Black needlerush (Juncus roemerianus Scheele; Juncaceae) is a clonal, gynodioecious macrophyte found in salt marshes from the mid-Atlantic in Maryland and Delaware to the western coast of the Gulf of Mexico in Texas (Godfrey and Wooten, 1979). The species has a high salt tolerance and dominates areas of low tidal flux, such as the Gulf Coast, forming large monotypic stands through sexual and clonal reproduction (Eleuterius, 1984). Juncus roemerianus is an ecosystem engineer and forms the foundation of the salt marsh, creating habitat for other marsh species by accumulating and stabilizing sediment (Pennings and Bertness, 2001). Genetic diversity of foundation species has an elevated importance in maintaining ecosystem health and resiliency in monotypic ecosystems such as salt marshes (Reusch and Hughes, 2006; Hughes et al., 2008). Restored macrophyte populations with higher genetic diversity are more resilient and have greater overall restoration success (Reynolds et al., 2012). Across the Gulf Coast, J. roemerianus habitat has been fragmented by human development and is vulnerable to future losses and degradation from pollution and sea level rise. Information on the genetic diversity and population structure of J. roemerianus is essential for salt marsh conservation.
While transplant studies suggest the existence of distinct populations of J. roemerianus, no molecular population genetic studies have been conducted on the species (Eleuterius, 1989). We address this need by developing and characterizing 19 microsatellite markers for J. roemerianus suitable for population studies. Microsatellites are highly variable and useful in characterizing the scale of population structure necessary for successful restoration and management.
METHODS AND RESULTS
Microsatellite markers were developed using an Illumina NextSeq sequencing system (Illumina, San Diego, California, USA). Genomic DNA was extracted from a leaf sample collected from the same site as the voucher specimen at the Grand Bay National Estuarine Research Reserve (NERR) in Moss Point, Mississippi, USA, using a QIAGEN DNeasy Plant Maxi Kit (QIAGEN, Hilden, Germany) (Appendix 1). Library preparation was completed using a KAPA LTP Library Preparation Kit (KAPA Biosystems, Wilmington, Massachusetts, USA) for Illumina platforms following the manufacturer's protocol. Reads totaling 872,449 sequences were paired by name using Geneious version 8.1.2 (Kearse et al., 2012) and archived in the GenBank Sequence Read Archive (SRR5076849). Illumina TruSeq adapters and bases with an error probability limit above 0.04 were trimmed, and de novo assembly was performed on sequences greater than 150 bases. Unused reads were extracted to MSATCOMMANDER version 1.0.8beta (Faircloth, 2008) and queried for microsatellite loci. MSATCOMMANDER identified 4237 loci with perfect repeats of 3–6 nucleotides using default minimum lengths and melting temperatures, and combining loci less than 50 bp apart. Five hundred and two loci had unique sequences surrounding the repeats with sufficient length for primer design. We selected 96 primer pairs with a pair penalty assigned by Primer3 (Rozen and Skaletsky, 1999) below six that had a diversity of repeat lengths and nucleotide motifs. One primer for each locus was tagged with either a CAG (CAGTCGGGCGTCATCA) or M13 (GGAAACAGCTATGACCAT) sequence addition to the 5′ end. Identical nucleotide matches between the 3′ end of the tag sequence and the 5′ end of the locus-specific primer were not duplicated. The corresponding primer for each locus was tagged with a GTTT PIG-tail (Schable et al., 2002).
Amplification through PCR was performed on individual loci in 10-µL reactions containing 0.05 µM CAG (CAGTCGGGCGTCATCA)-tagged or M13 (GGAAACAGCTATGACCAT)-tagged locus-specific primer, 0.5 µM GTTT PIG-tailed locus-specific primer (both from Integrated DNA Technologies, Coralville, Iowa, USA), 0.45 µM fluorescently labeled CAG- or M13-tagged primer, 0.125 mM dNTPs, 0.1 µg/µL bovine serum albumin (both from New England Biolabs, Ipswich, Massachusetts, USA), 15 mM Tris (pH 8.4), 50 mM KC1, 1.5 mM MgCl2, 0.5 units AmpliTaq Gold DNA polymerase (all from Applied Biosystems, Foster City, California, USA), and 1–10 ng of template DNA. Fluorophores used to label CAG- and M13-tagged primers included VIC, PET, NED (Applied Biosystems), and FAM (Integrated DNA Technologies) (Table 1). Thermal cycling parameters were: 95°C for 2 min; 25 cycles of 95°C for 3 min, 60°C for 30 s, and 72°C for 20 s; then 25 cycles of 95°C for 3 min, 52°C for 30 s, and 72°C for 20 s; with a final extension of 72°C for 5 min. Two microliters of PCR product was diluted in 50 µL USB nuclease-free water (Affymetrix, Santa Clara, California, USA). A 3130xl DNA Analyzer (Applied Biosystems) at the Georgia Genomics Facility and GeneScan 500 LIZ Size Standard (Applied Biosystems) were used to analyze amplicon sizes. GENEMAPPER version 4.0 (Applied Biosystems) was used to score allele sizes.
Traits and thermal cycling conditions for 19 microsatellite loci for Juncus roemerianus.a
The 96 primer pairs were initially tested for amplification in two individuals from the Grand Bay NERR, and 48 amplified and were polymorphic. These 48 primer pairs were subsequently screened for amplification consistency and polymorphism using 24 individuals that were collected throughout the Grand Bay NERR (Table 2). Nineteen loci consistently amplified, did not significantly deviate from Hardy-Weinberg equilibrium, and had a frequency of null alleles below 0.1 (Table 1). The 19 loci were then tested across two additional populations on the Gulf of Mexico from the Apalachicola NERR and Choctawatchee Bay area to ensure consistent amplification across the range in which the species is dominant (Table 2). No clonal replicates were used to test the microsatellite markers. Three loci consistently cross-amplified in 24 samples of the related taxon J. effusus L., collected from Perdido Bay, Alabama, USA (Table 3).
Allelic data from GENEMAPPER was formatted for analysis using GMCONVERT (Faircloth, 2006). CERVUS version 3.0.7 (Kalinowski et al., 2007) was used to calculate allelic richness, observed heterozygosity, expected heterozygosity, deviations from Hardy-Weinberg equilibrium, and frequency of null alleles (Table 2). GENEPOP version 4.2 (Raymond and Rousset, 1995) was used to calculate linkage disequilibrium. No loci exhibited linkage disequilibrium across or within populations following sequential Bonferroni correction except one pair of loci (Jr03 and Jr29) in the Choctawhatchee Bay population. Allelic richness ranged from 2–13 alleles per locus with an average of 5.737. The panel of 19 microsatellites had a combined nonexclusion probability of identity of 1.009 × 10−15.
Genetic diversity metrics for three populations of Juncus roemerianus located in the northern Gulf of Mexico.a
The 19 polymorphic nuclear microsatellite markers are useful for investigating genetic diversity and population structure in J. roemerianus for conservation and restoration efforts. The markers provide sufficient resolution to identify clonal replicates and to examine the roles of clonal and sexual reproduction in natural populations of J. roemerianus.
Characteristics of three primers that cross-amplified in the related species Juncus effusus.
The authors thank E. T. Jones for assistance during field collection and laboratory experimentation and J. Feura for assistance with field collection at the Grand Bay National Estuarine Research Reserve (NERR). The authors thank the University of Georgia Graduate School for funding awarded to H.R.T. and acknowledge resources provided by the Warnell School of Forestry and Natural Resources, the University of Georgia Genomics Facility, the Grand Bay NERR, and Apalachicola NERR.