Map of a portion of the North American Atlantic Coast depicting the general location and sample size of 17 river and estuary collections of shortnose sturgeon ( Acipenser brevirostrum ) surveyed at 11 polysomic microsatellite DNA loci. Sample sizes are in parentheses. doi:10.1371/journal.pone.0102784.g001 

Context in source publication

Context 1

... combined with genome doubling (allopolyploidy), or a combination of these processes [20]. Following the polyploid events that gave rise to extant sturgeon species, the random and gradual diploidization process [21] is assumed to have resulted in functional diploidy [22]; however, the degree to which their various polyploid nuclear genomes exhibit disomic inheritance is unknown. The shortnose sturgeon Acipenser brevirostrum is an amphidromous species endemic to the large coastal rivers of eastern North America. This species is distinguished among all the living acipenserids by exhibiting the largest number of chromosomes, 372 [18]. A. brevirostrum was listed as an ‘‘endangered species’’ under the US Endangered Species Preservation Act of 1967 and remains so despite re-assessment in response to a 1994 petition to de-list populations in tributaries to the Gulf of Maine. Of significant note is that of 19 putative population units identified based on the species’ perceived strong fidelity to natal rivers [23], [24] some river populations continue to exist, although much reduced, but in other rivers, the species has been extirpated. In most instances, spawning status is either unknown or indicated to be of limited extent [25], [23] further complicating the prediction of biological units that could respond to conservation measures. To date, all published information on phylogeographic- and population-level structuring in A. brevirostrum has been assessed through nucleotide sequence variation detected in the maternally- inherited mtDNA. This is presumably due to the difficult nature of interpreting allelic data from the functionally polyploid (putatively hexaploid) nuclear genome [26]. The mtDNA research has primarily been focused on a moderately polymorphic 440 base pair segment of the control region (CR) adjacent to the tRNA proline gene. These findings are well documented in the peer- reviewed literature [27], [28], [29], [26], [30], [31] and are consistent both among studies and between researchers. Although results reflect a shallow gene genealogy (gene tree) for the A. brevirostrum mtDNA CR, analyses of haplotype frequencies at the level of putative individual populations showed significant differences among nearly all river/estuarine systems in which reproduction is known to occur. One prior study [31] concluded that although higher level genetic relationships exist (e.g., Northeast vs. Mid-Atlantic; Northeast vs. Southeast; Mid-Atlantic vs. Southeast; and other Mid-Atlantic regional subdivisions), A. brevirostrum appear to function in discrete populations, and that relatively low female-mediated gene flow exists between the majority of populations. This implies that effective dispersal among drainages within regions has been sufficient to prevent deep divergence within this species over evolutionary time scales. Acipenser brevirostrum has been shown to possess the highest number of chromosomes (N = 362–372) among all the Acipenseriformes karyotyped to date [18]. These authors, however, could not determine the species’ exact level of polysomy (hexaploid or dodecaploid). Contemporary cytogenetic techniques (including signals from fluorescent in situ hybridization) suggest A. brevirostrum is a hexaploid species [19]. While immensely complex, nuclear DNA-based approaches to A. brevirostrum conservation could identify significant levels of informative genetic variation because certain duplicated loci and repetitive DNA may lack functional constraints, thus allowing rapid accumulation of differentiation in DNA sequences [32]. Moreover, if the observed patterns of nuclear DNA diversity and variation differed from those empirically determined for the maternally-inherited mtDNA, this would inform biologists of the degrees of site philopatry or sex-biased dispersal for A. brevirostrum . However, no phylogeographic or population informative nuclear markers have been identified for A. brevirostrum [33]. To address this important research need and for the first time, allow an extensive assessment of the phylogeographic structure of A. brevirostrum from a multilocus nuclear DNA perspective, we characterized the inheritance of polysomic microsatellite DNA loci in shortnose sturgeon collected throughout the species’ range using loci derived specifically from this species [34]. Because of the complex modes of inheritance underlying the putatively hexaploid genome, we scored each allele (fragment) as a dominant marker with two states, presence or absence, resulting in the production of a binary character matrix. We report on the findings of an extensive statistical comparison of the patterns in allelic variation to identify and assess the reproductive status of populations and to delineate functional units of management to aid in recovery planning. While no tissue sampling was collected as part of this study, Acipenser brevirostrum ( n = 561) were sampled from 17 river and estuarine systems representing the species’ range (Figure 1) by researchers approved and permitted by the Institutional Animal Care and Use Committee of the U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Fisheries Service. Sampling followed the guidelines mandated under NOAA Technical Memorandum NMFS-OPR- 18 or NOAA Technical Memorandum NMFS-OPR-45. The mandated non-invasive procedures were that tissue (1.0 cm 2 fin- clip) taken from soft pelvic fin was stored in 95% absolute ethanol or SDS/urea. The ambiguous reproductive status of A. brevirostrum in the Potomac and Merrimack Rivers affected categorization of specimens. We chose to treat the Potomac River collection ( n = 2) both as of unknown origin and as part of the large Chesapeake Bay-proper collection. The Merrimack River sample consisted of males collected at the same location and time; however, because eggs ( n = 4) and embryos ( n = 2) were collected in Spring 2009, we considered the Merrimack River collection as a reproducing population. Genomic DNA from ethanol preserved samples was extracted with the Gentra Puregene DNA kit (Qiagen, Valencia, CA) following the manufacturers guidelines for whole non-mammalian blood and resuspended in TE (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). Previously extracted genomic DNA from other researchers was extracted using the methods described in [28]. DNA concentrations were determined by fluorescence assay [35], and the integrity of the DNA was visually inspected on 1% agarose gels [36]. SDS-urea preserved samples were also processed using these procedures with the exception that cell lysis was not necessary and samples were subjected directly to protein precipitation and alcohol purification. All DNA samples were quantified as described above and diluted to 100 ng/ m l for use in PCR amplification. A suite of 11 microsatellite loci previously identified from A. brevirostrum [34] was surveyed: Abr B438, Abr D10, Abr D114, Abr D135, Abr D141, Abr D193, Abr D236, Abr D332, Abr D345, Abr D379, and Abr D557. Optimized PCR mixes consisted of the following: 100–200 ng of genomic DNA, 1 6 PCR buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl), 2.0 mM MgCl 2 , 0.2 mM dNTPs, 0.4 mg/ml Bovine Serum Albumin, 2.0 U Taq DNA polymerase (Promega Corporation, Madison, WI, USA), 0.25 m M forward and 0.5 m M reverse primer, and 0.3 m M fluorescent labeled M13 primer in a total volume of 25 m l. Amplifications were carried out on either PTC-200 or PTC-225 Thermal Cyclers (Bio-Rad Laboratories, Hercules, CA) using the following cycling conditions: initial denaturing at 94 u C for 15 min; 29 cycles of 94 C for 1 min; annealing temperature (56–66 C) for 45 sec, 72 C for 45 sec; 5 cycles of 94 C for 1 min; 53 C for 45 sec; 72 C for 45 sec; and final extension at 72 u C for 10 min. Fluorescently labeled fragment analysis was performed using an ABI 3130XL Genetic Analyzer and binning of alleles was performed using GENESCAN 2.1 Analysis software and GENOTYPER 3.6 fragment analysis software (Applied Biosystems, Grand Island, NY). Each locus was independently scored, and each amplicon meeting the signal strength conditions specified (at least 10% relative to the strongest allele) and fitting into the appropriate size category (based on repeat motif and an assumed step-wise mutational pattern) was classified as an allele. For polyploid individuals, gene duplication, multiple alleles, and the mode of inheritance can lead to practical and statistical complications in allelic identification and interpreting summary and population-level statistics [37]. Genetic stock identification studies on other Acipenserids using polysomic (tetrasomic) markers [38], [39] estimated gene dosages by relative peak intensity from electropherograms. Because of the large number of alleles (fragments) observed in the putatively hexaploid A. brevirostrum , allele dosage could not be reliably estimated from GENESCAN runs preventing the application of standard population genetic diversity statistics that require genotype or allele frequencies for their calculation. Two prior groups [39], [40] provided a validated approach to this dilemma and as in those studies, we scored each allele (fragment) as its own psuedodominant locus with one of two states, presence or absence, resulting in the production of a binary character (or allele) matrix. When codominant markers are screened in higher order polyploid species, and scored as psuedodominant loci (i.e., as binary character state), it is not possible to estimate either allele frequencies or heterozygosities directly. Allele (loci) frequencies at 11 polysomic microsatellite DNA markers were estimated on the A. brevirostrum allelotype matrix using the method of [41] as implemented by GenAlEx 6.3 [42]. Only alleles with an overall frequency of . 1% were used for these and other statistical analyses. Allelotypes were analyzed for patterns of population ‘genetic’ structure and for regions of genetic discontinuity at both the individual and ...