Introduction
The genus Narcissus L. (Amaryllidaceae) comprises about 60–80 species (Marques et al., Reference Marques, Aguilar, Martins-Louçao, Moharrek and Feliner2017) and is one of the most important floricultural crops in which more than 27,000 cultivar names have been registered (Hanks, Reference Hanks and Hanks2002; Kington, Reference Kington2008). Furthermore, Narcissus is a model system in the study of floral evolution. This genus has been used to understand the role of perianth traits, pollinators and self-incompatibility in the evolution and maintenance of stylar polymorphisms (Herrera, Reference Herrera1995; Barrett and Harder, Reference Barrett and Harder2005; Santos-Gally et al., Reference Santos-Gally, Gonzalez-Voyer and Arroyo2013; Simón-Porcar et al., Reference Simón-Porcar, Meagher and Arroyo2015). Narcissus dubius Gouan, N. cuatrecasasii Fern. Casas & al., N. assoanus Dufour ex Schult. & Schult. fil. and N. rupicola Dufour ex Schult. & Schult. fil. are stylar-dimorphic species that occur in the Iberian Peninsula and south east of France (online Supplementary Fig. S1). These species exhibit differences in perianth traits, pollinators, breeding systems and variation in morph ratios (Baker et al., Reference Baker, Thompson and Barrett2000; Pérez-Barrales et al., Reference Pérez-Barrales, Vargas and Arroyo2006; Santos-Gally et al., Reference Santos-Gally, Gonzalez-Voyer and Arroyo2013). Here, we characterized 20 microsatellite loci that will be used to study the stability of stylar dimorphism in these species through paternity analyses. The estimation of mating patterns will be critical for testing models on the evolution of stylar dimorphism (Lloyd and Webb, Reference Lloyd, Webb and Barrett1992; Simón-Porcar et al., Reference Simón-Porcar, Meagher and Arroyo2015).
Experimental
We selected 40 new primer pairs from microsatellite libraries developed for N. papyraceus in Simón et al. (Reference Simón, Picó and Arroyo2010) and we also tested 12 microsatellite loci characterized for N. papyraceus in the same study (52 primer pairs overall). Microsatellite libraries were provided by Genetic Identification Services (www.genetic-id-services.com), according to Jones et al. (Reference Jones, Levine and Banks2002). We collected samples from one population of N. dubius (Sant Mateu, Castellón, Spain; 40°26′N, 0°12′E; SEV 287078), N. cuatrecasasii (Sierra Mágina, Jaén, Spain; 37°44′N, 3°31′W; SEV 287080) and N. rupicola (Plataforma de Gredos, Ávila, Spain; 40°17′N, 5°13′W; SEV 287081), and two distant populations of N. assoanus (Sierra de Rute, Córdoba, Spain; 37°19′N, 4°20′W; SEV 287077, and Saint-Pierre-de-la-Fage, Hérault, France; 43°47′N, 3°29′E; SEV 287079) to test the primer pairs. For a preliminary test of amplification and polymorphism, we used 10 individuals per population (50 individuals overall). Afterwards, we further investigated the variability of polymorphic loci by genotyping 427 more individuals, for a total of 63–121 individuals per population (Table 1).
Results of primer screening in a population of Narcissus dubius, N. cuatrecasasii and N. rupicola, and two populations of N. assoanus
N, sample size; A, number of alleles; H o, observed heterozygosity; H e, gene diversity.
*Significant departure from Hardy–Weinberg equilibrium (P < 0.05).
+Primer developed for N. papyraceus in Simón et al. (Reference Simón, Picó and Arroyo2010) that presented polymorphism.
Silica-dried leaves of Narcissus spp. were used to extract genomic DNA with the Insorb spin plant mini kit (Stratec Molecular, Berlin, Germany). Polymerase chain reactions were performed in 20 µl of mixture containing 50 ng/μl of template genomic DNA, 1 × My Taq Red Reaction Buffer (Bioline, London, UK), 0.5 U taq polymerase, 0.40 dye-labelled M13 primer (FAM, VIC, NED or PET dyes; Invitrogen, Madrid, Spain), 0.40 µM PIG-tailed reverse primer, 0.04 µM M13-tailed forward primer (Boutin-Ganache et al., Reference Boutin-Ganache, Raposo, Raymond and Deschepper2001) and 0.01% bovine serum albumin (Promega, Madison, WI, USA). Amplification of DNA was performed in a Touch-Down PCR on a Veriti 96-Well Fast Thermal Cycler (Applied Biosystems, Foster city, CA, USA), with an initial denaturalization at 94°C for 5 min; nine cycles with denaturalization at 94°C for 30 s, annealing at 59–51°C for 30 s (1°C decrease in each cycle) and extension at 72°C for 30 s; 36 cycles at 94°C for 30 s, 50°C for 30 s, 72°C for 30 s; and a final extension at 72°C for 5 min. Polymerase chain reaction products were analysed in ABI prism 3130 and 3730 systems (STAB VIDA, Portugal) and sized using Gene Marker 2.4 (SoftGenetics, State College, PA, USA) and Gene Scan 500 LIZ size standard.
A total of 20 primer pairs amplified (Table 1), 17 of them were from the set of new primer pairs (online Supplementary Table S1) and three of them were transferred from the set available for N. papyrareus (Table 1). The number of alleles per locus (A), observed heterozygosity (H o), gene diversity (H e) and P-values for the Hardy–Weinberg equilibrium (HWE) were estimated using GenAlEx v.6.5 (Peakall and Smouse, Reference Peakall and Smouse2012). Tests for linkage disequilibrium were performed with FSTAT v.2.9.3 software with 100,000 permutations (Goudet, Reference Goudet1995).
Discussion
Four loci were polymorphic in one Narcissus species at least. In N. dubius, eight loci were polymorphic, with mean A = 5.63 (range 2–13), H o = 0.377 (range 0.107–0.667) and H e = 0.374 (range 0.103–0.864). In N. cuatrecasasii, there were five polymorphic loci, with mean A = 10.8 (range 7–11), H o = 0.351 (range 0.143–0.560) and H e = 0.652 (range 0.399–0.871). In N. assoanus, there were five polymorphic loci, with mean A = 10.8 (range 5–19), H o = 0.526 (range 0.311–0.729) and H e = 0.663 (range 0.380–0.848), for the population in Córdoba (Spain), and with mean A = 7.4 (range 4–16), H o = 0.306 (range 0.111–0.426) and H e = 0.616 (range 0.294–0.894) for the population in Hérault (France). In N. rupicola, there were four polymorphic loci, with mean A = 13.7 (range 11–16), H o = 0.347 (range 0.312–0.562) and H e = 0.799 (range 0.776–0.828) (Table 1). Tests for linkage disequilibrium showed that all the loci are independent (P < 0.0008 for all the loci). According to Simón-Porcar et al. (Reference Simón-Porcar, Meagher and Arroyo2015), in which four polymorphic loci were used for paternity analyses in N. papyraceus (A = 10–15 and H e = 0.482–0.885), the polymorphic loci characterized in the present study can be used to perform paternity analyses in N. dubius, N. cuatrecasasii, N. assoanus and N. rupicola. Such studies will be essential to understand the role of pollinators, perianth traits and self-incompatibility on the stability of stylar polymorphism. Other studies aiming similar prospects have used proxies for estimating assortative/disassortative mating patterns (Hodgins and Barrett, Reference Hodgins and Barrett2008; Pérez-Barrales and Arroyo, Reference Pérez-Barrales and Arroyo2010). Most of the polymorphic markers presented significative departure from the HWE, which might confirm a prevalence of assortative or disassortative mating instead of random mating in stylar dimorphic species (Simón-Porcar et al., Reference Simón-Porcar, Meagher and Arroyo2015). Finally, these markers can be used to conduct studies of genetic diversity and probably they could also be transferred to closely related Narcissus species (at least species from sections Apodanthi, Juncifolii and Dubii), providing valuable resources for studies of floral evolution in the genus.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262118000412.
Acknowledgements
The authors thank A. de Castro, D. Álvarez, J. Barranco and R. Santos-Gally for field and laboratory assistance and the Molecular Laboratory facilities of the Herbarium service (CITIUS-2) at the University of Seville. This research was supported by grants of the Spanish Ministry of Economy, Industry and Competitiveness (CGL2013-45037-P, CGL2009-12565). DB was supported by fellowships from the Spanish Ministry of Economy, Industry and Competitiveness (BES-2014-067795).
