Tilia L. (lime or basswood) is a genus of large trees with approximately 20–25 species in the family Malvaceae of the order Malvales (Angiosperm Phylogeny Group, 2002). These species are distributed in temperate regions of the northern hemisphere, throughout Europe, and in parts of Asia and the North American continent. Tilia cordata Mill. (small-leaved lime) and T. platyphyllos Scop. (large-leaved lime) are the two species that are widely distributed in Europe. They can hybridize, which results in the common lime, T. ×europaea (Pigott, 1969). Species relationships within the genus are unknown and are complicated by frequent hybridization. Although there are some morphological differences between T. cordata and T. platyphyllos, it is difficult to identify species in the absence of flowers and from characteristics of leaves at ground level; in particular, hybrids can vary in morphology (Pigott, 1991). In the United Kingdom, lime trees were a common species 5000 yr ago (Huntley and Birks, 1983), and trees alive today may be several hundred or even 1000 yr old.
There are no nuclear markers suitable for population genetics currently available within this genus. Here, we report on the development of the first set of microsatellite markers and their transferability to other species in the genus for future population genetic studies.
METHODS AND RESULTS
Genomic DNA was extracted from a fresh leaf bud of a mature T. platyphyllos tree at Chanstone Wood (52°0′52.36″N, 2°56′37.98″W) using a cetyltrimethylammonium bromide (CTAB) procedure (Morgan-Richards and Wolff, 1999). A microsatellite enrichment protocol was used, based on Edwards et al. (1996) and Squirrell and Wolff (2001). In short, 100 ng of purified DNA was digested with MboI (New England Biolabs, Beverly, Massachusetts, USA) and ligated to SauLA and SauLB linkers (Squirrell and Wolff, 2001). The ligated DNA was PCR amplified, and the product was hybridized at 47°C to nylon membranes with dot blots of (GA)15 and (CA)15 oligos. The eluted enriched DNA was PCR amplified and enriched for a second time, following the same method. The doubly enriched DNA was digested with MboI and ligated in pUC19 vector (QIAGEN, Hilden, Germany). The plasmids were transformed into competent E. coli cells (Bioline, London, United Kingdom). Recombinant colonies were selected by blue/white screening, and M13 PCR amplification was used to estimate the size of the inserts. Inserts with a size between 300 and 700 bp were sequenced using BigDye Terminator version 3.1 (Applied Biosystems, Foster City, California, USA). A total of 104 clones from the enriched library were sequenced on an ABI PRISM 3130 Genetic Analyzer (Applied Biosystems), and 96 clones contained microsatellite repeats with a minimum of five repeats. Inserts with 9–20 dinucleotide repeats were chosen for primer design. Thirty-one primer pairs were designed using Primer3 software version 0.4.0 using the default setting (Rozen and Skaletsky, 2000) but selecting primers longer than 21 nucleotides. The PCR amplification was performed in 10-µL multiplex reactions containing 5 ng of DNA, 1× reaction buffer, 2 mM MgCl2, 0.2 mM of each dNTP, 0.1–0.2 mM of each primer, and 0.5 U of Taq DNA polymerase (Bioline). Forward primers were labeled with FAM, HEX (Integrated DNA Technologies, Coralville, Iowa, USA), or NED (Applied Biosystems) (see Table 1 for multiplex sets, specific concentration of primers, and fluorescent label). The PCR reaction for all multiplex sets was as follows: an initial predenaturation step at 95°C for 5 min, followed by 15 cycles of 95°C for 15 s, 54°C for 15 s, and 72°C for 15 s, and 20 cycles of 89°C for 20 s, 52°C for 20 s, 72°C for 20 s, and a final extension step of 72°C for 30 min. Complete PCR reactions were diluted 1 : 10 with distilled water, and 1 µL of this dilution was mixed with 10 µL of Hi-Di Formamide (Applied Biosystems) and 0.1 µL of ROX-500 size standard before analysis on an ABI 3130xl Genetic Analyzer (Applied Biosystems). The data were analyzed by GeneMapper software (Applied Biosystems) and GenAlEx version 6 (Peakall and Smouse, 2006). Fifteen out of 31 primer pairs provided good patterns with the expected product size and were used for further characterization. The other 16 primer pairs failed to amplify targets, amplified nontarget sequences, or had nonspecific band patterns in preliminary tests and were discarded.
Polymorphism was studied in two populations of T. platyphyllos in France, one from Issole (Alpes de Haute Provence; 44°2′N, 6°49′E; N = 20) and the other from Gorges de la Carança (Pyrénées Orientales; 42°51′N, 2°22′E; N = 20). In addition, for testing transferability of microsatellite primers, leaves were collected from 23 species of this genus (52 individuals) from a living collection in Cartmel, Cumbria, United Kingdom (Pigott, 2002). No voucher specimens have been deposited because only leaves from ground level were available.
Characterization of 15 microsatellite markers developed in the genus Tilia.
All 15 loci were highly polymorphic in the samples analyzed. One locus (Tc943) was monomorphic within the Issole population, but was polymorphic in the Gorges de la Carança population. In these two populations, the number of alleles of the 15 loci ranged from one to 15, with a mean of 8.96. The observed and expected heterozygosities varied from 0.25 to 1.00 (average: 0.71) and 0.18 to 0.90 (average: 0.70), respectively. Significant departures from Hardy–Weinberg equilibrium (HWE) were detected at Tc918 and Tc963 in the Issole population (P < 0.05). For the Gorges de la Carança population, Tc4 and Tc920 deviated from HWE (P < 0.01), which may indicate population substructure or the presence of null alleles for those loci (Table 2). However, we did not find homozygous nulls in these populations.
Results of 15 microsatelite markers in two populations from France (Issole and Gorges de la Carança).
These 15 loci were also tested for their amplification in 23 species in the genus Tilia. Most of the loci (12 out of 15) were transferable and polymorphic in most of the related species (Appendix 1). Tc915 failed to amplify in two species (T. americana and T. caroliniana), and Tc920 failed to amplify in three species (T. ×euchlora, T. dasystyla, and T. endochrysea). Tc918 was successfully amplified in only five species (T. ×euchlora, T. ×europaea, T. dasystyla, T. paucicostata, and T. platyphyllos). Because Tc918 amplifies in T. platyphyllos but not in T. cordata, this locus may be useful for the identification of the two species and their hybrid.
These 15 highly polymorphic microsatellite markers will be useful tools for the study of population structure in T. platyphyllos and increase our understanding of their phylogeography and of the hybridization between Tilia species. Furthermore, the success of cross-amplification in related species will assist the future study of genetic diversity across the genus.