. de Montesinho
(sites 4 and 5)) and 12 (40%) are from populations at the northern limit of the
lineage distribution (Galicia
and Asturias, sites 1 and 2 respectively). This
geographic pattern of a more northern distribution for derived haplotypes suggests an
expansion from southern populations, where ancestral haplotypes are more frequent.
All samples located to the south of Douro River that belong to lineage L3
(apart from 2 samples that correspond to haplotype 57) represent either haplotype 46
or one of its immediate descendants, which are all represented in the network with a
star-like genealogy. The remaining haplotypes forming this star-like genealogy (53,
54 and 55) are all found in the sampling sites of Macedo de Cavaleiros (site 7) and P.
36
N. de Montesinho (site 5). This pattern suggests that populations located
immediately to the south of Douro River and the population of P. N. de Montesinho
and Macedo de Cavaleiros were formed by a range expansion most probably from
around the Douro River gorges. This implies that the range expansion from the
refugial area was primarily northward, but also involved some range expansion to
the south.
2.4.3.3. Contact zone between L3 and L5
Lineages L3 and L5 are geographically distinct, but fine scale sampling
around the putative contact zone between both lineages has allowed for the
identification of admixed populations, revealing the approximate location of a
mtDNA contact zone (Fig. 2.5.). Within transect site C and sampling sites of Castro
Daire and Caramulo (sites 16 and 17, respectively) haplotypes from both mtDNA
lineages occur in sympatry. The contact zone is located to the south of river Douro
from coastal Portugal to Spain and seems to extend further south in the westernmost
part of its distribution, in coastal Portugal.
2.4.3.4. Phylogeographic relationships of Numts
In order to determine the number of mutational steps separating each Numt
from any given haplotype of the lineages under study a network was constructed
with all samples obtained in this study and the Numt sequences identified (Fig. 2.6.).
The phylogenetic relatedness of Numts to mtDNA sequences suggests three different
events for the incorporation of mtDNA fragments into the nuclear genome. Numts I
and II are both derived from mtDNA lineage L5: Numt I corresponds to the sampled
haplotype 37 and Numt II differs from I by only one mutation (Fig. 2.6.). The
simplest explanation is that, rather than two independent translocations, Numt II is
an allele derived from Numt I by a single point mutation, thus we refer to these as
Numt I alleles “a” and “b”. Numt III is connected to an unsampled or extinct
haplotype near the root of the network, and therefore has a separate mtDNA origin to
Numt I. Numt IV is the most divergent Numt and could not be connected to the
37
network under a 95% parsimony connection limit. Furthermore, it does not have an
open reading frame, suggesting that it represents an older and independent
translocation to the nucleus. Thus the identified Numts are descended from at least 3
independent transfers to the nucleus.
The geographic range of each Numt was assessed by examining the sequence
trace files for all seven polymorphic samples (Table 2.1.) found outside the contact
zone. The presence of each Numt was established by subtracting the true mtDNA
sequence, obtained by the amplification of the large amplicon, from the polymorphic
sequence trace for each sample. Diagnostic sites between Numts and mtDNA
sequences from lineage L3 can be seen in Table 2.4 and the geographic distribution
of Numts is represented in Appendix 2.1..The minimum number of mutations
separating Numt IV from lineage L3 is 107, thus its presence is easily recognizable
from the polymorphic trace files. When Numt III is present together with Numt IV,
the presence of Numt I cannot be detected with certainty as there are no independent
diagnostic sites that distinguish it (unless allele “b” is present, which has an unique
diagnostic site at bp 385 - Table 2.4.). Numt IV is present in all seven samples
examined and Numt III is present in five samples (BS38, BS39, BS41, BS42 and
BS43). Numt I is present in at least two samples (BS40 and BS44). From the
remaining five samples it was not possible to determine the presence of Numt I due
to the presence of both Numt III and IV (and the absence of allele “b”).
2.5. Discussion
2.5.1. Phylogeographic history of lineages L3 and L5
Recent phylogeographic studies within the Iberian Peninsula have revealed
that a number of species have survived in several different allopatric refugia during
glacial periods (see Gomez and Lunt, 2007 for a review), suggesting the existence of
different refugia inside this peninsula, an idea first put forward by Cooper and
Hewitt (1993). Our results conform to this scenario where in the northwestern range
of its distribution Lacerta lepida is structured into two divergent mtDNA lineages
38
which have geographically distinct distributions, forming a zone of secondary
contact. Although Lacerta lepida is typically considered a Mediterranean species, its
presence in the northwestern part of Iberia, a region predominately influenced by a
temperate climate, indicates that populations of this species could have survived
during glacial periods in refugia dominated by deciduous forests, common in
temperate regions. Indeed the geographic ranges of the L3 and L5 lineages suggest
this to be the case. Thus the phylogeographic patterns of Lacerta lepida L3 and L5
lineages can be compared with Iberian species that present a typically Atlantic
distribution, restricted to the northwestern part of the Peninsula. These Atlantic
species show remarkably concordant phylogeographic patterns characterized by the
presence of northwestern and southeastern lineages or sister-species (e.g.
Discoglossus galganoi (Martínez-Solano, 2004); Chioglossa lusitanica (Alexandrino
et al., 2002; Alexandrino et al., 2000 and ); Lacerta schreiberi (Paulo et al., 2001;
Paulo et al., 2002); Lissotriton boscai (Martínez-Solano et al., 2006); Podarcis spp.
(Pinho et al., 2007)) which are likely to have diverged in allopatry in northern and
southern refugia, although sometimes over very different temporal periods. The
northern refugia for some of the above mentioned species have been inferred to be
concentrated within or near the relicts of temperate forests during the last glacial
maximum (LGM) (Zagwijn, 1992), probably somewhere to the west of “Serra da
Estrela” in central Portugal (Alexandrino et al., 2000; Martínez-Solano et al., 2006;
Paulo et al., 2001; Paulo et al., 2002).
In the case of Lacerta lepida, the southern L5 lineage seems to extend further
north than the southern lineages within most of the species mentioned above. In
addition to this, the more northerly southern limit for the L3 lineage suggests a
refugium for this lineage further north than has been inferred for other species, most
probably in the inland gorges of the Douro River. Interestingly it is in the same
region (the “Duero Arribes”) that white oaks are also suggested to have had probable
refugia during the full glacial maxima (Olalde et al., 2002).
The geographic distributions of haplotypes within both Lacerta lepida
lineages are represented by ancestral haplotypes predominating at the southern
distribution limit of each lineage, and derived haplotypes distributed more to the
north. This replicated pattern in both lineages is consistent with the hypothesis of
range expansions from southern refugia: the Douro river valley for lineage L3 and
the Tagus River valley for lineage L5. The geographic mosaic of habitats generated
39
by the complex topographies of these regions has likely allowed the persistence of
populations during adverse climatic conditions, with suitable conditions persisting
along the deep gorges of both rivers towards the eastern part of Portugal.
From these refugial areas, both lineages are inferred to have expanded their
ranges when climatic conditions permitted, meeting at what is now a zone of
secondary contact. Due to the most likely earlier climate amelioration at southern
latitudes, the northward range expansion of lineage L5 is expected to have been
initiated earlier than the expansion of lineage L3. This is supported by the
geographic distribution of the L5 derived Numt I which provides further detail about
the establishment of the secondary contact zone between both lineages (see below).
2.5.2. Origin of the polymorphism
From the strategy of increasing amplicon size it was possible to exclude the
hypothesis of BI heteroplasmy in Lacerta lepida. Beyond this there are a number of
other potential explanations for the patterns observed, and these will be evaluated in
turn below. The first objective is to identify the genomic location (mitochondrial
versus nuclear) of the extra cytb sequences responsible for the polymorphic signal
and then explore the possible explanations for their origin and prevalence in L3.
Intra-mitochondrial gene duplication
The scenario of intra-mitochondrial gene duplication can be excluded for the
sequence classified as “Numt IV” as this sequence does not have an open reading
frame, and thus it would have to represent a non-functional gene in the
mitochondrial genome. The other cytb pseudogene sequences would also require
duplicated genes within the mitochondrial genome to have evolved in different ways.
This is highly improbable, particularly if we consider “Numt I” in lineage L3. In this
case, intra-mitochondrial gene duplication would have required one of the copies to
remain unaltered with the other copy undergoing convergent mutations for lineage
L5. The scenario of intra-mitochondrial gene duplication would also imply all
individuals in lineage L3 to be polymorphic, with two different copies of the cytb
40
gene, which is not the case. Additionally, intra-mitochondrial gene duplication
would result in two differently sized PCR products, and this was not observed. Thus
intra-mitochondrial gene duplication can be excluded as the source the detected
polymorphism.
Horizontal Gene Transfer (HGT) between mitochondrial genomes
With regard to “Numt I”, the fact that this extra cytb sequence is identical to
a mitochondrial haplotype of lineage L5, could suggest that it was horizontally
transferred from that genome into the mitochondrial genome of L3, through an
external vector. Within multicellular eukaryotes, mitochondrial gene exchange
through HGT has been suggested to occur in higher plants (Bergthorsson et al.,
2003; Bergthorsson et al., 2004) and, although recently suggested as a possible
explanation of data in bruchid beetles (Alvarez et al., 2006), it has not been reported
to occur in animals yet. Thus HGT between mitochondrial genomes must be seen as
a highly implausible explanation for the patterns observed within this study.
Transfer to the nuclear genome
The data presented in this study is clearly inconsistent with a mitochondrial
location for all the cytb sequences identified. Therefore the polymorphism detected
in some individuals must be generated by the existence of cytb sequences within the
nuclear genome of Lacerta lepida. Intergenomic transfer of mtDNA fragments into
the nuclear genome is a widespread phenomenon reported for a great number of taxa
(Zhang and Hewitt, 1996), and is the most plausible explanation for the
polymorphism observed in lineage L3. The phylogenetic relationships among the
three Numts and the mtDNA sequences provide some information about where
and/or when some of these translocations occurred. It is important to note that the
detection of Numts only in lineage L3 might be a function of primer specificity, and
Numts most likely exist within individuals belonging to lineage L5. In fact, Numt I
most probably originated within the geographic range of lineage L5 with subsequent
introgression into the nuclear genome of the L3 lineage through hybridization and
41
backcrossing when the ranges of these two mtDNA lineages came into secondary
contact. Numt III, on the other hand, is phylogenetically closer to the root of the
network (Fig. 2.6.) and thus may be the result of an older translocation. The very
high divergence between Numt IV and all mtDNA sequences indicates this
translocation to be a much older event, predating the divergence events that gave rise
to mtDNA genetic diversity sampled within this study.
2.5.3. Phylogeographic utility of Numts
Once correctly identified, Numts can be used as important tools in
evolutionary biology, providing a unique window on past evolutionary events
(reviewed in Bensasson et al., 2001; Zhang and Hewitt, 1996). The geographic
distribution of Numt I within lineage L3 provides valuable information regarding the
demographic history of the two mtDNA phylogeographic lineages under study,
complementing information obtained through the analysis of mtDNA sequence data.
The data presented provides strong evidence for the introgression of a sequence
originated from the mitochondrial genome of lineage L5 into the nuclear gene pool
of lineage L3. The fact that Numt I is found as far north as Geres (sampling point 3,
Fig. 2.2, Appendix 2.1..), which is approximately 120 km north of the contact zone,
suggests that admixture between both lineages was established before, or coincident
with, the expansion of lineage L3 northwards. This supports the hypothesis of an
earlier, climatically mediated, northward range expansion of lineage L5 from the
more southerly refugia. The zone of secondary contact between lineages L3 and L5
is therefore consistent with the leading edge of lineage L5 expanding north and
contacting lineage L3 in the vicinity of the L3 refugia, prior to or coincident with the
northward range expansion of L3.
2.6. Conclusion
42
This study provides a detailed understanding of the recent demographic
history of Lacerta lepida in the north-western part of the Iberian Peninsula. The data
presented here suggests that the two L. lepida mtDNA lineages L3 and L5 have
diverged in allopatry in two different refugia. A secondary contact zone located to
the south of Douro River valley was formed between both lineages as a consequence
of a range expansion, predominantly in lineage L5. It was shown that the
polymorphism previously detected within lineage L3 is caused by the presence of at
least three different Numts. Through the phylogeographic analysis of Numts within
lineage L3 it was possible to conclude that genetic exchange between the lineages
occurred at the time of secondary contact and before, or coincident with, the
northwards expansion of lineage L3. In addition to the many evolutionary features
of Numts (see Bensasson et al., 2001), this study has shown that in the context of
phylogeographic analysis Numts can provide evidence for past demographic events.
This is an exciting prospect for the field of phylogeography.
43
Table 2.1. Sampled localities with name, site number and country of origin. For each sampling site the total number of samples collected, the
respective sample labels and the cytochrome b haplotypes detected are presented. Polymorphic samples are denoted by bold underline font.
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Location
Galicia
Asturias
Geres
Parque Natural de
Montesinho (West)
Parque Natural de
Montesinho (East)
Miranda do Douro
Macedo de Cavaleiros
Foz Coa (North)
Foz Coa
Penedono
Foz Coa (South)
Fig. Castelo Rodrigo
Pinhel
Satao
Serra Liomil
Castro Daire
Serra do Caramulo
Celorico da Beira
Lousa
Serra da Estrela
Sabugal
Nisa
Serra Sao Mamede
Country
No Samples
Sample labels
Haplotype
Spain
Spain
Portugal
11
1
10
BS49, BS50, BS51, BS52, BS53, BS54, BS55, BS56, BS57, GB1, GB2
GB3
BS34, BS40, BS45, BS46, BS48, BS61, BS62, BS63, GB4, GB5
41, 43
42
41, 44, 45, 46, 50, 51, 52, 58
Portugal
3
BS47, BS58, GB6
41, 46, 48
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
3
1
2
1
1
1
1
3
1
1
1
4
3
1
1
8
1
3
2
BS59, BS60, GB7
BS43
BS39, BS42
BS41
BS31
BS44
GB8
BS14, BS32, BS33
BS13
BS38
GB9
BS22, BS26, BS35, BS37
BS9, BS23, BS36
BS8
BS21
BS11, BS12, BS15, BS16, BS20, BS25, GB10, GB11
BS19
BS1, BS24, BS17
BS18, GB12
42, 54, 55
46
46, 53
49
65
40
47
27, 63, 64
27
56
40
17, 21, 46, 48
19, 32, 46
35
22
17, 23, 26, 29, 30
24
1, 17, 25
25
44
Table 2.1. - Continuation
Site
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
A
B
C
D
E
Location
Monforte
Paul Boquilobo
Coruche
Alcochete
Samarra
Peniche
Arrabida
Pegoes
Evora
Elvas
Moura
Huelva
Arcos de la Frontera
Ronda
Cordoba
Sierra Morena
Sierra Madrona
Pontes Rodrigo
Montes Toledo
Navahermosa
Soria
Crau
Oleron
Peso da Regua
Nagosa
Penedono (South)
Vila Franca das Naves
Lamegal
Country
No Samples
Sample labels
Haplotype
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
France
France
Portugal
Portugal
Portugal
Portugal
Portugal
1
1
1
1
3
4
1
1
1
2
1
1
1
1
1
1
2
1
2
2
1
2
3
7
19
20
14
24
BS30
GB13
BS28
GB14
BS29, GB15, GB16
BS10, GB17, GB18, GB19
GB20
BS7
GB21
BS2, BS27
GB22
GB23
GB24
GB25
GB26
GB27
GB28, GB29
GB30
GB31, GB32
BS4, GB33
GB34
BS3, BS35
BS5, BS6, GB36
A1, A2, A3, A4, A5, A6, A7
B1 to B6, B7 to B19
C1, C2, C3 to C20
D1 to D14
E1 to E24
13
18
15
7
14, 17, 20
17, 30, 33
4
1
2
1, 16
6
10
68
67
9
8
3, 5
1
11, 12
1
1
1
1
46, 57, 59
46, 66
40, 46, 52, 60, 61, 62
25, 28, 31, 36, 39
25, 27, 31, 34, 37, 38, 39, 63
45
Table 2.2. Primer sequences for the amplification of a fragment of cytb gene and respective specificity with Lacerta lepida mitochondrial
lineages L3 and L4. Dots represent nucleotide matches with primers sequences.
Primer Lineage
CYTBF
L3
L4
CYTBR
L3
L4
Sequence (5' - 3')
C C A T C
. . . C .
. . . C .
A A A T
. . . G
. . . G
C A A C A T C T C
A . . . . . . . .
A . . . . . . . .
A G G A A G T A T
. . . . . A . . C
. . . . . . . . .
A G C
C .
.
C .
.
C A C*
. . Y
. .
.
A T G A T G A A A
T . . . . . . . .
T . . . . . . . .
T C T G G T T T
. . . . . . . .
. . . . . . . .
46
Table 2.3. Number of clones sequenced per sample and identification of each type of sequence obtained (mitochrondrial DNA and/or Numts).
Sample codes are the same as in Table 2.1. Bold italic sample codes represent control homoplasmic samples. For each sample the mtDNA
haplotype is identified (mtDNA Hap.) and the divergence between the Numts and the mtDNA is shown (Divergence (%)).
Numt clones
o
Sample
N Clones
mtDNA clones
A7
B8
B10
B12
B13
B7
B18
B19
C3
C4
C5
8
9
7
8
7
8
10
8
8
8
6
4
C6
C7
C8
C9
C10
B11
8
8
8
8
8
7
C16
B3
B5
C1
Total
6
10
7
9
166
Numt I
6
4
4
3
5
a
Numt IIb
Numt III
Numt IV
Recombinants
mtDNA Hap.
Divergence (%)
1
2
3
3
3
3
2
5
1
4
2
2
57
46
46
46
46
46
66
46
40
40
46
2.1 - 17.0
2.2
2.2
2.2 - 17.0
2.2 - 2.4
2.2 - 2.4
n.a.
2.2 - 2.4
2.6 - 16.6
2.6 - 2.7
2.2
2
1
5
4
3
3
?c
46
40
62
46
46
?c
2.2 - 2.4
2.6
2.6 - 17.0
1.8 - 17.0
2.2
2
?c
46
66
25
?c
n.a.
n.a.
n.a.
1
1
1
5
2
2
3
5
2
5
4
2
1
1
5
5
1
1
1
2
1
1
1
1
4
4
10
7
9
42
54
9
5
5
50
47
Table 2.4 Diagnostic sites between each Numt identified in Lacerta lepida mitochondrial lineage L3. Dots represent nucleotide matches to
Lineage L3. Light grey shaded base pairs represent the diagnostic sites for Numt IIb and Numt III (see text for detailed explanation).
bp
L3
3 21 33 72 78 171 243 264 348 385 391 442 443 507 510 541 567 577
A G C T C C
T
C
C
G
T
G
C
C
C
A
G
G
Numt Ia
T
A
T
C
T
.
C
.
T
.
C
A
.
T
T
T
A
A
b
T
T
A
A
T
T
C
C
T
T
.
T
C
.
.
T
T
.
A
.
C
C
A
.
.
T
T
.
T
T
T
T
A
.
A
.
Numt IVc T
.
.
C
.
T
C
T
T
.
C
A
.
T
.
C
A
A
Numt II
Numt III
48
PARENTAL FORMS
HYBRIDIZATION
F1
a) Biparental Inheritance (BI) Heteroplasmy
Paternal leakage
Lineage A
sperm
egg
Lineage B
b) Numts
Numt carrier
*Numt
Lineage A
Lineage A
Lineage B
sperm
egg
Numt carrier
Fig. 2.1 (a) Biparental Inheritance (BI) heteroplasmy - heteroplasmy is generated
through paternal leakage at the time of fertilization. Hybridization of individuals
from two different mitochondrial lineages results in F1 carrying two types of
mitochondrial DNA. (b) Numts - incorporation of mitochondrial DNA in the nuclear
genome of lineage A. When Numt carrying males from lineage A hybridize with
females from lineage B, F1 will be polymorphic for the transferred mitochondrial
fragments, as they harbour the complete mitochondrial genome from lineage B
(through maternal inheritance of mtDNA) and fragments of mitochondrial DNA
from lineage A as Numts. Somatic cells are represented as pink squares. Within
each cell, mitochondria are shown as ellipses and mitochondrial DNA is represented
as coloured ellipses inside mitochondria. Blue mtDNA represents mitochondrial
lineage A while red mtDNA represents mitochondrial lineage B. Nuclei are
represented as white circles within each cell with nuclear genomes shown as green
helices.
49
10E
a) Broad scale sampling
ITALY
Lineage L1
46
!
Lineage L2
Lineage L3
45N
Lineage L4
Lineage N
!
Broad scale sampling points
!
Transect sampling points
FRANCE
45!
1
!
2
!
5E
L3
SPAIN
?
5
4
10W
!! !
! ! !!
17!
40N
!
!!
!
6
0
7!
!
!
?
!
!
3!
44
!
?
?
L1
?
!
18 ! !
20 ! 21!
19 !
43
?
PORTUGAL
22
25
!
29
!
!
28 27
!
!
! 31
!
30
24 !
! 26
?
!
!
42
23
!
!
41
!
33
L4
! 32
!
40
!
39
?
b) Fine scale sampling
! 34
?
5W
L2
! 35
38
!
?
N
8
!
10 ! 9
11
! !
B
!
12
!
! ! C
!
?
!
! 13
16 15 14
D
!
E
!
17
18
!
!
A
!
! 36
!
37
0
50 100
200 Km
ALGERIA
35N
20 !
21
?
!
MOROCCO
Fig. 2.2. Map of the Iberian Peninsula, southern France and north-western Italy showing the distribution of Lacerta lepida mitochondrial
lineages as described in Paulo et al. (2008). Numbers represent broad scale sampling sites and letters in b) represent transect sampling sites
regarding the fine scale sampling. Sampling site numbers are the same as in Table 2.1.
50
Base pair (bp)
3
21
33
72
78
243
391
442
507
510
541
577
Lineage L3
A
G
C
T
C
T
T
G
C
C
A
G
Lineage L5
T
A
T
C
T
C
C
A
T
T
T
A
Fig. 2.3. Polymorphic sites at cytochrome b gene between Lacerta lepida mitochondrial lineage L3 and L5 and respective
polymorphic sequence trace files.
51
51
50
41
43
45
42
40
46
58
57
55
59
56
60
48
46
62
54
61
47
53
49
L3
52
66
63
64
65
7
5
2
67
9
6
68
1
3
8
L4
11
12
10
4
13
15
14
16
20
18
24
23
17
21
22
19
38
26
39
27
25
25
28
37
29
30
32
36
31
33
35
34
L5
Fig. 2.4. Statistical Parsimony network of Lacerta lepida cytochrome b haplotypes.
Haplotype colours are the same as used in Fig. 2.5. to describe lineages distribution
area. Dashed lines represent ambiguities in the network. White circles with no
numbers represent unsampled or extinct haplotypes.
52
FRANCE
a)
L3
1
L5
SPAIN
L1
L4
PORTUGAL
2
N
L2
b)
5
4
3
L3
Douro River
6
7
8
A
16
15
B 10
14 C
17
9
11
12
D 13
E
21
20
18
1
19
28
(MRCA)
Tagus River
22
25
23
24
26
33
25
29
L5
N
0
50
100 Km
Fig. 2.5. a) Geographic distribution of ancestral and derived haplotypes within each Lacerta
lepida lineage (L3 and L5). Pie charts represent the proportion of ancestral and derived
haplotypes from each lineage found in each sampling site. Ancestral and derived haplotypes
from each lineage are represented by different colours: red represents L5 ancestral
haplotypes and bright green represents L3 ancestral haplotypes. Derived haplotypes are
coloured in light orange for lineage L5 and dark green for lineage L3. Numbers and letters
inside pie charts represent sampling sites as in Table 2.1. b) Statistical parsimony network
reduced to show haplotypes from lineages L3 and L5 and the most recent common ancestor
(MRCA) of both lineages only.
53
51
50
41
43
45
Numts
42
40
46
58
57
55
59
56
60
48
46
62
54
61
47
53
49
L3
52
66
III
63
64
65
7
5
2
67
9
6
68
1
3
8
L4
11
12
10
4
13
15
14
16
20
18
24
23
17
21
22
19
38
26
39
27
225
5
II b
28
37
Ia
29
30
32
36
31
33
35
34
L5
Fig. 2.6. Statistical Parsimony network of cytochrome b haplotypes and Numts I, II
and III. Numts are represented as red circles, while cytochrome b haplotypes are
represented as in Fig. 2.4. (a) Numt I is also mentioned in the text as Numt I allele
“a”. (b) Numt II is also mentioned in the text as Numt I allele “b”.
54
2.7. References
Alexandrino J, Arntzen JW, Ferrand N (2002) Nested clade analysis and the genetic
evidence for population expansion in the phylogeography of the golden-striped
salamander, Chioglossa lusitanica (Amphibia: Urodela). Heredity 88, 66-74.
Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic subdivision, glacial
refugia and postglacial recolonization in the golden-striped salamander,
Chioglossa lusitanica (Amphibia: Urodela). Molecular Ecology 9, 771-781.
Aljanabi SM, Martinez I (1997) Universal and rapid salt-extraction of high quality
genomic DNA for PCR-based techniques. Nucleic Acids Research 25, 46924693.
Alvarez N, Benrey B, Hossaert-McKey M, Grill A, McKey D, Galtier N (2006)
Phylogeographic support for horizontal gene transfer involving sympatric
bruchid species. Biology Direct 1, 21.
Arctander P (1995) Comparison of a Mitochondrial Gene and a corresponding Nuclear
Pseudogene. Proceedings of the Royal Society B: Biological Sciences 262, 1319.
Bandelt HJ, Forster P, Rohl A (1999) Median-joining networks for inferring
intraspecific phylogenies. Molecular Biology and Evolution 16, 37-48.
Bergthorsson U, Adams KL, Thomason B, Palmer JD (2003) Widespread horizontal
transfer of mitochondrial genes in flowering plants. Nature 424, 197-201.
Bergthorsson U, Richardson AO, Young GJ, Goertzen LR, Palmer JD (2004) Massive
horizontal transfer of mitochondrial genes from diverse land plant donors to the
basal angiosperm Amborella. Proceedings of the National Academy of Sciences
of the United States of America 101, 17747-17752.
Birky C (1995) Uniparental inheritance of mitochondrial and chloroplast genes:
mechanisms and evolution. Proceedings of the National Academy of Sciences 92,
11331-11338.
Birky CW (2001) The inheritance of genes in mitochondria and chloroplasts: Laws,
Mechanisms, and Models. Annual Review of Genetics 35, 125-148.
55
Brown WM, Prager EM, Wang A, Wilson AC (1982) Mitochondrial DNA sequences of
primates: Tempo and mode of evolution. Journal of Molecular Evolution 18,
225-239.
Burbrink FT, Lawson R, Slowinski JB (2000) Mitochondrial DNA Phylogeography of
the Polytypic North American Rat Snake (Elaphe obsoleta): A Critique of the
Subspecies Concept. Evolution 54, 2107-2118.
Cassens I, Mardulyn P, Milinkovitch MC (2005) Evaluating intraspecific network
construction methods using simulated sequence data: do existing algorithms
outperform the global Maximum Parsimony approach? Systematic Biology 54,
363 - 372.
Ciborowski KL, Consuegra S, Garcia de Leijniz C, Beaumont MA, Wang J, Jordan WC
(2007) Rare and fleeting: an example of interspecific recombination in animal
mitochondrial DNA. Biology Letters 3, 554-557.
Clement M, Posada D, Crandall KA (2000) TCS: a computer program to estimate gene
genealogies. Molecular Ecology 9, 1657-1659.
Collura RV, Stewart C (1995) Insertions and duplications of mtDNA in the nuclear
genomes of Old World monkeys and hominoids. Nature 378, 485-489.
Cooper SJB, Hewitt GM (1993) Nuclear DNA sequence divergence between parapatric
subspecies of the grasshopper Chorthippus parallelus. Insect Molecular Biology
2, 185-194.
Crandall KA, Templeton AR (1993) Empirical tests of some predictions from coalescent
theory with applications to intraspecific phylogeny reconstruction. Genetics 134,
959-969.
Fukuda M, Wakasugi S, Tsuzuki T (1985) Mitochondrial DNA-like sequences in the
human nuclear genome. Characterization and implications in the evolution of
mitochondrial DNA. Journal of Molecular Biology 186, 257-266.
concordance in the Iberian Peninsula. In: Phylogeography of Southern European
Refugia (eds. Weiss S, Ferrand N). Springer, Dordrecht.
56
Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and
analysis program for Windows95/98/NT. Nucleic Acids Symposium Series 41,
95–98.
Hoarau G, Holla S, Lescasse R, Stam WT, Olsen JL (2002) Heteroplasmy and Evidence
for Recombination in the Mitochondrial Control Region of the Flatfish
Platichthys flesus. Molecular Biology and Evolution 19, 2261-2264.
Kaneda H, Hayashi J, Takahama S, Taya C, Lindahl K, Yonekawa H (1995) Elimination
of paternal mitochondrial DNA in intraspecific crosses during early mouse
embryogenesis. Proceedings of the National Academy of Sciences 92, 45424546.
Kocher TD, Thomas WK, Meyer A, Edwards SV, Paabo S, Villablanca FX, Wilson AC
(1989) Dynamics of Mitochondrial-DNA Evolution in Animals - Amplification
and Sequencing with Conserved Primers. Proceedings of the National Academy
of Sciences of the United States of America 86, 6196-6200.
Kondo R, Satta Y, Matsuura ET, Ishiwa H, Takahata N, Chigusa SI (1990) Incomplete
Maternal Transmission of Mitochondrial-DNA in Drosophila. Genetics 126,
657-663.
Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS, Clayton DA, Vissing J,
Khrapko K (2004) Recombination of human mitochondrial DNA. Science 304,
981-981.
Kvist L, Martens J, Nazarenko AA, Orell M (2003) Paternal leakage of mitochondrial
DNA in the great tit (Parus major). Molecular Biology and Evolution 20, 243247.
Lopez JV, Yuhki N, Masuda R, Modi W, O'Brien SJ (1994) Numt, a recent transfer and
tandem amplification of mitochondrial DNA to the nuclear genome of the
domestic cat. Journal of Molecular Evolution 39, 174-190.
Lu X-M, Fu Y-X, Zhang Y-P (2002) Evolution of mitochondrial cytochrome b
pseudogene in genus Nycticebus. Molecular Biology and Evolution 19, 23372341.
encrasicolus) indicates incidental biparental inheritance of mitochondrial DNA.
Molecular Biology and Evolution 10, 319-325.
Martínez-Solano I (2004) Phylogeography of Iberian Discoglossus (Anura:
Discoglossidae). Journal of Zoological Systematics & Evolutionary Research 42,
298-305.
57
Martínez-Solano I, Teixeira J, Buckley D, Garcia-Paris M (2006) Mitochondrial DNA
phylogeography of Lissotriton boscai (Caudata, Salamandridae): evidence for
old, multiple refugia in an Iberian endemic. Molecular Ecology 15, 3375-3388.
fertilization of honeybee (Apis mellifera L.) eggs. Current Genetics 24, 539-543.
Moritz C, Schneider CJ, Wake DB (1992) Evolutionary relationships within the
Ensatina-Eschscholtzii complex confirm the ring species interpretation.
Systematic Biology 41, 273-291.
Olalde M, Herrán A, Espinel S, Goicoechea PG (2002) White oaks phylogeography in
the Iberian Peninsula. Forest Ecology and Management 156, 89-102.
Pliocene populations through the Pleistocene climatic cycles: evidence from the
phylogeography of an Iberian lizard. Proceedings of the Royal Society B:
Paulo OS, Jordan WC, Bruford MW, Nichols RA (2002) Using nested clade analysis to
assess the history of colonization and the persistence of populations of an Iberian
Lizard. Molecular Ecology 11, 809-819.
Pinho C, Harris DJ, Ferrand N (2007) Contrasting patterns of population subdivision
and historical demography in three western Mediterranean lizard species inferred
from mitochondrial DNA variation. Molecular Ecology 16, 1191-1205.
Posada D, Crandall KA (2001) Intraspecific gene genealogies: trees grafting into
networks. Trends in Ecology & Evolution 16, 37-45.
Richly E, Leister D (2004) NUMTs in sequenced Eukaryotic genomes. Molecular
Biology and Evolution 21, 1081-1084.
Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich
HA (1988) Primer-directed enzymatic amplification of DNA with a thermostable
DNA polymerase. Science 239, 487-491.
58
Shitara H, Hayashi J, Takahama S, Kaneda H, Yonekawa H (1998) Maternal inheritance
of mouse mtDNA in interspecific hybrids: Segregation of the leaked paternal
mtDNA followed by the prevention of subsequent paternal leakage. Genetics
148, 851-857.
Skibinski DOF, Gallagher C, Beynon CM (1994) Sex-limited mitochondrial DNA
transmission in the marine mussel Mytilus edulis. Genetics 138, 801-809.
Smith MF, Thomas WK, Patton JL (1992) Mitochondrial DNA-like sequence in the
nuclear genome of an akodontine rodent. Molecular Biology and Evolution 9,
204-215.
Sorenson MD, Quinn TW (1998) Numts: a challenge for avian systematics and
population biology. Auk 115, 214-221.
Steinborn R, Zakhartchenko V, Jelyazkov J, Klein D, Wolf E, Müller M, Brem G (1998)
Composition of parental mitochondrial DNA in cloned bovine embryos. FEBS
Letters 426, 352-356.
Sunnucks P, Hales DF (1996) Numerous transposed sequences of mitochondrial
cytochrome oxidase I-II in aphids of the genus Sitobion (Hemiptera: Aphididae).
Molecular Biology and Evolution 13, 510-524.
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G (1999)
Development: Ubiquitin tag for sperm mitochondria. Nature Genetics 402, 371372.
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G (2000)
Ubiquitinated sperm mitochondria, selective proteolysis, and the regulation of
mitochondrial inheritance in mammalian embryos. Biology of Reproduction 63,
582-590.
Sutovsky P, Van Leyen K, McCauley T, Day BN, Sutovsky M (2004) Degradation of
paternal mitochondria after fertilization: implications for heteroplasmy, assisted
reproductive technologies and mtDNA inheritance. Reproductive Biomedicine
Online 8, 24-33.
Swofford DL (2002) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other
Methods). Sinauer Associates, Sunderland, Massachusetts.
Templeton AR, Crandall KA, Sing CF (1992) A cladistic analysis of phenotypic
associations with haplotypes inferred from restriction endonuclease mapping and
DNA sequence data. III. Cladogram estimation. Genetics 132, 619-633.
59
Tindall KR, Kunkel TA (1988) Fidelity of DNA synthesis by the Thermus aquaticus
DNA polymerase. Biochemistry 27, 6008-6013.
Van Leeuwen T, Vanholme B, Van Pottelberge S, Van Nieuwenhuyse P, Nauen R, Tirry
L, Denholm I (2008) Mitochondrial heteroplasmy and the evolution of
insecticide resistance: Non-Mendelian inheritance in action. Proceedings of the
National Academy of Sciences 105, 5980-5985.
Zagwijn WH (1992) Migration of vegetation during the Quaternary in Europe. Courier
Forschungsinstitut Senckenberg 153, 9-20.
Zhang D-X, Hewitt GM (1996) Nuclear integrations: challenges for mitochondrial DNA
markers. Trends in Ecology & Evolution 11, 247-251.
Zischler H, Geiser H, Haeseler A, Paabo S (1995) A nuclear "fossil" of the
mitochondrial D-loop and the origin of modern humans. Nature 378, 489-492.
Zouros E, Ball A, Saavedra C, Freeman K (1994) An unusual type of mitochondrial
DNA inheritance in the blue mussel Mytilus. Proceedings of the National
Academy of Sciences 91, 7463-7467.
60
10E
ITALY
a)
Lineage L1
Lineage L2
!
Lineage L3
45N
Lineage L4
Lineage L5
Lineage N
FRANCE
!
Sampling points without NUMTs
!
Sampling points with NUMTs
!
!
!
L3
5E
SPAIN
0
!
!
!
10W
!
! !
!
! ! !!
!
! ! ! ! !!
!
!
!
!
!
!
L1
40N
b)
!
L5 !
PORTUGAL
!
!
!
!
!
!
!
!
!
!
!
!
!
L4
!
3 I
!
!
III, IV
!
5W
L2
!
N
8 III, IV
!
!
!
!
I, IV !
C
!
!
I, III, IV
! ! !
14
!
III, IV
!
!
IV
!
50 100
6
!
7
!
III, IV
!
0
!
!
!
A
!
B
!
10 I
!
200 Km
!
35N
0
25
50 Km
!
Appendix 2.1. Geographic distribution of Numts..a) Distribution of Lacerta lepida mtDNA lineages represented by different colours in the map
and sampling sites represented by black and pink dots. Pink dots also represent sites where Numts were detected The type of Numts detected in
each site is shown in b) where. numbers inside dots represent sampling sites as in Table 2.1.
Note: In sampling sites 6, 7, 8 and 14 the presence of Numt I (allele a) cannot be excluded (see text for detailed explanation).
61
Chapter 3
Phylogeography of Lacerta lepida in the Iberian
Peninsula
Photos by Brent Emerson
Sampling Lacerta lepida nevadensis in Cabo de Gata, Andalucia, Spain*
3. Phylogeography of Lacerta lepida in
the Iberian Peninsula
3.1 Abstract
Here a detailed phylogeographic study of a lizard species (Lacerta lepida) with a
distribution encompassing the entire Iberian Peninsula was carried out to better understand the
role of Quaternary climatic changes in generating and maintaining the phylogeographic histories
of typically Iberian species. Mitochondrial and nuclear sequence data support the existence of 6
evolutionary lineages within L. lepida. The strong association of mtDNA genetic variation with
geography suggests a history of allopatric divergence in different refugia. Using a coalescence
approach and exploring the geographic distribution of ancestral and derived alleles the refugia
for each lineage were identified. A concordant pattern of spatial and demographic expansions
within the lineages most probably associated with the last post-glacial climatic oscillation was
detected. Inferences of expansion routes from the refugia, together with the detection of
divergent nuclear and mitochondrial alleles in narrow zones of sympatry allowed the
identification of secondary contact zones. Although divergences between the mitochondrial
phylogroups seem to have a Mio-Plio-Pleistocene origin, a strong influence of later Pleistocene
events is registered, with ages for each phylogroup estimated to range from 0.45 to 0.85 Mya.
Results are compared with several published phylogeographic studies in the region.
Key words: phylogeography, Pleistocene, range expansions, contact zones, refugia.
62
3.2. Introduction
Phylogeography, as first named and described by Avise et al. (1987), is a
discipline that studies the role of historical factors in shaping the geographical
distribution of genealogical lineages at the intraspecific level. Of the historical processes
that have influenced the current distribution of genetic variation within species, the
cyclic climatic oscillations and environmental changes during the late Quaternary are
probably the most important. The ice-ages of the Pleistocene are generally believed to
have had a considerable influence on the genetic structure of populations and in species
survival across the world (reviewed in Hewitt 2000). In Europe, the cold and drier
conditions characteristic of the Pleistocene glacial periods have led to the contraction of
species distribution ranges towards the southern regions due to the advance of the icesheet from northern latitudes (Hewitt, 1996; Hewitt, 1999; Hewitt, 2000). Evidence
from a number of phylogeographic studies suggests that the southern peninsulas of
Iberia, Italy and the Balkans, as well as areas near the Caucasus and the Caspian Sea,
have functioned as refugial areas and as species survival pockets during periods of
adverse climatic conditions (Hewitt, 2004). The process of southward range contractions
and shifts into the so-called ‘glacial refugia’ has led to species contraction and
fragmentation into allopatric populations, promoting the diversification of evolutionary
lineages among the southern refugial areas. During warmer interglacial periods the
isolated populations expanded from the different refugia to re-colonize central and
northern Europe. This recolonization of northern latitudes by populations isolated in
different allopatric glacial refugia has led to the establishment of hybrid zones, where
divergent genomes came into contact. Clusters of hybrid zones in Northern Europe and
broad patterns of recolonization routes have been described across different species
(Hewitt, 1999; Taberlet et al., 1998). The cyclic processes of population contractions to
southern refugia and range expansions to northern territories have left strong signatures
in the geographic distribution of genetic diversity within many species. Species or
63
species-complexes that are widely distributed across Europe often exhibit a latitudinal
gradient in the distribution of genetic diversity. Within such species southern
populations present high diversity due to long-term persistence while northern
populations are typically less diverse, a consequence of extensive and rapid range
expansions from southern refugial populations. This pattern of “northern purity versus
southern richness” (Hewitt, 1996; Hewitt, 2000) is well illustrated in early European
phylogeographic studies from a variety of taxa (e.g. the hedgehog, Erinaceus
europaeus/concolor (Santucci et al., 1998; Seddon et al., 2001); the grasshopper,
Chorthippus parallelus (Cooper et al., 1995); the bear, Ursus arctus (Taberlet and
Bouvet, 1994); the crested newt Triturus spp, (Wallis and Arntzen, 1989) amongst
others). The importance of southern refugial areas as survival pockets and sources of
species recolonization to more northern regions is now widely accepted. Nevertheless,
recent studies also emphasize the important role that these refugial areas had in shaping
the evolutionary history of species that have persisted within these regions for several
ice ages. As anticipated by early studies (Cooper and Hewitt, 1993; Hewitt, 1996) the
topographic complexity and geographic mosaic of habitats in southern refugial
peninsulas have most probably favoured the occurrence of multiple isolated refugia,
allowing the persistence of isolated populations within them during glacial periods. The
cyclic fragmentation of species into different allopatric populations within the main
refugial areas allowed for complex demographic and evolutionary histories. For the
Iberian Peninsula these complex histories are well researched and described for a wide
variety of taxa, with some of the species showing remarkable patterns of
phylogeographic concordance (see Gomez and Lunt, 2007 and references therein)
involving deep genetic subdivisions, high haplotype richness, and long-term hybrid
zones. Indeed, the complexity of the evolutionary histories that have been revealed
within the Iberian Peninsula highlights the important role that this region has played.
Not only has the Iberian Peninsula facilitated the northern redistribution of species after
climatic cooling, but it has also facilitated diversification through patterns of repeated
population
fragmentation,
contraction,
expansion
and
admixture.
Detailed
phylogeographic studies at the species level for the golden-striped salamander,
Chioglossa lusitanica (Alexandrino et al., 2002; Alexandrino et al., 2000; Sequeira et
64
al., 2005) and the Schreiber’s Lizard, Lacerta schreiberi (Godinho et al., 2008; Paulo et
al., 2001) amongst others, are good examples of the type of complexity that most likely
typifies many species within this major Peninsular glacial refugium.
The response of a given species to climatic cycles, and the extent to which it
undergoes lineage diversification and extinction, will depend on the sharpness of the
climatic change, the latitude and the topography of the region and the dispersa1,
reproductive, and adaptive capabilities of the species itself (Nichols and Hewitt, 1994).
The topographical variation across the Iberian Peninsula (Fig. 3.1.) has most probably
influenced the way widely distributed species within this peninsula have responded to
the Quaternary climatic cycles. In northern regions of Iberia large mountain systems
prevail, allowing the survival of populations by altitudinal shifts while tracking their
suitable habitat as the climate changes. These mass altitudinal movements are expected
to result in less unstable demographic populations, and therefore smaller reductions of
genetic variability are expected to occur. This is partly due to the type of colonization
involved in these slow movements, with a high proportion of individuals dispersing only
short distances (the "phalanx" type of colonization as described in Nichols and Hewitt
1994). Furthermore mountain systems promote a metapopulation type of structure in
species distribution, which also helps to preserve variability. In northern European
latitudes it is generally accepted that the effects of glaciations have been more severe.
Within northern latitudes fewer areas of suitable habitat were available to serve as
refugia during glacial periods, leading to higher probabilities of local extinction
resulting in decreased regional genetic diversity. Thus, although multiple populations
distributed within the mountainous regions of the Iberian Peninsula can be expected to
maintain genetic diversity, the way individual populations respond to climatic changes
within those regions will be influenced by their latitudinal position. Perhaps
surprisingly, the latitudinal influence on the responses of species to climatic change
seems to be detectable even when we consider the limited latitudinal range across the
Iberian Peninsula. This pattern of lower diversity and lower number of genetic lineages
within northern latitudes was recently described for a complex of sister species of
Podarcis spp distributed across a latitudinal gradient in the Iberian Peninsula (Pinho et
al., 2007).
65
Even though the Iberian Peninsula is the most well studied glacial refugia in
terms of phylogeography, the majority of the phylogeographic studies have focused on
species that either have a narrow distribution within the region (e.g Chioglossa
lusitanica (Alexandrino et al., 2000), Lacerta scheriberi (Paulo et al., 2001), Lissotriton
boscai (Martínez-Solano et al., 2006)) or involve species-complexes that, although
distributed across the entire region, generally present a genetic structure that relates to
older cladogenic events (e.g Podarcis spp. (Harris and Sá-Sousa, 2002; Pinho et al.,
2008), Alytes spp (Martínez-Solano et al., 2004)). In order to better understand the
complex phylogeographic history of Iberian species, and the way they have responded
to Pleistocene climatic oscillations, a species with a distribution encompassing the entire
Iberian Peninsula should be studied in detail. For this purpose the ocellated lizard,
Lacerta lepida, is chosen as a model to study the impact of Pleistocene climatic changes
in generating and structuring intraspecific genetic diversity across the Iberian Peninsula.
The species is typically Mediterranean, with a distribution encompassing all the Iberian
Peninsula, and shows apparent phylogeographic structure across the region (Paulo et al.,
2008). Several mitochondrial lineages which appear to have non-overlapping geographic
ranges were recently described, suggesting a history of allopatric differentiation in
multiple refugia during the Plio-Pleistocene (Paulo et al., 2008).
Recent detailed
analysis of two of these lineages in the northwest corner of Iberia (chapter 2) has
revealed the probable glacial refugial areas for each, and the region of secondary contact
between them. Here this analysis is extended to assess the broader phylogeographic
patterns within Lacerta lepida with the specific aims to i) clarify the distribution of
mtDNA phylogroups; ii) identify refugial areas within these phylogroups during the
glacial periods; iii) identify postglacial expansion routes iv) date the main demographic
and evolutionary events within Lacerta lepida; and finally v) identify contact zones
between the different phylogroups.
It is generally acknowledged and accepted that phylogeographic histories
recovered using only mtDNA as a marker are constrained to reveal only one genealogy
which mainly reflects the maternally inherited natural history of an organism.
Relationships among phylogroups inferred through mtDNA might be discordant with
the inferences made based on nuclear genes (Harrison 1991; Avise 2000) and such
66
discordances have been illustrated in several recent phylogeographic studies (e.g.
Dowling et al., 2008; Leaché and McGuire, 2006; Lindell et al., 2008b; McGuire et al.,
2007; Thorpe et al., 2008b; Ujvari et al., 2008; Zink and Barrowclough, 2008).
Discordance is expected when time since divergence between phylogroups is
not enough to allow the achievement of reciprocal monophyly for more slowly evolving
nuclear DNA sequences. Discordances can also be explained both by the retention of
ancestral variation among populations, and/or more recent hybridization events (Avise,
2004). Within the Iberian Peninsula several studies have emphasized the importance of
using different types of markers to fully recover the complex evolutionary and
demographic scenarios that most likely characterize the species that have persisted there
across the Quaternary. For example, in Lacerta schreiberi (Godinho et al., 2008)
evidence for gene flow and ancestral introgression between apparently allopatric
mtDNA lineages was only detected by the use of nuclear markers. Further, within the
well defined mtDNA species boundaries of Podarcis spp. (Pinho et al., 2008) the
analysis of nuclear gene genealogies allowed for the detection of extensive nuclear
introgression between the species, which after detailed analysis was identified as due to
incomplete lineage sorting and not to recent gene flow. In this study, both mtDNA and
nDNA derived genealogies were used as their contrasting molecular and population
properties (principally uniparental versus biparental mode of inheritance and contrasting
population sizes) are valuable when opportunities for secondary contact, gene flow and
hybridization between diverging populations have most likely occurred.
3.3. Materials and methods
3.3.1. Sampling strategy collection
Lizards were captured under licence during the years 2005, 2006 and 2007. The
sampling strategy was devised in order to sample the entire distribution area of Lacerta
67
lepida in Portugal, Spain, and France covering all mitochondrial lineages previously
described (Paulo et al., 2008). Sampling intensity was concentrated in regions of high
genetic divergence within western and south-eastern part of Iberia (Paulo et al., 2008).
Lizards were captured using tomahawk traps or by hand, and tissue samples were taken
by clipping 1cm of the tail tip that was subsequently preserved in 100% ethanol. After
tissue sampling, animals were immediately released back into the wild in the place of
capture. Geographic coordinates of sampling sites were recorded with a GPS.
3.3.2. Laboratory procedures
DNA extraction, amplification and sequencing
A fragment of 627 base pairs of the mtDNA cytb gene was amplified using
primers CYTBF and CYTBR (chapter 2). DNA extractions, PCR amplifications and
sequencing conditions were the same as described in section 2.3.2.
In chapter 2 it was shown that primers CYTBF and CYTBR may also co-amplify
cytb Numts in some Lacerta lepida samples, therefore all cytb chromatograms were
visually assessed for sequence quality and for the presence of double peaks using
BioEdit Sequence Alignment Editor 7.01 (Hall, 1999). For all samples detected to be
polymorphic (with at least one double peak), the authentic mitochondrial sequence for
the cytb fragment was obtained through the amplification of the complete gene (1143
bp) using the primers TRNAGLU and TRNATHR designed in chapter 2. PCR
amplifications were conducted as before, but using 52ºC for primer annealing. Purified
PCR products were then sequenced with the same internal primers (CBF and CBR) used
in section 2.3.4. and sequencing conditions were also the same as before.
Intron 7 of the β-fibrinogen gene (β-fibint7) has been successfully used as a
nuclear marker in several vertebrate phylogeographic and phylogenetic studies (e.g.
Dolman and Phillips, 2004; Godinho et al., 2006; Pinho et al., 2008; Prychtko and
Moore, 1997; Sequeira et al., 2006). Specifically it was recently employed for a
phylogenetic study of the genus Lacerta (Paulo et al., 2008) where it revealed to have
68
sufficient variation within Lacerta lepida for phylogeographical inference. Initially, the
β-fibint7 amplifications were performed using primers FIB-B17U (5’- GGA GAA AAC
AGG ACA ATG ACA ATT CAC - 3’) and FIB-B17L (5’ – TCC CCA GTA GTA TCT
GCC ATT AGG GTT - 3’) (Prychtko and Moore, 1997) and the conditions described in
Paulo et al. (2008). However, due to low amplification and sequencing success a nested
PCR approach as suggested by Sequeira et al. (2006) was subsequently adopted. A
fragment of 788bp was first amplified from genomic DNA using primers FIB-B17U and
FIB-B17L (PCRa). The product of this reaction (1 µl) was then used as a template for a
subsequent PCR of 691bp (PCRb) using primer BFXF (5’ - CAG YAC TTT YGA YAG
AGA CAA YGA TGG - 3’) (Sequeira et al., 2006) and BFX8 (5’ - CAC CAC CGT
CTT CTT TGG AAC ACT G - 3’) (Pinho et al., 2008). Both amplifications were
performed in a total volume of 25µl, and included reagents in the same concentrations as
those specified for cytb gene fragment (see section 2.3.2.). PCR cycle conditions were
the same as described for cytb fragment but the primer annealing temperatures were
55ºC and 56ºC, for PCRa and PCRb respectively. Negative controls (no DNA) were
included for all amplifications. Purified PCRb products were then sequenced with
primers BFBX and BFX8 using identical sequencing conditions as for the mtDNA cytb
sequencing (see section 2.3.2.).
3.3.3. Phylogeographic and historical demographic analysis
DNA sequences were aligned by eye using BioEdit Sequence Alignment Editor
7.01 (Hall, 1999). β-fibint7 alleles of heterozygous individuals were inferred using
PHASE version 2.1 (Stephens and Scheet, 2005; Stephens et al., 2001). Several tests
implemented in the software RDP3 (Recombination Detection Program, Martin et al.,
2005) were used to detect recombination in this nuclear gene: RDP (Martin and Rybicki,
2000), GENECONV (Padidam et al., 1999), Maximum Chi Square (Posada and
Crandall, 2001a; Smith, 1992), Chimaera (Posada and Crandall, 2001a) and Sister
Scanning (Gibbs et al., 2000). Due to the small size of the fragment used in the analyses
69
(315 bp), the window size for the recombination detection methods was set to 20bp,
whenever possible.
Haplotype network construction
To correctly represent intraspecific gene genealogies biological phenomena
relevant at the population genetic level need to be taken into account. Processes such as
the persistence of ancestral haplotypes in populations and lower levels of divergence
usually lead to a lack of phylogenetic resolution. Representing this uncertainty is
important if one wants to depict all the possible evolutionary pathways that might
explain the data. Furthermore, recombination is expected at the intraspecific level, and if
present will lead to reticulate relationships, thus complicating the representation of a
genealogy (Cassens et al., 2005; Posada and Crandall, 2001b). Although recently it has
been shown that the Maximum Parsimony (MP) method, which is a standard tree
approach, outperforms most of the network approach methods (Woolley et al., 2008)
one major advantage of the latter over phylogenetic trees is that they allow the
representation of alternative genealogical pathways in a single graphical representation.
The ability to reveal ambiguities due to homoplasy and/or recombination, which cannot
be revealed by a strict consensus tree, justifies networks as a more appropriate approach
to represent intraspecific evolutionary relationships (Cassens et al., 2005). In this study,
intraspecific gene genealogies were inferred using the median-joining (MJ) (Bandelt et
al., 1999) and the statistical parsimony (SP) (Templeton et al., 1992) network
construction approaches (see section 2.3.5. for an explanation of both methods). The
MJ network was computed with the program NETWORK 4.5.0 (www.fluxusengineering.com) and the SP network was inferred using the program TCS 1.21
(Clement et al., 2000). For the MJ approach the parameter ε was set to 0 which does not
allow less parsimony pathways to be included in the analysis. The SP network was
inferred with a parsimony confidence limit of 95%, allowing therefore the inclusion of
less parsimonious alternatives whenever those alternatives cannot be excluded at the
confidence limit chosen. Ambiguities within networks were resolved following the
criteria of Crandall & Templeton (1993).
70
Neutrality tests and demographic analyses
Relatively recent demographic events, such as a population growth or a range
expansion, leave genetic footprints that can be detected through the analysis of DNA
sequences. In order to detect departures from a constant population size under the
neutral model the Tajima’s D (Tajima, 1989), Ramos-Onsins & Rozas R2 (RamosOnsins and Rozas, 2002) and Fu’s Fs (Fu, 1997) tests were applied to both types of
DNA datasets, mtDNA and nDNA. It is important to stress that departures from the null
hypothesis could be due either to an effect of natural selection on the markers under
study or the result of past demographic expansions. Both Tajima’s D and Ramos-Onsins
& Rozas’s R2 use information from the mutation (segregating sites) frequencies, but the
latter also takes into account the average number of nucleotide differences between
sequences.
Fu’s Fs (1997) is a different type of statistic test based on haplotype
distribution information. Both R2 and Fs statistics have been shown to be the best
statistical tests to detect population growth (R2 has been suggested to behave better for
small sample sizes whereas Fs is better for bigger ones) (Ramos-Onsins and Rozas,
2002).
Population expansions have also been shown to leave particular signatures in the
distribution of pairwise sequence differences (Rogers and Harpending, 1992; Slatkin
and Hudson, 1991). We capitalized upon this by employing statistics based on the
mismatch distribution to test for demographic expansions. The observed distribution of
pairwise differences between haplotypes within each mtDNA phylogroup was compared
with the expected results under a sudden-demographic and a spatial-demographic
expansion model. Statistically significant differences between observed and expected
simulated distributions were evaluated with the sum of the square deviations (SSD) and
the Harpending’s raggedness index (hg) (Harpending, 1994; Harpending et al., 1993).
Tests were performed with ARLEQUIN version 3.11 (Excoffier et al., 2005) for Tajimas’
D, Fu’s Fs, SDD and hg, and with DNASP version 4.50 (Rozas et al., 2003) for R2 and
expected values for the mismatch distribution.
71
Geographical distribution of alleles and refugial areas
Using predictions from coalescent theory (haplotypes at the tips of a tree are
younger than the interior haplotypes to which they are connected) ancestral and derived
haplotypes within each phylogroup were identified, thus obtaining a temporal
framework for haplotype origin within phylogroups. The null hypothesis of random
geographic distribution of haplotypes was also tested using GEODIS version 2.5 (Posada
et al., 2000) to perform statistical tests and assess their significance through permuting
the data 106 times. When non-random associations of haplotypes with geography were
detected the geographic distribution of ancestral versus derived haplotypes (interior and
tip haplotypes, respectively) was further explored to identify possible refugial areas and
the directionality of previously detected demographic and spatial expansions.
3.3.4. Estimation of divergence times
Divergence times within and between phylogroups were estimated from the
cytochrome b dataset using BEAST version 1.4.2 (Drummond and Rambaut, 2007).
BEAST performs Bayesian Statistical inferences of parameters, such as divergence times,
by using Markov Chain Monte Carlo (MCMC) as a framework. Input files were
generated with BEAUTI version 1.4.2 (Rambaut and Drummond, 2007). The nucleotide
substitution model and its parameter values were selected according to the results of
MODELTEST version 3.7 (Posada and Crandall, 1998), with upper and lower bounds
around the values defined as 120% and 80% respectively (Emerson, 2007). Mutation
rates were not fixed and an uncorrelated lognormal relaxed molecular clock was used
(Drummond et al., 2006). A mean mutation rate of 0.01 (the same mutation rate
described for a close lizard species, Gallotia spp. (Paulo et al., 2001)) with a standard
deviation of 0.0015, assuming a normal distribution, was used as prior information and
implemented in BEAST. No tree was selected at the start of the analysis and a constant
population size tree prior was assumed. Two runs were each executed for 106
72
generations, sampling every 500 generations and discarding the first 10% as burn in.
Results of the two runs were displayed and combined in TRACER (Rambaut and
Drummond, 2005) to check for stationarity and ensure that effective samples sizes
(ESS) were above 100. For all analyses one sequence of Lacerta pater (Genebank
accession number: AF378963) was included as an outgroup.
In a second approach to estimate divergence times within phylogroups the
method of Saillard et al. (2000) was employed where each extant haplotype descending
from the most recent common ancestor (MRCA) represents a time interval between the
present and the MRCA. Average distances from the MRCA within each phylogroup are
calculated from the number of mutation steps separating each haplotype sampled from
the MRCA. The absolute timing of divergence is then calculated by multiplying the
observed values by the average mutational changes per lineage per million years (Myr).
The molecular clock for cytochrome b sequences of a reptile species has been
previously calculated, using Gallotia spp and the geological origin of the most recently
emerged Canary Island, El Hierro (see Paulo et al., 2001 for details). The mean pairwise
sequence divergence obtained was approximately 2%. Based on this information, three
mutation rates were used: 0.01, as representing the average mutation rate; a faster
mutation rate of 0.0125, assuming an underestimation of the mean calibration mutation
rate due to the assumption of immediate island colonization that was used in the
molecular clock calibration in the work of Paulo et al. (Paulo et al., 2001); and finally a
slower mutation rate of 0.0085 to account for the longer generation time (3 years
approximately) and larger body size of Lacerta spp. group when compared to Gallotia
spp (1 year generation time). Variation in rates of nucleotide substitution among
divergent taxonomic groups have been shown to be associated with differences in body
size and generation time (Martin and Palumbi, 1993). Generation time is the time it
takes for germ-line DNA to reproduce itself. If most mutations are the result of errors in
this process and if species have similar number of cell divisions per generation, then it is
expected that species with longer generation times will accumulate fewer substitutions
than those with shorter generations simply because there will be fewer opportunities for
replication errors in the former. Furthermore, the results of Martin and Palumbi (1993)
indicate that DNA substitutions accumulate at a slower rate in large animals than in
73
small animals and that this is intimately associated with metabolic rates. Metabolic rates
and the rate of germ cell division are higher for smaller species and therefore they
usually have fast rates of molecular evolution.
3.4. Results
A total of 422 lizards were sampled from 129 different sites across the
distribution area of Lacerta lepida. Sampling sites and number of samples per site are
shown in Fig. 3.1. and Table 3.1., respectively. A total of 390 cytb and 104 β-Fibrinogen
intron 7 sequences were obtained.
3.4.1. Mitochondrial DNA data
All cytb sequences represented uninterrupted open reading frames, with no gaps
or premature stop codons, suggesting they are functional mitochondrial DNA copies.
One hundred and fifty two (152) unique haplotypes were obtained from the 390
sequences analysed. Of a total of 627 sites sequenced, 171 were variable, from which
123 were parsimony informative.
Pairwise genetic distances (uncorrected p-values) between haplotypes ranged
from 0.16% to 13.2%. According to the Bayesian Information Criterion (BIC) and the
hierarchical Likelihood Ratio Tests (hLRT’s), the model of nucleotide substitution
identified as the best fit to the data is the HKY (Hasegawa et al., 1985) with a gamma
distribution (Γ) for substitution rates across sites (shape parameter, α = 0.2889) and no
category of invariable sites. According to the Akaike Information Criterion (AIC) this
was not the first ranked model, nevertheless it has an AIC difference (delta) of 1.6 and it
is included within the 95% confidence limit, thus having substantial support. Pairwise
74
genetic distances among sequences corrected with the above mentioned model ranged
between 0.16% and 24.96 %.
The genealogical relationships between haplotypes inferred by the two
approaches for network construction (MJ and SP) are highly congruent. While the MJ
approach connected all haplotypes in a single network (as expected by the nature of the
method) the SP approach failed to do so, due to the high number of mutational steps
separating groups L2 and N from the main network. Therefore, at the 95% confidence
limit, TCS calculated three unconnected networks (results not shown). Nevertheless, the
three SP networks were connected in one single network when the confidence limit was
reduced (to 92% to include group L2 and to 65 mutational steps for group N) (Fig.
3.2.a). The relationships inferred by the two approaches (MJ and SP) when considering
the single network were identical and included 12 loops, from which 7 were easily
resolved by applying the criteria of Crandall and Templeton (1993).
The network reveals two very divergent groups of haplotypes (phylogroups),
separated by 65 mutational steps with an average pairwise uncorrected genetic distance
between groups of 11.7% (see Table 3.2. for both corrected and uncorrected genetic
distances between phylogroups). The geographic distribution of phylogroup N is
coincident with the Betic Mountains in south-eastern Spain while phylogroup L
occupies the remaining area of the species distribution. Within phylogroup L, five
geographically distinct groups of haplotypes can be identified (Fig. 3.2.a and Fig. 3.3.),
which include the four mitochondrial phylogroups (L1-L4) identified by Paulo et al.
(2008) and phylogroup L5 identified in chapter 2. Average genetic distances
(uncorrected p distances) between these phylogroups range from 1.1% (between
phylogroup L4 and L5) and 3.28% (between phylogroups L2 and L5; and L2 and L1).
Phylogroup L1 is distributed mainly across the Central Mountain system in-between the
Douro and Tagus river basins in Spain. Phylogroup L2 is distributed in southern
Portugal, occupying the entire region of Algarve and the south-western part of Alentejo.
This phylogroup is clustered together in the network with phylogroup L3 (Fig. 3.2.a)
forming a monophyletic group. L3 is distributed several hundred kilometres (300km) to
the north of L2 occupying the north-western corner of Iberia, mainly the regions to the
north of Douro River in Portugal and the regions of Asturias and Galicia in Spain. The
75
area in between phylogroups L3 and L2 is occupied by two other phylogroups, L4 and
L5. Phylogroup L5 is restricted to central Portugal, occupying the region between Tagus
and Douro River. L4 has the widest distribution of any phylogroup, and occupies the
remaining areas of southern Portugal and Spain; passing through the Ebro valley to
reach the Atlantic and Mediterranean coasts of France and is possibly also present in
north-western Italy.
The root of the network can be inferred to be located somewhere along the
branch that connects the very divergent clades L and N, allowing for the inference of the
most recent common ancestor (MRCA) among the sampled haplotypes for each
phylogroup (Fig. 3.2.a). Although this identification is straightforward for clade N
(haplotype 133), the networks reveal two probable ancestral haplotypes (haplotypes 134
and 111) within clade L, which are connected to haplotype 133 through a loop. SP and
MJ networks constructed with 0-fold degenerate sites only, thus reducing homoplasy
within the data set (Fig. 3.2.b) result in the collapse of this reticulation, and haplotype
133 (clade N) connects unambiguously to haplotype 134 (Clade L), supporting 133 as
the ancestral haplotype within phylogroup L.
Within phylogroups, statistically significant associations between genetic
variation and geographical distribution were detected for L1, L3, L5 and N (Table 3.3.).
Significant deviations from neutrality that could reflect past population expansion events
were detected for all phylogroups with Tajima’s D and with the exception of L1 the
same was true for all phylogroups when more powerful statistics were applied (Table
3.3.). The distributions of pairwise differences within each phylogroup were also found
to be consistent with sudden-expansion and spatial-expansion models (as seen by the
SDD and hg p values in Table 3.3.), with signatures of population growth being
exhibited by the bell shaped mismatch distributions (Fig. 3.4.). However, for L1 and N it
is possible to detect slightly bimodal and ragged shaped curves, suggestive of population
size constancy within these two groups. Interestingly, phylogroups L1 and N are the
ones which have a distribution mainly associated with mountainous areas (Betic Central
System and Sierra Nevada, respectively).
76
3.4.2. Nuclear DNA data
β-fibint7 sequences were trimmed to 315 bp in order to eliminate gaps within the
sequences. From the 315 bp, 15 sites were variable of which 6 were parsimony
informative. Twenty unique haplotypes (B1 to B20) were identified among the 208
alleles analysed and recombination was not detected by any of the tests applied. The
relationships among haplotypes inferred by the two network construction approaches
(MJ and SP) were identical, resulting in one single network with 4 unresolved
reticulations (Fig. 3.5.). In order to root the network a sequence of the sister species
Lacerta pater (Genebank accession number: EU: 365413) was incorporated. Haplotype
B15 is inferred to be the ancestral haplotype as it connects unambiguously to the
outgroup. This haplotype is restricted to the south-west of the species distribution area
(sampling sites 11, 56, 72, 94 and 112). Haplotype B1 is the most common haplotype
and has the widest distribution within the group, occurring in 83% of samples. This
haplotype is connected to several low frequency haplotypes, generating a star-like
genealogy, which suggests a possible past range expansion for which signatures of
expansion were formally detected by the mismatch distributions and neutrality tests
results (Table 3.3.). Within the 50 southernmost samples, 16 alleles are registered,
representing 80% of the nuclear allele diversity. From those 16 alleles, 8 are restricted to
that area, not occurring further north, suggesting this southern region as the probable
source for the northwards range expansion.
The nuclear data failed to recover the phylogroups detected by the mtDNA
dataset. Nevertheless, when the nuclear dataset is grouped according to the mtDNA
phylogroups previously identified structure in the distribution of alleles can be detected.
For this analysis each of the mtDNA phylogroups was considered as a geographic
region and GEODIS was used to test for geographical structure amongst the nuclear
genetic variation. The two most ancestral haplotypes, haplotype B1 and B15, occur in
all phylogroups. Additional nuclear haplotypes are also shared among some of the
mtDNA phylogroups (with the exception of phylogroup N): all haplotypes from L3 (B7
and B20) occur either in L1 (B7) or in L5 (B20); all haplotypes from L2 (B4, B6, B13
and B14) occur in L4 (with B13 also occurring in L5); all haplotypes from L5 (B5, B13
77
and B20) occur either in L4 (B5 and B13), either in L3 (B20) or in L2 (B13) and finally
haplotypes from L1 (B7 and B17) occur either in L3 (B7) or in L4 (B17). It is important
to note that only geographically close phylogroups share haplotypes. Nevertheless, it is
also possible to identify private haplotypes within the geography of several mtDNA
phylogroups: B2 is restricted to phylogroup L1; haplotypes B3, B9, B10, B14, B16 and
B18 to L4 and haplotypes B8, B11, B12 and B19 to phylogroup N. This pattern of
ancestral haplotypes (B1 and B15) being shared between all phylogroups suggests
incomplete lineage sorting. Incomplete lineage sorting may also explain the pattern of
geographically close phylogroups sharing more derived haplotypes, but current gene
flow between phylogroups is equally plausible.
3.4.3. Divergence times
Mean ages and 95% highest posterior density (HPD) of mtDNA phylogroups are
shown in Table 3.4. Divergence within the group is estimated to have started
approximately 9.4 million years ago (Ma) (5.58-13.66) in the mid-late Miocene,
corresponding to the cladogenic event between phylogroups N and L. Although
divergence within phylogroup L is estimated to have started in the late Pliocene (1.96
Ma; 1.13-2.91) the majority of the phylogroups within this clade are estimated to have
Pleistocene origins, with divergence times younger than 1.0 Ma. The oldest split within
this group refers to the divergence of the monophyletic lineage composed of
phylogroups L2 and L3 from the remaining phylogroups, followed by the emergence of
phylogroup L1 and finally the divergence of the more recent phylogroup L5.
3.5. Discussion
The strong association of mtDNA genetic variation with geography suggests a
history of allopatric divergence in different refugia within the Iberian Peninsula, a
78
pattern that has been described across several taxa within the region (see Gomez and
Lunt, 2007, and references therein). Although this pattern of differentiation of distinct
evolutionary units in allopatry was less evident from the analysis of the nuclear data, the
distribution of nuclear haplotypes is not in conflict with the mtDNA phylogroups. The
cytb genealogy clearly defines 6 geographic phylogroups that, in accordance to the βFibint7 genealogy, are inferred to have diverged in allopatry in southern refugia,
followed by demographic and spatial range expansions. The range expansions have
resulted in the establishment of secondary contact zones between the phylogroups,
where divergent mitochondrial haplotypes co-occur in the same populations.
Detailed analysis of the distribution of mitochondrial genetic variation within
Lacerta lepida across the Iberian Peninsula revealed a complex phylogeographic history
for the species. Lacerta lepida is structured into six mitochondrial phylogroups with
generally non-overlapping geographic distributions.
This pattern of distribution of
genetic variation associated with geography suggests the former isolation of at least six
populations in different allopatric refugia. The existence of contact zones with very
divergent haplotypes occurring in some populations (Table 3.1.) could be explained by a
scenario of differentiation in allopatry followed by contact due to range expansions.
This pattern has been well described for phylogroups L3 and L5 in the previous chapter
(chapter 2), and it also seems to be the case for the other phylogroups. Within Lacerta
lepida, usually older haplotypes are found in the southern limits of phylogroups’
distributions with derived haplotypes having more northerly distributions. The neutrality
tests and mismatch distribution tests reveal signs of demographic and spatial expansions
in almost all phylogroups, suggesting that populations must have been contracted in
smaller range refugial areas previously to expansion. Nevertheless, slightly bimodal
curves of the mismatch distribution (Fig. 3.4.a) in phylogroups that currently have a
distribution area more associated with mountainous regions (L1, L3 and N) reveal some
signs of population stasis, indicating that populations have probably persisted in these
areas for longer periods. The slightly bimodality registered, which is more apparent in
79
phylogroup L1, could probably reflect different expansions most likely associated with
the effect of different ice ages. This emphasizes the importance of mountainous regions
as areas that allow long-term survival of populations, which is also suggested by the
high endemicity levels usually associated with these regions (García-Barros et al.,
2002). For example, the Sierra Nevada Mountains have been evaluated as encompassing
the highest number of plants endemics in Europe (Gomez-Campo et al., 1984) and also
the main Iberian hotspot in biodiversity (Castro-Parga et al., 1996); the northern
mountain ranges of Iberia, as the Sistema Central, have been designated as the principal
areas of monocotyledons endemism (Saiz et al., 1998) as well as for certain animal
species. Mountainous regions located at northern latitudes within the Iberian Peninsula,
as the Iberian Central System and the regions of northern Portugal, play an even more
important role, as they allow the survival of populations at higher latitudes where the
impact of climatic oscillations should be more pronounced. It was recently shown that
populations at northern latitudes, even across the small latitudinal range observed within
the Iberian Peninsula, usually show lower diversity and lower number of genetic
lineages (Pinho et al., 2007) than southern ones. This was shown for a complex of
Podarcis species distributed through a latitudinal gradient from Northern Africa to
North-western Iberia. The data presented here suggests that this pattern probably does
not hold true for populations that are able to persist in mountainous regions at northern
latitudes, which seems to be the case for phylogroups L1 and L3. This pattern of
persistence is evident in the haplotype network within phylogroups L3 and L1, which
show long internal branches with several missing haplotypes (Fig. 3.2.a), suggesting
long term persistence with diversification and further extinction of ancestral haplotypes.
Failure of the nuclear gene genealogy to reveal concordant genetic structure with
the mitochondrial genealogy can be expected if we take into account the fact that
nuclear genes take on average four times longer to reach monophyly than mitochondrial
ones. In fact, most of the intraspecific differentiation within Lacerta lepida is of
80
relatively recent origin, with the majority of phylogroups (L1 to L5) estimated to have
diversified within the Pleistocene, increasing therefore the possibility of mitochondrial
lineages not being monophyletic when nuclear markers are considered. Nevertheless,
lineage N is estimated to have diverged from the remaining phylogroups during the PlioMiocene (5.6-13.7 Ma) representing a much older cladogenic event within the group.
This older split provides longer isolation periods, therefore allowing for a more
complete lineage sorting at the nuclear level. This pattern is clearly evident in the
composition of nuclear haplotypes of lineage N, which are almost all private with the
exception of the ancestral haplotypes (which are shared among almost all phylogroups).
Thus, although lineages have not reached monophyly at the nuclear level, some level of
differentiation between lineages is detected by the existence of private alleles. In fact
phylogroups that represent older cladogenic events (L1, L4 and N) are the ones that
show private nuclear alleles. Although not conclusive due to the very low level of
variation within the nuclear genealogy, the fact that some derived alleles, which are at
the tip of the nuclear network, are shared between geographically close phylogroups
could be indicative of the existence of gene flow between the differentiated
phylogroups. For example allele B20 occurs only in the very divergent phylogroups L3
and L5 near the zone of contact, but it was not detected in the ancestral phylogroup L4,
suggesting that gene flow is occurring between the lineages. The same is true for allele
B4 which although more widespread within phylogroup L4, occurs as well in L2
individuals, near the zone of contact, but does not occur in phylogroup L3, which is
phylogenetically closer to phylogroup L2.
Due to the slower evolving nature of nuclear genes when compared to the
mitochondrial genome, nuclear genealogies should record older demographic events
(Avise, 2004). Therefore, analysing the distribution of the oldest nuclear haplotypes
allows us to access a greater temporal depth on the evolutionary history of Lacerta
lepida. Haplotype B15 is the root of the nuclear gene network, representing therefore the
most ancestral allele in the dataset. Its distribution is currently registered in the southern
regions of Iberia, where it occurs with higher frequency than in northern regions (70%
of samples with haplotype B15 are from southern latitudes) , but it is also found as far
north as the Spanish Central System and Douro River basin, although much less
81
frequently. The high nuclear haplotype richness found in the southern regions of Iberia,
together with the high frequency of B15 in this region suggest that southern populations
are older and are the source for the demographic and spatial northerly expansions
registered.
3.5.3. Historical biogeography of Lacerta lepida
Divergence within Lacerta lepida is estimated to have started in the Miocene,
approximately 9.4 Mya (5.58-13.66), with divergence of phylogroup N. Within
phylogroup L, estimated divergence times are much younger and are inferred to have
occurred in the Plio-Pleistocene, approximately 1.96 Mya (1.13-2.91 Mya).
Interestingly, although divergences between the mitochondrial phylogroups seem to
have a Plio-Pleistocene (within group L) or a late Miocene (for group N) origin,
haplotype diversity within each phylogroup indicates a strong influence of later
Pleistocene events, with ages for each phylogroup estimated to range from 0.45 to 0.85
Ma. The importance of the Pleistocene climatic oscillations in promoting species
differentiation in the Iberian Peninsula has been emphasized by previous studies (see
Gomez and Lunt, 2007, for a recent review) and this clearly also seems to be the case
for Lacerta lepida.
Phylogroup N
The earliest divergence within Lacerta lepida has lead to the establishment of
two very divergent mitochondrial lineages, phylogroups N and L. Phylogroup N is
distributed across the Betic Mountain range in south-western Spain and its distribution
roughly coincides with the described subspecies Lacerta lepida nevadensis. Paulo et al
(2008) have inferred that divergence between the phylogroups must have started due to
overseas dispersal between what was then the Iberian mainland and the Betic Massif
that at that time existed as an island between Iberia and North Africa. Under this
82
scenario contact between the phylogroups must have been initiated after the merging of
the Betic Massif with Iberian mainland, due to the closing of the Betic corridor 7.6-7.8
Ma, (see Paulo et al., 2008, and section 2.1.1. from this thesis for a detailed explanation
of the kinematics of the western Mediterranean basin). The fact that these populations
have remained genetically distinct at the mitochondrial level since the closing of the
Betic corridor suggests reproductive isolation of the forms upon subsequent secondary
contact. The Betic Mountain range is a region with high numbers of plant and animal
endemics and has been pointed out as a refugium for several taxa (see Gomez and Lunt,
2007). Interestingly, some taxa show similar divergence times and distribution areas as
phylogroup N (e.g. Salamandra salamandra longirostris (Garcia-Paris et al., 1998) and
Alytes dickhiller (Arntzen and Garcia-Paris, 1995)) emphasizing the striking similar
responses of species to the history of this region. Within phylogroup N, the mismatch
distribution shows a negative binomial curve, with a slightly ragged shape at the end
(Fig. 3.4.a). The negative binomial curve can be indicative of recent population
expansions, nevertheless the ragged shape seems to indicate population stasis with the
extra peaks representing older ancestral polymorphisms. A history of a past geographic
substructuring with restricted gene flow followed by recent population expansions could
result in such pattern (Marjoram and Donnelly, 1994).
Phylogroups L2 and L3
The monophyletic group composed of phylogroups L2 and L3 is estimated to
have started diverging from the remaining phylogroups in the early Pleistocene,
approximately 1.5 Mya (0.82-2.27). Interestingly these two phylogroups currently have
a disjunct distribution, with phylogroup L2 occupying the south of Portugal whereas L3
occupies the north-western parts of the Iberian Peninsula. The intervening region
between phylogroups L2 and L3 is occupied by phylogroup L5. A vicariant event during
the Plio-Pleistocene transition (0.82-2.27 Mya) triggering divergence between the L2-L3
lineage and the remaining populations of Lacerta lepida seems probable. Interestingly
most phylogeographic studies within Iberia reveal similar phylogenetic breaks
associated with the same period (e.g. Chioglossa lusitanica (Alexandrino et al., 2002;
83
Alexandrino et al., 2000), Oryctolagus cuniculus (Biju-Duval et al., 1991), Lacerta
schreiberi (Paulo et al., 2001; Paulo et al., 2002)), suggesting a common vicariant
history. Such a vicariant event was most likely climatically mediated as no apparent
geographical barrier exists within the western Iberian Peninsula.
The fact that phylogroup L2 and L3 are part of a monophyletic group indicates a
shared glacial refugium for at least part of their early evolutionary history. The region
of western Algarve in southern Portugal has been pointed as the evolutionary centre for
several species and also as a main refugial area (Fritz et al., 2006; Mesquita et al.,
2005). The region harboured relicts of temperate forests during the Last Glacial
Maximum (Zagwijn, 1992), probably providing suitable conditions for species survival
through glacial periods. Thus the Algarve is a potential refugial area for the ancestor of
L2 and L3, although the uncertain geological and ecological history of the region means
this must be treated as speculative.
The high genetic distances found between phylogroup L2 and L3 can be
explained by further range fragmentation and divergence within this group. As L2 is
inferred to descend from L3 haplotype 40 (or 44), this reveals an older age for L3, which
likely once had a more extensive distribution area than now. The ancestral form, most
closely related to L3, became disjunct most likely associated to climatic cooling. This
disjunction has promoted divergence between L2 and L3. These phylogroups ought to
have remained effectively isolated for a period of time long enough to allow the levels
of divergence that are observed today. When climatic conditions allowed, probably
during an interglacial period, the intervening region between L2 and L3 was colonized
by L5, expanding from a nearby refugium (Tagus river basin, see chapter 2).
Interestingly, a similar pattern of distribution of genetic variation is found within
Discoglossus galganoi across Portugal, with two phylogenetically closer phylogroups
distributed in the south and north and a less related phylogroup bisecting their
distribution (Martínez-Solano, 2004). The distribution of Discoglossus galganoi
phylogroups and the evolutionary relationship between them is remarkably similar to the
ones just described for Lacerta lepida.
84
Phylogroups L1, L4 and L5
The refugium for Lacerta lepida phylogroups L1, L4 and L5 was probably
located in the south-eastern side of the Guadalquivir basin. Support for this comes from
the distribution of ancestral haplotypes 134, 147 and 8 within phylogroup L4, the oldest
within lineage L, that are found only in this region (localities 10, 12, 32, 36, 37, 38, 39,
40, 41). It is also to this region that the two very divergent haplotypes within L4 (67 and
68) are restricted. The occurrence of deeply differentiated taxa isolated on the southern
side of the Guadalquivir basin (e.g. Arntzen and Garcia-Paris, 1995; Garcia-Paris et al.,
1998) emphasize the importance that this region may have played in divergence
processes inside the Iberian Peninsula. The widespread distribution of the most
frequently sampled haplotype 1 suggests that the spatial and demographic expansion
detected within L4 was of a “leading edge” type, with few individuals rapidly colonizing
adjacent regions, leading to a decrease in genetic diversity on the newly colonized areas.
Expansion was likely facilitated by the extensive low altitudinal plains that characterize
most of the distribution area of L4. The different phylogroups are most likely the result
of three different expansions from the southern refugia dominated by different ancestral
haplotypes, followed by further divergence in allopatry.
3.6. Conclusion
Mitochondrial and nuclear gene genealogies in Lacerta lepida provided evidence
for a history of isolation and divergence in allopatry resulting in the diversification of
six genetically and geographically distinct lineages. Although diversification within the
group is largely concordant with the onset of the major glaciations at the beginning of
the Pleistocene (approximately 2 Mya) an earlier event, associated with the Miocene,
was also registered. This event (9 Mya), which marks the divergence of lineage N,
seems to be associated with geological events related to the evolution of the
Mediterranean basin. The detailed analyses of the distribution of ancestral and derived
85
alleles within each lineage allowed the identification of six geographically distinct
refugia distributed throughout the Iberian Peninsula. Signs of recent demographic and
spatial expansions were registered in all lineages. As a result of spatial expansions after
periods of divergence in allopatry most lineages have established zones of secondary
contact. Further analysis of these zones should provide insights into the mechanisms
involved in speciation and divergence in this lizard.
86
!
49
a)
b)
France
L. l. iberica
48
!
!
90
L. l. lepida
L. l. nevadensis
5W
!
91
!
118
92
!
!
47
!
87
!
89
!
88
!
86
!
117
Spain
84!
83
!
57
!
!
85
!
59
56
!
!
! 80
! 79
! 81
82
!
106
58
!
78
!
!
104
!
103
102
!
105
!
!
101
Portugal
!
93
!
55
!
54
46
!
!
94
!
108
44
!
98
!
! 99
100
!
97
!
!
95
76
!
!
45
43
!
!
42
!
2
1
!
10W
22
! 21
!
!
! 19
20
!
53
!
18
96! ! !
14
! 15! 16
17
74
!
!
50
3
!
75
!
!
72
!
4
13
!
5
!
6 !
!
9
!
12
!
52
8
!
7 !
!
73
!
11
!
70
!
! 107
!
69
!
64 23
!
26
!
24
65
!
! 67
68
!
! ! 62
! 66
!
61
27
!
!
25
!
77
!
71
!
41
40
!
!
10
!
38
!
119
!
39
!
122
37
!
!
124
!
129
!
125
!
36
!
30
51
!
!
115
! 128
!
! 116
35
!
34
!
121
60
!
32
!
!
33
!
109
!
110
!
113
112
!
111
!
!
127
!
114
!
28
0
50
100
200
Kilometers
!
126
!
29
123
! 31
!
Fig. 3.1. a) Distribution area of Lacerta lepida and recognized continental subspecies (L. l. iberica, L. l. lepida and L. l. nevadensis).
b) Sampled localities. Numbers are the same as in Table 3.1. Shaded areas denote altitude gradients, with darker areas representing
higher altitudes.
87
a)
b)
102
51
109
108
110
124
123
125
41
43
45
42
103 *
100
107 104
L2
50
57
53
43
109
C2
C0
40
44
105
106
48
L3
58
C1
57
61
47
98
101
59
122
53 47
60
48
46
62
99
52 *
A0
D0
A1
*
133
126
127
128
130
//
150
80
19
27
38
131
E0
133
134
16
A2
97
L4
*
134
149
148 151 92 147146
121
89
137
94 12 11
8
5 80
150
10
81
93
9 88
76
118
152
71 3
72 77
84
85
143
86
120
7
75 4
1
135
6
70
2
91
142
119
90
73
117
78
69
79
83
140 141
87
139
82
114
68
136
74
67
*
15
13
14
B2
117
N
129
131
132
B1
114
85
84
144
A3
90
64
145
139
4
55
61
83
67
54
56
49
138
B0
147
29
22
96
36
32
115
66
65
63
*
111
112 64
A0: 1, 2, 3, 5, 6, 8, 9, 10, 11, 67, 68, 60, 70, 71, 72, 73, 75,
113
76, 78, 81, 86, 88, 89, 91, 92, 93, 115, 118, 119, 120, 121,
L1
135, 136, 137, 140, 141, 142, 143, 148, 149, 150, 151 and
152 A1: 12, 79, 87 and 94; A2: 74, 82 and 146; A3: 7 and
16
20
77; B0: 17, 18, 20, 21, 23, 24, 25, 26, 28, 30, 31, 33, 34, 35,
18
19
22
21
24
95
96
97
116
39
L5
23
17
40, 41, 42, 44, 45, 46, 49, 50, 52, 54, 55, 56, 58, 60, 62, 98,
26
27
28
25
38
30
37
37, 38, 95, 97 and 116; B1: 13 and 39; B2: 14 and 15; C0:
29
99, 100, 103, 104, 105, 106, 107, 108, 122, 123, 124, 125
and 138; C1: 101, 51 and 59; C2: 102 and 110; D0: 63, 65,
66, 111, 112 and 113; E0: 126, 127, 128, 129, 130, 131, 132
36
32
31
35
33
and 144.
34
Fig. 3.2. Statistical Parsimony network of Lacerta lepida cytochrome b haplotypes (a) using all 627 bp sites and (b) using 0-fold degenerate sites only. Dashed
lines represent ambiguities in the networks. White circles with no numbers represent unsampled or extinct haplotypes. L1, L2, L3, L4, L5 and N represent
different mitochondrial phylogroups. The ancestral haplotype within each phylogroup is marked with *. Phylogroup N connects to the main network through 65
mutations and the connection is represented by an interrupted line. In (b) numbers inside circles represent haplotypes from network (a) and circles with letters
represent groups of haplotypes from network (a) that show no differences in 0-fold degenerate sites, being thus represented as a single haplotype in the network
(b). Composition of each group of haplotypes in network (b) is shown under the network.
88
Lineage L1
Lineage L2
ITALY
FRANCE
Lineage L3
Lineage L4
Lineage L5
Lineage N
L3
SPAIN
59
!
!79 !56
!58
!82
!78
L5
22
!!21
19
!18
53!20
!
!
96
!!15
PORTUGAL
L1
!94
44
!
L4
!6
L2
!25
0
150
300 Km
N
23
!24
!
!33
ALGERIA
Fig. 3.3. Distribution of Lacerta lepida mitochondrial phylogroups based on cytochrome b gene. Colours are the same as in Fig. 3.2.
Red areas represent contact zones between phylogroups. Contact zones were inferred according to sampling sites where haplotypes
from different phylogroups were detected in sympatry (see Table.3.1). Sampling sites used for inferring the spatial distribution of
contact zones are represented by numbers.
89
Theta I = 1.35; Theta F = 1000; t = 0.75
Theta I = 0.72; Theta F = 1000; t = 2.38
L1
Theta I = 0; Theta F = 1000; t = 2.17
L4
Theta I= 0.6; Theta F = 1000; t = 1.51
L5
L2
Theta I = 1.04; Theta F = 1000; t = 1.32
Theta I = 1.51; Theta F = 1000; t = 0.54
L3
N
Fig. 3.4. Mismatch distribution of mtDNA haplotypes for each of the 6 Lacerta lepida
phylogroups. The expected frequency is based on a population growth-decline model,
determined using DnaSP v4.50 (Rozas et al., 2003) and is represented by a continuous
green line. The observed frequency is represented by a red dotted line
90
B6
B4
B5
B3
B2
B7
B1
B8
B16
B9
B10
B19
B18
B14
B12
B17
B11
B20
B13
B15
L1
L2
L3
L4
L5
N
L. pater
Fig. 3.5. Statistical Parsimony network of Lacerta lepida β-Fibrinogen intron 7 alleles.
Dashed lines represent ambiguities in the network; white circles with no numbers
represent unsampled or extinct haplotypes and the black circle represents the outgroup
(Lacerta pater). The proportion of each allele found within each mitochondrial
phylogroup is represented through pie charts. Colours in pie charts are the same as the
ones used to represent the mitochondrial phylogroups in Fig. 3.2.
91
Table 3.1 Number of Lacerta lepida samples per site (n) and correspondent number of
sequences amplified for cytb (Cytb) and β-Fibrinogen intron 7 (β-Fib) genes. For each
site, haplotypes found regarding each gene are shown. Grey shaded rows indicate sites
were haplotypes of two different phylogroups were found in sympatry, revealing the
location of secondary contact zones (see Fig. 3.3.).
No of sequences
mtDNA
Site
n
Cytb
β-Fib
Cytb haplotypes
β-Fibint7 alleles
phylogroup
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
2
1
1
1
3
3
3
3
2
1
5
1
1
3
5
3
9
2
1
1
1
3
3
3
3
2
1
4
1
1
3
5
3
9
2
0
1
0
0
1
0
3
2
0
4
0
0
0
0
0
0
B13, B14
L4
L4
L4
L4
L4
L2, L4
L4
L4
L4
L4
L4
L4
L4
L4
L4, L5
L4
L4, L5
18
19
20
21
22
23
24
6
5
2
5
5
3
9
6
5
2
5
5
3
9
0
1
0
0
2
0
0
25
26
27
28
8
3
6
1
8
3
5
1
2
3
4
0
1
7
2
1
1, 73, 90
6, 70, 104
1, 69, 70
70, 84, 143
10, 85
8
10, 70, 93
8
80
1
1, 25, 79, 83
69
1, 6, 16, 69, 86,
121
1, 25, 69
25, 74, 79, 87
25, 72
1, 25, 79
1, 17, 15, 116, 118
107, 120, 125
6, 75, 76, 81, 88,
98, 117, 119
70, 78, 98, 100,
135, 139, 142
10, 68, 91, 137
67
B1
B1, B4
B1, B3, B4, B13
B1, B6
B1,B 3, B15, B16
B1
1, 13
B1, B13, B14
B1, B5, B16
B1, B5
L4, L5
L4, L5
L4, L5
L4, L5
L4, L5
L2, L4
L2, L4
L2, L4
L4
L4
L4
92
Table 3.1. Continuation
No of sequences
mtDNA
Site
n
Cytb
β-Fib
Cytb haplotypes
β-Fibint7 alleles
phylogroup
29
30
5
3
2
3
5
0
140, 141
9, 89, 114
B1, B4, B5, B13
L4
L4
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
4
1
2
1
2
4
6
3
4
3
1
1
3
2
1
3
3
2
3
1
1
3
3
6
2
5
1
25
3
1
2
1
1
4
6
3
4
3
1
1
3
2
1
3
3
2
3
1
1
3
3
6
2
5
1
25
3
0
0
1
1
0
0
0
0
1
0
0
1
0
0
2
3
0
0
0
0
0
2
3
1
1
0
2
B1, B4, B18
L4
L4
L4, N
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4
L4, L5
L1
L1
L1, L5
L1
L1, L5
59
60
61
62
63
19
1
4
5
1
19
1
4
5
1
1
1
1
0
0
136, 140, 141
8
126
89
89
89, 147, 148
8, 89, 94, 149
8, 151, 152
8, 147, 150
3, 5, 8
134
1
1, 11, 12
1
69
1, 69
1, 71
1
1
4
92
1, 92, 146
25, 77, 97
111, 112
111
27, 63, 64
65
25, 27, 31, 34, 37,
38, 39, 63
46, 66
115
98, 104, 106, 108
98, 122
100
B1
B1, B9
B1, B17
B1, B19
B1
B1, B17
B1
B1
B1
B1, B15
B1
B1
B1, B17
B1
L1, L3
L1
L2
L2
L2
93
No of sequences
mtDNA
Site
n
Cytb
β-Fib
Cytb haplotypes
β-Fibint7 alleles
phylogroup
64
65
3
1
3
1
1
1
100, 104, 123
105
B4
B1
L2
L2
66
67
68
69
70
71
72
3
1
1
1
2
1
6
3
1
1
1
2
1
6
1
0
1
1
0
1
2
B1
L2
L2
L2
L2
L2
L2
L2
73
74
75
76
77
78
79
1
1
1
1
1
2
18
1
1
1
1
1
2
18
0
1
1
0
1
0
0
80
81
82
83
84
85
86
87
88
89
1
1
4
7
1
1
2
3
1
10
1
1
4
7
1
1
2
3
1
10
0
0
0
0
0
0
0
0
0
0
90
91
92
93
94
95
11
1
3
5
4
3
11
1
3
5
1
3
1
0
1
0
3
1
100, 124
98
98
103
100, 103
110
98, 100, 101, 102,
105
107
109
100
100
99
32, 46
40, 46, 52, 60, 61,
62
56
40
17, 21, 46, 48
46, 57, 59
49
47
46, 53
42, 54, 55
46
41, 44, 45, 46, 50,
51, 52, 58
41, 43
42
41, 46, 48
49
138
14, 17, 20
B1
B1, B13
B1
B1, B15
B1
B1, B6
B1
L2
L2
L2
L2
L2
L3, L5
L3
L3
L3
L3, L5
L3
L3
L3
L3
L3
L3
L3
B1
B1
B1, B2, B7, B15
B1
L3
L3
L3
L3
L1, L3
L5
94
No of sequences
mtDNA
Site
n
Cytb
β-Fib
Cytb haplotypes
96
97
98
99
2
1
1
8
2
1
1
8
0
0
0
2
13, 86
15
18
17, 33, 95, 96
100
101
102
103
104
105
106
107
108
109
110
111
112
2
7
1
2
1
1
14
1
1
1
1
5
11
2
7
1
2
1
1
14
1
1
1
1
4
6
0
3
1
1
0
1
0
0
1
0
0
1
7
30
23, 26, 29, 30
17
19, 32
35
24
25, 28, 31, 36, 39
25
22
133
131
127, 128, 130, 132
126
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
1
3
3
3
1
3
1
1
1
1
2
1
1
1
1
1
1
1
3
3
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
3
1
1
1
1
2
1
1
1
1
1
1
129
126
126, 144
126, 145
β-Fibint7 alleles
phylogroup
B1, B20
L4, L5
L5
L5
L5
B1, B5
B1
B1
B13, B20
B1
B1, B8
B1,B8, B11, B12,
B15, B19
B7, B20
B1
B1, B17
B1, B17
B1,B 17
B1
B1
B1, B10
B4, B5
B1, B5
B1
B1
B5, B6
L5
L5
L5
L5
L5
L5
L5
L5
L5
N
N
N
N
N
N
N
N
L3*
L4*
L4*
L4*
L4*
L4*
L4*
L4*
L4*
L4*
N*
N*
L4*
* When sequences were not available the phylogroup was inferred using the geographic
location of samples.
95
Table 3.2 Pairwise genetic distances (corrected and uncorrected) between Lacerta lepida mitochondrial
phylogroups.
mtDNA
phylogroup
L1
L2
L3
L4
L5
N
uncorrected p distancesa (%)
L1
L2
L3
L4
L5
N
a
3.28
0.32 (0.55)
3.74
0.55 (0.53)
2.80
3.13
1.57
3.29
2.00
3.77
19.07
22.38
2.53
1.48
1.85
2.75
2.92
3.28
2.10
2.44
0.67 (0.64)
2.30
1.11
0.74 (0.71)
2.72
1.17
0.61 (0.59)
21.20
20.61
21.07
b
HKY corrected (%)
11.30
12.44
12.02
11.74
11.86
0.72 (0.69)
values above diagonal and values in diagonal inside brackets represent uncorrected genetic distances (p distances).
b values under diagonal and values in diagonal outside brackets represent genetic distances corrected using the HKY
+ Γ model for nucleotide substitution
96
Table 3.3 Results from mismatch distribution and neutrality tests for cytb mtDNA phylogroups and for β-fibint7 nuclear gene.
(p(SDD) = sum of square deviations; p(hg) = Harpending’s raggedness index; Tajima’s D (D) and respective p value; Fu’s Fs test (Fs)
and respective p value; Ramos-Onsis R2 (R2) and respective p value). Results for the spatial genetic structure estimated with GEODIS
are also shown. Statistics that do not suggest range expansion are shown in bold font.
Mismatch Distribution
Locus
Cytb
β-Fib
Neutrality tests
Spatial genetic
structure
Sudden-expansion
model
Spatial-expansion
model
Phylogroup
χ2
p
p (SDD)
p (hg)
p (SDD)
p (hg)
L1
83.60
0.00
0.40
0.76
0.36
L2
358.07
0.24*
0.10
0.31
L3
754.10
0.00
0.64
L4
3494.93
0.09*
L5
1280.03
N
...
105.14
148.25
D
p
Fs
p
R2
p
0.67
-1.66
0.03
-2.32
0.07*
0.11
0.26*
0.31
0.38
-1.72
0.02
-9.92
0.00
0.05
0.00
0.84
0.51
0.88
-1.77
0.01
-14.78
0.00
0.04
0.01
0.92
0.64
0.94
0.70
-2.35
0.00
-26.33
0.00
0.02
0.00
0.00
0.10
0.09
0.06
0.11
-2.19
0.00
-27.19
0.00
0.03
0.00
0.03
0.02
0.64
0.03*
0.74
0.96
0.89
0.58
0.88
0.85
-2.00
-1.58
0.01
0.03
-3.90
-16.52
0.01
0.00
0.06
0.04
0.00
0.02
97
Table 3.4. Divergence time estimates in million years for each Lacerta lepida phylogroup using the method of Saillard et
al. (2000) and for each monophyletic group using Beast (see text for explanation of each method).
Saillarda
Mutation rate
2%
mtDNA phylogroups
L1
L2
L3
L4
L5
N
L2+L3
L1+L2+L3+L4+L5
All (L+N)
1.70%
Beastb
2.50%
(mean ± s.d.)
0.64 ± 0.12
0.45 ± 0.21
0.54 ± 0.34
0.92 ± 0.30
0.59 ± 0.19
0.85 ± 0.34
n.a
n.a
n.a
0.75 ± 0.14
0.53 ± 0.24
0.63 ± 0.40
1.08 ± 0.35
0.70 ± 0.22
0.99 ± 0.41
n.a
n.a
n.a
1%
Low HPD Mean Upper HPD
0.51 ± 0.10
0.36 ± 0.16
0.43 ± 0.27
0.73 ± 0.24
0.47 ± 0.15
0.68 ± 0.28
n.a
n.a
n.a
0.28
0.21
n.a
n.a
0.29
n.a
0.82
1.13
5.58
0.76
0.47
n.a
n.a
0.61
n.a
1.50
1.96
9.43
1.32
0.78
n.a
n.a
0.98
n.a
2.27
2.91
13.66
98
3.7. References
Alexandrino J, Arntzen JW, Ferrand N (2002) Nested clade analysis and the genetic
evidence for population expansion in the phylogeography of the goldenstriped salamander, Chioglossa lusitanica (Amphibia: Urodela). Heredity 88,
66-74.
Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic subdivision, glacial
refugia and postglacial recolonization in the golden-striped salamander,
Chioglossa lusitanica (Amphibia: Urodela). Molecular Ecology 9, 771-781.
Arntzen JW, Garcia-Paris M (1995) Morphological and allozyme studies of midwife
toads (genus Alytes), including the description of two new taxa from Spain.
Contributions to zoology 65, 5-34.
Biju-Duval C, Ennafaa H, Dennebouy N, Monnerot M, Mignotte F, Soriguer R, El
Gaied A, El Hili A, Monoulou JC (1991) Mitochondrial DNA evolution in
Lagomorphs: origin of systematic heteroplasmy and organization of diversity
in European rabbits. Journal of Molecular Evolution 33, 92-102.
Cassens I, Mardulyn P, Milinkovitch MC (2005) Evaluating intraspecific network
construction methods using simulated sequence data: do existing algorithms
outperform the global Maximum Parsimony approach? Systematic Biology
54, 363 - 372.
Castro-Parga I, Moreno JC, Christopher S, Humphries J, Williams PH (1996)
Strengthening the Natural and National Park system of Iberia to conserve
vascular plants. Botanical Journal of the Linnean Society 121, 189-206.
Clement M, Posada D, Crandall KA (2000) TCS: a computer program to estimate
gene genealogies. Molecular Ecology 9, 1657-1659.
Cooper SJB, Hewitt GM (1993) Nuclear DNA sequence divergence between
parapatric subspecies of the grasshopper Chorthippus parallelus. Insect
Molecular Biology 2, 185-194.
99
Cooper SJB, Ibrahim KM, Hewitt GM (1995) Postglacial expansion and genome
subdivision in the European grasshopper Chorthippus parallelus. Molecular
Ecology 4, 49-60.
Crandall KA, Templeton AR (1993) Empirical tests of some predictions from
coalescent theory with applications to intraspecific phylogeny reconstruction.
Genetics 134, 959-969.
Dolman G, Phillips B (2004) Single copy nuclear DNA markers characterized for
comparative phylogeography in Australian wet tropics rainforest skinks.
Dowling DK, Friberg U, Lindell J (2008) Evolutionary implications of non-neutral
mitochondrial genetic variation. Trends in Ecology & Evolution 23, 546-554.
Drummond AJ, Ho SYW, Phillips MJ, Rambaut A (2006) Relaxed Phylogenetics
and Dating with Confidence. PLoS Biology 4, e88.
Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by
sampling trees. BMC Evolutionary Biology 7, 214.
Emerson BC (2007) Alarm bells for the Molecular Clock? No support for Ho et al.'s
model of time-dependent molecular rate estimates. Systematic Biology 56,
337 - 345.
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: An integrated software
package for population genetics data analysis. Evolutionary Bioinformatics
Online 47-50.
Fritz U, Barata M, Busack SD, Fritzsch G, Castilho R (2006) Impact of mountain
chains, sea straits and peripheral populations on genetic and taxonomic
structure of a freshwater turtle, Mauremys leprosa (Reptilia, Testudines,
Geoemydidae). Zoologica Scripta 35, 97-108.
Fu YX (1997) Statistical tests of neutrality of mutations against population growth,
hitchhiking and background selection. Genetics 147, 915-925.
García-Barros E, Gurrea P, Luciáñez MJ, Cano JM, Munguira ML, Moreno JC,
Sainz H, Sanz MJ, Simón JC (2002) Parsimony analysis of endemicity and
its application to animal and plant geographical distributions in the IberoBalearic region (western Mediterranean). Journal of Biogeography 29, 109124.
Garcia-Paris M, Alcobendas M, Alberch P (1998) Influence of the Guadalquivir
river basin on mitochondrial DNA evolution of Salamandra salamandra
(Caudata: Salamandridae) from southern Spain. Copeia 1998, 173-176.
Gibbs MJ, Armstrong JS, Gibbs AJ (2000) Sister-Scanning: a Monte Carlo
procedure for assessing signals in recombinant sequences. Bioinformatics 16,
573-582.
100
Godinho R, Crespo EG, Ferrand N (2008) The limits of mtDNA phylogeography:
complex patterns of population history in a highly structured Iberian lizard
are only revealed by the use of nuclear markers. Molecular Ecology 17,
4670-4683.
Godinho R, Mendonca B, Crespo EG, Ferrand N (2006) Genealogy of the nuclear
beta-fibrinogen locus in a highly structured lizard species: comparison with
mtDNA and evidence for intragenic recombination in the hybrid zone.
Heredity 96, 454-463.
Gomez-Campo C, Bermudez-de-Castro L, Cagiga MG, Sanchez-Yelamo MD (1984)
Endemism in the Iberian Peninsula. Webbia 38, 709-714.
analysis program for Windows95/98/NT. Nucleic Acids Symposium Series
41, 95–98.
Harpending HC (1994) Signature of ancient population growth in a low-resolution
mitochondrial DNA mismatch distribution. Human Biology 66, 591-600.
Harpending HC, Sherry ST, Rogers AR, Stoneking M (1993) The Genetic Structure
of Ancient Human Populations. Current Anthropology 34, 483-496.
Harris DJ, Sá-Sousa P (2002) Molecular phylogenetics of Iberian Wall lizards
(Podarcis): Is Podarcis hispanica a species complex? Molecular
Phylogenetics and Evolution 23, 75-81.
Hasegawa M, Kishino H, Yano T-A (1985) Dating of the human-ape splitting by a
molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22,
160-174.
247-276.
Hewitt GM (1999) Post-glacial re-colonization of European biota. Biological
Journal of the Linnean Society 68, 87-112.
Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405, 907913.
Hewitt GM (2004) Genetic consequences of climatic oscillations in the Quaternary.
Philosophical Transactions of the Royal Society B: Biological Sciences 359,
183-195.
101
Leaché AD, McGuire JA (2006) Phylogenetic relationships of horned lizards
(Phrynosoma) based on nuclear and mitochondrial data: Evidence for a
misleading mitochondrial gene tree. Molecular Phylogenetics and Evolution
39, 628-644.
Lindell J, Méndez-de la Cruz FR, Murphy RW (2008) Deep biogeographical history
and cytonuclear discordance in the black-tailed brush lizard (Urosaurus
nigricaudus) of Baja California. Biological Journal of the Linnean Society
94, 89-104.
Marjoram P, Donnelly P (1994) Pairwise comparisons of mitochondrial DNA
sequences in subdivided populations and Implications for early Human
evolution. Genetics 136, 673-683.
Martin AP, Palumbi SR (1993) Body size, metabolic rate, generation time, and the
molecular clock. Proceedings of the National Academy of Sciences of the
United States of America 90, 4087-4091.
Martin DP, Rybicki E (2000) RDP: detection of recombination amongst aligned
sequences. Bioinformatics 16, 562-563.
Martin DP, Williamson C, Posada D (2005) RDP2: recombination detection and
analysis from sequence alignments. Bioinformatics 21, 260-262.
Martínez-Solano I (2004) Phylogeography of Iberian Discoglossus (Anura:
Discoglossidae). Journal of Zoological Systematics & Evolutionary Research
42, 298-305.
Martínez-Solano I, Gonçalves HA, Arntzen JW, García-París M (2004) Phylogenetic
relationships and biogeography of midwife toads (Discoglossidae: Alytes).
Journal of Biogeography 31, 603-618.
Martínez-Solano I, Teixeira J, Buckley D, Garcia-Paris M (2006) Mitochondrial
DNA phylogeography of Lissotriton boscai (Caudata, Salamandridae):
evidence for old, multiple refugia in an Iberian endemic. Molecular Ecology
15, 3375-3388.
McGuire JA, Linkem CW, Koo MS, Hutchison DW, Kristopher A, David L, Orange
I, Lemos-Espinal J, Riddle BR, Jaeger JR (2007) Mitochondrial introgression
and incomplete lineage sorting through space and time:phylogenetics of
Crotaphytid lizards. Evolution 61, 2879-2897.
Mesquita N, Hanfling B, Carvalho GR, Coelho MM (2005) Phylogeography of the
cyprinid Squalius aradensis and implications for conservation of the endemic
freshwater fauna of southern Portugal. Molecular Ecology 14, 1939-1954.
Nichols RA, Hewitt GM (1994) The genetic consequences of long distance dispersal
during colonization. Heredity 72, 312-317.
102
Padidam M, Sawyer S, Fauquet CM (1999) Possible emergence of new
geminiviruses by frequent recombination. Virology 265, 218-225.
Pliocene populations through the Pleistocene climatic cycles: evidence from
the phylogeography of an Iberian lizard. Proceedings of the Royal Society B:
Paulo OS, Jordan WC, Bruford MW, Nichols RA (2002) Using nested clade analysis
to assess the history of colonization and the persistence of populations of an
Iberian Lizard. Molecular Ecology 11, 809-819.
Pinho C, Harris DJ, Ferrand N (2007) Contrasting patterns of population subdivision
and historical demography in three western Mediterranean lizard species
inferred from mitochondrial DNA variation. Molecular Ecology 16, 11911205.
Pinho C, Harris DJ, Ferrand N (2008) Non-equilibrium estimates of gene flow
inferred from nuclear genealogies suggest that Iberian and North African
wall lizards (Podarcis spp.) are an assemblage of incipient species. BMC
Evolutionary Biology 8, 63.
Posada D, Crandall KA (1998) MODELTEST: testing the model of DNA
substitution. Bioinformatics 14, 817-818.
Posada D, Crandall KA (2001a) Evaluation of methods for detecting recombination
from DNA sequences: computer simulations. Proc Natl Acad Sci USA 98,
13757-13762.
Posada D, Crandall KA (2001b) Intraspecific gene genealogies: trees grafting into
networks. Trends in Ecology & Evolution 16, 37-45.
Posada D, Crandall KA, Templeton AR (2000) GeoDis: a program for the cladistic
nested analysis of the geographical distribution of genetic haplotypes.
Prychtko TM, Moore WM (1997) The utility of DNA sequences of an intron from
the beta-fibrinogen gene in phylogenetic analysis of woodpeckers (Aves:
Picidae). Molecular Phylogenetics and Evolution 8, 193-204.
Rambaut A, Drummond A (2005)
http://beast.bio.edsac.uk/Tracer.
Tracer
v1.3.
Available
from
Rambaut A, Drummond A (2007) BEAUTi v1.4.2. In: Bayesian Evolutionary
Analysis Utility. Available from: http://beast.bio.edsac.uk/BEAUTi.
103
Ramos-Onsins SE, Rozas J (2002) Statistical properties of new neutrality tests
against population growth. Molecular Biology and Evolution 19, 2092-2100.
Rogers AR, Harpending H (1992) Population growth makes waves in the
distribution of pairwise genetic differences. Molecular Biology and Evolution
9, 552-569.
Rozas J, Sanchez-DelBarrio JC, Messeguer X, Rozas R (2003) DnaSP, DNA
polymorphism analyses by the coalescent and other methods. Bioinformatics
19, 2496-2497.
Saillard J, Forster P, Lynnerup N, Bandelt H-J, Nørby S (2000) mtDNA variation
among Greenland Eskimos: the edge of the Beringian expansion. The
American Journal of Human Genetics 67, 718-726.
Saiz JCM, Parga IC, Ollero HS (1998) Numerical analyses of distributions of Iberian
and Balearic endemic monocotyledons. Journal of Biogeography 25, 179194.
Santucci F, Emerson BC, Hewitt GM (1998) Mitochondrial DNA phylogeography of
European hedgehogs. Molecular Ecology 7, 1163-1172.
Seddon JM, Santucci F, Reeve NJ, Hewitt GM (2001) DNA footprints of European
hedgehogs, Erinaceus europaeus and E. concolor: Pleistocene refugia,
postglacial expansion and colonization routes. Molecular Ecology 10, 21872198.
Sequeira F, Ferrand N, Harris DJ (2006) Assessing the phylogenetic signal of the
nuclear β-Fibrinogen intron 7 in salamandrids (Amphibia: Salamandridae).
Amphibia-Reptilia 27, 409-418.
Slatkin M, Hudson RR (1991) Pairwise comparisons of mitochondrial DNA
sequences in stable and exponentially growing populations. Genetics 129,
555-562.
Smith JM (1992) Analyzing the mosaic structure of genes. Journal of Molecular
Stephens M, Scheet P (2005) Accounting for decay of linkage disequilibrium in
haplotype inference and missing-data imputation. The American Journal of
Human Genetics 76, 449-462.
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.
104
Taberlet P, Bouvet J (1994) Mitochondrial DNA polymorphism, phylogeography,
and conservation genetics of the brown bear Ursus arctos in Europe.
Taberlet P, Fumagalli L, Wust-Saucy A-G, Cosson J-F (1998) Comparative
phylogeography and postglacial colonization routes in Europe. Molecular
Ecology 7, 453-464.
Tajima F (1989) Statistical method for testing the neutral mutation hypothesis by
DNA polymorphism. Genetics 123, 585-595.
associations with haplotypes inferred from restriction endonuclease mapping
and DNA sequence data. III. Cladogram estimation. Genetics 132, 619-633.
Thorpe RS, Surget-Groba Y, Johansson H (2008) The relative importance of ecology
and geographic isolation for speciation in anoles. Philosophical Transactions
of the Royal Society B: Biological Sciences 363, 3071-3081.
Ujvari B, Dowton M, Madsen T (2008) Population genetic structure, gene flow and
sex-biased dispersal in frillneck lizards (Chlamydosaurus kingii). Molecular
Ecology 17, 3557-3564.
Wallis GP, Arntzen JW (1989) Mitochondrial-DNA variation in the crested newt
superspecies: limited cytoplasmic gene flow among species. Evolution 43,
88-104.
Woolley SM, Posada D, Crandall KA (2008) A comparison of phylogenetic network
methods using computer simulation. PLoS ONE 3, e1913.
Zagwijn WH (1992) Migration of vegetation during the Quaternary in Europe.
Courier Forschungsinstitut Senckenberg 153, 9-20.
Zink RM, Barrowclough GF (2008) Mitochondrial DNA under siege in avian
phylogeography. Molecular Ecology 17, 2107-2121.
105
Chapter 4
Genetic analysis of a secondary contact zone between
Lacerta lepida lepida and Lacerta lepida nevadensis
Photos by Andreia Miraldo
Pattern of dorsal scales in Lacerta l. lepida and L. l. nevadensis*
*Lacerta lepida lepida from sampling site 5 and nevadensis from sampling site 1 in chapter 4
4. Genetic analysis of a secondary contact
zone between Lacerta lepida lepida and
Lacerta lepida nevadensis
4.1 Abstract
Lacerta lepida has endured repeated range fragmentation that has promoted
diversification within the species, and a very old split separating lineages N (subspecies
Lacerta lepida nevadensis) and L (subspecies Lacerta lepida lepida) has been identified in
the previous chapter. Using mtDNA and microsatellite data a population genetic analysis of
an area of secondary contact between the two mitochondrial lineages was performed. Levels
of gene flow across the zone were assessed to clarify if the divergent lineages are
independently evolving or if they will coalesce due to high levels of gene flow.
Hybridization between the lineages was detected by the presence of F1 hybrids but the
overall coincidence of mitochondrial and nuclear loci and the generality of the observed
narrow clines in relation to dispersal support the idea that this contact zone is acting as a
barrier to gene flow. Population genetic structure within each lineage was assessed and
estimates of genetic diversity inferred from mitochondrial DNA are relatively higher than
the ones revealed from nuclear markers, especially within lineage L. This suggests that
there might be low female dispersal relative to males. Despite some low levels of similarity
in allele frequencies between the lineages, results suggest that lineages N and L are in
independent evolutionary trajectories.
Key words: gene flow, tension zone, hybrids, selection, dispersal, clines, speciation
106
4.2. Introduction
The Biological Species Concept (BSC) defines species as “groups of actually
or potentially interbreeding natural populations which are reproductively isolated
from other such groups” (Mayr, 1963). Under this concept, speciation results from
the development of reproductive isolation between diverging taxa, which as a
general rule, involves the disruption of genic interactions (a process commonly
known as the “Dobzhansky-Muller model” (Bateson, 1909; Dobzhansky, 1937;
Muller, 1942)). The disruption of genic interactions might occur when populations
become geographically isolated for a certain period of time, during which mutations
arise and become fixed through natural selection or genetic drift. The new
accumulated mutations can lead to epistatic incompabilities between genes
responsible for ecological, physiological and/or behavioral differentiation, which are
revealed when the differentiated populations meet and hybridize (Coyne and Orr,
2004).
With growth in the use of molecular markers over recent years an increasing
number of such studies have reported subspecific parapatric contacts within species
(Avise, 2000; Avise et al., 1987). Species that reveal these parapatric contacts within
their distributions may well be representative of different stages in speciation, and
their study can provide for the quantification of the genetic differences that may
underlie this (Coyne and Orr, 2004; Hewitt, 1988). As an example, assessing the
level of reproductive isolation between evolutionary lineages at regions of contact
may help us to understand if they will remain isolated, coalesce, or sustain some
degree of gene flow between full isolation and full coalescence. Thus, examining the
degree of reproductive isolation that populations have reached through the process of
differentiation is an important issue from both evolutionary and conservation
perspectives (Avise, 2000, Crandall, 2000, Moritz, 2002). According to the genic
view of speciation (Wu, 2001) if the incompatibilities accumulated during the period
of divergence involve only small parts of the genome, the rest may mix freely upon
contact, allowing for high levels of gene flow between the diverging populations. At
107
this point populations might still fuse and the process of differentiation can be
reversed, with the result being a single genetic entity. However, if differentiation is
more extensive, even upon contact populations continue to diverge and there will be
a point when gene flow is impeded. Only after this point (“point of no return”,
following Wu (2001)), complete reproductive isolation is achieved and therefore
speciation (according to the BSC) is complete. It should be noted that Wu’s view on
speciation requires that reproductive isolation precedes adaptive differences, which
might not always be the case (see Alphen and Seehausen, 2001; Bridle and Ritchie,
2001; Mayr, 2001; Rieseberg and Burke, 2001; Vogler, 2001 for examples and other
views), but it seems to offer a plausible explanation for speciation events initiated by
a period of allopatric isolation. In fact with increasing allopatric divergence a gradual
decrease in gene flow at the time of contact occurs with gene flow being
progressively impeded with increasing heterozygote unfitness (due to increasing
mutations), until a threshold is reached when the genome is strongly linked and
introgression greatly reduced (Barton and Hewitt, 1983; Barton and Hewitt, 1985;
Barton and Hewitt, 1989).
Stages of incipient speciation are being reported at parapatric contact zones,
with different degrees of genetic isolation being found. Measuring the diffusion of
genes between evolutionary units through zones of secondary contact allows the
detection of gene flow by means of hybridization and backcrossing, providing
insight into the extent and nature of the reproductive isolation that has been achieved
(Barton and Hewitt, 1985; Harrison, 1990; Harrison, 1993; Hewitt, 1988). When
genetically distinct populations meet and if hybrids are produced selection can act on
the new combinations of alleles that result from recombination. This will create
clines at specific loci, whose shape and width will be determined by a balance
between dispersal into the zone and the strength of selection acting on them. If there
is variation on selection pressures among loci, different levels of introgression
through the hybrid zone are expected for different loci, resulting in clines that are
neither coincident nor concordant (e.g. Butlin and Hewitt, 1985). Clines associated
with neutral mixing will be wider than clines associated with selection, thus it is
possible to detect genes that might be under selection as they will introgress less than
neutral markers.
However, if the degree of genetic differentiation achieved is
extensive and selection against hybrids is strong, changes in allele frequencies across
the zone of contact will be steep and a pattern of cline coincidence and concordance
108
among loci is expected (e.g. Szymura and Barton, 1986). With time, due to epistatic
effects between loci, and genome wide linkage disequilibria, cline concordance and
coincidence will spread across to independent loci, resulting in a strong tension zone
(Barton, 1983; Barton and Gale, 1993; Barton and Hewitt, 1985; Barton and Shpak,
2000; Gavrilets, 1997). In this case, populations in the hybrid zone act as genetic
barriers to gene flow. Therefore, narrow hybrid zones with coincident and
concordant clines are more likely to effectively maintain isolation and further
promote divergence between already diverged forms.
In previous chapters it has been shown that Lacerta lepida has endured
repeated range fragmentation that has promoted diversification within this species.
Mitochondrial DNA lineages identified within Lacerta lepida have non-overlapping
geographic ranges supporting the idea of allopatric differentiation in multiple
refugia, during the Mio-Plio-Pleistocene (chapter 3). The oldest split within the
group was estimated to have occurred around 9 Mya (chapter 3) and corresponds to
the split between lineage N (subspecies Lacerta lepida nevadensis) and lineage L
(subspecies Lacerta lepida lepida). These evolutionary lineages are inferred to have
undergone range expansions following the last ice age, probably establishing zones
of secondary contact (chapter 3). Recently Paulo et al. (2008) suggested elevating
lineage N to a new species, due to the high levels of mitochondrial genetic
differentiation detected between lineages L and N and the existence of
morphological differences between these (Mateo and Castroviejo, 1990; Mateo and
López-Jurado, 1994; Mateo et al., 1996) (Appendix 1 shows pictures of lizards from
the two lineages). Furthermore, differences between the lineages seem to be also
supported by high levels of allozyme differentiation (Mateo et al., 1996). Despite the
apparently deep historical subdivision within the species, as revealed by
mitochondrial genealogy and differentiation at allozyme markers, it remains unclear
whether these divergent lineages are independently evolving or if they will in fact
coalesce due to high levels of gene flow at zones of secondary contact. Recent
studies focusing on reptiles have revealed that deep mtDNA divergence not always
corresponds to nuclear genomic divergence (Lindell et al., 2005a; Lindell et al.,
2008a; Rassmann et al., 1997; Stenson et al., 2002; Thorpe et al., 2008b; Ujvari et
al., 2008), with male biased dispersal been invoked as the main explanation for such
discrepancies. In such cases, males act as the main agents of gene flow leading to
109
reduced genetic structuring in biparentaly inherited markers compared to maternally
inherited mtDNA.
The description of individuals with seemingly hybrid morphology across a
putative zone of secondary contact between the lineages (Mateo and López-Jurado,
1994) suggests hybridization is a feature in the interaction between these two taxa. In
this chapter a population genetic analysis of an area of secondary contact between
the two Lacerta lepida mitochondrial lineages, N and L, is undertaken using both
mitochondrial and nuclear markers. The specific objectives of this work are: 1) to
indentify the exact geographic location of the putative contact zone between the
mitochondrial lineages; 2) to investigate if a barrier to gene exchange exists at the
contact zone, by analyzing clinal patterns of genetic variation (shape, coincidence
and concordance of clines); 3) to assess levels of gene flow across the contact zone,
describing the geographical extent of introgression; and finally 5) to clarify if both
lineages can be considered as good species.
4.3. Materials and methods
4.3.1. Sampling strategy collection
Sampling was conducted in 2007 along a Southeast-Northwest transect
perpendicular to the location of the putative contact zone between mitochondrial
lineage L4 and N (hereafter described as Lepida and Nevadensis lineages
respectively) (Fig. 4.1.). In order to identify the exact location of contact, nine
populations were sampled along this transect, with the first sampled population
corresponding to the south-eastern limit of the geographic distribution of
Nevadensis. Additional populations sampled were located to the north-west of the
first sampled population with each population being located 20 to 50 km away from
the closest sampled population. Lizards were captured using tomahawk traps or by
hand, and tissue samples were taken by clipping 1cm of the tail tip that was
subsequently preserved in 100% ethanol. After tissue sampling, animals were
immediately released back into the wild in the place of capture. Geographic
110
coordinates of sampling sites were recorded with a GPS. All lizards were captured
under appropriate license.
4.3.2. Laboratory procedures
Total genomic DNA was extracted from ethanol-preserved muscle tissue
using the same protocol described in section 2.3.2. The entire (1143bp) cytochrome
b gene was amplified with primers TRNAGLU and TRNATHR (see chapter 2).
These primers were shown to be specific for the mitochondrial cytb gene and do not
amplify Numts. Protocols for the amplification and sequencing of the entire cytb
gene were as in section 2.3.4.
Microsatellites
To obtain genotypic profiles for the lineages under study, eight polymorphic
microsatellite loci were amplified for all samples. As there are no microsatellites
specifically characterized for Lacerta lepida, microsatellite loci characterized for
other lacertid lizard species that were previously shown to be polymorphic within
Lacerta lepida (Nunes, unpublished data) were amplified: loci C9, B4 and D1, which
were characterized in Podarcis muralis (Nembrini and Oppliger, 2003); loci PB66
and PB73, characterized in Podarcis bocagei (Pinho et al., 2004); locus LV-4-72,
from Lacerta vivipara (Boudjemadi et al., 1999); locus LVIR17 characterized in
Lacerta viridis (Böhme et al., 2005) and locus LIZ24 in Lacerta schreiberi (Paulo,
unpublished data).
PCR’s were carried out in a final volume of 10 µL containing 5x PCR buffer,
2.0 µL of 10x Go Taq® Buffer, 2.0 mM of MgCl2, 0.2 mM of each dNTP, 0.5 µM
of each primer, 2 µg of BSA, 0.5 units of Go Taq® DNA polymerase and
approximately 50 ng of DNA. Locus C9 was amplified together with locus PB66,
and locus D1 with locus PB73 in two different PCR duplexes. All other loci were
amplified individually in independent PCRs. All PCR reactions were performed in a
DNA engine tetrad 2, Peltier thermocycler, using the following profile: initial
111
incubation at 94°C for 3 min followed by 30 cycles of denaturing at 94°C for 30 s,
annealing for 30 s (with temperature according to each locus, Table 4.1.) and
extension at 72°C for 30 s; plus a final extension incubation at 72°C for 30 min.
PCR products of the 8 loci were combined in two different mixes that allow
distinguishing loci according to fluorescent dye and allele size. Mix A included
locus D1, LV-4-72, LVIR17 and PB73; and Mix B included locus B4, C9, LIZ24
and PB66. To determine fragment length, 1 µL of either Mix A or Mix B was added
to 8.9 µL of Hi-Di Formamide
TM
and 0.1 µL of GeneScan
TM
-500 RoxTM size
standard. Each cocktail mix was run in an automated ABI Prism 377 and peaks were
visualized with Genemapper Software version 4.0 (Applied Biosystems).
4.3.3. Data analyses
Mitochondrial DNA data
DNA sequences were aligned by eye using BioEdit Sequence Alignment
Editor 7.01 (Hall, 1999). All sequences were trimmed to 627bp before further
analysis. In order to assign each sample to the correct mitochondrial lineage (Lepida
versus Nevadensis), median joining (MJ) (Bandelt et al., 1999) and statistical
parsimony (SP) (Templeton et al., 1992) networks were constructed. The MJ
network was computed with the program NETWORK 4.5.0 (www.fluxusengineering.com) keeping the parameter ε = 0, which does not allow less
parsimonious pathways to be included in the analysis. The SP network was inferred
using the program TCS 1.21 (Clement et al., 2000) with a connection limit of 70
mutational steps. In order to identify which haplotypes were new, sequences from
lineage L4 and N (chapter 3) were added to the dataset and new networks were
constructed.
For each sampling site the frequency of each haplotype and the
frequency of Lepida and Nevadensis haplotypes were calculated.
For each
population haplotype (H) and nucleotide diversity (π) were also calculated.
Population structure within each mtDNA lineage was assessed by estimating values
of ΦST (a mtDNA analogue for FST; Excoffier et al (1992)) for each locality and
performing hierarchical analysis of molecular variance (AMOVAs), using localities
112
as groups. Pairwise ΦST values between localities were also calculated as a measure
of population genetic differentiation. All tests were performed in ARLEQUIN
version 3.11 (Excoffier et al., 2005).
Microsatellite data
Overall assessment of microsatellite variability
As this is the first time that this set of 8 microsatellite loci has been applied to
a population level study in Lacerta lepida, and as almost each locus has been
characterized in a different species through different independent studies, tests for
non-random associations between diploid genotypes at each pair of loci were carried
out. Tests for linkage disequilibrium were performed through a log-likelihood ratio
statistic (G-test) using the Markov chain algorithm as implemented in GENEPOP 4.0
(Rousset, 2008). For each locus levels of polymorphism, across and within
populations, were determined by assessing allele number and frequency. Deviations
from expectations of Mendelian inheritance were tested for each population at each
locus using exact tests (Raymond and Rousset, 1995) to check for the presence of
heterozygote deficits. Departures from Hardy Weinberg Equilibrium (HWE) can be
due to biological factors such as population structure, non-random mating and
selection against hybrids. Nevertheless, a departure from HWE can also be due to
technical issues that occur during the process of microsatellite amplification, such as
the presence of null alleles. Null alleles arise when mutations in the microsatellite
flanking regions occur leading to the non-amplification of certain alleles. This can
result in heterozygotes being wrongly identified as homozygotes due to the nonamplification of one of the alleles, leading to an apparent excess of homozygotes and
departures from HWE. It has been shown that the frequency of null alleles in a
congeneric species rapidly increases with increasing phylogenetic distance from a
focal species (e.g. Li et al., 2003). As the microsatellite used in this study were not
specifically characterized for L. lepida, flanking region mutations are more likely to
have occurred, increasing the probability of non-amplification of alleles. To address
this issue the presence of null alleles was assessed using the program MICROCHECKER version 2.2.1 (Van Oosterhout et al., 2004). When null alleles were
detected, their frequency was estimated using the methodology of Dempster et al
113
(1977), assuming that any detected heterozygote deficit is due to the presence of null
alleles and not to population structure (Wahlund effect.).
Population based analysis
Allele richness, gene diversity (expected heterozygosity) and the indicator of
inbreeding within populations (FIS) were calculated. Levels of genetic differentiation
amongst pairs of populations were assessed by multilocus estimates of FST. All tests
were carried out in GENEPOP 4.0 (Rousset, 2008). Furthermore, isolation by
distance within each mitochondrial lineage was also assessed by regressing a
pairwise matrix of FST values (linearised as FST/(1-FST)) against the geographic
distance between localities. The statistical significance of the association was
determined with a Mantel test using the online version of the software IBD
(Bohonak, 2002).
Admixture estimation
STRUCTURE version 2.2 (Pritchard et al., 2000) was used to evaluate the
extent of admixture between the two mitochondrial lineages under study. Two
datasets were analysed: one dataset comprising all loci and a reduced dataset with
loci that were not in HWE for the majority of localities removed. In an exploratory
analysis to infer the number of genetically homogenous groups of individuals
(clusters, K) along the transect, several analyses were run changing the value of K
from 2 to 9. The true value of K was chosen using information from the posterior
probability (Ln P (D)) given by the software and by ∆K, a quantity based on the rate
of change of the posterior probability with respect to the number of clusters, as
defined by Evano et al. (2005). According to these authors, in most cases the
posterior probability given by STRUCTURE does not provide the correct estimation
of the number of clusters in the data, while ∆K always shows a clear peak at the true
value of K. Analyses were run 3 times for each K with a burn in period of 50,000, to
minimize the effect of the starting parameter settings, followed by 500,000
repetitions per run. Consistency and convergence of parameter estimates were
checked by visualizing the plots of the parameters. After choosing the true value of
K an analysis with all loci was performed to evaluate levels of admixture along the
transect. For all analyses the admixture model was assumed. According to this model
114
individuals can have a mixed ancestry, inheriting a fraction of their genome from
different ancestors. The posterior mean estimate of those fractions was used to
estimate the proportion of membership of each sampling locality in each of the
inferred clusters.
Cline analysis
Mitochondrial cline analysis was performed using information about the
frequency of Lepida haplotypes in each sampled locality. For the purpose of nuclear
genome cline analysis the mean proportion of membership of each sampling locality
to Lepida using all microsatellite loci data was first calculated in STRUCTURE as
described above. This data was used to estimate a multilocus cline from which levels
of nuclear genome introgression across the contact zone could be inferred. Single
locus clines were also estimated for loci that showed diagnostic shifts in allele
frequencies across the transect. Those loci were collapsed into a two allele system,
representing Lepida and Nevadensis alleles. All rare and low frequency alleles were
allocated either to Lepida or Nevadensis groups according to their occurrence along
the transect.
Maximum likelihood clines were fitted independently and plotted with
ANALYSE (Barton and Baird, 1995). Cline fitting was performed by adding
geographic information of sampling localities to the allele frequency data. Localities
were collapsed into a one-dimensional transect, with geographic distances measured
from the southernmost sampling site (Locality 1) (Fig. 4.1.). As sampling sites 7 and
8 do not fit well with the one-dimensional transect, they were not used for cline
fitting purposes. Clines were fitted to the Symmetric Tanh curve (Barton and Gale,
1993), and the two parameters describing each curve (center, c, and width, w) were
estimated by the program. Estimation of both parameters started from approximate
values of c calculated from the data and incorporated in the program. The centre of
the cline is the point where the frequency of alleles switches above 0.5. Cline width
was calculated as the inverse of the maximum of the slope of the cline curve
(1/maximum slope) as described in Szymura and Barton (1986). The proportion of
membership to Lepida (or the frequency of Lepida alleles in the case of single locus
clines), p, was allowed to vary between the pmin and pmax (minimum and maximum
115
gene frequency at the tail ends of the cline) estimated from the data and incorporated
in the program.
4.4. Results
A total of 200 samples were collected with 19 to 30 samples collected per
sampling site. Sampling sites and number of samples per site are shown in Fig. 4.1.
All cytb sequences represented uninterrupted open reading frames, with no
gaps or premature stop codons, suggesting they are functional mitochondrial DNA
copies. To allow comparisons with sequences from the chapter 3, all sequences were
trimmed to 627 bp. From the 627 bp analysed, a total of 104 were variable from
which 85 were parsimony informative. Fifty eight unique haplotypes were obtained.
The genealogical relationships between haplotypes inferred by the two approaches
for network construction (MJ and SP) were identical (Fig. 4.2.), with two very
divergent groups of haplotypes identified. The two haplotype clusters correspond to
the two mitochondrial lineages under study, Lepida and Nevadensis, and are
connected through 67 mutational steps. From the 58 haplotypes sampled 17 have
been previously sampled in a broader phylogeographic analysis of Lacerta lepida
(chapter 3), 21 represent new Lepida haplotypes and the remaining 20 represent new
Nevadensis haplotypes. No populations were found to be admixed for both mtDNA
lineages, with populations 1 to 4 being fixed for Nevadensis haplotypes while
populations 5 to 9 were fixed for Lepida haplotypes (Fig. 4.1.). Haplotype
frequencies within each lineage are represented in Fig. 4.3.
Overall genetic differentiation among mtDNA lineages is shown in Table
4.2. Within Lepida some level of genetic structure was detected, with 11.5%
(ΦST=0.1145, p=0.0) of the genetic variation occurring between sampled localities.
116
This pattern was not found in Nevadensis (ΦST=0.014, p=0.16). Pairwise ΦST values
between localities are shown in Table 4.3.
Overall assessment of microsatellite variability
From all microsatellite loci used in this study only locus LV-4 failed to
amplify consistently across all populations and was therefore eliminated from further
analyses. All loci had moderate to high levels of polymorphism with the number of
alleles over all populations ranging from 8 (LIZ24) to 23 (B4 and LVIR17) (Table
4.1.). Changes in allele frequencies over all localities and per locality are shown in
Fig. 4.4. and Fig. 4.5. respectively. Generally, the most frequent allele at each locus
is different for the two mitochondrial lineages (Fig. 4.4.), with the exception of the
least polymorphic locus (LIZ24), where allele 115 has a gene frequency of 85% to
99 % in Lepida and Nevadensis, respectively. The differences in allele frequencies
per locus within each mitochondrial lineage are shown in Fig. 4.6. The patterns of
allele frequency and allele sharing between lineages differ across loci; nevertheless,
each mitochondrial lineage exhibits private alleles at each locus. Three loci (Locus
C9, LVIR17 and PB73) exhibit a clear difference in allele frequencies between the
lineages, revealing a clear clinal transition. The clearest picture of clinal variation
occurs at locus C9 where lineages show almost non-overlapping allele size ranges,
with few shared alleles of intermediate size and low frequency (Fig. 4.6.). A similar
pattern can be detected at loci LVIR17 and PB73. Nevertheless at locus PB73 the
frequency of private alleles from each lineage is very low when compared with the
frequency of intermediate size shared alleles, which have the highest frequency in
each of the lineages. The remaining loci show a broad overlap in allele sizes between
Nevadensis and Lepida lineages with the majority of high frequency alleles being
shared among them. Although some alleles in these loci show evidence of clinal
variation across the transect (for e.g. alleles represented by dark green and turquoise
colour in locus B4, Fig. 4.5.), the majority of alleles do not (for e.g. alleles
represented by grey and mustard colour in locus B4, Fig. 4.5.).
Despite both
lineages sharing the same most frequent allele at locus LIZ24, some level of
117
differentiation is still detectable, as all other alleles, although occurring at very low
frequencies, are private for the Nevadensis lineage.
Significant linkage disequilibrium was detected for 11 pairs of loci in 6
localities, with most of the non-random associations being detected at locality 4
(Table 4.4.). Generally, pairs of loci in linkage disequilibrium were not detected at
more than one locality. High deviations from HWE were detected in the majority of
loci when considering all populations, with the exception of locus LIZ24 (χ2=13.6,
df=10, p=0.19). Tests of HWE for each locus/locality combination revealed 16 cases
of significant heterozygote deficit (FIS>0; p<0.05) relative to what is expected under
HWE (Table 4.5.). Loci LVIR17 and D1 showed the highest percentage of localities
with a heterozygote deficit. HW disequilibrium in locus LVIR17 occurred mainly
within the Lepida lineage while HW disequilibrium in D1 occurred only within the
Nevadensis lineage. With the exception of localities 6 and 9, which show
heterozygote deficits for the majority of loci, heterozygote deficits were usually
detected at only a single locus within each locality. As HW deviations are not
distributed consistently across loci or localities they are possibly the result of local
population effects. Nevertheless, the presence of null alleles was detected which
could explain some of the observed heterozygote deficits. In all cases where null
alleles were detected, the estimated frequency was always lower than 0.16 (Table
4.5.). The estimation of null alleles assumes that there is no population structure in
the data, and therefore these can be viewed as overestimations. Even though the
values estimated are relatively low when compared to allele frequencies within each
locus, and therefore null alleles most probably did not influence the data analysis.
Population based analysis
FST values between pairs of localities are shown in Table 4.3. Two groups of
localities can be identified that exhibit significant differentiation from each other.
One group consists of localities 1 to 4 and the other is comprised of localities 5 to 9,
with pairwise FST values between localities from the two groups ranging from 0.09
to 0.15. These two groups correspond to Nevadensis and Lepida mitochondrial
lineages respectively. Similar levels of overall multilocus estimates of FST were
detected for each mtDNA lineage (FST=0.022 for Nevadensis and FST=0.024 for
118
Lepida). Localities within each group show little genetic differentiation between
them, with pairwise FST values ranging from 0.01 to 0.04 (Table 4.3.) Generally,
higher pairwise FST values are reported within Nevadensis. This lineage also shows a
trend towards isolation by distance (r2=0.27) (Fig. 4.7.), although not statistically
significant (Mantel test, r=0.52; p=0.13).
Admixture estimation
Similar results were obtained when analysing the microsatellite data with
STRUCTURE, which revealed K=2 as representing the most likely number of
clusters. The highest values of L(K) were obtained when K=2 and K=9, nevertheless
there was a clear peak in ∆K when K=2 (Fig. 4.8.), which according to Evanno et al.
(2005) reveals the true value of K. Both analysed datasets (dataset with all loci and
reduced dataset with LVIR17 and D1 removed due to deviations from HW
expectations in the majority of localities) gave similar results. The proportion of each
locality assigned to each cluster is shown in Fig. 4.1. Localities 1 to 4 were assigned
to cluster 1 (corresponding to Nevadensis mitochondrial lineage), while localities 5
to 9 were assigned to cluster 2 (corresponding to Lepida mitochondrial lineage).
These results are concordant with the mitochondrial data.
For each individual analysed, the proportion of assignment to each lineage
can be seen in Fig. 4.9. Generally, the majority of individuals have a high proportion
of assignment to one of the lineages (higher than 95%), but with a few individuals
showing some level of admixture (Fig. 4.9.). In individuals with admixed ancestry,
usually a high proportion of their genome is assigned mainly to one of the lineages
(80 to 95%). Nevertheless in locality 5 two individuals were identified as having
more extensive admixture levels. One of them represents an F1 hybrid between
Lepida and Nevadensis, having half of its nuclear genome assigned to each lineage.
The other lizard likely represents a backcross of an F1 hybrid with a pure Lepida
form, having 24% of its genome assigned to Nevadensis and the remaining assigned
to Lepida.
4.4.3. Cline analysis
119
Loci LVIR17, C9 and PB73 showed diagnostic shifts in allele frequencies
across the transect and were therefore used for a single locus cline analysis.
Maximum likelihood fitted clines are presented in Fig. 4.10. Cline centers are
generally coincident, being located between locality 4 and 5 with distance from
locality 1 ranging from 103 to 110km depending on the marker used (Table 4.6.).
Single locus cline centers derived from microsatellite data were more distant from
locality 1 than the center of the mtDNA cline, with the exception of locus LVIR17
(Fig. 4.10.b and Table 4.6.). Although cline centers do not differ much between loci,
cline widths do exhibit a greater extent of variation, with locus PB73 having the
widest cline (40 Km) and mitochondrial DNA the steepest (2.7 Km). Although
microsatellite single locus clines differ from the mtDNA cline in terms of width, this
difference is modest overall (3km) when a multilocus cline approach is taken (Fig.
4.10.a and Table 4.6.).
4.5. Discussion
In chapter 3 it was shown that Lacerta lepida has been subject to historical
range fragmentation that promoted diversification within the species. The oldest split
within the group represents the divergence between lineages N and L and was
estimated to have occurred around 9 Mya, during the Miocene. Both evolutionary
lineages are inferred to have undergone range expansions following the last ice age
establishing a zone of secondary contact. Evidence for the existence of a secondary
contact zone between both lineages emerged with the discovery of one population
containing both mitochondrial lineages (sampling site 33 in chapter 3) located just to
the west end of Sierra Nevada Mountains. Results presented here indicate that the
hybrid zone occurs in the valley north of the Sierra Nevada Mountains, between
locality 4 and 5. It is plausible to infer that the contact zone runs from the coastal
area of Granada and contours the northern side of the Sierra Nevada Mountains in a
north-east direction reaching the western part of Sierra de Baza (Locality 4). Further
inferences about the eastern extent of the contact zone cannot be made due to lack of
sampling. Nevertheless previous studies suggest that individuals with what
120
resembles hybrid morphology between the lineages occur in the limits between the
provinces of Murcia and Albacete and also in Valencia (Mateo and López-Jurado,
1994).
Several taxa show phylogenetic breaks associated with the Betic mountains
caused by an history of allopatry in this region (see Gomez and Lunt, 2007 and
references therein), which also seems to be the case for Lacerta lepida. The Sierra
Nevada is the highest altitudinal limit to the distribution of Lacerta lepida, where it
reaches 2,400m. At the peak of the last glacial maximum (LGM), conditions
throughout the Sierra Nevada and adjacent mountains were most likely unsuitable
for the persistence of Lacerta lepida, resulting in a distribution restricted to much
lower altitudes. It is therefore probable that contact between the lineages has
occurred after the LGM, when an increase in temperature allowed populations to
expand their ranges from refugial areas. Contact is thus estimated to have occurred
approximately 15,000 ya (years ago), or perhaps even more recently due to the effect
of the Younger Dryas. The Younger Dryas was a period of rapid climatic change
during the interglacial characterized by a dramatic fall in temperature re-establishing
conditions similar to the last glacial period. The Younger Dryas ended
approximately 10,000 ya, with the start of the pre-boreal when the climate warmed
markedly. Although the climatic changes during the YD are thought to have been
less dramatic in southern Spain, an increase in steppe type vegetation in the region is
registered during this period, especially at higher altitudes (Carrión et al., 1998;
Carrión and Dupre, 1996). It is therefore likely that the Younger Dryas had an
impact on the distribution of Lacerta lepida in the region, likely delaying or
interrupting contact between the lineages L and N until the end of the climatic
reversal, around 10,000 ya.
4.5.1. Genetic structure of the contact zone: tension zone vs
neutral diffusion
Clines through the hybrid zone are narrowly coincident with cline centres
being located 104-110 km northwest of locality 1. Coincidence of clines is expected
after secondary contact that seems to be the case for this contact zone. At the time of
121
contact genetic introgression is initiated, but any co-adaptation of lineage specific
alleles may enhance the effect of divergence between the lineages through epistatic
interactions among loci. New recombinants generated from hybridisation between
the lineages may be less fit, under these conditions epistasis and linkage can promote
cline coincidence (Barton and Hewitt, 1989). Nevertheless cline widths are not
concordant, with mtDNA cline being extremely narrow (2.7 km for mtDNA) when
compared to the consistently wider nuclear clines (between 10 and 40 km) (Table
4.6.). Interestingly, when all nuclear loci are analysed in a multilocus cline approach,
the width of the nuclear cline narrows, approximating the mitochondrial one.
Frequencies used when generating the multilocus nuclear cline represent the
proportion of membership of each sampling locality in each of the inferred clusters
(lineages) and it is estimated using information from all microsatellite loci.
Therefore the cline calculated using this approach is a representation of what is
happening in the nuclear genome as a whole, rather than at single locus. Although it
is known that upon contact different parts of the genome will introgress differently
depending on the selection forces that act directly or indirectly on them leading to
differences in cline widths (Butlin and Hewitt, 1985; Hewitt, 1993), the single locus
cline widths should be interpreted with caution. Collapsing allele frequencies to a
two allele system, with alleles being classified as either belonging to Lepida or
Nevadensis might be a source of error, particularly considering alleles where
differences in the frequency across lineages are not very pronounced. Alleles that
represent persistent ancestral variation due to incomplete lineage sorting may be
represented equally in both lineages, making it difficult to collapse them into the two
allele system. These factors might have affected the subsequent single locus cline
analysis undertaken, most likely widening the clines. Therefore, the multilocus
approach is considered as a better estimate of nuclear genome cline width.
The fact that only two hybrids were found in the populations near the centre
of the contact zone, and that mtDNA and nuclear clines are coincident suggests that
selection against hybrids is occurring in the zone. As is the case for many hybrid
zones, the Lacerta lepida hybrid zone conforms to the “tension zone” model where
clines are maintained by a balance between selection and dispersal (Barton and
Hewitt, 1985). Selection forces that are influencing cline shape are probably
endogenous as there is no clear evidence for clinal environmental variation through
the zone. Hybrid fitness is therefore most probably determined by genome
122
interactions, such as heterozygote disadvantage and epistasis, independent from the
environment. Prezygotic mechanisms can also be responsible for the observed
pattern of steep cline widths. In fact, differences in the reproductive activity between
both lineages have been identified. Nevadensis shows an extended reproductive
period in concordance with the longer period of male sexual activity and has the
ability of producing two clutches per year, while Lepida only produces one (Castilla
and Bauwens, 1989; Mateo, 1988; Mateo and Castanet, 1994). Nevertheless the
presence of F1 hybrids in locality 5 and the occurrence of individuals with
intermediate morphological characters between both lineages (Mateo and LópezJurado, 1994) suggest hybridization between lineages to be relatively frequent.
The time since the Younger Dryas corresponds approximately to 3,300 generations
for Lacerta lepida, where the generation time is estimated to be on average 3 years
(Mateo, 1988; Mateo and Castanet, 1994). To generate a cline width of 10km
assuming neutral diffusion, a dispersal rate of 100m per generation would have to be
invoked (using the equation: T = 0.35 (d/w)2 (Endler, 1977); where T represents time
since contact, d represents dispersal rate and w represents cline width). One hundred
metres dispersal per generation (3 years) is a small distance for Lacerta lepida.
Dispersal rates for Chioglossa lusitanica, a relatively small salamander, have been
estimated to be 120m per generation (Sequeira et al., 2005). This species’ dispersal
might be restricted due to habitat requirements such as high dependence of juveniles
on water streams, which is not the case for Lacerta lepida. Furthermore, Lacerta
lepida territories were estimated to be on average 3500 m2 for females and 11000 m2
for males (Salvador et al., 2004). These large territories suggest that the species
dispersal rates (variance in parent/offspring distance) might be higher than 100 m.
This is supported by the overall multilocus FST values within each lineage which are
relatively small (0.02), suggesting high levels of gene flow. Taking into account
these FST values, higher dispersal between the lineages would be expected implying
wide nuclear clines, which is not observed. It seems likely that further sampling
between localities 4 and 5 (25km apart) will probably reveal much steeper cline
widths, requiring even smaller dispersal rates for the nuclear genome to conform to
neutral diffusion. Although our evidence is indirect, it seems likely that selection
against hybrids is responsible for the observed cline widths.
Interestingly, overall ΦST values within lineages inferred from mitochondrial
DNA are relatively higher than the ones revealed from nuclear markers. ΦST values
123
from mitochondrial data suggest some level of genetic structuring within the Lepida
lineage (Table 4.2.). Pairwise ΦST values between localities are generally one order
of magnitude higher in Lepida than in Nevadensis, suggesting higher levels of
female mediated gene flow in the latter. However, the haplotype frequencies within
populations of Nevadensis suggest that the lower ΦST values may be a consequence
of the very high frequency of the same haplotype in all populations (Fig. 4.3.).
Interestingly this haplotype is that identified as the ancestral haplotype within this
lineage (Fig. 4.2.). Thus the lower ΦST value observed within the Nevadensis lineage
may be a consequence of incomplete lineage sorting giving a signature of low
population differentiation. Indeed some level of differentiation is indicated by the
existence of a number of private haplotypes (although at low frequency) within each
locality (Fig. 4.3.). The evidence for greater structuring in the maternally inherited
mtDNA marker compared to the biparentaly inherited nuclear markers in the Lepida
lineage suggests low female dispersal relative to males, which has been increasingly
reported in other reptilian studies (e.g. Lindell et al., 2005b; Lindell et al., 2008b;
Stenson et al., 2002; Thorpe et al., 2008a; Ujvari et al., 2008). The absence of this
signature among Nevadensis populations is consistent with either a higher level of
female dispersal or mutation-drift non-equilibrium conditions amongst these
populations.
4.5.2. The historical dynamics of lineages contact and
introgression
It should be noted that there may have been episodes of introgression during
earlier interglacial periods of contact between the lineages, especially if during those
previous contacts selection against hybrids was weaker or if the contact was
maintained for longer. The deeply divergent mtDNA lineages, corresponding to a
divergence time of approximately 9 Mya, suggest a substantial period of geological
time has been available for climatically mediated allopatry and parapatry. Earlier
contact and hybridisation may have allowed for the exchange of alleles between
lineages, therefore some degree of similarity in allele frequencies between the
lineages is expected, concomitant with the “evolutionary filter” role played by the
124
contact zone. This is corroborated by the FST values, which are much higher in
comparisons between lineages than within lineages. It is more likely that those
similarities are the result of ancestral gene flow and do not indicate contemporary
gene flow amongst lineages. As postulated by Hewitt (1988; 1996) species that have
persisted in southern European refugia through the ice ages have most probably
established long term hybrid zones through complex patterns of contraction and
expansion. Lacerta lepida lineages have most probably established contact
repeatedly during the Quaternary. Earlier contacts provided the possibilities for
exchange of alleles between the lineages through older hybridization events.
Although the low frequency shared alleles observed in this study might be the
reflection of incomplete lineage sorting, they most likely represent past introgression
from one lineage to the other at the time of earlier contacts.
4.5.3. Taxonomic and conservation implications
Despite some apparent evidence for the existence of gene flow as revealed by
the morphological intermediacy found across the hybrid zone (Mateo and LópezJurado, 1994), the existence of clear significant morphological differences between
the pure forms (Mateo and Castroviejo, 1990; Mateo and López-Jurado, 1994; Mateo
et al., 1996), clinal variation in genotype frequencies and indirect indication of
hybrid inferiority (as revealed by the very low numbers of hybrids detected) suggest
that Lepida and Nevadensis are on independent evolutionary trajectories.
Divergence between lineages seems to have passed the threshold whereby
introgression is greatly reduced and therefore coalescence is unlikely. The two
mitochondrial lineages should be considered as different evolutionary units and
conservation efforts should be put in place to protect them.
Lacerta lepida is widely distributed across Spain and Portugal and there are
no specific conservation measures for its protection. The IUCN considers the
existence of only one species within the group which is generally classified as in
significant decline mainly due to habitat loss. In the last Mediterranean Red list
assessment Lacerta lepida was classified as Near Threatened (NT), a status that is
more alarming if the existence of two species within it is to be considered. More
125
worrying is the case of “Nevadensis” lineage (Lacerta lepida nevadensis) which
presents a very restricted distribution area, associated with zones of high touristic
pressure and where current changes in land use, (e.g. the increasing density of
greenhouses in the province of Almeria) are most likely to be detrimental for the
species survival, and the threats posed by habitat loss might be more alarming.
126
Table 4.1. Primer sequences, annealing temperature (TA), number of alleles (NA) and allele size range for each locus in
Lacerta lepida. Further information regarding each locus can be found in the papers where they were first characterized
(Source).
Locus
Source
Primers
LIZ24
Paulo (not
published)
B4
Nembrini &
Oppliger (2003)
C9
Nembrini &
Oppliger (2003)
Pb73
Pinho et al
(2004)
Pb66
Pinho et al
(2004)
D1
Nembrini &
Oppliger (2003)
LVIR17
Bohme et al.
(2005)
LV4-72
Boudjemadi et
al. (1999)
F: FAM - TCAGTCCAAATATCTCTACAGG
R: AGATGAGCAGCATATAGTGATG
F: HEX - AATCTGCAATTCTGGGATGC
R: AGAAGCAGGGGATGCTACAG
F: FAM - CATTGCTGGTTCTGGAGAAAG
R: CCTGATGAAGGGAAGTGGTG
F: FAM - GCCCATGTCACTTCAGGTAGAAGC
R: GAAAACTAGGAGTTAGGGAGAAGG
F: NED - GGACAGCTAGTCCCATGGCTTAC
R: GGATTGCTGTCACCAGTCTCCCC
F: NED - GAGTGCCCAAGACAGTTGTAT
R: GAGGTCTTGAATCTCCAGGTG
F: NED-AGCTCTGGATCGAGACAACCTGG
R: TCTCTGAAGGAGACCGGCTCC
F: HEX - CCCTACTTGAGTTGCCGTC
TA (oC)
NA
Allele Size Range
50
8
115-139
61
23
122-166
58
14
130-169
58
17
120-152
58
21
148-192
58
22
134-209
61
23
221-265
63
...
...
R: CTTTGCAGGTAACAGAGTAG
127
Table 4.2.. Number of samples (N), number of haplotypes (H) and nucleotide diversity
(π) for each Lacerta lepida sampled locality using information from mitochondrial DNA
cytb gene sequences. ΦST with respective p values for each mitochondrial lineage is also
shown.
Nevadensis
Loc 1
Loc 2
Loc 3
Loc 4
Lepida
Loc 5
Loc 6
Loc 7
Loc 8
Loc 9
N
79
19
22
18
20
99
21
23
18
21
16
H
24
7
8
9
8
34
7
8
8
11
10
π
...
0.0038
0.0024
0.0057
0.0052
...
0.0024
0.0028
0.0039
0.0029
0.0040
ΦST
0.0141
p
0.1603
0.1145
0.0000
Table 4.3. Pairwise FST values between nine Lacerta lepida localities (Loc.) for 627 bp
of mtDNA cytb gene (below diagonal) and for 7 microsatellite loci. Statistically
significant pairwise FST values (p<0.05) are denoted with grey shading.
Loc.
1
2
3
4
5
6
7
8
9
1
...
0.021
0.037
0.013
0.974
0.972
0.967
0.971
0.967
2
0.019
...
0.011
0.025
0.979
0.977
0.974
0.977
0.974
3
0.016
0.008
...
0.007
0.966
0.965
0.959
0.964
0.958
4
0.039
0.036
0.014
...
0.967
0.966
0.961
0.965
0.960
5
0.089
0.091
0.086
0.091
...
0.087
0.227
0.125
0.166
6
0.115
0.126
0.107
0.117
0.022
...
0.093
0.012
0.107
7
0.136
0.146
0.126
0.144
0.041
0.024
...
0.135
0.132
8
0.133
0.140
0.117
0.130
0.032
0.023
0.011
...
0.059
9
0.095
0.111
0.092
0.105
0.016
0.025
0.027
0.022
...
128
Table 4.4. Results of tests for linkage disequilibrium for each pair of 7 microsatellite
loci from Lacerta lepida, in each sampled locality. Only significant non-random
associations between pairs of loci are shown.
Locality
1
2
4
4
4
4
4
4
4
4
6
6
8
8
9
Locus 1 Locus 2
PB66
D1
LVIR17 PB73
B4
LVIR17
B4
C9
LVIR17
C9
LVIR17 PB73
PB73
C9
B4
PB66
LVIR17 PB66
C9
PB66
LVIR17 PB73
C9
PB66
C9
PB73
B4
CYTB
LIZ24
C9
P
0.029
0.030
0.036
0.000
0.000
0.044
0.034
0.000
0.006
0.004
0.014
0.029
0.042
0.002
0.045
129
Table 4.5.. Measures of genetic diversity at 7 microsatellite loci in Lacerta lepida: expected (HE) and observed (HO) heterozygotes,
FIS values and Null allele frequency for each locality (Loc.)/locus combination. Shaded values are statistically significant (p<0.05) and
denote significant heterozygote deficits. The presence of null alleles detected by MICROCHECKER is denoted with bold font.
Locus
Loc.
1
2
3
4
5
6
7
8
9
B4
HE (HO)
14.3 (13)
15.4 (14)
16.14 (13)
16.1 (13)
21.1 (22)
20.6 (17)
15.2 (12)
25.2 (26)
16.3 (14)
FIS
0.10
0.09
0.20
0.20
-0.04
0.18
0.21
-0.03
0.15
Null
0.00
0.00
0.07
0.07
0.00
0.07
0.06
0.01
0.07
LVIR17
HE (HO)
FIS
15.0 (12)
0.20
15.8 (17) -0.08
14.5 (11)
0.25
17.1 (15)
0.12
20. 8 (18) 0.14
29.8 (16)
0.20
18.7 (19) -0.01
27.2 (21)
0.25
17.0 (11)
0.36
Null
0.07
0.00
0.10
0.06
0.05
0.09
0.00
0.12
0.16
LIZ24
HE (HO)
FIS
0 (0)
0 (0)
1 (1)
0 (0)
3.8 (4)
-0.05
4.6 (4)
0.14
9.4 (8)
0.15
7.4 (8)
-0.08
5.5 (4)
0.28
C9
Null
0.00
0.00
0.14
0.11
0.00
0.00
HE (HO)
17.1 (16)
19.0 (19)
17.1 (15)
16.4 (17)
17.4 (15)
16.2 (12)
14.1 (12)
12.9 (13)
12.8 (9)
FIS
0.06
0.00
0.12
-0.04
0.14
0.26
0.15
-0.01
0.30
Null
0.07
0.08
0.04
0.00
0.06
0.11
0.05
0.00
0.11
Locus
Loc.
1
2
3
4
5
6
7
8
9
PB73
HE (HO)
FIS
16.9 (16)
0.06
18.3 (19)
-0.04
15.8 (15)
0.05
16.5 (16)
0.03
19.8 (21)
-0.06
17.3 (12)
0.31
18.9 (18)
0.05
24. 5 (23)
0.06
16.1 (13)
0.20
Null
0.04
0.00
0.00
0.01
0.00
0.13
0.00
0.01
0.10
PB66
HE (HO)
FIS
17.2 (17) 0.01
18.2 (17) 0.07
16.2 (18) -0.12
17.8 (17) 0.05
20.3 (22) -0.09
20.9 (20) 0.04
18.8 (19) -0.01
26.8 (24) 0.11
16.7 (16) 0.04
Null
0.00
0.00
0.00
0.03
0.00
0.00
0.00
0.04
0.00
HE (HO)
16.1 (11)
18.2 (15)
16.4 (11)
16.1 (10)
20.4 (20)
19.7 (20)
19.5 (19)
27.2 (24)
17.9 (17)
D1
FIS
0.32
0.18
0.34
0.39
0.02
-0.02
0.03
0.12
0.05
Null
0.12
0.06
0.14
0.16
0.00
0.00
0.00
0.05
0.04
130
Table 4.6. Maximum likelihood estimates of cline centers (c) and widths (w) for 3 nuclear loci and cytochrome b (mtDNA), estimated
independently, and multilocus cline parameters estimated using all 7 microsatellites loci (nDNA).
Locus
C (Km)
W (Km)
Log
C9
PB73
LVIR17
mtDNA
nDNA
106.75
110.18
103.77
106.87
109.72
32.94
40.38
22.27
2.70
10.73
-6.28
-2.59
-5.50
0.00
-0.47
131
a)
b)
!
L3
L1
L5
L4
!
!!
!!!!
!
!! !!
!
!!
!
!! !!! !!!
!
!!
!
!!!
!
!!
!!!
!
!
! !!!
!
!!
!
!!!!
!!
!!!
L2
0
50 100
!
!
!!
!!
!
!!!
!
!
! !
!
!!
!
!!
!
!
! !
! !!!
!!
!!
!!!
!
N
!
!!
!
!!!
9
!
! !!
!
!
!
! !
!
!
!
! !!
!
!
!!
!!
!
!
! !! ! !
!
!
!!!!
Lepida
8
!
!
?
!
!
200 Kilometres
!
!
!
! !
!
! !
!!
7
!
!!
!
?
!
! !
Nevadensis
?
6
!
!!
!
!
!
!
!
!
!!!
!
!!
!!
!
!
!
5
!!
!
!!
3
!
!
!
! !
!
!
!
!
!
4
!!
!!
!!
!
2
!
!
!!! !
33
!
!
!
!
!
!
!
!
!
!!
!
!
!
!
1
0
Locality
n
25
50 Kilometres
1
2
3
4
5
6
7
8
9
(19)
(21)
(19)
(20)
(23)
(23)
(23)
(30)
(19)
mtDNA
nDNA
Fig. 4.1. Distribution of Lacerta lepida mitochondrial lineages as in chapter 3 (a) and the study area (b). Shaded areas denote altitude gradients, with darker
areas representing higher altitudes. In b) numbers represent sampling localities along the transect (dashed line) and samples are represented by red dots. The
putative zone of secondary contact between both mitochondrial lineages is indicated by question marks (?). Yellow numbered dot represents sampling site of
chapter 3 where mitochondrial haplotypes from both phylogroups were found. Pie charts represent: mtDNA - the proportion of mtDNA haplotypes at each site
derived from Lepida (red) and Nevadensis (blue) mtDNA lineages; nDNA - the proportion of each site assigned to Lepida (red) and Nevadensis (blue) estimated
with the software STRUCTURE using 7 microsatellite loci. The number of samples (n) at each sampling site is also indicated.
132
Lepida
67 mutations
Nevadensis
Fig. 4.2. Statistical Parsimony network of cytochrome b haplotypes. Black circles represent unsampled
or extinct haplotypes. White circles represent haplotypes already sampled in the previous chapters while
coloured circles represent new haplotypes. Size of circles does not correspond to frequency.
133
Pop 1
F
0
1
Pop 2
Pop 3
Pop 4
Pop 5
Pop 6
Pop 7
Pop 8
Pop 9
Fig. 4.3. Frequency of mitochondrial DNA haplotypes (cytb gene) in each Lacerta
lepida sampled population. Each bar represents one haplotype. Haplotypes from
Nevadensis are represented in blue and from Lepida in red.
134
0.2
0.2
1
Locus B4
Locus LVIR17
N L
Locus LIZ24
N
0.1
L
0.1
0
0.3
N; L
0.5
0
L
0
L
0.2
0.2
Locus C9
Locus PB73
Locus PB66
L
L
Frequency
0.2
0.1
N
0.1
N
0.1
N
0
0
0
0.2
Locus D1
0.1
N
L
0
Fig. 4.4. Allele frequencies per locus over all Lacerta lepida sampled localities. Alleles with overall frequency less than 1% are not
represented, apart from locus LIZ24, with all alleles included. Each bar represents one allele. The most common allele in each
mitochondrial lineage is represented by a letter above the bar (N, for Nevadensis lineage and L for Lepida lineage).
135
1
1
Locus B4
Locus LVIR17
0.5
0.5
0
0
1
2
3
4
5
6
7
8
9
Locus LIZ24
1
1
2
3
4
5
6
7
8
9
6
7
8
9
6
7
8
9
1
Frequency
Locus C9
0.5
0.5
0
0
1
2
3
4
5
6
7
8
1
9
1
2
3
4
5
1
Locus PB73
Locus PB66
0.5
0.5
0
0
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
Sa mpling locality
7
8
4
5
1
Locus D1
0.5
0
1
2
3
9
Fig. 4.5. Allele frequencies per locus for each Lacerta lepida sampled locality. Alleles with frequency less than 1% are not
represented, apart from locus LIZ24, with all alleles included. Each bar represents one allele. Colours are the same as in Fig. 4.3.
136
0.4
0.4
1
Locus B4
Locus LVIR17
0.2
0.2
0
0.5
0
1 2 3 4 5 6 7 8 9 1011121314151617181920212223
0.6
Locus LIZ24
0
1 2 3 4 5 6 7 8 9 1011121314151617181920212223
1
2
3
4
5
6
7
8
0.3
0.3
Locus C9
Locus PB73
Locus PB66
0.2
0.2
0.1
0.1
0.3
0
0
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
0.3
Locus D1
0.2
0.1
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Fig. 4.6. Allele frequencies per locus for each Lacerta lepida mtDNA lineage. Red bars represent Lepida (populations 5 to 9) and
blue bars represent Nevadensis (populations 1 to 4). Alleles are ordered according to allele size.
137
a)
Fig. 4.7.
r2 = 0.0012
b)
r2 = 0.27
Isolation by distance analysis showing association between genetic and
geographic distance in two Lacerta lepida mtDNA lineages. Genetic distances
represent pairwise FST values (linearised as FST /1- FST) calculated using data from 7
microsatellite loci. Geographic distances represent distances in Km between sampled
localities within each mitochondrial lineage. a) represents Lepida mtDNA lineage and
b) represents Nevadensis mtDNA lineage.
35
30
∆K
25
20
15
10
5
0
2
3
4
5
K
6
7
8
9
Fig. 4.8. Magnitude of ∆K, as defined by Evanno et al. (2005), as a function of K.
138
Proportion of ancestry
1.00
0.80
0.60
0.40
0.20
0.00
1
2
3
4
5
6
7
Sampled individuals (columns) and sampled localities (numbers)
8
9
Fig. 4.9. Proportion of ancestry of each sampled individual of Lacerta lepida (columns) as inferred with STRUCTURE for 7 microsatellite loci,
assuming the admixture model. Dark grey represents Nevadensis ancestry whereas light grey represents Lepida. Individuals are sorted by
sampled localities.
139
b)
a)
P – proportion of membership to Lepida
1.0
mtDNA
*
*
**
**
*
*
1.0
1.0
*
LVIR17
nDNA
*
C9
PB73
*
*
0.5
0.5
P
*
*
*
*
*
*
0.0
47
mtDNA
** *
83 93 118
*
150
Distance from Locality 1 (km)
226
0
*
47
**
83 93 118
150
226
Distance from Locality 1 (km)
Fig. 4.10. Best fitted Tanh curves showing the clinal transition of mitochondrial and nuclear markers through the contact zone of two
mitochondrial DNA lineages of Lacerta lepida. a) Changes in proportion of membership (P) to Lepida mitochondrial lineage along the transect
based on 7 microsatellite loci (black stars) and changes in frequency of Lepida mitochondrial haplotypes (red stars in both graphs). b) Changes
in northern allele frequencies along the transect, for PB73 (black stars), C9 (triangles) and LVIR17 (circles).
140
4.6. References
Alphen JJMV, Seehausen O (2001) Sexual selection, reproductive isolation and the
genic view of speciation. Journal of Evolutionary Biology 14, 874-875.
Avise JC (2000) Phylogeography Harvard University Press, Cambridge, MA.
Barton NH (1983) Multilocus Clines. Evolution 37, 454-471.
Barton NH, Baird SJE (1995) Analyse: an application for analysing hybrid zones.
Available
at
www.helios.bto.ed.ac.uk/research/institutes/evolution/software/Mac/Analyse/
index.html, Edinburgh.
Barton NH, Gale KS (1993) Genetic analysis of hybrid zones. In: Hybrid zones and
the evolutionary process (ed. Harrison RG), pp. 13-45. Oxford University
Press, Oxford.
Barton NH, Hewitt GM (1983) Hybrid zones as gene barriers to gene flow. In:
Protein polymorphism: adaptive and taxonomic significance (eds. Oxford
GS, Rollinson D), pp. 341-359. Blackwell, Oxford, UK.
Barton NH, Hewitt GM (1985) Analysis of hybrid zones. Annual Review of Ecology
and Systematics 16, 113-148.
Barton NH, Hewitt GM (1989) Adaptation, speciation and hybrid zones. Nature 341,
497-503.
Barton NH, Shpak M (2000) The effects of epistasis on the structure of hybrid zones.
Genetical Research 75, 179-198.
Bateson W (1909) Mendel's Principles of Heredity Cambridge University Press,
Cambridge, Massachusetts.
Böhme MU, Berendonk TU, Schlegel M (2005) Isolation of new microsatellite loci
from the Green Lizard (Lacerta viridis viridis). Molecular Ecology Notes 5,
45-47.
141
Bohonak AJ (2002) IBD (Isolation by Distance): A Program for Analyses of
Isolation by Distance. J Hered 93, 153-154.
Boudjemadi K, Martin O, Simon J-C, Estoup A (1999) Development and crossspecies comparison of microsatellite markers in two lizard species, Lacerta
vivipara and Podarcis muralis. Molecular Ecology 8, 513-525.
Bridle JR, Ritchie MG (2001) Assortative mating and the genic view of speciation.
Journal of Evolutionary Biology 14, 878-879.
Butlin RK, Hewitt GM (1985) A hybrid zone between Chorthippus parallelus
parallelus and Chorthippus parallelus erythropus (Orthoptera: Acrididae):
morphological and electrophoretic characters. Biological Journal of the
Carrión J, Munuera M, Navarro C (1998) The palaeoenvironment of Carihuela Cave
(Granada, Spain): a reconstruction on the basis of palynological
investigations of cave sediments. Review of Palaeobotany and Palynology
99, 317-340.
Carrión JS, Dupre M (1996) Late Quaternary vegetational history at Navarres,
Eastern Spain. A two core approach. New Phytologist 134, 177-191.
Clement M, Posada D, Crandall KA (2000) TCS: a computer program to estimate
gene genealogies. Molecular Ecology 9, 1657-1659.
Coyne JA, Orr HA (2004) Speciation Sinauer, Sunderland, Massachusetts.
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete
data via the EM algorithm. Journal of the Royal Statistical Society B. 39, 138.
Dobzhansky T (1937) Genetics and the Origin of Species Columbia University
Press, New York.
Endler JA (1977) Geographic variation, speciation, and clines Princeton Universty
Press, Princeton.
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of
individuals using the software structure: a simulation study. Molecular
Ecology 14, 2611-2620.
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: An integrated software
package for population genetics data analysis. Evolutionary Bioinformatics
Online 47-50.
142
Excoffier L, Smouse PE, Quattro JM (1992) Analysis of molecular variance inferred
from metric distances among DNA haplotypes: application to Human
mitochondrial DNA restriction data. Genetics 131, 479-491.
Gavrilets S (1997) Hybrid zones with Dobzhansky-type epistatic selection. Evolution
51, 1027-1035.
analysis program for Windows95/98/NT. Nucleic Acids Symposium Series
41, 95–98.
Harrison RG (1990) Hybrid zones: windows on the evolutionary process. Oxford
Surveys in Evolutionary Biology 7, 69-128.
Harrison RG (1993) Hybrid zones and the Evolutionary Process Oxford University
Press, New York.
Hewitt GM (1988) Hybrid zones - natural laboratories for evolutionary studies.
Trends in Ecology & Evolution 3, 158-167.
Hewitt GM (1993) After the Ice: parallelus meets Erythropus in the Pyrenees. In:
Hybrid zones and the evolutionary process (ed. Harrison RG). Oxford
University Press, New York.
247-276.
Li G, Hubert S, Bucklin K, Ribes V, Hedgecock D (2003) Characterization of 79
microsatellite DNA markers in the Pacific oyster Crassostrea gigas.
Lindell J, Mendez-de la Cruz FR, Murphy RW (2005a) Deep genealogical history
without population differentiation: Discordance between mtDNA and
allozyme divergence in the zebra-tailed lizard (Callisaurus draconoides).
Lindell J, Mendez-De La Cruz FR, Murphy RW (2008a) Deep biogeographical
history and cytonuclear discordance in the black-tailed brush lizard
(Urosaurus nigricaudus) of Baja California. Biological Journal of the
Lindell J, Méndez-de la Cruz FR, Murphy RW (2005b) Deep genealogical history
without population differentiation: Discordance between mtDNA and
allozyme divergence in the zebra-tailed lizard (Callisaurus draconoides).
143
Lindell J, Méndez-de la Cruz FR, Murphy RW (2008b) Deep biogeographical
history and cytonuclear discordance in the black-tailed brush lizard
(Urosaurus nigricaudus) of Baja California. Biological Journal of the
215-229.
Mayr E (1963) Animal species and evolution. Harvard University Press, Cambridge,
MA.
Mayr E (2001) Wu's genic view of speciation. Journal of Evolutionary Biology 14,
866-867.
Muller HJ (1942) Isolating mechanisms, evolution and temperature. Biology
Symposium 6, 71-125.
Nembrini M, Oppliger A (2003) Characterization of microsatellite loci in the wall
lizard Podarcis muralis (Sauria: Lacertidae). Molecular Ecology Notes 3,
123-124.
Pinho C, Sequeira F, Godinho R, Harris DJ, Ferrand N (2004) Isolation and
characterization of nine microsatellite loci in Podarcis bocagei (Squamata:
Lacertidae). Molecular Ecology Notes 4, 286-288.
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using
multilocus genotype data. Genetics 155, 945-959.
144
Rassmann K, Tautz D, Trillmich F, Gliddon C (1997) The microevolution of the
Galapagos marine iguana Amblyrhynchus cristatus assessed by nuclear and
mitochondrial genetic analyses. Molecular Ecology 6, 437-452.
Raymond M, Rousset F (1995) An Exact Test for Population Differentiation.
Evolution 49, 1280-1283.
Rieseberg LH, Burke JM (2001) A genic view of species integration. Journal of
Evolutionary Biology 14, 883-886.
Rousset F (2008) Genepop'007: a complete re-implementation of the genepop
software for Windows and Linux. Molecular Ecology Resources 8, 103-106.
Salvador A, Veiga JP, Esteban M (2004) Preliminary data on reproductive ecology
of Lacerta lepida at a mountain site i central Spain. Herpetological Journal
14, 47-49.
Stenson AG, Malhotra A, Thorpe RS (2002) Population differentiation and nuclear
gene flow in the Dominican anole (Anolis oculatus). Molecular Ecology 11,
1679-1688.
Szymura JM, Barton NH (1986) Genetic analysis of a hybrid zone between the firebellied toads, Bombina bombina and B. variegata, near Cracow in Southern
Poland. Evolution 40, 1141-1159.
associations with haplotypes inferred from restriction endonuclease mapping
and DNA sequence data. III. Cladogram estimation. Genetics 132, 619-633.
Thorpe RS, Surget-Groba Y, Johansson H (2008a) The relative importance of
ecology and geographic isolation for speciation in anoles. Philosophical
Transactions of the Royal Society B-Biological Sciences 363, 3071-3081.
Thorpe RS, Surget-Groba Y, Johansson H (2008b) The relative importance of
ecology and geographic isolation for speciation in anoles. Philosophical
Transactions of the Royal Society B: Biological Sciences 363, 3071-3081.
Ujvari B, Dowton M, Madsen T (2008) Population genetic structure, gene flow and
sex-biased dispersal in frillneck lizards (Chlamydosaurus kingii). Molecular
Ecology 17, 3557-3564.
Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) Micro-checker:
software for identifying and correcting genotyping errors in microsatellite
data. Molecular Ecology Notes 4, 535-538.
145
Vogler AP (2001) The genic view: a useful model of the process of speciation?
Journal of Evolutionary Biology 14, 876-877.
Wu CI (2001) The genic view of the process of speciation. Journal of Evolutionary
Biology 14, 851-865.
146
Appendix
147
Chapter 5
Testing for the presence of heteroplasmy in Lacerta
lepida through single molecule PCR
Photo by Andreia Miraldo
Photo taken to mark the capture of lizard number 44*
* Sampling site E in chapter 2
5.
Testing
heteroplasmy
for
in
the
presence
Lacerta
of
lepida
through single molecule PCR
5.1. Abstract
In the last decade general assumptions regarding the inheritance of
mitochondrial DNA in animals have been questioned mainly as a result of
accumulating evidence for the existence of bi-parental inheritance and recombination
of mitochondrial DNA across several taxa.
In this chapter, polymorphic
mitochondrial DNA sequences detected in several individuals from a zone of contact
between two Lacerta lepida mitochondrial lineages (chapter 2) are re-analysed using
a single molecule PCR approach to test if heteroplasmy and mitochondrial DNA
recombination are features of this contact zone. Results indicate that low levels of
heteroplasmy occur in some individuals. Strong evidence for mitochondrial DNA
recombination was also detected. The origins of heteroplasmy and mitochondrial
DNA recombinant haplotypes within Lacerta lepida are discussed in detail.
Key words: heteroplasmy, mitochondrial DNA, recombination, smPCR,
contact zone
148
5.2. Introduction
Mitochondrial DNA has been the most employed molecular marker for
phylogeographic inference in animals (Avise, 2004). It has become the tool of choice
in phylogeographic studies due to several of its properties, not found in nuclear
genomes. It has a maternal, non-recombining mode of inheritance that enables
evolutionary histories to be reconstructed without the complexities introduced by
biparental recombination, and it has a high mutation rate, that generates enough
signal to make inferences about population history over short time frames.
Nevertheless some of these assumptions have been questioned in the last few years.
For instance, although the standard paradigm postulates that mtDNA is strictly
maternally inherited, it has been increasingly apparent that more than one mtDNA
type can be associated with an individual or cell, a condition known as heteroplasmy.
Recent findings have shown that in organisms that normally transmit mtDNA
through the female line only, heteroplasmy can be produced by occasional paternal
leakage that is usually associated with interspecific crosses (Arunkumar et al., 2006;
Ciborowski et al., 2007; Fontaine et al., 2007; Sherengul et al., 2006). Although
intraspecific paternal leakage is thought to be less probable, as the recognition
mechanisms of paternal mitochondria are more efficient when genetic divergence
between individuals is small (Kaneda et al., 1995; Shitara et al., 1998; Sutovsky et
al., 2000), recent studies have reported the occurrence of paternal leakage within
species (Gantenbein et al., 2005; Sherengul et al., 2006; Ujvari et al., 2007).
The discovery of heteroplasmy through paternal leakage in a wide range of
taxa raises questions about the consistency of the other important property of
mtDNA, the absence of recombination in this molecule. Recombination in mtDNA
is thought to be absent in animals, mainly because of a failure to observe clear cases
of recombinant haplotypes in natural populations. However, whether mtDNA
recombination occurs is a different issue from whether it produces new haplotypes.
In fact, the lack of recombination in animal mtDNA is becoming very controversial
149
as evidence accumulates suggesting that both intra and intermolecular recombination
may occur (for a review in inheritance and recombination of mitochondrial genomes
in other systems see Barr et al., 2005). There is evidence that human mitochondria
have the required enzymatic machinery for homologous recombination to occur
(Thyagarajan et al., 1996; Yaffe, 1999). Moreover, mitochondria are extremely
dynamic organelles that are constantly fusing and dividing, and it has been shown
that after fusion matrix contents are mixed providing the possibility for homologous
recombination to occur (see Detmer and Chan, 2007 for a recent review on the
subject).
Furthermore,
intramolecular
mtDNA
recombination
has
been
experimentally demonstrated in the nematode Meloidogyne javanica (Lunt and
Hyman, 1997) providing evidence that animal mtDNA can self-recombine. Further
indirect evidence for intramolecular recombination comes from the detection of
mitochondrial rearrangements (“sublimons”) found at very low levels in healthy
human tissues that are suggested to be the result of homologous recombination (Holt
et al., 1997; Kajander et al., 2000; Tang et al., 2000). Morevover, intermolecular
recombination occurs frequently in mussels (Burzynski et al., 2006; Ladoukakis and
Zouros, 2001), which are known to have a unique mitochondrial inheritance system
(“doubly uniparental inheritance”), and incidentally in humans (Kraytsberg et al.,
2004b). More recently, strong evidence for the occurrence of intermolecular
recombination has been found in fish (Ciborowski et al., 2007) and reptiles (Ujvari
et al., 2007). The conclusion that animal mitochondrial DNA does not recombine
based on the absence of recombinant haplotypes in natural populations does not
consider the probability of a mtDNA recombination event producing a detectable
recombinant haplotype. This probability is likely to be very small, mainly as a
consequence of the typically strict maternal inheritance of mtDNA. In the case of a
recombination event occurring, it will most likely occur in homoplasmic cells,
making the detection of recombinants very difficult, if not impossible, unless it
results in size heteroplasmy (reviewed in Rokas et al., 2003). Interestingly almost all
reported cases of mitochondrial recombination were detected because recombination
occurred between divergent mtDNA co-occurring in the same cell. An exception is
the recombination events registered in the nematode Meloidogyne javanica, where
intramolecular mtDNA recombination was detected as it resulted in size
heteroplasmy and variability in sequence organization (Lunt and Hyman, 1997). The
detection of mtDNA recombination seems to require an initial state of heteroplasmy
150
that can only be achieved either by paternal leakage or by mutations in the
mitochondrial genome of germ-line cells. Given the increasing evidence for paternal
leakage in both inter and intraspecific crosses (see above), it would seem that hybrid
zones may be likely areas to detect mtDNA recombination events. This idea is
further supported by two recent studies which report evidence for recombination of
mtDNA in contact zones: the study of Jaramillo-Correa and Bousquet (2005) reports
recombination of mtDNA in a zone of contact between two hybridizing conifers
while Ujvari and collaborators (2007) report evidence of mitochondrial
recombination in an hybrid zone between two mitochondrial lineages of the
Australian frillneck lizard (Chlamydosaurus kingii).
The phylogeographic study of a zone of secondary contact between two
divergent mitochondrial lineages (L3 and L5) of Lacerta lepida revealed the
existence of clear polymorphic trace files when sequencing the cytochrome b gene
for single individuals (chapter 2). Although it was shown that the mixed signal was
most probably generated by the presence of Numts, the existence of low levels of
heteroplasmy within the species could not be discarded completely. Furthermore, the
quantification of intra-individual variation in chapter 2 was achieved through a PCRcloning procedure, which is known to have some inherent and significant
disadvantages. The disadvantages are mainly associated with the amplification step
where, PCR derived mutations, template jumping and allelic preference are known to
occur (Lin et al., 2002; Paabo et al., 1990). These disadvantages become especially
problematic when PCR-cloning procedures are used to describe mutations that
distinguish different gene copies. While PCR induced errors (in vitro errors) will not
be detectable upon sequencing, as at the most they will affect 25% of all molecules
synthesized, upon cloning in vitro polymerase errors will become indistinguishable
from in vivo mutations since each of the errors will affect all the molecules (100%)
of a clone, just as a genuine in vivo mutation does. Results from chapter 2 show that
35% of the sequenced clones represent probable recombinants between the two
divergent mitochondrial lineages. Such recombinant molecules could have originated
either due to rearrangements from mixed templates, via jumping PCR, or through
intramolecular recombination. Although the data suggests that recombinants could
in fact have originated in vitro, the question that still remains is if the recombinant
molecules originated through recombination between two divergent mitochondrial
151
molecules (true heteroplasmy) or between a homoplasmic mitochondrial genome and
Numts.
In this chapter these issues will be further analyzed using a single molecule
PCR (smPCR) approach. SmPCR has been used in several types of study, being
most commonly employed for sequencing and genotyping purposes (e.g. Konfortov
et al., 2007; Krause et al., 2006; Kraytsberg and Khrapko, 2005; Lukyanov et al.,
1996). A single molecule PCR is essentially a normal PCR but where the template
DNA is diluted to very low concentration. If Numts are the only source of the
heteroplasmic signal, and are indeed the origin of the recombinant molecules found
in chapter 3, performing a smPCR by limiting DNA dilution to one amplifiable
mitochondrial genome should only result in the amplification of mitochondrial
fragments. Therefore using this approach it should be possible to identify if true
heteroplasmy and recombination occur in Lacerta lepida.
5.3. Material and methods
5.3.1. Sample selection and DNA extraction
SmPCR was performed using DNA of four individuals from a population at
the centre of the contact zone between lineages L3 and L5, where haplotypes from
both lineages were detected (individuals C3, C4, C8 and C9 from population C,
chapter 2). Previous amplification of 627 bp fragment of cytb gene in these samples
revealed the existence of clear polymorphic trace files and cloning of the amplified
fragments resulted in the detection of several recombinant molecules (chapter 3).
DNA extraction, amplification of cytb fragment and cloning procedures are
explained in detail in chapter 3.
5.3.2. Estimation of the number of template copies
152
DNA concentration was measured using a NanoDrop® ND-1000
spectrophotometer and was diluted to a concentration of approximately 103
mitochondria /µl. Assuming that 1 million base pairs weighs 1pg (1x10-3ng), 1
mitochondrial genome of approximately 17,000 base pairs weighs 1.7x10-3ng. Fifty
microtitre aliquots (single use aliquots) at a concentration of 103 mitochondria/µl
(hereafter described as stock DNA) were stored at -20oC. Stock DNA was serially
diluted and dispensed in a 96-well microtitre plate in order to obtain 5 different DNA
concentrations (16 wells per concentration and 16 negative controls). In order to
assess which DNA concentration conforms to expectations from smPCR, DNA was
subjected to the hemi nested smPCR protocol (see below) and the proportion of
positive wells for each DNA concentration was estimated. According to a Poisson
distribution, 36.8% of the amplifications from DNA at a concentration of a single
molecule are not expected to contain a molecule of the desired template, another
36.8% are expected to contain a single molecule, and the remainder are expected to
contain multiple molecules (Stephens et al., 1990). In order to decrease the number
of false positives (positive amplifications resulting from multiple molecules),
template with a final concentration of 0.3 amplifiable molecules should be used,
meaning that approximately one third of the amplifications will yield a product
derived from a single molecule template and less than 5% of positive amplifications
will be derived from multiple molecules (Kraytsberg et al., 2004a). Therefore, only
amplifications derived from DNA dilutions that resulted in a proportion of positive
amplifications conforming to a DNA concentration of 0.3 amplifiable molecules are
sequenced.
5.3.3. Selection of loci and design of PCR primers
In order to increase PCR specificity a hemi nested PCR approach was used,
which involves the use of two sets of primers employed in two successive PCR
reactions (hereafter described as Phase I and Phase II PCR). During Phase I PCR the
first set of primers, that includes the forward-external (FEXT) and the reverse primer
(REXT), are used to generate a DNA product that is longer than the final target
sequence (Fig. 5.1.). The product from Phase I PCR is then used to start a second
153
PCR (Phase II) using a set of primers that involves the same reverse primer used in
Phase I and a new forward primer (forward-internal, FINT). The forward-internal
primer binding site is located within the first amplified sequence, in a region nearby
the forward-external primer binding site (Fig. 5.1.).
To assure that the low numbers of positive amplifications (which are
expected in a smPCR approach) are the result of single molecule amplifications and
not due to PCR inefficiency, two fragments (amplimers) from different regions of
the mitochondrial genome were amplified. Co-segregation of the two markers would
re-assure PCR effectiveness; therefore if an aliquot gives a positive amplification for
one of the markers, then the same aliquot should give a positive amplification for the
other marker. If the two markers segregate independently then low number of
amplifications might be due to PCR inefficiency. The markers chosen were a
fragment of the cytochrome b gene (CYTB amplimer) and a fragment that spans part
of the 12S and 16S ribosomal genes (12S amplimer) (Fig. 5.2.).
To design specific Lacerta lepida primers to amplify the CYTB amplimer the
entire cytb gene was first amplified with modified versions of primer L14919
(TRNAGLU, 5’- AAC CAC CGT TGT ATT TCA ACT - 3’) and L16064
(TRNATHR, 5’- CTT TGG TTT ACA AGA ACA ATG CTT TA - 3’) (Burbrink et
al., 2000) using the conditions described in chapter 3. After aligning the entire cytb
sequences specific primers for Lacerta lepida were designed (CYTB-FEXT, 5’-TTA
CAA AAT TAT TAA CTC CTC CT - 3’; CYTB-FINT, 5’ - GCC TAT GTC TTA
TTA TTC AAG - 3’ and CYTB-REXT, 5’ - GGT TTA CAA GAA CAA TGC TTT
A - 3’).
The final CYTB amplimer is 1143 bp long. For the 12S amplimer,
published (Paulo et al., 2008) partial sequences of 12S and 16S genes from Lacerta
lepida were aligned and specific primers were designed in order to amplify a
fragment of similar size to that of the CYTB amplimer. As there are no published
sequences for the entire 12S and 16S genes from Lacerta lepida the published
mitochondrial genome of Lacerta viridis (Böhme et al., 2007) was used to estimate
the approximate final size of the amplimer. The primers designed for the
amplification of the 12S amplimer were: 12S-FEXT (5’ - GCA AAT GTT AGG
GAA GAG AT - 3’), 12S-FINT (5’ - CTA TTT TAA CAA CGC TCT GGG - 3’)
and 16S-REXT (5’ - GAG TCA CTG GGC AGG CAA GA - 3’). The final 12S
amplimer is 933 bp long.
154
5.3.4. PCR amplifications, scoring and sequencing
Phase I PCR consisted of the amplification of both markers using a multiplex
approach in a mix containing 1x PCR Gold buffer (Perkin-Elmer), 4 mM MgCl2,
200 mM each of dATP, dCTP, dGTP and dTTP (Applied Biosystems), 0.2 mM of
forward-external and reverse external primers (Operon Technologies) for both
markers (see above primer selection for each loci) and 0.5U Taq Gold DNA
polymerase (Perkin-Elmer). Five microlitres of this mix were added to 5µl of
template DNA which was previously prepared by performing serial dilutions from
stock DNA and dispensed in a 96 well plate under 1 drop of oil (see section 6.3.2.).
Negative controls (no DNA) were included for all amplifications. In order to avoid
contamination, Phase I PCR was prepared in a room separated from any PCR
products and DNA solutions at high concentration. Amplifications were conducted
in the normal lab as follows: 93oC for 9min, then 28 cycles of 94oC for 20s, 50oC for
30s, 72oC for 90s. Phase I PCR products were diluted to 1000µl with bi-distilled
water (ddH20) and 5µl aliquots of it were used to perform the Phase II PCR. Phase II
PCR consists of amplifying the two markers independently in two monoplex PCRs.
The 5µl aliquots of diluted Phase I PCR products are supplemented to give a 10µl
final volume containing 1 mM each of the relevant forward-internal and reverse
primers, 1x PCR Gold buffer, 4 mM MgCl2, 200 mM each dNTP and 0.2U Taq
Gold DNA polymerase. Amplifications were conducted as follows: 93oC for 9min,
then 33 cycles of 94oC for 20 s, 54oC for 30 s, 72oC for 90 s.
PCR products were analyzed by agarose gel electrophoresis (2%), scoring
presence and absence of the expected PCR product. Positive products that could be
identified as obtained from a single molecule were purified by filtration through
QIAquick® columns (Qiagen) following manufacturer’s recommendations and the
CYTB amplimer was sequenced in both directions using Phase II PCR primers.
Sequencing reaction mixes consisted of 6.35µl of ddH2O, 1.5µl of primer at 3.5µM,
1µl of BigDye Terminator v3.1TM (Applied Biosystems) and 1µl of PCR product.
Sequence reactions were performed as follows: initial incubation at 96ºC for 1min;
25 cycles of incubation at 90ºC for 10s, 50ºC for 5s and 60ºC for 4min. PCR and
155
sequencing reactions were performed in a DNA engine tetrad 2, Peltier
thermocycler, and sequences were obtained using an ABI 3700 capillary sequencer.
5.4. Results and discussion
SmPCR was successfully implemented in all samples. The percentage of
positive amplifications was always lower (between 0.29 and 0.43, Table 5.1.) than
what was expected from a single molecule PCR protocol. Both amplimers were
successfully amplified supporting the efficacy of the protocol and no contamination
was detected as revealed by negative amplification of controls.
To infer the phylogenetic relatedness of all smPCR sequences a statistical
parsimony network (see chapter 3, section 3.3.3. for a detailed explanation of the
method) using all 68 mitochondrial haplotypes from chapter 2 and the 55 sequences
obtained by smPCR was constructed (Fig. 5.3.). Forty eight smPCR sequences
correspond to haplotype 40, six sequences correspond to haplotype 62, one sequence
to haplotype 1 and one sequence corresponds to a new haplotype (153) not found
before. All smPCR sequences from individuals C3 and C4 correspond to the
expected mitochondrial DNA haplotype (haplotype 40) previously identified by the
amplification of the entire cytb gene (chapter 3). This was not the case for
individuals C8 and C9. Although the majority of smPCR sequences in individuals
C8 and C9 also correspond to the expected mitochondrial DNA haplotype (haplotype
40 in C8 and 62 in C9), one sequence in each individual corresponds to a different
haplotype. Individual C8 carries haplotypes 40 and 1 while individual C9 carries
haplotypes 62 and 153.
5.4.1. Ruling out Numts
It is highly unlikely that the sequences achieved through smPCR represent
Numts instead of real mitochondrial copies, due to the level of DNA dilution that
each sample was subjected to prior to the amplification process. The DNA in each
156
sample was diluted to a level that only allows fragments of DNA smaller than 17.000
bp to be present in each aliquot, excluding therefore the nuclear genome. Although
the single molecularity of smPCR protocol refers to “a single amplifiable molecule,
which means a continuous DNA with no impassable adducts/modifications”
(Kraytsberg et al., 2004a), additional broken, “non-amplifiable” molecules might be
present in the DNA mixture. Nevertheless, it is highly improbable that through
dilution an aliquot with a fragment of nuclear DNA representing exactly the portion
under analysis (cytb Numt) would be obtained. It is therefore more likely that the
sequences obtained by smPCR do in fact represent true mitochondrial copies that
exist in heteroplasmy in the individuals analysed.
5.4.2. Ruling out contamination
SmPCR is highly prone to contamination and therefore during the smPCR
preparation several measures were carried out to avoid it. The smPCR step most
prone to contamination is the preparation of Phase I PCR, as it is during this step that
DNA is highly diluted and therefore more prone to contamination. All smPCR Phase
I preparations were done in a separated lab (clean lab), which is located in a different
building from the main lab. Furthermore the clean lab was never exposed to high
DNA concentrations, PCR products or any work involving reptile DNA, therefore
substantially reducing the possibility of contamination. In order for contamination to
be the source of heteroplasmy detected, DNA representative of haplotype I or the
new haplotype 153 would have to be carried to the clean lab, through the handling of
reagents and materials or by contamination of the diluted DNA previously to its
transfer to the clean lab. All reagents and materials used in the clean lab were
specifically bought for this purpose and were never in contact with the main lab
where contamination could occur. DNA carried to the clean lab was DNA diluted
from the 4 individuals analysed which do not correspond to haplotype 1 neither to
the new haplotype (153) detected. Therefore no direct sources of contamination are
present in the clean lab. Furthermore, each Phase I PCR preparation was done under
a UV hood, which after the assemblage of Phase I PCR plate, was turned on so that
any DNA present in the hood was eliminated, thus avoiding cross contamination.
157
The best evidence against contamination in the clean lab is that no amplifications
were ever detected in the negative controls, which always represented 16.6% of the
wells of each plate. These facts suggest that if contamination was the source of
heteroplasmy then DNA would have to be contaminated previously to the phase I
PCR set up, either during DNA extraction or dilution in the main lab. This
explanation is only applicable regarding haplotype 1 as haplotype 153 was never
amplified before. Nevertheless, DNA carried to the clean lab was already extremely
diluted and therefore if DNA was already contaminated higher frequency of
amplifications of haplotype 1 would be expected, close to the frequency of the other
amplified haplotype. The very low frequency of one of the haplotypes in both
individuals is therefore more consistent as representing true low levels of
heteroplasmy.
5.4.3. Heteroplasmy and mtDNA recombination
The co-occurrence of mitochondrial DNA haplotypes 1 and 40 in individual
C8, and haplotypes 153 and 62 in C9 confirms the existence of low levels of
heteroplasmy in these individuals. An increasing number of species have been
shown to harbour some level of heteroplasmy (see section 6.2) and Lacerta lepida
seems to be no exception.
Haplotype 153 is connected in the network to haplotype 40 by 4 mutations
(Fig. 5.3., branch a) and to an ancestral unsampled haplotype * by 3 mutations (Fig.
5.3., branch b). The mutations involved in branch a and b are the same 7 mutations
that occur from the ancestral unsampled haplotype * to haplotype 40, resulting in the
loop. The phylogenetic relationship of haplotype 153 with the remaining haplotypes
seems to suggest that it resulted from a recombination event between two divergent
mitochondrial haplotypes from the network. Another explanation for the origin of
haplotype 153 is the occurrence of homoplasies. This would imply that all sites
involved either in branch a or in branch b have suffered re-current mutations, which
would seem less likely. The origin of the recombinant haplotype153 most likely
derived from a recombination event via paternal leakage resulting in the fusion of
paternal and maternal mitochondrial DNA. Recombination through paternal leakage
158
has also been inferred as the most likely explanation for the recombinants detected in
the contact zone of two conifers, black spruce (Picea mariana) and red spruce (Picea
rubens) (Jaramillo-Correa and Bousquet, 2005) and in the contact zone between two
mitochondrial lineages of the Australian frillneck lizard (Chlamydosaurus kingii)
(Ujvari et al., 2007).
5.4.4. Origin of heteroplasmy and recombination in Lacerta
lepida
In animals heteroplasmy can be achieved through the accumulation of
somatic mutations (e.g. Khrapko et al., 1997), paternal leakage (e.g. Fontaine et al.,
2007) or through intramolecular recombination (e.g. Kajander et al., 2000; Lunt and
Hyman, 1997). In the case of Lacerta lepida the heteroplasmy documented in
individuals C8 and C9 is most consistent with paternal leakage. In both individuals
the differences between the heteroplasmic haplotypes is too large to be explained by
the accumulation of somatic mutations within an individual (11 mutational steps
between haplotypes present in C8 and 9 in C9). Moreover, it is very unlikely that the
mutations accumulated would result in a previously sampled haplotype, as it is the
case of haplotype 1 in individual C8. In fact most cases of heteroplasmy reported to
date in animals represent heteroplasmy originated through paternal leakage and it has
been reported in birds (Kvist et al., 2003), insects (Fontaine et al., 2007; Kondo et
al., 1990; Meusel and Moritz, 1993; Sherengul et al., 2006; Van Leeuwen et al.,
2008), fish (Hoarau et al., 2002; Magoulas and Zouros, 1993) and mammals
(Gyllensten et al., 1991; Kaneda et al., 1995; Shitara et al., 1998; Steinborn et al.,
1998; Sutovsky et al., 2000; Zhao et al., 2004) including humans (Kraytsberg et al.,
2004b; Schwartz and Vissing, 2002).
It is not possible to determine if leakage responsible for the detected
heteroplasmy and recombination occurred from the father of the heteroplasmic
individuals at the time of fertilization or if it occurred several generations ago and
was transmitted to the individuals from the maternal line. The latter hypothesis
would imply that heteroplasmy persisted in the population for a long period of time.
Heteroplasmy can be resolved within one or few generations through a reduction of
159
mtDNA copies during early oogenesis, as firstly reported in bovines (Ashley et al.,
1989; Hauswirth and Laipis, 1982; Koehler et al., 1991). Nevertheless, the reestablishment of homoplasmy seems to differ amongst taxa and it is influenced by
the type of mutations involved in the heteroplasmy and, in the case of neutral
polymorphisms, on the effective population size. For example, reports show that in
mice neutral heteroplasmy can persist for as long as 14 generations (Gyllensten et
al., 1991) while in insects the number of generations to re-establish homoplasmy
might reach 500 (Rand and Harrison, 1986; Solignac et al., 1984).
In mammals it is known that mitochondrial genotypes segregate differently in
the offspring due to a mitochondrial bottleneck and random segregation of organelles
into early embryonic cells, which is seen as a tool to prevent the accumulation of
deleterious mutations and “mutational meltdown” that would otherwise occur via
Muller’s ratchet (Bergstrom and Pritchard, 1998). This decrease in mtDNA per cell
during embryogenesis is followed by a dramatic increase during oogenesis, which
means that only a sub-set of maternal mtDNA will populate the next generation. This
can lead to a return to homoplasmy from a heteroplasmic state but it can also lead to
strong founder effects (Bergstrom and Pritchard, 1998).
So if heteroplasmy in
Lacerta lepida was generated in the past and has persisted in the population through
several generations we should expect to detect haplotype 1 and 153 at higher
frequencies in the sampled area due to random segregation of mtDNA, which is not
the case. If heteroplasmy in Lacerta lepida persists across multiple generations, this
would imply that some form of selection is maintaining haplotype 1 and 153 at low
frequencies in the population. Recently, evidence for strong purifying selection has
been found in heteroplasmic mice (Stewart et al., 2008), although this is still very
controversial. It seems, therefore, more plausible that heteroplasmy is recent,
resulting from hybridization of diverged mitochondrial phylogroups. The occurrence
of haplotype 1 and 153 in the surrounding areas of phylogroup L3 cannot be
completely excluded, and thorough sampling could reveal their presence nearby,
allowing the occurrence of paternal leakage through hybridization.
5.5. Conclusion
160
In this study several important issues regarding the inheritance of
mitochondrial DNA in Lacerta lepida were disclosed. It was shown that paternal
leakage occurs in this species originating low frequency heteroplasmy. Furthermore
evidence for recombination of the mitochondrial genome of Lacerta lepida was also
detected. Screening of more individuals using smPCR is likely to increase the
number of heteroplasmic and recombinant molecules and therefore allow for a better
understanding of these phenomena in Lacerta lepida.
Despite the widespread
occurrence of heteroplasmy reported in the literature and the incidental cases of
mitochondrial DNA recombination this is the first case that both phenomena are
reported to occur in the same species and in a natural population. Therefore, Lacerta
lepida seems to be an excellent system to further investigate issues related to
mitochondrial DNA heteroplasmy and recombination.
161
Template DNA
REXT
FEXT
Phase I PCR
Phase I PCR product and Phase II PCR template
FINT
REXT
Phase II PCR
Final amplimer
Fig. 5.1. Schematic representation of smPCR nested design.
Bp 2540
12S
943bp
Fext_12S
(starts at bp702)
65bp
tRNA-val
Bp 0
16S
1532bp
Fint_12S
(starts at bp811)
Rext_16S
(starts at bp1873)
1062 bp
Bp 1143
Bp 0
CytB
1143 bp
Fint_CytB
(starts at bp116)
Rext_CytB
(starts at bp1049)
Fext_CytB
(starts at bp33)
933 bp
Lacerta lepida sequences available at Genebank
Fragment to be amplified by smPCR (amplimers)
Fig. 5.2. Schematic representation of two amplimers to be amplified by smPCR and
the position of the primers used in the nested PCR. Base pair numbers in the first
scheme are set according to Lacerta viridis publised 12S and 16S genes, and in the
second scheme are set according to Lacerta lepida cytb gene.
162
51
50
41
43
45
42
40
46
40
58
351
57
510
264
55
59
46
56
60
48
21
52
b
46
62
391
54
61
47
b
510
53
49
L3
72
52
391
33
72
33
153
351 264
21 *
*
153
a
a
66
L1
63
64
65
7
5
67
9
2
6
68
1
3
8
4
L4
11
12
10
13
15
14
16
20
18
24
23
17
21
22
19
38
26
39
smPCR haplotypes
27
25
28
37
29
30
Extinct/unsampled haplotypes
32
36
31
33
35
34
L5
Fig. 5.3. Statistical parsimony network of Lacerta lepida cytochrome b haplotypes.
Dashed lines represent ambiguities in the network. White circles with no numbers
represent unsampled or extinct haplotypes and yellow circles represent haplotypes
detected by smPCR. Grey shaded area shows the loop that connects haplotype 153 to
the network. The mutations involved in that loop are shown to the left of the
network, where a and b represent alternative branches to connect haplotype 153 to
the network.
163
Table 5.1. Number of smPCR amplifications performed in Lacerta lepida samples
(Total), with scoring of positive (+) and negative (-) amplifications and respective
percentage of positive amplifications (%). The absolute frequency of each
mitochondrial haplotype detected in each sample is also shown.
Sample
code
C8
C9
mtDNA
Haplotype
H40
H62
C4
C3
H40
H40
13
6
22
15
35
21
0.37
0.29
smPCR Haplotypes
(frequency)
H40 (12), H1 (1)
H62 (6), H153 (1)
15
21
20
42
35
63
0.43
0.33
H40 (15)
H40 (21)
(+) (-) Total % (+)
164
5.6. References
Arunkumar KP, Metta M, Nagaraju J (2006) Molecular phylogeny of silkmoths
reveals the origin of domesticated silkmoth, Bombyx mori from Chinese
Bombyx mandarina and paternal inheritance of Antheraea proylei
mitochondrial DNA. Molecular Phylogenetics and Evolution 40, 419-427.
Ashley MV, Laipis PJ, Hauswirth WW (1989) Rapid segregation of heteroplasmic
bovine mitodiondria. Nucl. Acids Res. 17, 7325-7331.
Barr CM, Neiman M, Taylor DR (2005) Inheritance and recombination of
mitochondrial genomes in plants, fungi and animals. New Phytologist 168,
39-50.
Bergstrom CT, Pritchard J (1998) Germline bottlenecks and the evolutionary
maintenance of mitochondrial genomes. Genetics 149, 2135-2146.
Böhme MU, Fritzsch G, Tippmann A, Schlegel M, Berendonk TU (2007) The
complete mitochondrial genome of the Green Lizard Lacerta viridis viridis
(Reptilia: Lacertidae) and its phylogenetic position within squamate reptiles.
Gene 394, 69-77.
Burbrink FT, Lawson R, Slowinski JB (2000) Mitochondrial DNA Phylogeography
of the Polytypic North American Rat Snake (Elaphe obsoleta): A Critique of
the Subspecies Concept. Evolution 54, 2107-2118.
Burzynski A, Zbawicka M, Skibinski DOF, Wenne R (2006) Doubly uniparental
inheritance is associated with high polymorphism for rearranged and
recombinant control region haplotypes in Baltic Mytilus trossulus. Genetics
174, 1081-1094.
Ciborowski KL, Consuegra S, Garcia de Leijniz C, Beaumont MA, Wang J, Jordan
WC (2007) Rare and fleeting: an example of interspecific recombination in
animal mitochondrial DNA. Biology Letters 3, 554-557.
Detmer SA, Chan DC (2007) Functions and dysfunctions of mitochondrial
dynamics. Nature 8, 870-879.
165
Gantenbein B, Fet V, Gantenbein-Ritter IA, Balloux Fo (2005) Evidence for
recombination in scorpion mitochondrial DNA (Scorpiones: Buthidae).
Hauswirth WW, Laipis PJ (1982) Mitochondrial DNA polymorphims in a maternal
lineage of Holstein cows. Proc Natl Acad Sci USA 79, 4686-4690.
Hoarau G, Holla S, Lescasse R, Stam WT, Olsen JL (2002) Heteroplasmy and
Evidence for Recombination in the Mitochondrial Control Region of the
Flatfish Platichthys flesus. Molecular Biology and Evolution 19, 2261-2264.
Holt IJ, Dunbar DR, Jacobs HT (1997) Behaviour of a population of partially
duplicated mitochondrial DNA molecules in cell culture: segregation,
maintenance and recombination dependent upon nuclear background. Human
Molecular Genetics 6, 1251-1260.
Jaramillo-Correa JP, Bousquet J (2005) Mitochondrial genome recombination in the
zone of contact between two hybridizing conifers. Genetics 171, 1951-1962.
Kajander OA, Rovio AT, Majamaa K, Poulton J, Spelbrink JN, Holt IJ, Karhunen
PJ, Jacobs HT (2000) Human mtDNA sublimons resemble rearranged
mitochondrial genomes found in pathological states. Hum. Mol. Genet. %R
10.1093/hmg/9.19.2821 9, 2821-2835.
Kaneda H, Hayashi J, Takahama S, Taya C, Lindahl K, Yonekawa H (1995)
Elimination of paternal mitochondrial DNA in intraspecific crosses during
early mouse embryogenesis. Proceedings of the National Academy of
Sciences 92, 4542-4546.
Khrapko K, Coller HA, André PC, Li X-C, Hanekamp JS, Thilly WG (1997)
Mitochondrial mutational spectra in human cells and tissues. Proc Natl Acad
Sci USA 94, 13798-13803.
Koehler CM, Lindberg GL, Brown DR, Beitz DC, Freeman AE, Mayfield JE, Myers
AM (1991) Replacement of bovine mitochondrial DNA by a sequence
variant within one generation. Genetics 129, 247-255.
Genetics 126, 657-663.
Konfortov BA, Bankier AT, Dear PH (2007) An efficient method for multi-locus
molecular haplotyping. Nucl. Acids Res. 35, e6-.
Krause J, Dear PH, Pollack JL, Slatkin M, Spriggs H, Barnes I, Lister AM,
Ebersberger I, Paabo S, Hofreiter M (2006) Multiplex amplification of the
166
mammoth mitochondrial genome and the evolution of Elephantidae. Nature
439, 724-727.
Kraytsberg Y, Khrapko K (2005) Single-molecule PCR: an artifact-free PCR
approach for the analysis of somatic mutations. Expert Review of Molecular
Diagnostics 5, 809-815.
Kraytsberg Y, Nekhaeva E, Chang C, Ebralidse K, Khrapko K (2004a) Analysis of
somatic mutations via long-distance single molecule PCR. In: DNA
amplification. Current technologies and applications (eds. Deminov VV,
Broude NE), p. 335. Horizon bioscience, Wymondham.
Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS, Clayton DA,
Vissing J, Khrapko K (2004b) Recombination of human mitochondrial DNA.
Science 304, 981-981.
Kvist L, Martens J, Nazarenko AA, Orell M (2003) Paternal leakage of
mitochondrial DNA in the great tit (Parus major). Molecular Biology and
Ladoukakis ED, Zouros E (2001) Direct evidence for homologous recombination in
mussel (Mytilus galloprovincialis) mitochondrial DNA. Molecular Biology
Lin MT, Simon DK, Ahn CH, Kim LM, Beal MF (2002) High aggregate burden of
somatic mtDNA point mutations in aging and Alzheimer's disease brain.
Human Molecular Genetics 11, 133-145.
Lukyanov KA, Matz MV, Bogdanova EA, Gurskaya NG, Lukyanov SA (1996)
Molecule by molecule PCR amplification of complex DNA mixtures for
direct sequencing: an approach to in vitro cloning. Nucleic Acids Research
24, 2194-2195.
Lunt DH, Hyman BC (1997) Animal mitochondrial DNA recombination. Nature
Genetics, 247.
encrasicolus) indicates incidental biparental inheritance of mitochondrial
DNA. Molecular Biology and Evolution 10, 319-325.
fertilization of honeybee (Apis mellifera L.) eggs. Current Genetics 24, 539543.
Paabo S, Irwin DM, Wilson AC (1990) DNA damage promotes jumping between
templates during enzymatic amplification. Journal of Biological Chemistry
265, 4718-4721.
167
Rand DM, Harrison RG (1986) Mitochondrial DNA transmission in crickets.
Genetics 114, 955-970.
Rokas A, Ladoukakis E, Zouros E (2003) Animal mitochondrial DNA
recombination revisited. Trends in Ecology & Evolution 18, 411-417.
Shitara H, Hayashi J, Takahama S, Kaneda H, Yonekawa H (1998) Maternal
inheritance of mouse mtDNA in interspecific hybrids: Segregation of the
leaked paternal mtDNA followed by the prevention of subsequent paternal
leakage. Genetics 148, 851-857.
Solignac M, Génermont J, Monnerot M, Mounolou J-C (1984) Genetics of
mitochondria in Drosophila: mtDNA inheritance in heteroplasmic strains of
D. mauritiana. Molecular and General Genetics MGG 197, 183-188.
Steinborn R, Zakhartchenko V, Jelyazkov J, Klein D, Wolf E, Müller M, Brem G
(1998) Composition of parental mitochondrial DNA in cloned bovine
embryos. FEBS Letters 426, 352-356.
Stephens JC, Rogers J, Ruano G (1990) Theoretical underpinning of the singlemolecule-dilution (SMD) method of direct haplotype resolution. American
Journal of Human Genetics 46, 1149-1155.
Stewart JB, Freyer C, Elson JL, Wredenberg A, Cansu Z, Trifunovic A, Larsson N-G
(2008) Strong purifying selection in transmission of mammalian
mitochondrial DNA. PLoS Biology 6, e10.
Sutovsky P, Moreno RD, Ramalho-Santos J, Dominko T, Simerly C, Schatten G
(2000) Ubiquitinated sperm mitochondria, selective proteolysis, and the
regulation of mitochondrial inheritance in mammalian embryos. Biology of
Reproduction 63, 582-590.
Tang Y, Manfredi G, Hirano M, Schon EA (2000) Maintenance of Human
rearranged mitochondrial DNAs in long-term cultured transmitochondrial
cell lines. Molecular Biology of the Cell 11, 2349-2358.
Thyagarajan B, Padua RA, Campbell C (1996) Mammalian mitochondria possess
homologous DNA recombination activity. Journal of Biological Chemistry
271, 27536-27543.
168
Van Leeuwen T, Vanholme B, Van Pottelberge S, Van Nieuwenhuyse P, Nauen R,
Tirry L, Denholm I (2008) Mitochondrial heteroplasmy and the evolution of
insecticide resistance: Non-Mendelian inheritance in action. Proceedings of
the National Academy of Sciences 105, 5980-5985.
Yaffe MP (1999) The machinery of mitochondrial inheritance and behavior. Science
283, 1493-1497.
169
Appendix
Schematic representation of smPCR protocol
Part I
Template DNA dilution and smPCR PHASE I – “Clean lab”
Step 1: Prepare a 96 well plate (Plate A) with template DNA serially diluted starting
from the stock solution (103G/µl). Five different concentrations will be tested, with 16
wells per concentration and 16 negative control wells.
Step 1: Plate A
DNA template
(serially diluted)
(Serial DNA dilution scheme: Rows 1 & 2 = 200 µl at 25G/µl (195 µl ddH2O + 5 µl Stock DNA); Rows 3
& 4 = 150 µl at 5G/µl (120 µl ddH2O + 30 µl of A); Rows 5 & 6 = 150 µl at 1G/µl (120 µl ddH2O + 30 µl of
B); Rows 7 & 8 = 150 µl at 0.20G/µl (120 µl ddH2O + 30 µl of B); Rows 9 & 10 = 150 µl at 0.04G/µl (120 µl
ddH2O + 30 µl of B); Rows 11 & 12 = 200 µl ddH2O)
Step 2: In a new plate (Plate B) dispense 1 drop of mineral oil in each well.
Step 3: With a multi-channel pipette transfer 5 µl of DNA template from Plate A to Plate
B, to obtain Plate B1 (Release the DNA template underneath the oil)
Step 3
Step 2: Plate B
Mineral oil
Step 4: Prepare Phase I reaction mix for 120 reactions in an eppendorf tube and
dispense 75µl of this solution to the first column wells of a new plate (Plate C).
Step 5: With a multi-channel pipette dispense 5µl of PCR Phase I mix to the wells of
Plate B1, obtaining Plate B2. Release the PCR mix at the top of the walls without
touching the mineral oil. (Start dispensing from the lowest concentration to the highest concentration
wells)
Step 6: PCR plate B2.
Plate B1
Mineral Oil and DNA template
Step 5
Step 4: Plate C
Phase I PCR mix
Plate B2
Mineral oil + DNA template + PCR mix
Part II
Step 7: Plate B3
Phase I PCR product + 60µl ddH2O
smPCR Dilution of Phase I PCR products – Normal lab
Step 7: Add 60µl of ddH2O to the wells of Phase IPCR product to obtain Plate B3.
Centrifuge.
Step 8: In a deep well plate (Plate D) dispense 250 µl of ddH2O in each well.
Step 9: With a multi-channel pipette transfer 15 µl of DNA template from Plate B3 to
Step 9
Step 8: Plate D
Dispense 250µl of ddH2O
Plate D, obtaining a final 100x dilution of Phase I PCR products (Plate D1).
Plate D1
100x dilution of Phase I PCR products
170
Appendix I Continuation
Part III
smPCR Phase II – Normal lab
Step 10: Dispense 1 drop of mineral oil in a new plate (Plate E).
Step 11: With a multi-channel pipette transfer 5 µl of Phase I diluted PCR products
Plate D1
100x dilution of Phase I PCR products
from Plate D1 to the bottom of plate E, obtaining Plate E1. Centrifuge.
Step 12:. Prepare Phase II reaction mix (for one marker) for 120 wells in an
eppendorf tube and dispense 75µl of this solution to the first column wells of a new
plate (Plate F).
Step 13:. With a multi-channel pipette dispense 5µl of PCR Phase II mix to the
wells of Plate E1, obtaining plate E2. Release the PCR mix at the top of the walls
Step 11
Step 10: Plate E
Mineral oil
without touching the mineral oil.. Centrifuge.
Step 14: PCR plate E2.
Step 15: repeat steps 11-14 for the second marker, changing the primers used in
the reaction mix
Plate E1
Mineral oil + diluted Phase I PCR products
Step 13
Plate E2
Mineral oil + DNA template + PCR mix
Step 12: Plate F
Phase II PCR mix
171
Chapter 6
General discussion and conclusions
Phots by Andreia Miraldo
Car and traps used during 3 years of fieldwork
6. General discussion and conclusions
By studying a species with a distribution that encompasses the entire Iberian
Peninsula it was possible to have a broader and more complete picture about the role
of this peninsula as a diversification hotspot. Using mitochondrial and nuclear
genealogies it became clear that Lacerta lepida, like other species in the region, has
endured repeated processes of fragmentation that have promoted the diversification
of six genetically and geographically distinct lineages. Estimating the dates of
divergence between the different evolutionary lineages revealed that diversification
within Lacerta lepida is largely concordant with the onset of the major glaciations at
the beginning of the Pleistocene approximately 2 Mya. The earliest divergence,
during the Miocene, represents a deep split within the species marking the
divergence of a lineage (lineage N) associated with the Betic Mountains in southeastern Spain. Both climatically mediated events during the Quaternary, and
geological events associated with the evolution of the Mediterranean basin, are
inferred to have triggered intraspecific diversification within Lacerta lepida.
The majority of phylogeographic studies within Iberia reveal similar
diversification events across several taxa that are usually attributed to allopatric
differentiation in several refugia within the Peninsula, although sometimes at
different temporal scales. These studies are however typified by the absence of
detailed analyses of the distribution of ancestral and derived alleles within each
lineage. This approach has been shown to be extremely valuable for the delimitation
of refugial areas (Emerson and Hewitt, 2005) and in the context of this work it has
identified six geographically distinct refugia within the Iberian Peninsula. The
172
identified refugia occur throughout the region: in north-western Iberia, around the
gorges of the Douro River; in central Spain around the central mountain system; in
inland central Portugal in the Tagus River region; in the south-western corner of
Portugal, in the Algarve region; in southern Spain around the Guadalquivir area and
finally in the Betic Mountains in south-eastern Spain.
Of particular interest are the refugia detected around the gorges of the Douro
River, in the region between Portugal and Spain and in the central system mountains.
The detection of such northerly located refugia, for what is considered a
Mediterranean species, suggests that suitable ecological conditions have existed at
these northern latitudes during glacial maxima. Although northern refugia in Iberia
have been previously detected they are typically associated with species with
ecological requirements intimately associated with Atlantic influences (e.g. Lacerta
schreiberi and Chioglossa lusitanica). This thesis reveals that Lacerta lepida is
likely to have persisted in these northerly refugia as well, emphasizing the
importance of these regions in the survival of species with very different ecological
requirements throughout adverse climatic conditions.
The phylogeographic analysis of Lacerta lepida has also revealed areas of
secondary contact between divergent lineages, formed mainly as a result of
demographic range expansions. Detailed analysis of two different contact zones
between Lacerta lepida mitochondrial lineages was carried out revealing very
different dynamics for each. The contact zone in the north-western part of Iberia is
relatively recent. Evidence for hybridization was inferred by the detection of a Numt
within one of the lineages that originated from the mitochondrial genome of the
other. Detection of additional Numts from different introgression events are
consistent with other mitochondrial lineages that are now extinct. Although Numts
have been described in a wide range of taxa their function in the genome, if any, is
unknown. However, their utility as a tool in evolutionary biology is recognized, as
they provide a unique window on past evolutionary events (Bensasson et al., 2001).
Despite their potential as important sources of information, very few studies to date
take advantage of Numts. Once detected, Numts are typically discarded from further
analysis. This thesis has demonstrated that Numts can be extremely valuable in the
context of phylogeographic analysis, as they can provide evidence for past
173
demographic events. The indiscriminate discarding of Numts from analysis may
result in researchers losing a valuable source of information, and in this study Numts
reveal that within L. lepida hybridization between the lineages has occurred.
An unexpected outcome from the detailed analysis of the northern contact
zone was the detection of low levels of heteroplasmy and mitochondrial DNA
recombination in one of the mitochondrial lineages. These findings are a new
addition to the already extensive list of studies reporting evidence for exceptions to
the general assumptions regarding mitochondrial DNA inheritance in animals.
Heteroplasmy and mtDNA recombination were only detected in one of the lineages
and it remains unknown whether these phenomena are widespread in Lacerta lepida.
A wide range of mechanisms are responsible for controlling the strict maternal
inheritance of mtDNA in animals which can act at any stage of the reproductive
process (for a review on the subject see Birky, 1995). The mechanisms vary from the
complete lack of mitochondria in the sperm to the active elimination of paternally
derived mitochondria at fertilization. For example, in some tunicates paternal
mitochondria fail to enter the egg whereas in honey bees more than a quarter of all
mitochondria in one egg are reported to be paternally derived, although their mtDNA
has defective replication, making it undetectable in the larvae stage. In most animals
though, it seems that the strictly maternal inheritance of mtDNA is derived due to a
combination of factors (Birky, 1995; Birky, 2001) involving the limited number of
mitochondria from the sperm cell that enter the oocyte during fertilization and their
active elimination by a ubiquitin-dependent mechanism (Sutovsky et al,. 1999). This
process secures the homoplasmy of the embryo. Nevertheless, the recognition of
paternal mtDNA apparently depends on phylogenetic relatedness. As the degree of
genetic divergence between species increases, the probability that sperm
mitochondria are recognized and eliminated decreases, but also reduces the
probability that F1 hybrids are viable and fertile. Therefore, the path to producing
heteroplasmy and recombinant haplotypes in a population through hybridization
might be narrow. Earlier studies in Drosophila suggested that paternal leakage is
more likely to occur if the genetic divergence (uncorrected) between taxa is
approximately 2.5% or higher (Kondo et al., 1990), but more recent studies detected
leakage between Drosophila subspecies which show much lower divergence levels
(Sherengul et al., 2006). In cicadas leakage was also demonstrated to occur between
174
crosses that show a wide range of genetic divergence, from almost no divergence to
8% (Fontaine et al., 2007). Divergence levels detected between Lacerta lepida
lineages range from 1% to almost 13% so it is possible that paternal leakage could
occur between most lineages that form zones of secondary contact.
The findings of this thesis have implications for evolutionary analyses using
mtDNA, but their significance certainly depends on the capacity of the detected
heteroplasmy and recombination to leave a footprint at the population level. To be
consequential for future generations heteroplasmy must persists via the germ line
and remain in the oocyte long enough for recombination to occur. The establishment
of a fertile female hybrid carrying a recombinant haplotype may then result in the
transmission of such a haplotype to the next generation by backcrossing with a male
from either species. Similar events of repeated backcrossing may then potentially fix
the recombinant mtDNA haplotype against the nuclear background of one or the
other parental species. In summary, even if biparental recombination is detected in
an individual, there are other prerequisites for this recombination to leave a footprint
at the population level.
The dynamics of gene flow was also assessed in a contact zone between
Lepida and Nevadensis mitochondrial lineages in south-eastern Spain. The
microsatellite analysis of this contact zone revealed very restricted gene flow
amongst the lineages and it was postulated that the lineages are on independent
evolutionary paths and therefore should be considered as two different species. For
future work it would be interesting to assess the mechanisms that are driving
speciation in these lizards. According to Jiggins and Mallet (2000) premating
isolation is likely to be more effective than hybrid incompability in maintaining
species differences despite gene flow. In Heliconius butterflies it seems that
speciation occurred after the evolution of ecological divergence and mate choice
differences and well before hybrid unfitness (Mallet et al., 1998). Although F1
hybrids were detected in the contact zone between Lepida and Nevadensis, data
suggests that some form of hybrid incompability might be responsible for the
reduced gene flow observed. Nevertheless prezygotic mechanisms are also quite
likely to be involved in the dynamics of this contact zone. The lineages show
extreme variation in colour patterns (Mateo and Castroviejo, 1990; Mateo and
175
López-Jurado, 1994; Mateo et al., 1996) and visual cues in these lizards may play an
important role in mate recognition as this has been observed for other lacertid lizards
(Molina-Borja, 1987). Courtship behaviour in Lacerta lepida usually involves overt
displays of lateral blue spots (Paulo, 1988; personal observation) which differ
substantially between the lineages in contact. Other prezygotic mechanisms between
the lineages could result from differences in the reproductive activity between them
as Nevadensis shows an extended reproductive period (Castilla and Bauwens, 1989;
Mateo, 1988; Mateo and Castanet, 1994). Both prezygotic and postzygotic
mechanisms could well be important for maintaining the isolation between these two
lineages, and distinguishing which are more important will require detailed analysis.
This thesis represents a major contribution to our understanding of the
evolutionary history of Lacerta lepida, providing extensive information about the
history, distribution and dynamics of genetic variation within the species. It is also
an important contribution to the understanding of the evolutionary dynamics of
Iberian Peninsula biota in general, in particular by describing areas of importance for
species diversification and survival in response to historical and contemporary
events. This study also provides the first detailed analysis of secondary contact zones
within the species, providing insights into hybridization and speciation processes that
are relevant for the evolutionary history of Lacerta lepida. The detection of Numts
originated as a result of hybridization between divergent lineages and the realization
of their utility in elucidating phylogeographic studies is an exciting prospect for the
field of phylogeography. Also exciting is the strong evidence for mitochondrial
DNA recombination, which until now was rarely reported for natural populations in
the literature.
176
6.1. References
Birky C (1995) Uniparental inheritance of mitochondrial and chloroplast genes:
mechanisms and evolution. Proceedings of the National Academy of Sciences
92, 11331-11338.
Birky CW (2001) The inheritance of genes in mitochondria and chloroplasts: Laws,
Mechanisms, and Models. Annual Review of Genetics 35, 125-148.
Emerson BC, Hewitt GM (2005) Phylogeography. Current Biology 15, 367-371.
Jiggins CD, Mallet J (2000) Bimodal hybrid zones and speciation. Trends in Ecology
& Evolution 15, 250-255.
Genetics 126, 657-663.
Mallet J, McMillan WO, Jiggins CD (1998) Mimicry and warning color at the
boundary between races and species. In: Endless forms: species and
speciation (eds. Howard D, Berlocher SH), pp. 390-403. Oxford University
Press, Oxford.
215-229.
177
Molina-Borja (1987) Spatio-temporal distribution of aggressive and courting
behaviors in the lizard Gallotia galloti from Tenerife, the Canary Islands.
Journal of Ethology 5, 11-15.
Paulo OS (1988) Estudo eco-etologico da populacao de Lacerta lepida (Daudin
1802) (Sauria, LAcertidae) da ilha da Berlenga, Universidade de Lisboa.
178

Andreia Miraldo - University of Helsinki

Transcrição

Documentos relacionados

mitochondrial dna and human evolution

Using Molecular Biology Techniques to Characterize the Diversity of

A standard procedure for accommodating forensic anthropological

Binuclear Copper(II) Complexes Modified with Pyrene: DNA

Genetica populacional e conservag8.0 das especies de tartarugas

Number 17 – April 2009 - Geoscience Research Institute

Detecting multiple DNA human profile from a mosquito blood meal

Number 26 – July 2011 - Geoscience Research Institute

GENETIC VARIABILITY FOR 25 AUTOSOMAL STR

lista de inglês