Lobelia inflata L. (Campanulaceae) is a herbaceous plant native to North America in the cosmopolitan genus Lobelia L., which contains more than 400 species (Lammers, 2011). Lobelia inflata is monocarpic and is capable of expressing either an annual or biennial life history; in the latter case, the plant overwinters as a frost-hardy rosette. Systematists have placed L. inflata in Lobelia sect. Lobelia, along with 20 other (primarily) North American species (Murata, 1992). Of these species, L. cardinalis L., L. siphilitica L., L. kalmii L., L. nuttallii Roem. & Schult., L. spicata Lam., and L. dortmanna L. all coexist sympatrically with L. inflata in the northeastern United States and Canada. The phylogenetic relationships between these species are not well understood, but morphological analysis has suggested that L. inflata is most closely related to L. cardinalis, L. siphilitica, and L. dortmanna (Lammers, 2011).

Populations of L. inflata consist of myriad distinct genetic lineages, which are expected to be reproductively isolated from one another. This is because L. inflata is assumed to be obligately autogamous—i.e., it is incapable of outcrossing— because it possesses a closed anther tube, which permits pollen release only onto the stigma of the same flower. As such, plants will produce offspring that are genetically identical to the parent, and therefore heterozygosity is expected to be zero in natural populations (Simons and Johnston, 2006). Although quantitative genetic variation is present for some traits, the extent of genetic variation among L. inflata ecotypes is unknown (Simons and Johnston, 2000). Highly variable molecular labels (i.e., polymorphic microsatellite loci) would permit the genotyping of reproductive lineages and the tracking of gene flow among populations. In this study, we characterized 28 microsatellite loci for L. inflata, tested these markers in 58 individuals from three eastern North American populations, and assessed cross-amplification success in two congeneric species within Lobelia sect. Lobelia: L. cardinalis and L. siphilitica.

METHODS AND RESULTS

Genomic DNA was extracted from a single L. inflata specimen collected from the Petawawa Research Forest (individual DAO-887897; see Appendix 1). To obtain high-quality DNA for enrichment, total genomic DNA was extracted using the cetyltrimethylammonium bromide (CTAB) extraction method of Murray and Thompson (1980). We prepared a CT microsatellite-enriched DNA library using the method of Hamilton et al. (1999) and linker sequences of Glenn and Schable (2005). Ninety-six clones from this library were sequenced using the BigDye 3.1 kit on an ABI 3730 DNA analyzer (Applied Biosystems, Carlsbad, California, USA). Twenty-eight clones contained dinucleotide motifs with at least seven repeat units and were used to design primer pairs using the Primer3Plus software package (Untergasser et al., 2007). Selection of the final 28 primer sets was based on: (1) reliable and repeatable amplification by PCR (using the PCR parameters listed below); and (2) distinct banding pattern visualization on 3% agarose gel. All 28 primer pairs met these inclusion criteria and were used to assess genetic diversity.

To assess amplification, we performed PCR in 20-µL reaction volumes, using the Phire II Direct PCR Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Unlike two-step protocols that require separate DNA isolation and PCR steps, direct PCR allows amplification directly from plant material (Bellstedt et al., 2010). Here, the genomic DNA template for the direct PCR reaction was taken from a circular punch of dried leaf or fruit capsule material using a 0.35-mm Harris Uni-Core Micro-Punch (Thermo Fisher Scientific). Each 20-µL reaction contained 0.5 µM primers and 1.5 mM MgCl₂. A standard three-step PCR protocol was used, including an initial denaturation at 98°C for 5 min, followed by 30 cycles of denaturation at 98°C for 5 s, annealing at 50–58°C for 5 s, extension at 72°C for 20 s, and a final extension step of 72°C for 10 min. The annealing temperatures for each primer set used in this study are given in Table 1. The PCR reactions were performed using a T-3000 ThermoCycler (Biometra, Goettingen, Germany). We initially screened primer sets for polymorphism using 58 individuals from three L. inflata populations: Petawawa, Ontario (n = 43); Martock, Nova Scotia (n = 8); and Petersham, Massachusetts (n = 7). We then tested cross-amplification of these primer sets in L. cardinalis (n = 12) and L. siphilitica (n = 3) (see Appendix 1 for collection location and voucher deposition information). Amplicon size was determined using a GeneRuler 100-bp DNA Ladder (Thermo Fisher Scientific). Gel images for all samples were compared via Sequentix GelQuest (Sequentix Digital DNA Processing, Klein Raden, Germany) from 3% agarose gel electrophoresis (run at 60 V for 90 min).

Genetic diversity parameters for the three populations are presented in Table 2. Among L. inflata individuals, 24 loci were polymorphic, with two to four alleles per locus. Four of the loci (Linflatal, Linflata6, Linflata11, and Linflata24) were monomorphic in all L. inflata individuals tested. Notably, some loci showed substantial differences in allele size; for example, two alleles were found for Linflata5 in the Petawawa population—one was 215 bp and the other 331 bp. Although the variability of simple sequence repeats (SSRs) undermines their usefulness as precise molecular clocks, in the absence of outcrossing, differences in SSR allele size may be proportional to lineage divergence time (Neff, 2004). We used GENEPOP version 4.2 to test linkage disequilibrium and Hardy-Weinberg equilibrium (Rousset, 2008). Observed heterozygosity was zero across all loci in L. inflata, a finding that supports the hypothesis that outcrossing is rare or nonexistent in field populations of L. inflata.

Table 1.

Characteristics of microsatellite markers developed for Lobelia inflata.

Table 2.

Allele numbers and cross-amplification for Lobelia inflata microsatellite loci. Data for L. inflata includes 58 individuals from three populations.

Cross-amplification success of these 28 loci in L. cardinalis and L. siphilitica is also presented in Table 2. There was successful amplification at 16 loci for L. cardinalis, with two to three alleles per locus, and at 11 loci for L. siphilitica, with one to two alleles per locus. For L. cardinalis, we found three loci (Linflata2, Linflata12, and Linflata26) to be polymorphic. We found no polymorphism for any loci in L. siphilitica.

CONCLUSIONS

These microsatellite markers are the first to be developed for L. inflata, and offer a new opportunity to investigate allelic diversity and gene flow within and among populations of L. inflata. Observed homozygote excess is likely due to lack of gene flow between L. inflata reproductive lineages. Successful cross-amplification of these loci in both L. cardinalis and L. siphilitica suggests that these loci may also be used to assess genetic diversity in congeneric species.