Short communication. Phylogeny and genetic diversity within Iberian populations of Ornithopus L. and Biserrula L. estimated using ITS DNA sequences

Genetic diversity within Iberian populations of Ornithopus pinnatus, O. compressus, O. sativus and Biserrula pelecinus were assessed using ITS1 and ITS2 DNA sequences from sixty four specimens, and a phylogeny between Ornithopus species was estimated. Generally within-species variation was low, particularly within Ornithopus. The Mediterranean species of Ornithopus form a sister clade relative to the South American O. micranthopus. The sometimes considered a full species, O. sativus isthmocarpus, was not distinct from O. sativus. Between some species there is limited genetic divergence using these markers, although the situation of O. perpusillus requires additional specimens to be examined before firm conclusions can be drawn. Additional key words: Biserrula pelecinus; forage; Ornithopus compressus; Ornithopus pinnatus; Ornithopus sativus.

from the Mediterranean basin, and O. micranthus from South America (Fu et al., 1994).Ornithopus sativus isthmocarpus is also sometimes recognized as a distinct species (Allan & Porter, 2000).All the species are adapted to sandy soils and are highly palatable.Biserrula pelecinus and some Ornithopus (O. compressus, O. sativus and O. pinnatus) in particular have become used for forage in disparate regions with Mediterranean climate, especially Australia (Nichols et al., 2010) and Chile (Del Pozo & Ovalle, 2009).Sowing O. sativus pasture between cropping sequences in Australia has shown to benefit grain production through restoring soil fertility (Doole et al., 2009), and during the last 15 years both O. sativius and B. pelecinus have been commercialised in Australia (Nichols et al., 2010).
Given this agricultural importance surprisingly little is known regarding relationships between Ornithopus species or levels of genetic variation within species.Biodiversity of root-nodule bacteria has been assessed for both B. pelecinus (Vicente et al., 2009) and O. compressus (Loi et al., 1999).Determination of morphological differences between species of Ornithopus indicated considerable differences of important characteristics such as stem length or time of first flowering (Fu et al., 1994), so that Loi et al. (1997) concluded there was sufficient morphological variation to initiate a selection program for Southern Australia.Such programs benefit from an assessment of genetic variation within species, and of the phylogenetic relationships between them.
The aim of this study was to determine genetic variation within Biserrula pelecinus and the species of Ornithopus from the Iberian Peninsula using ribosomal internal transcribed spacer (ITS) sequences.At the same time by incorporating data from GenBank for O. micranthus, as well as some additional samples of these species, a phylogeny for the genus can be determined.This gene region is widely used both in phylogeny reconstruction in plants (Allan & Porter, 2000), and in barcoding studies to differentiate species (Chen et al., 2010).It has also been recently used to assess variation within other species from the region, Scorpiurus muricatus and S. vermiculatus (Visnevschi-Necrasov et al., 2011).Thus direct comparisons of relative within-species genetic diversity can be made.This should help resolving the taxonomic issue, such as the distinctiveness or not of O. s. isthmocarpus, and to identify genetically divergent groups that might be important for breeding programs.
Sixty four specimens of Biserrula and Ornithopus were collected (Table 1) in Portugal and Western Spain (Fig. 1).In order to collect only wild germplasm seeds were collected in natural occurring plants from field borders and road sides.DNA was extracted using a cetyl trimethylammonium bromide (CTAB)-based protocol following Wang et al. (1996).The ITS1 and ITS2 region was amplified by polymerase chain reaction (PCR) using standard primers (White et al., 1990).Amplifications were performed in 20 μL reactions consisting of approximately 10 ng DNA template, 1 μM of each primer, 200 μM of each dNTP, 0.5 U EcoTAQ DNA polymerase, 2 μL of 10X PCR buffer and 1.5 mM MgCl 2 .The amplification protocol consisted of an initial denaturation at 95°C for 2 min followed by 30 cycles of 95°C for 30s, 53°C for 30s and 72°C for 1 min.A final extension step at 72°C for 7 min was performed.PCR products were purified using the JetQuick (Genomed, Löhne, Germany) micro spin kit and sequenced using the same primers on an ABI 3730 DNA sequencer using BigDyeTerminator v3.1 from the same supplier.
Sequences were aligned with the available sequences from GenBank for these genera using Clustal W with default conditions in the program BioEdit v5.0.9 (Hall, 1999).Within closely related groups the program TCS v2.1 (Clement et al., 2000) was used to create a parsimonious network of the aligned haplotypes.To estimate phylogenetic relationships of Ornithopus all unique haplotypes were aligned.Maximum likelihood (ML) analysis with random sequence addition (100 replicate heuristic searches) was used to estimate their evolutionary relationships, using the program PAUP v4.0b10 (Swofford, 2002).Support for nodes was estimated using the bootstrap technique (Felsenstein, 1985) with 1000 replicates.The model of evolution employed was chosen using the Akaike Information Criteria carried out in Modeltest 3.06 (Posada & Crandall, 1998).Bayesian analysis was implemented using Mr. Bayes v.3.1 (Huelsenbeck & Ronquist, 2001) with parameters estimated as part of the analysis.The analysis was run for 1×10 7 generations, saving one tree every 1000 generations.The log-likelihood values of the sample point were plotted against the generation time and all the trees prior to reaching stationary were discarded as burn-in samples.Remaining trees were combined in a 50% majority consensus tree (Huelsenbeck & Ronquist, 2001).New haplotypes have been submitted to GenBank (JQ042900 toJQ042909).ITS genetic diversity within Biserrula and Ornithopus For Biserrula 16 new specimens were sequenced (625 bp aligned length) and compared with a single specimen from GenBank (AB287409, 50 bp shorter).Six haplotypes were recovered, with a single common haplotype (nine individuals), four unique haplotypes, and one haplotype shared by three individuals.The greatest distance between haplotypes was three differences (Fig. 2C).For Ornithopus 48 new specimens were sequenced.Within O. pinnatus seven individuals were sequenced (600 bp aligned length) and compared with a single specimen from GenBank (AY325278).All shared the same haplotype, distinct from the remaining Mediterranean Ornithopus.Within the remaining Ornithopus from the Mediterranean 41 new specimens were sequenced (600 bp aligned length), and compared with five sequences from GenBank (AF450226-8, AF218533-4, Fig. 2B).There was haplotype sharing between species, and a single haplotype network was recovered, with six unique haplotypes in total (Fig. 2B).
For the estimate of phylogenetic relationships, the unique haplotypes (seven from the Mediterranean species, plus one O. micranthopus -AY325277) were aligned with a specimen of Lotus wrangelianus (AF450174) as outgroup.The resulting alignment was 603 bp long.Both ML (GTR+I+G model of evolution) and Bayesian estimates of relationships strongly suggested the monophyly of the Mediterranean species relative to O. micranthopus.Within the Mediterranean species O. sativus (including O. s. isthmocarpus -AF218534) was closely related to, but distinct from, O. perspusillus (including AF450226) and O. compressus which share haplotypes (1, Fig. 2B).Estimates of relationships were well supported (Fig. 2A).
Our results in general revealed low levels of intraspecific genetic variation.Within Biserrula, where six haplotypes were recovered, this is considerably less than recovered in Scorpiurus muricatus (Visnevschi-Necrasov et al., 2011).On the other hand morphologically distinct North African populations (Loi et al., 1997) were not assessed.Within Ornithopus variation was even lower.This may mean that the extensive morphological variants known to occur in Ornithopus (Fu et al., 1994) have arisen in a relatively short evolutionary time, so that minimal neutral genetic variation (such as within the ITS region) has had time to occur.It may also be that population sizes of these species were considerably smaller during the Pleistocene, when climatic conditions were colder, and this has led to a genetic bottleneck effect.The estimate of phylogenetic relationships indicates, as expected, that the South To conclude, none of the examined species shows high levels of intraspecific variation.This implies that the morphological variants do not reflect deep genetic divergences, and that no genetic barriers to breeding programs were identified.Even between some species there is limited genetic divergence, although the situation of O. perpusillus requires additional specimens to be examined before firm conclusions can be drawn.It will also be important in the future to assess North African populations, especially for Biserrula.

Figure 2 .
Figure 2. A. Estimate of relationships within Ornithopus based on maximum likelihood analysis.Numbers above nodes indicate bootstrap support; those below nodes correspond to Bayesian posterior probabilities.B. Network of haplotypes within the clade of O. compressus and O. sativus.Numbers refer to haplotype codes in Table 1 and in Fig. 2A.Small circles indicate inferred missing haplotypes.C. Network of haplotypes within B. pelecinus.

Table 1 .
Identification (ID) number, species and coordinates of the collection sites for the 64 samples used in this study.

Table 1 (
cont.).Identification (ID) number, species and coordinates of the collection sites for the 64 samples used in this study.Haplotype refers to Figure2 Madrid Lisboa Figure 1.Map of the collections sites of the species of Ornithopus and Biserrula used in this study.Numbers refer to Table 1

.
ITS genetic diversity within Biserrula and Ornithopus