Experimental
Common lilac is an ornamental shrub cultivated in Western Europe since mid-16th century, mainly because of its fragrant, showy flowers. Lilac growers and breeders, aimed at increasing the size and colour assortment of flowers, have introduced nearly 2000 common lilac cultivars (Fiala and Vrugtman, Reference Fiala and Vrugtman2008), most of them without adequate description, which makes cultivar verification on the basis of morphological traits extremely difficult.
Molecular techniques, such as microsatellite (simple sequence repeat, SSR) markers, provide a complementary approach for characterization and identification of genotypes. In the Oleaceae family, SSR markers have been developed for several species (De La Rosa et al., Reference De La Rosa, James and Tobutt2002; Harbourne et al., Reference Harbourne, Douglas, Waldren and Hodkinson2005; Kodama et al., Reference Kodama, Yamada and Maki2008), but we found only one report on transferring these markers to the genus Syringa. De La Rosa et al. (Reference De La Rosa, James and Tobutt2002) published seven microsatellite primer pairs for olive (Olea europaea L.), four of which amplified markers also in common lilac, but only two showed some level of polymorphism. Rzepka-Plevneš et al. (Reference Rzepka-Plevneš, Smolik and Tańska2006) employed inter-SSR (ISSR) markers in a study of genetic distances between different Syringa species, but they did not investigate within-species variation. Furthermore, ISSR markers are not well suited for genotyping of cultivars due to their low repeatability.
The aim of our study was to develop cultivar-specific SSR markers for common lilac. The genome screening technique of Korpelainen et al. (Reference Korpelainen, Kostamo and Virtanen2007) was applied for finding the markers. The markers were tested by genotyping 75 common lilac samples, including 58 local shrubs previously studied for their morphological traits (Lindén et al., Reference Lindén, Hauta-aho, Temmes and Tegel2010) and 17 reference accessions that represented named cultivars and were obtained from botanical collections. The plant material is described in Table S1 (available online).
Genomic DNA was extracted from leaf tissue using the CTAB protocol of Doyle and Doyle (Reference Doyle and Doyle1990) and the concentration of DNA was quantified with a spectrophotometer. Polymerase chain reaction (PCR) was performed as described in Korpelainen et al. (Reference Korpelainen, Kostamo and Virtanen2007) with 31 single ISSR primers. After cloning and sequencing, specific SSR primers were designed for flanking regions of the microsatellite repeats found within the sequenced ISSR amplification products. Forward primers were fluorescently labelled with FAM™, HEX™ or TET™ dyes.
PCR amplifications were conducted in a volume of 10 μl containing 1 μl (about 20 ng) genomic DNA, 1 × reaction buffer with 1.5 mM MgCl2 −, 0.2 μl of 10 mM dNTP mixture, 0.3 μl DyNAzyme II DNA polymerase (Finnzymes) (2U/μl) and 5 pmol of both primers. The mastercycler gradient (Eppendorf AG, Hamburg, Germany) was programmed for 5 min with denaturation at 95°C, followed by 34 cycles of denaturation at 95°C for 40 s, annealing at 53–66°C for 40 s (Table 1), elongation at 72°C for 1 min and a final extension at 72°C for 10 min. The PCR products were diluted to 1:2–1:4, and detected by capillary electrophoresis using the MegaBACE™-1000 DNA sequencer (Amersham Biosciences Ltd., Little Chalfont, Buckinghamshire, UK), with ET400-R as an internal standard. The genetic distances were calculated with GenAlEx 6 (Peakall and Smouse, Reference Peakall and Smouse2006). Cluster analysis was conducted using MEGA version 4 (Tamura et al., Reference Tamura, Dudley, Nei and Kumar2007) with unweighted pair-group method algorithm (UPGMA).
Table 1 Characteristics of the nine novel microsatellite markers developed for common lilac (Syringa vulgaris L.)
T a, annealing temperature; H o, observed heterozygosity; H e, expected heterozygosity.
With 31 ISSR primers and one to three DNA samples, a total of 96 sequenced clones in the size range of 240–1100 bp were obtained. The number of sequences per primer was 3.8 for GA/AG, 3.6 for CA/AC, 2.5 for CT/TC and 1.5 for GT/TG. Microsatellite repeats were identified in 48 (50%) of the sequences. Nine (19%) of the repeats were perfect microsatellite repeats, while 39 (81%) were imperfect. Dinucleotide SSRs were the most prevalent and among them, the AG/CT repeats were the most frequent. Additional data on the SSR markers are given in Table S2 (available online).
We developed altogether 20 specific SSR markers, nine of which showed polymorphism. The marker SV2 (Table 1) amplified two loci, but one of them was discarded due to the diffuse amplification product. For each polymorphic locus, two to five alleles were detected. The most informative marker was SV9 (Table 1), which possessed five alleles and amplified successfully in all but one of the 75 accessions. SV2 and SV3 did not amplify in 9 and 15 accessions, respectively, while the number of failures for the remaining six markers was 0 to 4. In 69 accessions (92%), at least eight markers rendered amplification products.
Discussion
All reference samples were clearly separated and identified using the nine novel microsatellite markers. In the dendrogram (Fig. 1), the local shrubs were grouped in a manner that was generally consistent with the grouping based on morphological traits. The shrubs named as ‘Mme Lemoine’, ‘Prince Notger’, ‘Krasavitsa Moskvy’ or ‘Michel Buchner’ on the basis of morphological characters clustered well together with their reference cultivars. A few samples, assumed to represent the same genotype, were clustered apart in the dendrogram (Fig. 1). The discrepancies were mainly due to missing amplification products. It is also conceivable that some of the sample plants were misidentified or mislabelled.
Fig. 1 UPGMA dendrogram of the 75 common lilac (Syringa vulgaris L.) shrubs based on a genetic distance matrix calculated for codominant data (GenAlEx), using the nine SSR loci. The reference samples, obtained from botanical collections, are indicated in bold.
There are few molecular marker studies published on common lilac, which are mainly focused on cultivars other than those studied by us. However, in a study of Smolik et al. (Reference Smolik, Andrys, Franas, Krupa-Malkiewicz and Malinowska2010) on rDNA polymorphism in Syringa, three cultivars common to our study were included. Our results are in accordance with their findings showing that ‘Katherine Havemeyer’ and ‘Mme Lemoine’ group relatively close together, whereas ‘Krasavitsa Moskvy’ is placed in a different and more distant cluster.
The lilac monograph by McKelvey (Reference McKelvey1928) states following relationships between our reference cultivars: ‘Mme F Morel’ is a seedling of ‘Marlyensis’, ‘Michel Buchner’ belongs to the offspring of ‘Lemoinei’, ‘Mme Lemoine’ was born by crossing ‘Marie Legraye’ with a double form of common lilac and ‘Hugo Koster’ came out of a lot obtained by hybridizing ‘Marie Legraye’, ‘Président Grévy’ and ‘Andenken an Ludwig Späth’. These data match fairly well with the results obtained by the novel SSR markers, except for ‘Mme Lemoine’ (Fig. 1).
In conclusion, the nine novel microsatellite markers proved valuable for distinguishing common lilac cultivars.
Supplementary material
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S1479262113000166
Acknowledgements
The authors thank the Arnold Arboretum of Harvard University, the Helsinki University Botanic Garden and the Montreal Botanical Garden for kindly providing the reference samples for this study. We are also thankful to the Nikolai and Ljudmila Borisoff Horticultural Foundation as well as to the Maiju and Yrjö Rikala Horticultural Foundation for their financial support.
