GENETIC ISOLATION BETWEEN THREE CLOSELY RELATED TAXA

Get Complete Project Material File(s) Now! »

Evolution of mating systems in the Fucaceae

Introduction

Mating system evolution in marine algae is poorly understood despite their high variability. The family Fucaceae is an excellent model for studying evolution of mating systems at a macro-evolutionary scale because the character dioecy/hermaphroditism has undergone multiple switches (Serrão et al. 1999). The scattered phylogenetic positions of hermaphroditic and dioecious species along the evolutionary history of the Fucaceae suggests that either mating system has evolved independently several times, possibly by relatively simple mechanisms (Serrão et al. 1999) as in the flowering plant genus Silene (Desfeux et al., 1996), Wurmbea (Barrett and Case 2006), and in angiosperms generally (for review see Charlesworth 2002). However, while in land plants, based on modelling, empirical and phylogenetic studies, dioecy generally appears to be the derived state arising from cosexual ancestors (Charlesworth 1999), in the Fucaceae, partly due to the lack of discriminating markers, the ancestral state is still questionable. Yet, in the genus Fucus, the occurrence of relics of male function in hermaphrodites and females of the sister species F. vesiculosus and F. spiralis, suggests that hermaphroditism is the ancestral state (Billard et al. 2005b).
The family Fucaceae appears to have had its origin in the North Pacific Ocean, presumably resulting from a transequatorial crossing of an ancestor of Australasian origin, as suggested by several independent phylogenies (Serrão et al. 1999, Draisma et al. 2001, Cho et al. 2006). The closest living relatives of the Fucaceae are thus the two Australasian families Hormosiraceae (currently including only the dioecious species Hormosira banksii) and newly created (Cho et al. 2006) family Xiphophoraceae (comprising the species Xiphophora chondrophyla and Xiphophora gladiata, both monoecious hermaphrodites, which used to be classified in the family Fucaceae). Sequences of the internal transcribed spacer region of nuclear ribosomal DNA (ITS) have provided the most complete phylogeny of Fucaceae (Serrão et al. 1999). A Pacific origin in the Northern hemisphere has been followed by several switches between oceans, but the history of these is still unclear because some relationships were not clear or concordant between different studies. This is the case for the position of Ascophyllum nodosum and Pelvetia canaliculata, for example, for which different studies (Serrão et al. 1999, Cho et al. 2006) are not in agreement as to which of them represents the first divergence between the Pacific and Atlantic taxa in the Fucaceae. It is also the case for the relationships between the species of Fucus, where phylogenetic data reveal two distinct clusters (Serrão et al. 1999, Coyer et al. 2006), within which only the species Fucus serratus is clearly separable, and are unable to resolve taxa that are clearly different species when compared based on microsatellite allelic frequencies (Billard et al. 2005a; Engel et al. 2005) or a recently derived partially clonal form of F. vesiculosus in the Baltic Sea (Tatarenkov et al. 2005), which has been named F. radicans (Bergström et al. 2005). The alternate distribution of the character dioecy/hermaphroditism highlights that reproductive system switched several times during speciation in the genus Fucus. Of the two distinct clusters within the genus Fucus, one, hereafter called lineage 1, comprises the dioecious F. serratus distinct from the group of hermaphroditic F. distichus, F. gardneri and F. evanescens and the second one, hereafter called lineage 2, consists of the dioecious F. vesiculosus and F. ceranoides undistinguishable from the hermaphroditic F. spiralis and F. virsoides. This lineage 2 includes also asexual or partially asexual entities which may be additional species such as Atlantic F. cottonii (Wallace et al. 2004, 2006, Coyer et al. 2006, Engel et al. 2005) and F. radicans (Bergström et al. 2005; Tatarenkov et al. 2005). The lack of resolution within each of these Fucus clusters has been proposed to be associated to the recent and rapid radiation within the genus (Serrão et al. 1999, Coyer et al. 2006), and may be further complicated by hybridization being possible between taxa with contrasting mating systems, as revealed using microsatellites (Coyer et al. 2002, Engel et al. 2003). Coyer et al. (2006) defend the hypothesis of a north Pacific origin of Fucus followed by radiation in the north Atlantic, which would imply hermaphroditism as ancestral state in the genus, from which at least two independent switches to dioecy were derived. However, until now the question of the ancestral reproductive system in Fucus is still not clear, even if the evolution of hermaphroditism to dioecy seems the most parsimonious scenario (Billard et al. 2005b).
An additional question that still remains unanswered is why have so many switches between reproductive modes taken place along the evolutionary history of the family Fucaceae? Different ecological conditions are known to favour different mating systems (Takebayashi and Morrell 2001). Fixed abiotic stress might favour selfing of the best adapted genotypes thereby favouring the maintenance of local adaptation as well as reproductive assurance and colonising capacity (Baker 1955, Pannel and Barrett 1998), whereas biotic effects such as the need for competitive ability may favour outcrossing for maintenance of the adaptive capacity towards biotic interactions, maintaining high diversity and avoiding inbreeding depression (reviewed in Uyenoyama et al. 1993). In the Fucaceae, geographical isolation under different environmental conditions, such as when colonising a new ocean system, may have resulted in contrasting mating systems best adapted to each environment type, or on the other way round, it may have been the mating system itself rather than geographical isolation, that may have provided the opportunity for speciation.
The aim of this study is to investigate the evolutionary history of mating systems in the family Fucaceae, in order to test the hypothesis that evolution proceeded in all cases always from hermaphroditism to dioecy, and to assess whether shifts in mating system might be correlated with major events such as dispersing between different oceans. In order to achieve these goals we will revise existing molecular information and add new datasets from several intergenic spacer regions derived from the Fucus chloroplast genome (Pearson unpublished), for the family Fucaceae. Species mating systems will then be mapped on the phylogeny and together with geographic distributional information; these will be used to define evolutionary pathways for the mating systems and the relations between such pathways and important ecological or historical events.

Material & methods

Taxon sampling and DNA extraction

Nineteen species were sampled among the Fucaceae, and for outgroup its closest relatives which are the southern hemisphere families Hormosiraceae and Xiphophoraceae (Table 1). A northern hemisphere family Himanthaliaceae was initially used to compare its distance levels and potential usefulness but it was excluded afterwards because it was almost unalignable. When possible, samples used were the same individuals as in Serrão et al (1999) or as in Coyer et al (2006) or at least from the same locations. For samples that required new DNA extraction, 20 mg of dried tissue were used in the nucleospin column plant DNA extraction Kit (Macherey-Nagel Düren, Germany) according to the manufacturer’s protocol and diluted 1:100.

Chloroplast marker selection

Based on the completely sequenced chloroplast genome of F. vesiculosus (Pearson, unpublished) we identified intergenic spacer regions to test for phylogenetic usefulness. Primers were designed in the coding sequences flanking the regions of interest using Primer3 software (Rozen and Skaletsky 2000). The spacer regions that appeared most useful for distinction between species (Table 2) were then selected for sequencing analyses.
The first studied chloroplast region (ThiG-ycf54) was about 250bp long and localised between the genes thiG and ycf54. The second region (psbX-ycf66) was about 280 bp long and located between the genes psbX and ycf66. In addition to the polymorphism tests for phylogenetic purposes, a 550 bp long region including the Rubisco spacer and part of the flanking coding regions RbcL and RbcS was tested for species diagnostic purposes on several individuals of F. spiralis and F. vesiculosus, using a restriction enzyme (SspI) with a restriction site specific to a sequence found only in F. spiralis. In order to check the consistency of this distinction between F. spiralis and F. vesiculosus, this region was reamplified on 14 individuals of parapatric populations (populations which are not in contact with each other) from each species in Portugal. They were then submitted to restriction by the enzyme SspI, which was expected to cut only for F. spiralis.

DNA amplification and sequencing

Sequencing reactions were carried out directly on polymerase chain reaction (PCR) products. PCRs were performed in 20µL containing 0.1µg/µL bovine serum albumin, 75mM Tris-HCl, 20mM (NH4)2SO4, 0.01% Tween®20, 2.5mM MgCl2, 0.25µM of each forward and reverse primer, 200µM of each dNTP, 0.5U Thermoprime Plus Taq polymerase (ABgene) and 5µL of diluted DNA. PCRs were run on a PTC200 thermocycler (MJ Research). After an initial denaturation step (95°C, 5min), ‘touchdown’ PCR was carried out for 5 cycles of 30s at 95°C, 30s at 60°C, reduced by 1°C per cycle for 5 cycles, and 30s at 72°C, followed by 30 cycles of 95°C for 30s, 55°C for 30s and 72°C for 30s and a final 7 min elongation at 72°C. Purified PCR products (Millipore Multiscreen-PCR plates) were sequenced in both directions by using the amplification primers, purified and sequenced on an ABI 3100 capillary sequencer (Perkin-Elmer Applied Biosystems) using the BigDye kit (Perkin-Elmer Applied Biosystems), following the manufacturer’s protocol.

Sequence analysis

Sequences were aligned using CLUSTALW (Houssard et al. 1994) as implemented in BIOEDIT 6.0.6 (Hall 1999). Corrections were made by hand because of the numerous insertions/deletions (indels). Polymorphism of sequences within and between clusters was analysed with DnaSP 4.10 (Rozas et al. 2003).
For the phylogenetic reconstruction, intergenic sequences thiG-ycf54 and psbX-ycf66 were concatenated. Indels were coded using the simple coding model (Simmons and Ochoterena 2000) with SeqState (Müller 2005). Aligned sequences were analyzed with Bayesian, maximum likelihood, and parsimony methods. The best evolution model to use in the Bayesian and the Maximum Likelihood analyses was selected using the likelihood ratio test implemented in ModelTest (Posada and Crandall 1998). For ThiG-ycf54 + psbX-ycf66 dataset (referred to as thyGy-psbX), the best model was a Kimura three-parameters with unequal base frequencies K81uf (A=0.4390; C=0.0494; G=0.0724; T=0.4391; A↔C=1.0000; A↔G=1.7321; A↔T=0.2065; C↔G=0.2065; C↔T=1.7321; G↔T=1.0000). For the rubisco spacer dataset, the best model was a general time reversible model with shape parameter of the gamma distribution HKY + Γ (A= 0.36722; C= 0.11114; G= 0.17258; T= 0.34905; A↔C=1.1574; A↔G=1.7057; A↔T=0.1427; C↔G=2.8862; C↔T=3.5705; G↔T=1.0000). These models and substitution rates were used to construct maximum likelihood trees by heuristic searches with random sequence addition and 100 bootstrap values replicates, using PhyML (Guindon and Gascuel 2003). Bayesian analyses were conducted using MRBAYES 3.1 (Huelsenbeck and Ronquist 2003). Each analysis was initiated from a random starting tree and the program was set to run four chains of Markov chain Monte Carlo iterations simultaneously for 1,000,000 generations with trees sampled every 100th generation.
Maximum parsimony (MP) was conducted using PAUP* 4.0 (Beta) using a heuristic search, with tree bisection-reconnection branch swapping, and used 1000 bootstrap replicates (Swofford 2002). Bayesian estimates of ancestral character states for reproductive system were mapped on the combined thiGy-psbX dataset and on the rubisco spacer dataset using SIMMAP (Bollback 2006).

Results

Polymorphism of the cpDNA regions

The three chloroplastic DNA spacers were well conserved within the genus Fucus and less polymorphic at the species level than previously studied nuclear (ITS, Serrão et al. 1999) and mitochondrial (Coyer et al. 2006) DNA regions (Table 3). In the thiG-ycf54/psbX-ycf66 dataset (Table 4), the divergence between genera was at least ten times higher than within Fucus, ranging from 0.049 net number of substitution per site between Hesperophycus and Pelvetiopsis to 0.112 between Ascophyllum and Hesperophycus. This sequence could not be obtained for the Xiphophora genus, so for these loci we used only Hormosira banksii as outgroup.
The rubisco spacer was however useful for species diagnostic because it presented one mutational difference between the hermaphroditic F. spiralis and the two dioecious species F. vesiculosus and F. ceranoides. Over the 14 F. spiralis and 14 F. vesiculosus PCR products obtained for this locus, all individuals of F. spiralis presented a haplotype cut by the restriction enzyme Ssp1 as expected, whereas none of the F. vesiculosus was. Given the problematic distinction between the species Fv and Fspir based on morphology and the occurrence of intermediate morphologies and intermediate genotypes, we chose to screen the rubisco spacer for individuals whose species name we can identify based on microsatellite genotypes. One individual in genbank (AY246553) was not consistent with this difference but since it does not come from a publication and it is not reported where this individual was collected and identified, we choose to include only those that we could certify the genetic entity for.

Phylogenetic analyses

All methods of phylogenetic analyses gave the same tree topologies, thus only one tree is shown for the concatenated thiGy-psbX dataset (Fig. 1) and one for the rubisco spacer (Fig. 2) for which sometimes different individuals had been sequenced. The cpDNA intergenic markers confirmed the existence of the two clusters of Fucus already mentioned by Serrão et al (1999) and Coyer et al (2006) in their analyses based on nuclear and mitochondrial markers respectively. These clusters correspond to lineage 1 of Serrão et al (1999) comprising the dioecious F. serratus and the hermaphroditic F. evanescens, F. distichus, F. edentatus and F. gardnerii, and to lineage 2 comprising the dioecious F. vesiculosus and F. ceranoides and the hermaphroditic F. spiralis and F. virsoides. On the other hand the Bayesian analysis of the Rubisco shows the genus Fucus as not monophyletic, with the Pelvetiopsis/Hesperophycus cluster grouping with the F. serratus. However, this was very poorly supported with the ML analysis (50.8%) and was unresolved with the MP analysis. The hermaphroditic genus Silvetia, as described in Serrao et al (1999) and the dioecious species Ascophyllum nodosum are now clustering together revealing a common ancestor and a more recent divergence than what was suggested based on the ITS data (Serrão et al 1999). This result is well-to-maximally supported depending on the Maximum parsimony (MP bootstrap = 0.71), maximum likelihood (ML bootstrap = 0.92) or Bayesian analysis (posterior probability = 1.0). Moreover, this cluster is always very-well-to-maximally supported when the analysis is performed with the rubisco marker (Fig 2). The Pelvetia genus, from the Atlantic, branched with the cluster containing the genera Hesperophycus, Pelvetiopsis and Fucus, confirming the nuclear ITS results (Serrão et al. 1999) that show it as the first divergence within this group, a pattern that was not supported with the chloroplast gene psaA (Cho et al 2006).

Table of contents :

1 GLOSSARY
2 INTRODUCTION
Life cycle evolution
Mating system evolution
The biological Model
Thesis goals
3 EVOLUTIONARY HISTORY OF MATING SYSTEMS AMONG THE FUCACEAE (PHAEOPHYCEAE) INFERRED FROM A PHYLOGENETIC STUDY BASED ON INTERGENIC CHLOROPLAST SEQUENCES
3.1 Abstract
3.2 Introduction
3.3 Material & methods
3.3.1 Taxon sampling and DNA extraction
3.3.2 Chloroplast marker selection
3.3.3 DNA amplification and sequencing
3.3.4 Sequence analysis
3.4 Results
3.4.1 Polymorphism of the cpDNA regions
3.4.2 Phylogenetic analyses
3.4.3 Ancestrality of sexual phenotypes
3.5 Discussion
3.6 Acknowledgments
3.7 References
4 GENETIC ISOLATION BETWEEN THREE CLOSELY RELATED TAXA: FUCUS VESICULOSUS, F. SPIRALIS AND F. CERANOIDES
4.1 Absract
4.2 Introduction
4.3 Materials and Methods
4.3.1 Sampling
4.3.2 DNA extraction, PCR reaction and genotyping
4.3.3 Data analysis
4.4 Results and discussion
4.5 Acknowledgements
4.6 References
5 ANALYSIS OF SEXUAL PHENOTYPE AND PREZYGOTIC FERTILITY IN NATURAL POPULATIONS OF FUCUS SPIRALIS, F. VESICULOSUS (FUCACEAE, PHAEOPHYCEAE) AND IN THEIR PUTATIVE HYBRIDS
5.1 Abstract
5.2 Introduction
5.3 Materials and Methods
5.3.1 Sampling
5.3.2 Variation in sexual phenotype and male and female fertility within individuals 64
5.3.3 Variation for male and female fertilities between both parental taxa
5.3.4 Comparison between hybrids and parental species
5.4 Results
5.4.1 Variation of sexual phenotype within individuals
5.4.2 Variation in male and female fertility within individuals
5.4.3 Variation in male and female fertility between parental taxa
5.4.4 Comparison between hybrids and parental species
5.5 Discussion
5.6 Acknowledgements
5.7 References
6 A MOSAIC OF HYBRIDS BETWEEN SPECIES WITH ACONTRASTING REPRODUCTIVE SYSTEMS AT THE MICRO-SPATIAL-SCALE OF THE SHORE: PHENOTYPIC AND GENETIC ANALYSES
6.1 Abstract
6.2 Introduction
6.3 Materials and Methods
6.3.1 Sampling design
6.3.2 DNA extraction, PCR reaction and genotyping
6.3.3 Chloroplast DNA genotyping
6.3.4 Genetic analyses
6.3.5 Spatial auto-correlation
6.3.6 Phenotypic analyses
6.3.7 Sexual phenotype analysis
6.4 Results
6.4.1 Genetic analyses
􀀹 Quadrats
􀀹 Transects
6.4.2 Phenology
6.5 Discussion
6.5.1 Number of genetic clusters
6.5.2 Diagnostic marker in Chloroplast
6.5.3 Hybridization
6.6 Acknowledgements
6.7 References
7 CONCLUSION & PERSPECTIVES
7.1 Phylogenetic approaches
7.1.1 Development cytoplasmic markers
7.1.2 Evolution of dioecy from hermaphroditism in Fucus
7.2 Population genetic approaches
7.2.1 Genetic barriers within the species complex Fucus vesiculosus /spiralis / ceranoides
7.2.2 Two divergent entities within F. spiralis
7.2.3 The shore as a model of hybrid zone in Fucus
7.2.4 Reproductive system and mating system
7.2.5 Neutrality of microsatellite markers used?
7.3 Population Biology Approaches
7.3.1 Différences of resource allocation to male and female function between hermaphroditic and dioecious individuals
7.3.2 Phenological discrepancy and hybridization
7.4 Hybridization / selection in Fucus
7.5 Evolution of reproductive systems
8 REFERENCES
9 APPENDICES

GET THE COMPLETE PROJECT