Pokeweed, Phytolacca americana L. (Phytolaccaceae), is a large, weedy, perennial herb native to eastern North America but widely distributed in Asia and Europe. With high seed production and bird-dispersed fruits, the plant establishes readily in disturbed habitats (McDonnell et al., 1984). Pokeweed was likely intentionally introduced to China and first reported in 1935 in Zhejiang Province (Li and Xie, 2002; Xu et al., 2012). Recent reports in Asia indicate that P. americana may have emerged as a more aggressively invasive species where it has established (Kim et al., 2005). The species has recently become a significant threat to coastal forest ecosystems in China (Zhai et al., 2010; Fu et al., 2012). Furthermore, P. americana has largely displaced a Chinese native congener, P. acinosa Roxb., in parts of China. Although P. acinosa has been a historic part of Chinese pharmacopoeias for more than 2000 years, it is similar in appearance to P. americana, which in turn is considerably more toxic than P. acinosa (Kim et al., 2005), leading to the possibility of accidental poisonings. Despite its widespread distribution and emergence as an invasive species, many aspects of pokeweed evolutionary ecology are not well understood, including its breeding system. The microsatellites we have developed could be valuable tools to understand pokeweed population structure and invasion history.
METHODS AND RESULTS
We extracted genomic DNA from freshly collected leaves of one individual of P. americana from Memphis, Tennessee, using a QIAGEN DNeasy Plant Mini Kit (QIAGEN, Valencia, California, USA). We mined 1,100,538 Illumina MiSeq reads from genomic DNA from that individual for microsatellites and flanking primers using the PAL Finder pipeline (Castoe et al., 2012). We added a CAG sequence to the 5′ end of one of each of the primer pairs to facilitate use of a third, fluorescently labeled primer in the PCR.
To screen the primers, we collected leaf samples from populations both in the native and introduced range of P. americana: Florida, USA; Illinois. USA; Anhui, China; and Fukushima, Japan. We also collected leaf samples from a single population of the native P. acinosa in Jiangsu, China (GPS coordinates 32.06°N, 118.83°E). Collection coordinates and voucher information are provided in Appendix 1. Leaves were dried, frozen, and lysed before genomic DNA was extracted as described above. We tested 58 primer pairs for amplification; 11 produced consistent results for P. americana (Table 1). We were able to amplify only a single locus (PW65) for P. acinosa, where 12 of the 21 samples yielded products in the expected size range given data from P. americana. We amplified each locus in a 12.5-µL PCR reaction (10 mM Tris-HCl, 50 mM KCl, 0.1% Triton X-100, 100 µg/mL bovine serum albumin [BSA], 0.4 µM unlabeled primer, 0.04 µM tag-labeled primer, 0.36 µM universal dye-labeled primer, 4.0 mM MgCl2, 0.8 mM dNTPs, 0.25 units EconoTaq Polymerase [Lucigen, Middleton, Wisconsin, USA], and 10 ng DNA template) using a touchdown protocol where the annealing temperature decreased by 0.5°C for each of the first 20 cycles: 96°C (2.5 min), then 20 cycles of 96°C (30 s), initial 65°C (30 s), 72°C (30 s), then 20 cycles: 96°C (30 s), 50°C (30 s), 72°C (30 s), then 72°C (10 min). Initial annealing temperature was 60°C for PW182. After PCR, amplified fragments were diluted 1 : 5 and pooled in the following four groups: PW106, PW29, and PW43; PW54, PW46, and PW79; PW69, PW11, and PW223; and PW182 and PW65. Fluorescent dyes used to label each primer can be found in Table 1. Amplicons were visualized on an ABI 3730xl DNA sequencer (Applied Biosystems, Carlsbad, California, USA) and sized with an internal ROX-labeled size standard (GGF500R; Georgia Genomics Facility, Athens, Georgia, USA). Genotyping results were scored using GeneMarker software (version 2.4; SoftGenetics, State College, Pennsylvania, USA). Alleles were binned using the program MsatAllele (Alberto, 2009) according to the core repeat size of the microsatellite.
Characteristics of 11 microsatellite loci developed in Phytolacca americana.
Data were checked for errors and null alleles using MICRO-CHECKER (van Oosterhout et al., 2004). No errors were detected. Possible null alleles were detected in the Florida and Illinois populations for PW46, in the Illinois and Japanese populations for PW223 and PW106, and in the Illinois population for PW79. Deviation from Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were tested in GENEPOP (version 4.2; Rousset, 2008), both with default parameters and Bonferroni corrections. Genetic diversity measures, including the number of alleles and observed and expected heterozygosities, were estimated in GenAlEx (version 6.501; Peakall and Smouse, 2006, 2012).
Genetic diversity of 11 newly developed microsatellites in two native (United States) and two invasive (Asia) populations of Phytolacca americana.a
Because populations were largely monomorphic for one allele, only eight of the 44 possible tests were subject to both LD and HWE analyses. Overall, seven of the eight loci deviated from HWE, but none of the eight loci pairs were in LD (P > 0.05). The number of alleles per locus ranged from two to six and observed heterozygosity ranged from 0.00 to 1.00, although average observed heterozygosity was only 0.03 (Table 2). The breeding system of P. americana has never been adequately described. Armesto et al. (1983) speculated that the species was autogamous because of its high fruit set. Caulkins and Wyatt (1990) bagged inflorescences while still in bud and 60% of fruits set, but bagged emasculated flowers (which should not set any fruit) set 46% of their fruit, making their results difficult to interpret. A lack of within-population genetic variation does suggest that P. americana may be highly selfing. The allelic differences between populations are also consistent with autogamy. Although our markers cannot be used to distinguish between individuals of the same population, they can distinguish between individuals from different populations.
We report 11 microsatellite loci in P. americana from both the native range in the United States and the invasive range in Asia. The genetic diversity data suggest that this species may be highly selfing, although a more detailed examination of the breeding system of this species is warranted. Even though these markers will be of limited use within populations, they should be useful for studies across populations, including those tracing the invasion history of this species. Only one marker was able to amplify any product in the Chinese native congener, P. acinosa, suggesting high genetic differentiation between the two species.
 The authors thank M. M. Mauricio and H. A. Dahn for collecting field samples, H. Donaldson and M. Marks for assistance in screening primers, U. Bagal for help in analyzing data, and an anonymous reviewer for comments. This work was supported by the National Science Foundation Partnership for International Research and Education (PIRE) program (OISE 0730218). K.E.B. was supported by a National Science Foundation Graduate Research Fellowship (DGE0903734).