Hypochaeris incana (Hook. & Arn.) Macloskie (Asteraceae, Cichorieae) is a rosulate perennial herb that may propagate by underground stolons. It inhabits the Patagonian steppe of southern South America and extends its range to the subantarctic southernmost part of the continent in Tierra del Fuego. The species includes diploid, triploid, and tetraploid cytotypes. Peculiarly, diploids occur in the southern part of its range and tetraploids in the northern part of its range, but H. incana seems to have originated in the north (Tremetsberger et al., 2009). Tremetsberger et al. (2009) suggested that tetraploids may have repeatedly replaced their diploid progenitors in the northern part of the range. The factors involved in the establishment of polyploid cytotypes, however, are still poorly understood. We developed microsatellite primers for H. incana to investigate the competitive abilities of diploids and tetraploids in terms of their clonal growth strategies (discrimination between genets and ramets). We also tested the primers in the close relatives H. acaulis (J. Rémy) Britton, H. hookeri Phil., H. palustris (Phil.) De Wild., and H. tenuifolia (Hook. & Arn.) Griseb. to study the possible relationship between interspecific gene flow and the origin of the polyploid cytotypes.
METHODS AND RESULTS
We extracted genomic DNA from leaf material of H. incana and related species dried on silica gel in the field with the DNeasy Plant Mini Kit (QIAGEN, Hilden, Germany; Appendix 1). The ploidy level of all individuals of the Cerro La Buitrera population of H. incana and of a few other populations was determined by flow cytometry (C. König, unpublished data; see Appendix 1). The ploidy level of the remaining populations was retrieved from Weiss et al. (2003), Weiss-Schneeweiss et al. (2007), and Tremetsberger et al. (2009) and/or inferred from microsatellite peak patterns. One diploid individual of the Cerro La Buitrera population of H. incana was sequenced on a GS FLX Titanium sequencer (454 Life Sciences, a Roche Company, Branford, Connecticut, USA) at LGC Genomics (Berlin, Germany). The mean length obtained for the 180,338 sequences was 1048 bp (range = 50-1780 bp; National Center for Biotechnology Information [NCBI] Sequence Read Archive BioProject no. PRJNA314301). The methodology for primer development followed Böckelmann et al. (2015) with slight modifications as outlined below. MSATCOMMANDER version 0.8.2 (Faircloth, 2008) identified 2466 sequences with microsatellite motifs with the following options: di-, tri-, and tetranucleotide repeats ≥6 repeat units, combine multiple arrays within a sequence if within 50 bp distance. Primers for a total of 838 microsatellite loci were designed using Primer3 implemented in MSATCOMMANDER (Rozen and Skaletsky, 1999). A CAG or M13R tail (CAG: 5′-CAGTCGGGCGTCATCA-3′; M13R: 5′-GGAAACAGCTATGACCAT-3′) was added to the 5′ end of one primer (Schuelke, 2000) and a GTTT PIG-tail was added to the 5′ end of the other primer (Brownstein et al., 1996). OLIGO 7 (Rychlik, 2007) was used to reevaluate the quality of primers, and 75 primer pairs were selected for the subsequent preliminary trial on seven individuals of H. incana and eight individuals from the congeneric species (three individuals of H. hookeri, three individuals of H. tenuifolia, and two individuals of H. palustris; Appendix 1). The PCR mix for amplification (total volume 12.5 µL) contained: 6.25 µL of JumpStart REDTaq ReadyMix (Sigma-Aldrich, St. Louis, Missouri, USA), 0.25 µL of GTTT-tailed primer, 0.05 µL of CAG- or M13R-tailed primer. 0.25 µL of 5′ FAM-labeled universal CAG or M13R primer, and 0.5 µL of diluted DNA extract. The concentration of the primers was 10 pmol/µL (10 µM). A touchdown PCR protocol was used. The cycling conditions were: 95°C for 5 min (initial denaturation); 17 cycles with 95°C for 45 s (denaturation). 58–50°C for 90 s (annealing with a 0.5°C decrease per cycle), and 72°C for 60 s (extension); 20 cycles with 95°C for 45 s, 50°C for 90 s, and 72°C for 60 s; and 72°C for 5 min and 60°C for 30 min (final extension). PCR products were separated on a 3730xL DNA Analyzer (Applied Biosystems, Foster City, California, USA) at Microsynth (Balgach, Switzerland), and fragment sizes were estimated with GeneMarker 2.4 (SoftGenetics, State College, Pennsylvania, USA). Of the 75 microsatellite loci tested, 15 were clearly interpretable and polymorphic and were therefore selected for further study. The primers without the GTTT PIG-tails were labeled with a fluorescent dye at their 5′ end rather than with the previously used CAG or M13R tail and were used in multiplex PCR reactions (Table 1) to amplify a larger number of individuals of the five species. PCR was performed in a total volume of 20 µL containing 10 µL of JumpStart REDTaq ReadyMix, 0.4 µL of forward primer and 0.4 µL of reverse primer (each at a concentration of 10 µM) of each primer pair entering in the multiplex reaction, and 1 µL of diluted DNA extract, using the same cycling protocol described above. The PCR products were analyzed and scored as described above. In most cases, genotype assignment was unambiguous for diploid, triploid, and tetraploid cytotypes (Tables 2, 3).
Characteristics of the 15 polymorphic microsatellite markers developed for Hypochaeris incana and related species.a
Genetic variation of the 15 polymorphic microsatellite markers in three populations of Hypochaeris incana. a
Cross-species amplification of the 15 polymorphic microsatellite markers developed for Hypochaeris incana in four related species.a
We checked for the presence of null alleles in the two purely diploid populations as well as in the diploids of the mixed ploidy population (N = 14) of H. incana using the software MICRO-CHECKER version 2.2 with default settings (van Oosterhout et al., 2004). Three loci showed significant evidence of the presence of a null allele in all three populations (Table 2); for these loci, we adjusted diploid homozygous genotypes of H. incana by setting the state of the second allele to missing and adjusted tetraploid homozygous genotypes by setting the states of the third and fourth alleles to missing. One heterozygous triploid and one heterozygous tetraploid genotype demonstrated the suspected presence of a null allele based on peak heights; these were adjusted by setting one allele as missing in each case. Observed heterozygosity (Ho) and inbreeding coefficient (F IS) are not reported for these loci. The number of alleles per locus, Ho, expected heterozygosity (He), and FIS were calculated using SPAGeDi 1.5 (Hardy and Vekemans, 2002) by entering all (i.e., diploid, triploid, and tetraploid) individuals. All of the 15 microsatellite loci showed polymorphisms among the three populations of H. incana (Table 2). The number of alleles per locus and population ranged from two to 29. He and Ho ranged from 0.382 to 0.959 and 0.296 to 1.000, respectively. F IS ranged from −0.168 to 0.499. Most of the 15 newly developed markers were successfully amplified and scored in the four congeneric species (Table 3). To assess the power of the markers to discriminate among species, we produced a Neighbor-Net split network based on a matrix of Roussel's (2000) interindividual differentiation with the software SplitsTree4 version 4.14.5 (Huson and Bryant, 2006) and performed a Bayesian admixture clustering analysis using the software Structure version 2.3.4 (Pritchard et al., 2000) assuming independent allele frequencies among populations. For each K from 2 to 13, we requested five independent runs with a burn-in period of 100,000 and 500,000 subsequent repetitions of the simulation. A typical run with K = 6 perfectly distinguished among species as well as between the two southern populations and the northern population of H. incana, with some indication of admixture in H. tenuifolia ( Appendix S1 (apps.1700081_s1.pdf)).
We developed 15 polymorphic microsatellite markers for H. incana, which also worked well in some of the analyzed congeneric species. These 15 primer pairs will be suitable for studying the population clonal structure, genetic diversity, phylogenetic relationships, and interspecific hybridization in H. incana and its closest relatives.
The authors thank all collectors of plant material; A. Calvo (Bariloche) for permission to collect on his property; and J. Böckelmann, C. König (both Vienna), and A. López (San Isidro) for help with flow cytometric measurements and marker development. Financial support was provided by the Ministerio de Ciencia, Tecnología e Innovación Productiva (MINCyT, Argentina; project AU/10/16 to E.U.), the Österreichische Austauschdienst (OeAD, Austria; project AR 27/2011 to K.T.), a Eurasia-Pacific Uninet scholarship to P.W., and the University of Natural Resources and Life Sciences, Vienna.