GENETIC INVESTIGATIONS REVEAL NEW INSIGHTS INTO THE DIVERSITY, DISTRIBUTION, AND LIFE HISTORY OF FRESHWATER MUSSELS (BIVALVIA: UNIONIDAE) INHABITING THE NORTH AMERICAN COASTAL PLAIN By NATHAN ALLEN JOHNSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017
160
Embed
GENETIC INVESTIGATIONS REVEAL NEW INSIGHTS ......GENETIC INVESTIGATIONS REVEAL NEW INSIGHTS INTO THE DIVERSITY, DISTRIBUTION, AND LIFE HISTORY OF FRESHWATER MUSSELS (BIVALVIA: UNIONIDAE)
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
GENETIC INVESTIGATIONS REVEAL NEW INSIGHTS INTO THE DIVERSITY, DISTRIBUTION, AND LIFE HISTORY OF FRESHWATER MUSSELS (BIVALVIA:
UNIONIDAE) INHABITING THE NORTH AMERICAN COASTAL PLAIN
By
NATHAN ALLEN JOHNSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
2 USING DNA BARCODES TO RECALIBRATE TAXONOMY, TEST MISIDENTIFICATION RATES, AND UNCOVER PATTERNS OF GENETIC DIVERSITY IN FRESHWATER MUSSELS (BIVALVIA: UNIONIDAE) ................... 23
Methods .................................................................................................................. 27 Taxon Sampling and Data Collection ............................................................... 27 Data Analyses .................................................................................................. 30
Misidentifications .............................................................................................. 33 Barcode Gap Analyses ..................................................................................... 34 Shallow Interspecific Divergence and Non-monophyletic Species ................... 35
Cases of Deep Intraspecific Divergence and Putative Cryptic Diversity ........... 37 Anodontini .................................................................................................. 37
3 APPLYING DNA BARCODES TO INVESTIGATE ECOLOGICAL HOST ASSOCIATIONS AND SPECIES BOUNDARIES FOR FRESHWATER MUSSELS ............................................................................................................... 82
Methods .................................................................................................................. 85 Specimen Collection ......................................................................................... 85 DNA Sequencing and Data Analyses ............................................................... 86
Reference DNA Barcode Library ...................................................................... 88 Juvenile Mussel Identification and Host Fish Characterization ......................... 90
Table page 2-1 The number and percentage of freshwater mussel misidentifications revealed
using F-cox1 and M-cox1 DNA barcodes. .......................................................... 46
2-2 Frequency of occurrence for each original morphology-based identification corrected using F-cox1 barcodes. ...................................................................... 47
2-3 Sample sizes (n), mean and maximum uncorrected p-distance (%), and the distance to the nearest neighbor (NN) species based on 1551 Fcox1 sequences for 57 currently recognized freshwater mussel species. ................... 48
2-4 Sample sizes (n), mean and maximum uncorrected p-distance (%), and the distance to the nearest neighbor (NN) species based on 377 M-cox1 sequences representing 37 currently recognized freshwater mussel species. ... 50
2-5 BIN assignments based on 1551 F-cox1 DNA barcodes representing 57 freshwater mussel species in southeastern United States ................................. 51
3-1 Collection sites (site abbreviations) and sampling dates for the five fish surveys where metamorphosed juveniles were recovered and identified using DNA barcodes. ................................................................................................... 99
3-2 Sample identifiers, museum catalog number, and collection coordinates (latitude and longitude) for 124 freshwater mussel specimens. ........................ 100
3-3 Sample sizes (n), mean and maximum intraspecific p-distances (d), and distance to nearest neighbor species (NN) shown as percentages for taxa included in the F-cox1 barcode library. ............................................................. 104
3-4 Barcode index number (BIN) assignments based on 124 F-cox1 DNA barcode sequences representing 24 freshwater mussel species known from central Texas .................................................................................................... 105
3-5 Naturally infested host fishes that produced juvenile freshwater mussels. Site numbers follow Table 3-1. The total number of juveniles for each mussel species is shown in parenthesis. ...................................................................... 106
4-1 Taxa sampled, drainage of collection, and number of sequences for all individuals included in molecular analyses. ...................................................... 131
4-2 Analysis of molecular variance (AMOVA) among members of the Cyclonaias (Quadrula) petrina and Cyclonaias (Quadrula) pustulosa species complexes.. 132
9
LIST OF FIGURES
Figure page 1-1 Map delineating the North American Coastal Plain. ........................................... 22
2-1 Collection sites in the southeastern United States for the 1571 specimens analyzed in this study. Each point may represent several collection sites and multiple taxa collected from the same locality..................................................... 53
2-2 Circular phylogram based on 1551 F-cox1 gene sequences representing 23 genera and 57 species of freshwater mussels collected from the southeastern United States. ............................................................................... 54
2-3 Neighbor-joining tree based on 1551 F-cox1 sequences. .................................. 55
2-4 Neighbor-joining tree based on 373 M-cox1 sequences. .................................... 72
2-5 Frequency distribution histogram of uncorrected pairwise genetic distance based on 1551 Fcox1 sequences assigned to 57 currently recognized freshwater mussel species in the southeastern United States. .......................... 77
2-6 Neighbor-joining subtrees illustrating BIN sharing between Fusconaia burkei and Fusconaia escambia (left) and Elliptio chipolaensis and Elliptio nigella (right). ................................................................................................................. 78
2-7 Neighbor-joining subtrees illustrating examples of shallow interspecific divergence for which two or more species formed a single genetic cluster. ....... 79
2-8 Neighbor-joining subtrees illustrating examples of deep intraspecific divergence for which conspecifics were split into two or more BINS, indicating possible cases of cryptic diversity. ..................................................... 80
2-9 Neighbor-joining subtrees illustrating examples of high intraspecific divergence in which conspecific individuals were assigned to two or more BINS, indicating possible cases of cryptic diversity. ........................................... 81
3-1 Sampling locations in central Texas for fishes that produced juvenile freshwater mussels identified using DNA barcodes. ......................................... 107
3-2 Neighbor-joining tree based on 124 F-cox1 sequences. .................................. 108
3-3 Most likely topology generated in the BI analysis with indications of clades containing juvenile mussels recovered from naturally infested fishes. .............. 111
3-4 Intraspecific and interspecific uncorrected p-distances with cases of high intraspecific variation and low interspecific divergence indicated. .................... 112
10
3-5 Scatterplot illustrating the overlap of maximum intraspecific p-distances with the nearest neighbor distances. Points above the diagonal line indicate species with a barcode gap. ............................................................................. 113
4-1 Map showing sampled localities (dots) for members of the Cyclonaias (Quadrula) pustulosa species complex (left) and Cyclonaias (Quadrula) petrina species complex (right). ........................................................................ 133
4-2 Maximum likelihood (ML) phylogeny based on concatenated mtDNA and nDNA datasets for Quadrulini. .......................................................................... 134
4-3 Comparison of results for members of the Cyclonaias (Quadrula) petrina species complex.. ............................................................................................. 135
4-4 Comparison of results for members of the Cyclonaias (Quadrula) pustulosa species complex. .............................................................................................. 136
4-5 Histograms illustrating the distribution of all intraspecific and interspecific pairwise uncorrected-p distances for Cyclonaias (Quadrula) petrina complex (top) and Cyclonaias (Quadrula) pustulosa complex (bottom) .......................... 137
11
LIST OF ABBREVIATIONS
ACF Apalachicola-Chattahoochee-Flint River Basin
DUI Doubly uniparental inheritance
ESA Endangered Species Act
EYC
F-cox1
Escambia-Yellow-Choctawhatchee River Basin
Maternal copy of cytochrome oxidase subunit 1 gene
ITS1
M-cox1
Internal transcriber subunit 1 gene
Paternal copy of cytochrome oxidase subunit 1 gene
ND1
PCR
NADH dehydrogenase subunit 1 gene
Polymerase Chain Reaction
12
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
GENETIC INVESTIGATIONS REVEAL NEW INSIGHTS INTO THE DIVERSITY, DISTRIBUTION, AND LIFE HISTORY OF FRESHWATER MUSSELS (BIVALVIA:
UNIONIDAE) INHABITING THE NORTH AMERICAN COASTAL PLAIN
By
Nathan Allen Johnson
December 2017
Chair: James D. Austin Major: Fisheries and Aquatic Sciences
There is an urgent need to reevaluate species diversity in North American
freshwater mussels (Bivalvia: Unionidae) because of high rates of imperilment and
inherent difficulties with the delineation of species boundaries. Molecular data,
specifically mitochondrial gene sequences, are common in systematic studies of
freshwater mussels, but DNA barcoding methods have received little attention. As the
basis for my dissertation, I initiated a new biodiversity assessment for freshwater
mussels by developing a comprehensive barcode reference library as an important
taxonomic first-step toward an integrative and unified taxonomy. This effort resulted in
two regionally comprehensive DNA barcode libraries representing approximately 80
species and 1,700 specimens collected from nearly 300 locations across 54 river
basins. The vast majority of the collections were made from Gulf Coast rivers draining
the North American Coastal Plain (NACP), which is a known biodiversity hotspot. My
analyses of DNA barcodes revealed high levels of misidentification rates for several
genera, including Elliptio (44.3%), Villosa (23.9%), Utterbackia (18.2%), Uniomerus
(15.9%), and Toxolasma (6.8%). At the species-level, 15 taxa were genetically
13
indistinguishable, including several that were either federally listed or being considered
for listing under the US Endangered Species Act. In contrast, 16 species exhibited high
intraspecific divergence, suggesting possible cases of overlooked species-level diversity
that warrant further investigation. I subsequently used the DNA library to determine
ecological hosts for freshwater mussel larvae (glochidia) by providing molecular
identifications for juvenile mussels that completed metamorphoses on in situ parasitized
fishes. Using members of the Quadrulini as a case example, I demonstrated how DNA
barcodes represent an important taxonomic first-step where multiple independent lines
of evidence (e.g. morphology, genetics, behavior, geography) can be integrated to
make holistic decisions regarding evolutionary relationships. Specifically, I used data
from 3 genes (CO1, ND1, ITS1), external morphometrics, and geographic distributions
to revise generic placement of 11 taxa, synonymize 4 taxa, and provide evidence for a
previously unrecognized species. Finally, I advocated for expansion of the DNA library
to include other regional mussel faunas to assist with future evolutionary studies,
biodiversity assessments, accurate identifications, and development of properly
targeted conservation management programs.
14
CHAPTER 1 INTRODUCTION
A major focus of evolutionary biology involves the explanation of how, when, and
where new species evolve. This is an important challenge because understanding the
status of various groups of taxa underlies the need to predict responses to global
change and to identify schemes for in situ conservation (Gaston 2000). Global
biodiversity is not heterogeneously distributed and 36 biodiversity hotspots are formally
recognized (Noss et al. 2015), including the North American Coastal Plain (NACP)
(Figure 1-1). The NACP once harbored approximately 6200 plant, 424 freshwater fish,
291 reptile (Noss et al. 2015), and 225 freshwater mussel species (Williams et al. 1993;
2017). However, large-scale habitat modification and other anthropogenic activities
have caused extinction rates to increase, triggering a biodiversity crisis (Wake and
Vredenburg 2008; Burkhead 2012). New species are still being discovered and many
groups, especially invertebrates, are still poorly known. This scenario creates an
urgency to protect remaining diversity while racing to discover and describe existing
diversity before undocumented species go extinct.
Freshwater mussels of the family Unionidae are one of the best studied groups of
freshwater mollusks in the NACP. However, high rates of imperilment coupled with
difficulties discriminating between intraspecific variability and interspecific diversity using
traditional methods (e.g. conchology, soft anatomy characters, reproductive structures)
may impede conservation of remaining diversity. In recent years, analyses of mtDNA
sequence datasets have successfully resolved a number of important questions
regarding the validity of some unionid groups (Mulvey et al. 1997; King et al 1999;
Gangloff et al 2006; Jones et al 2006; Campbell et al. 2008; Pfeiffer et al. 2016; Perkins
15
et al. 2017), and helped identify cryptic species in some southeastern watersheds
(Mulvey et al. 1997; King et al. 1999; Jones and Neves 2010; Lane et al. 2016).
Integration of the molecular data generated by these studies with other character-based
datasets (e.g. shell morphology, soft anatomy, glochidial and conglutinate morphology,
periods of gravidity, fish host specificity, mantle displays) has greatly benefitted recent
taxonomic revisions and has led to a more holistic understanding of the biogeography
and conservation genetics of some freshwater mussel groups (King et al. 1999; Roe
and Hartfield 2005; Jones et al. 2006; Zanatta and Murphy 2008; Lane et al. 2016).
The principal goal of my dissertation research was to reevaluate previous
hypotheses regarding the distribution, classification, and diagnosis of unionids using an
integrative taxonomic framework with inferences drawn from multiple independent lines
of evidence (e.g. morphology, geography, genetics). The study focused on freshwater
mussels inhabiting the NACP, which is characterized by a high degree of endemism
and a rich geologic history, making it an excellent area to focus my systematic studies
and phylogeographic research on this highly imperiled group of animals.
The species problem as it relates to unionids has evolved throughout the
taxonomic history of this group. Early taxonomists followed a typological species
concept in which descriptions were based on a single or only a few specimens. Poor
understanding of intraspecific variation in morphological characters resulted in the
description of more than 4000 taxa (Haas 1969). As systematics advanced and more
collections were made, biogeographic patterns began to emerge and taxonomists
realized many descriptions were based on subtle differences and probably represented
variation within the same species (Ortmann 1920; Johnson 1970). Studying variation in
16
morphological characters triggered the application of a morphological species concept
that aimed to consider both between- and within-species variation. However, several
confounding factors still make application of this species concept difficult. For example,
the environment is known to impact the shell morphology of unionids, making it difficult
to distinguish between intraspecific variability, phenotypic plasticity, and interspecific
similarity (Ortmann 1920; Eagar 1954; Zieritz et al. 2010; Inoue et al. 2013; Bourdeau et
al. 2015; Fassatoui et al. 2015; Zajac et al 2017).
Interpretation and quantification of morphological variation is subjective and
extremely difficult because freshwater bivalves lack consistent landmarks for informative
and reproducible morphometric analyses. In other groups, geographic distributions are
sometimes helpful in delineating species boundaries. Unfortunately, the dispersal
ranges for most mussels were (and still are) largely unknown and difficult to predict
through space and time. Cryptic diversity and issues of convergence are additional
limitations that must be considered when relying solely on morphological characters.
The morphological species concept served as a temporary remedy to the typological
concept, but synonymy issues resulting from “over-splitting” and subsequent scrutiny of
“over-lumping” revisions still plague the taxonomy of this group today (Williams et al.
2014; 2017). Recent ecological, physiological, and molecular studies are being used to
develop a more integrative taxonomy. Researchers now recognize how these new
findings can provide a more holistic understanding of the evolutionary relationships
within and among taxa.
The problem now lies in how to interpret these new data within the established
taxonomy built on non-molecular characters. Is this left to the expert taxonomist or to
17
ecologists, physiologists, or geneticists? The immediate solution has been to adopt a
taxonomic species concept and let the expert taxonomist make the call. This is a
conservative approach that relies on authoritative interpretation by a regional expert
who typically has the most experience with the group. The adaptive nature of this
concept is attractive and allows the expert to consider multiple lines of evidence before
making the decision. However, there are several problems with this approach to
resolving our current taxonomic problems. First, it does not provide options for
specimens that experts are unable to identify, or for cases in which experts can’t agree
on the appropriate taxonomic classification. Second, it often forces taxonomists to
interpret findings from studies outside their area of expertise. Lastly, the highly
subjective nature of diagnosing species still remains and the taxonomic uncertainties
continue to fester.
To date, at least 32 species concepts have been proposed (Zachos 2016), all of
which have limitations, from demonstrating reproductive isolation (Mayr 1942) to a lack
of phylogenetic resolution despite clear morphological, ecological, or physiological
divergence (Funk and Omland 2003; McVay and Carstens 2013; Whelan and Strong
2015). In recent years, there has been a shift towards delimiting and describing taxa by
integrating information from different methods and data types. This multisource
framework, known as “integrative taxonomy” (Dayrat 2005; Will et al. 2005), increases
opportunities for linking decades of morphological scrutiny and other observations in
malacology with recent advancements in science and technology.
Here, I outline a new foundation for the future of mussel taxonomy that utilizes a
DNA barcoding approach to bring experts from multiple disciplines together to
18
objectively characterize species and delineate their boundaries objectively. The
objective criteria and vision of DNA barcoding has sparked biodiversity inventory
initiatives across the globe. Since 2003, several thousand papers related to DNA
barcoding have been published with a large number directly related to taxonomy and
systematics. Many are campaigns to inventory the biological diversity within a particular
taxonomic group (e.g. FISH-BOL, Mosquitoes of the World, ABBI-All birds barcoding
initiative). Another large body of literature revolves around interpretation of molecular
barcode data (and molecular data in general) debating the strengths and weaknesses of
delineating species using a limited amount of molecular data (Tautz et al. 2003; Blaxter
2004; Will et al. 2004; 2005; Hickerson et al. 2006; Kohler 2007; Packer et al. 2009) .
However, when coupled with complementary perspectives (e.g. morphology, ecology,
life history), the method provides a powerful approach for solving taxonomic problems
(Ward et al. 2009; Puillandre et al. 2012; Kekkonen and Hebert 2014; Hendrich et al.
2015; Chambers and Hebert 2016). To date, the method has seen little application for
freshwater mussel research and our efforts are the first to build comprehensive DNA
barcode reference libraries.
The DNA barcoding method differs from previous approaches using mtDNA by
incorporating standardized data collection to meet compliance of community standards
for DNA barcoding (Ratnasingham and Hebert 2007; Benson et al. 2012). Under this
framework, results can be combined and analyzed using combinations of distance- and
phylogenetic-based approaches. Identification and current taxonomy can be evaluated
objectively and distinct lineages revealed by DNA barcode analyses can be compared
against other independent lines of evidence (e.g. morphology, geography, ecology, etc.)
19
to determine the appropriate rank of classification. DNA barcoding is an attractive
application for refining our understanding of the diversity and distributions of North
American freshwater mussels.
In Chapter 2, I introduce the species problem as it pertains to freshwater mussels
and describe how I’ve created and applied a dual DNA barcoding approach (F-cox1 and
M-cox1) and phylogenetic analyses to identify and characterize unionid diversity in
Florida and contiguous drainages in Alabama and Georgia. This area is known as the
Greater Floridan Region (GFR) (Williams et al. 2014). Diversity in shell shapes has
resulted in >100 original species descriptions from specimens collected throughout the
state, yet only 61 species and 23 genera are currently recognized (Williams et al. 2014).
The current distribution of many species within the state remains unresolved because of
taxonomic ambiguities and poor understanding of conchological variation within and
among populations. Our analysis of 1551 maternal and 373 paternal cox1 sequences
(and counting) representing 57 species from 23 genera indicate that morphological
characters used in previous taxonomic assessments are largely congruent with results
from DNA barcode analysis. However, there are exceptions, especially within the genus
Elliptio. Results also reveal deeply diverged populations within several species,
suggesting overlooked cryptic diversity in the study area. Results from DNA barcoding
were compared to recent taxonomic assignments based largely on morphological
characters, and directions toward a unified taxonomy based on an integrative taxonomic
framework were outlined.
In Chapter 3, I expand spatial and taxonomic coverage of the DNA library to
incorporate freshwater mussels of central Texas. Specifically, I generated F-cox1
20
sequences for 124 adult specimens, which represent all 17 genera and 24 species of
mussels known from the Colorado and Guadalupe river basins in central Texas. In most
cases, mussel species exhibited low intraspecific variation (mean p-distance 0.50%)
when compared with distance to nearest neighbor species (mean p-distance 8.44%).
However, DNA barcode analyses highlighted three cases of misidentification and other
instances in which current taxonomy both overestimated and underestimated diversity
in this region. We subsequently used the mussel DNA barcode library to characterize
ecological host use by generating F-cox1 sequences for 137 juveniles recovered from
12 fish species infected in situ. All newly transformed juveniles were identified by
assignment to monophyletic clades corresponding to 8 mussel species, including 4 of 5
federal candidate species (Lampsilis bracteata, Quadrula aurea, Quadrula
houstonensis, and Quadrula petrina) that occur in the sampled drainages. Our efforts to
recalibrate taxonomy and characterize ecological hosts provide insights into population
processes such as recruitment (e.g. availability of suitable hosts), dispersal (e.g. vagility
of host), and resiliency (e.g. host generalist vs. specialist). This information is critical to
managers working to assess conservation status, extinction risks, and recovery options
for remaining freshwater mussel populations.
In Chapter 4, I focus on the taxonomic issues within the genus Quadrula and
demonstrate the utility of an integrative approach to species delimitation that considers
molecular, distribution, and morphology data to evaluate evolutionary relationships.
Specifically, I examined genetic relationships using three genes (CO1, ND1, and ITS1)
representing 8 genera and 20 species in the Quadrulini and evaluated morphological
variation throughout the ranges of 8 species in two species complexes. My results
21
support the assignment of 12 nominal taxa to the genus Cyclonaias: C. aurea, C.
asperata, C. houstonensis, C. infucata, C. kleiniana, C. mortoni, C. nodulata, C. petrina,
C. pustulosa, C. refulgens, C. succissa, and C. tuberculata. Additionally, congruence
across all lines of evidence (i.e. morphology, geography, and genetics) indicated that
current taxonomy overestimates species-level diversity in the C. pustulosa species
complex while underestimating diversity in the C. petrina species complex. I revised
species-level classifications by synonymizing four taxa (C. aurea, C. houstonensis, C.
mortoni, and C. refulgens) considered either species or subspecies under Cyclonaias
pustulosa and provide evidence for a previously unrecognized species from the
Cyclonaias petrina complex that is endemic to the Guadalupe River basin. These
findings have important implications regarding the conservation assessments and
pending legislative protections for several freshwater mussel species within western
Gulf of Mexico drainages.
In my fifth and final chapter, I present a synopsis of the major findings within my
dissertation and present a plan for continuing to move the field of systematic
malacology and conservation genetics forward through education, outreach, and
collaborative research. Finally, I announce my goal to lead the effort to build a
comprehensive DNA barcode library (UNIO-BARCODE) for all freshwater mussels in
North America and discuss several applications for such a dataset.
22
Figure 1-1. Map delineating the North American Coastal Plain.
23
CHAPTER 2 USING DNA BARCODES TO RECALIBRATE TAXONOMY, TEST
MISIDENTIFICATION RATES, AND UNCOVER PATTERNS OF GENETIC DIVERSITY IN FRESHWATER MUSSELS (BIVALVIA: UNIONIDAE)
Accurate and precise identification and classification provide the foundation for
understanding evolutionary relationships, are key to identifying biogeographic
processes, and facilitate the targeting of conservation programs (Dexter et al. 2010;
Shea et al. 2011). Our ability to identify distinct evolutionary lineages objectively in
some groups remains an important challenge for modern systematics research (Fujita et
al. 2012). Urgency increases for assessments involving imperiled species, particularly
those fraught with taxonomic instability, because conservation efforts are typically
based on species-level designations (Chambers and Hebert 2016; Huang and Knowles
2016; Sukumaran and Knowles 2017).
Freshwater mussels of the family Unionidae, also known as naiads, pearly
mussels, freshwater clams, or unionids, are a diverse group of bivalve mollusks that are
distributed on every continent except Antarctica. Approximately 300 species are known
from the United States, with the majority of this diversity residing in rivers of the
Southeast where many endemic taxa have evolved (Turgeon et al. 1988; 1998; Williams
et al. 1993; 2008; 2014; 2017). This fauna is highly imperiled, with about 78% of
currently recognized species considered either extinct, endangered, threatened, or of
special concern (Williams et al. 1993; Turgeon et al. 1998; Lydeard et al. 2004; Haag
2012; Haag and Williams 2014). At least 10% of North American mussel taxa became
extinct in the past 100 years (Neves et al. 1997; Haag 2012; Haag and Williams 2014),
which is comparable to extinction rates observed in the rainforest (Ricciardi and
Rasmussen 1999) and for other freshwater organisms (e.g. Burkhead 2012). Despite
24
the realization that native unionid populations are in peril, taxonomic uncertainties
remain for many species and limit the development of effective conservation
management strategies (Pfeiffer et al. 2016). Furthermore, the identification of many
unionid species is complicated by the lack of discrete morphological characters that are
useful for diagnosing species accurately, delimiting species boundaries, or determining
evolutionary relationships (Shea et al. 2011; Pfeiffer et al. 2016; Perkins et al. 2017).
The conservation of remaining North American mussel populations would benefit greatly
from improved methods for identification and delineation of species boundaries, which
would also enable a better understanding of the biogeographic processes responsible
for creating this biodiversity.
DNA barcoding has gained popularity in recent years as an effective method for
rapid biodiversity assessment, specimen identification, and taxonomic exploration
(Hebert et al. 2003). DNA barcoding differs from past applications of genetics regarding
identification and taxonomic revision by encouraging a transparent and reproducible
method based on a well-curated collection and reference barcode library, which allows
the testing of falsifiable hypotheses using an objective foundation. The intention is to
supplement and test conventional taxonomic classifications, not discard evidence
presented in studies relying on traditional approaches. This creates an alliance between
molecular and morphological taxonomists to offer a more objective, reliable, and rapid
species identification and classification scheme.
In additional to high levels of diversity, imperilment, and taxonomic uncertainty,
freshwater mussels have several other intrinsic characteristics that make them an
attractive group for broad-scale DNA barcode efforts. Perhaps the most captivating is
25
the parasitic larval stage that requires attachment to a host fish before metamorphosis
into a free-living individual. Adult mussels are relatively sedentary and the larval stage
represents the primary mechanism for dispersal, making it difficult to determine
distribution limits when host fishes are unknown. DNA barcodes have the potential to
link different life stages and have been used to provide species-level identifications of
larvae encysted on fishes (Boyer et al. 2011; Chapter 3). Additionally, freshwater
mussels, along with at least six marine bivalve families, have a peculiar way of mtDNA
inheritance called doubly uniparental inheritance (DUI). The DUI system is
characterized by two independently inherited mtDNA genomes: the female-transmitted
(F-genome), which is carried in somatic tissues of both sexes, and the male-transmitted
(M-genome), which is transcribed not only in male gonads, but recently discovered to be
heteroplasmic in male and female soma as well (for review, see Breton et al. 2007;
Breton et al. 2017). The F genome exhibits strict maternal inheritance (SMI), as in all
other members of the animal kingdom, but M genomes are transmitted paternally (i.e.
from fathers to male offspring). Studies investigating intra/interspecific variation have
consistently shown that levels of differentiation were higher in M genomes compared to
F genomes indicating the M genome evolves at a faster rate (Liu et al. 1996; Krebs
2004; Doucet-Beaupré et al. 2012; Krebs et al. 2013). This allows creation of dual DNA
barcode libraries based on independently evolving mtDNA genomes, which alleviates
some limitations associated with single-gene approaches (Rubinoff and Holland 2005;
Toews and Brelsford 2012).
The effectiveness of DNA barcoding as a tool for identification of freshwater
mussels has received relatively little attention. Despite using DNA barcoding to describe
26
the methods in previous studies (Campbell et al. 2008; Boyer et al. 2011; Moyer and
Diaz-Ferguson 2012), none have included required documentation to meet community
standards for DNA barcode compliance. This is problematic given the limitations and
errors of mussel identifications (Shea et al. 2011) and need for reproducible results in
biological research (Vink et al. 2012). The potential for errors associated with using data
from public repositories is well known (Harris 2003), especially for groups in which
species are difficult to identify or taxon sampling is sparse (Santos and Branco 2012).
Multiple studies have highlighted errors in GenBank accessions for freshwater mussels
(Campbell et al. 2005; Boyer et al. 2011; Campbell and Lydeard 2012; Williams et al.
2017).
Here, we establish a comprehensive DNA barcode library for freshwater mussels
using both male- and female-specific copies of the mtDNA gene COI (F-cox1 and M-
cox1). We analyzed the barcode sequences to compare molecular and morphological
based species assignments and evaluated rates of misidentification. Specifically, we
tested the effectiveness of the barcode genes to delineate intraspecific variation from
interspecific divergence while identifying cases for which: 1) barcode clusters showed
high congruence with morphology-based identification and current taxonomy; 2)
morphologically indistinguishable, but genetically and geographically diagnosable
lineages suggested overlooked or cryptic diversity; and 3) groups of currently
recognized species exhibited low genetic divergence that prevents their discrimination
using DNA barcodes. Finally, we discuss the implications of our findings and utility of
the DNA reference library for facilitating future research and conservation by providing
the accurate identification of unidentified specimens, discovering potentially cryptic taxa,
27
and formally incorporating molecular divergences with past and present taxonomic
assessments.
Methods
Taxon Sampling and Data Collection
To build our DNA barcode library, we aimed to collect and sequence at least 3
individuals for all 57 extant freshwater mussel species recently recognized as occurring
in the Greater Floridan Region (GFR), comprised of Florida and contiguous drainages in
Alabama and Georgia (Williams et al. 2011; 2014). Sampling focused on type localities
and drainages from where species were originally described (Figure 2-1). In some
cases, sampling type drainages required collecting individuals outside the GFR. Sample
size for each species varied as a function of rarity and distribution. We analyzed
additional individuals from multiple watersheds throughout the range of each species
whenever possible.
All specimens were sampled either non-lethally using DNA swabs (Henley et al.
2006) or preserved in 95-100% non-denatured ethanol (EtOH) after both anterior and
posterior adductor muscles were severed. Specimens were preserved in volumes of
EtOH approximately five times the volume of tissue (Nagy 2010; Evans and Paulay
2012). The concentration of EtOH was measured after 24 hours using a hydrometer
(Fisherbrand 11-590). In most cases, a total volume change of EtOH was required to
maintain concentrations ≥ 95%. After initial preservation, a subsample of mantle tissue
was placed in 95% EtOH and stored at -80°C for long-term preservation. To ensure the
highest levels of reproducibility of our findings, we deposited all vouchered specimens in
public museums and provided all metadata related to the voucher specimens, including
collection locality data, specimen images, nucleotide sequences, PCR primers, and
28
.ab1 trace files, to the Barcode of Life Data (BOLD) system (www.barcodinglife.org;
Ratnasingham and Hebert 2007) under the project ‘UNIO-BARCODE’. Specimens
sampled using swabs were tagged externally with cyanoacrylic adhesive and
individually coded Hallprint shellfish tags (Hallprint Inc., Hindmarsh Valley, South
Australia), photographed, and released to the site of capture. Specimens were
considered ‘barcode compliant’ when DNA sequences were > 500 nucleotides and all
metadata, photographs, and trace files were supplied.
Initial specimen identifications were based on formal diagnostics using a
combination of morphological characters, including shell and soft anatomy, and the
geographic location consistent with original species descriptions from recent taxonomic
treatments (Williams et al. 2011; 2014). Regional freshwater malacologists assigned
specimens to the lowest taxonomic level possible after consulting the primary literature.
The initial identifications and taxonomic assignments were recorded and captured in the
10-character sample ID as follows: first character represents the first letter of the genus,
characters 2-4 correspond to the first three letters of the species epithet, characters 5-7
represent the drainage of collection, and 8-10 are numeric values to make each sample
ID unique. We labeled ambiguous generic or species-level designations as “Unknown”
or “species.” Each specimen was measured, labeled with the unique sample ID, and
photographed to capture morphological characters contained within the inside of the left
valve and outside of the right valve. Only external images were available for swabbed
specimens.
DNA was isolated from tissue (mantle or gonadal) and Isohelix swabs (Boca
Scientific) using one of the following three protocols: (1) Puregene extraction kit
29
(Qiagen); (2) organic extractions at the Smithsonian Institution using robotic facilities;
and (3) a modified plate extraction protocol (Ivanova et al. 2006). Stock and diluted DNA
were stored at -80°C for long-term preservation. For the maternal or female genome
copy of COI (F-cox1), we first used the primer sets dgLCO-1490 5’-
GGTCAACAAATCATAAAGAYATYGG-3’ and dgHCO-2198 5’
TAAACTTCAGGGTGACCAAARAAYCA-3’ (Meyer 2003) and used a variety of other
primers and primer cocktails, including COIFCamp 5’-
GTTCCACAAATCATAAGGATATTGG-3’ and COIRCamp 5’-
TACACCTCAGGGTGACCAAAAAACCA-3’ (Campbell et al. 2005), all of which amplify
the same gene segment. For the paternal or male genome COI (M-cox1), we used
HCO700dy2 5’-TCAGGGTGACCAAAAAAYCA-3’ and MCO1_22F 5’-
RTGCGTTGRRYDTTTTCBACTA-3’ (Walker et al. 2007). PCR amplifications were
conducted in 96-well plates containing 10 µL reactions with the following reagents and
volumes: 1 µL of DNA template (~20 ng), 0.25 µL BSA (10mg/mL), 0.5 µL dNTPs
(10mM; 2.5 mM each dNTP), 2µL 5X PCR buffer, 0.3 µL of each primer (10 µM), 0.3 µL
MgCL2 (2mM), 0.1 µL GoTaq DNA Polymerase (Promega), and 5.25 µL dH2O. Thermal
cycling profiles were as follows: for F-COX1, an initial denaturation at 95ºC for 3 min
followed by 5 cycles of 95°C for 30 s, 45°C for 40 s, 72°C for 45 s, then 35 cycles of
95°C for 30 s, 51°C for 40 s, 72ºC for 45 s, with a final elongation at 72°C for 10 min,
and hold at 4°C for 30 min followed by 15°C forever; for M-COX1, we used an initial
denaturation at 95°C for 2 min followed by 40 cycles of 95°C for 60 s, 45°C for 60 s,
72°C for 60 s, with a final elongation at 72°C for 10 min, and hold at 4°C for 30 min
followed by 15°C forever.
30
All PCR products were visualized on 1.5% agarose gels stained with ethidium
bromide. Unincorporated PCR products were removed using 1µL of ExoSAP-IT (USB,
Santa Clara, CA, USA) per 10 µL of PCR product, following manufacturer’s protocols.
Cycle sequencing was performed on both forward and reverse strands using the BigDye
Terminator v3.1 Cycle Sequencing Kit. Sequencing reactions of 5 µL were cleaned
using spin column filtration through Sephadex pellets and Millipore plates before
electrophoresing products on an ABI 3130xl or ABI 3730 DNA analyzer (Applied
Biosystems, Inc). Complementary DNA sequences were assembled, edited, and
exported as consensus sequences using Geneious v6.1.2 (http://www.geneious.com,
Kearse et al. 2012). DNA sequences were aligned to published sequences in Mesquite
v 3.04 (Madison and Madison 2015) using MUSCLE (Edgar 2004) and translated to
protein using standard invertebrate mitochondrial code amino acid to ensure no gaps or
stop codons were present in the alignments. Both DNA sequence alignments were
submitted to BOLD.
Data Analyses
We assessed misidentification rates and tested the ability of gender-associated
mtDNA barcodes to distinguish between currently recognized freshwater mussel
species, using a combination of genetic distances, phylogenetic methods, and Barcode
Index Numbers (BINs). For all distance-based analyses, we calculated the uncorrected
p-distance instead of various models of sequence evolution to avoid high levels of
variance that often result from parameterized models (Lefebure et al. 2006; Collins et al.
2012) and assumptions related to nucleotide evolution models (Ratnasingham and
Hebert 2013). Initial neighbor-joining (NJ) trees were calculated to evaluate
morphology-based identifications, and to assign unidentified specimens to haplotype
31
clusters for subsequent analyses. Identifications were assessed for all haplotypes that
nested within a monophyletic cluster containing specimens with a priori morphology-
based assignment to the species level. Misidentifications were not counted for
individuals representing species in shared BINs because in these cases currently
recognized species could not be delineated using DNA barcodes (see section 3.3).
Additionally, labeling errors or incorrect use of names in synonymy (e.g. Elliptio
hazelhurstianus synonymized under Elliptio ahenea) were not counted as
misidentifications. Based on these conditions, we consider our calculations of
misidentification rates to be fair and conservative.
After correcting identifications, intra/interspecific pairwise genetic distances
(hereafter referred to as genetic distances) were calculated using Mega 7 (Kumar et al.
2016) and distance to nearest neighbor (NN) species was calculated using the ‘Barcode
Gap Analysis’ tool in BOLD (Ratnasingham and Hebert 2007). We assessed the
number of instances when maximum intraspecific divergence exceeded the nearest
neighbor distance, which represent the absence of a barcode gap.
In addition to assessing putative species limits based on monophyly, all
sequences were assigned to a BIN using the 3-step Refined Single Linkage (RESL)
algorithm (Ratnasingham and Hebert 2013) implemented in BOLD. The BIN system
was used to establish an interim taxonomic system to identify operational taxonomic
units (OTUs) and reveal conflicts between morphological and molecular assignments.
Each taxon was scored into one of four categories (MATCH, MERGE, SPLIT, or
MIXTURE) as defined by Ratnasingham and Hebert (2013). For example, BINs
containing multiple species may indicate the need for taxonomic revision or highlight
32
cases in which barcode divergence is insufficient to allow species discrimination using
COI alone (MERGES). Alternatively, a signal for cryptic diversity occurs when members
of a species are assigned to multiple BINs, revealing overlooked species-level diversity
during application of traditional (morphology-based) taxonomic approaches (SPLITS).
After alignment of the input sequences, all individual sequences with < 2.2% genetic
distance were clustered into initial BINS. Any sequence or set of sequences with a
genetic distance to NN > 4.4% were assigned to a newly formed BIN. In the final
‘cluster refinement’ step, clusters that exhibit high sequence variation and clear
partitions within the BIN were further split into separate clusters, even when genetic
distance was < 2.2%. Clusters of sequences without a clear demarcation and <2%
genetic distance remained as a single OTU. The RESL method benefits from not relying
on a single distance-based threshold and is computationally less demanding when
compared to other methods used to evaluate the ‘barcode gap’ (Meyer and Paulay
2005; Meier et al. 2006; Pons et al. 2006; Puillandre et al. 2012; Yang and Rannala
2017).
Results
Newly-generated mtDNA barcode entries were created for 1571 freshwater
mussels representing 1 family, 2 subfamilies, 5 tribes, 23 genera, and 57 species
currently recognized from the southeastern United States (Figure 2-2). The dataset
represents 56 of the 57 extant taxa known from the Greater Floridan Region (Williams
et al. 2011; 2014), including 12 species protected under the US Endangered Species
Act (ESA). The dataset also includes two species, Ligumia subrotunda, not previously
recognized from the study area, and Toxolasma parvum, both of which have been
introduced outside their native range (Williams et al. 2014). The only extant species not
33
included was Medionidus simpsonianus, which is federally designated as endangered
and was recently rediscovered in the Ochlockonee River (Holcomb et al. 2015).
Geographic coverage included 257 localities sampled within 25 independent river
drainages that flow directly into the Gulf of Mexico or Atlantic Ocean (Figure 2-1). Most
species were represented by multiple specimens (mean specimens per species for F-
cox1 = 27.3; M-cox1 = 10.2) and broad geographic sampling to better characterize
intraspecific variability. Date of collection ranged from 1934 to 2013 (mean = 2008) for
specimens from which full barcode sequences were recovered. Both DNA alignments
were free of indels and stop codons. A total of 1924 DNA sequences were analyzed for
1571 specimens in the library; 1551 were represented by F-cox1 and 373 for M-cox1.
Mean sequence length of F-cox1 was 644 nucleotides (range: 522-648) and included
sequences for 98.3% of currently recognized taxa. For M-cox1, the mean sequence
length was 656 nucleotides (range: 561-678) and provide coverage for 63.8% of taxa.
Both Fcox1 and Mcox1 sequences were generated for 353 specimens. Barcode
compliance was achieved for 96.7% of individuals (1519 of 1571) with 8 specimens
lacking successful trace files, 29 lacking photographs, and 20 represented only by M-
cox1 sequences.
Misidentifications
All sequenced individuals nested within monophyletic clusters containing
specimens with a priori morphology-based assignment to the species level (Figure 2-3;
Figure 2-4). This included 172 specimens without species designations prior to DNA
sequencing. F-cox1 barcodes revealed that morphology-based identifications were
incorrect at the generic and species level at rates of 1.9% and 5.7%, respectively (Table
2-1), for a total of 88 misidentified specimens (Table 2-2). The number of
34
misidentifications revealed using M-cox1 barcodes was slightly lower (genus 1.3%;
species 2.4%). Misidentifications involved 27 species belonging to 13 genera. Members
of Elliptio were involved with 44.3% of the misidentifications, including 19 individuals
misidentified as either E. chipolaensis or E. nigella. High percentages of
misidentifications also included members of Villosa (23.9%), Utterbackia (18.2%),
Uniomerus (15.9%), and Toxolasma (6.8%).
Barcode Gap Analyses
The majority of both sex-linked COI barcodes were congruent with current
taxonomic designations; however, intra/interspecific genetic distances showed the
absence of a barcode gap (Figure 2-5). In fact, 11 species (19.3%) had F-cox1 barcode
sharing in which two or more currently recognized species had matching haplotypes
and exhibited a NN distance of zero (Table 2-3). We observed 7 additional cases for
whichF-cox1 NN distances were < 2.2% and max intraspecific distance exceeded the
distance to NN. The remaining 39 taxa (68.4%) could be unambiguously assigned using
F-cox1. Intraspecific genetic distance for F-cox1 ranged from 0% to 12.65% and
distance to the nearest neighbor ranged from 0% to 11.42%. Performance of the M-
cox1 gene for discriminating between currently species was similar, although samples
sizes were much lower. A total of 7 currently recognized species shared M-cox1
haplotypes and 6 additional species were characterized by NN distances < 1% (Table 2-
4). NN distances ranged from 3.53-18.44 for the 24 remaining species represented by
M-cox1 barcodes.
The 1551 F-cox1 barcodes were represented by 72 BINs (Table 2-5), 6 of which
were considered to be taxonomically discordant, impacting 426 UNIOBARCODE
records among 18 currently recognized species. These shared BINs highlight cases for
35
which barcode divergence was insufficient to allow species discrimination using F-cox1
alone and warrant further investigation. In contrast, a total of 66 BINs were considered
taxonomically concordant (1125 records), 5 of which were represented by only a single
record. These concordant BINs involved 39 currently recognized species with 24 BINs
scored as Match and 39 as Split. High intraspecific divergence was observed in 15
cases in which conspecific individuals were assigned to 2 or more BINs, suggesting
possible cryptic diversity in these lineages (Table 2-5).
Shallow Interspecific Divergence and Non-monophyletic Species
Analyses of our barcode data exposed several cases in which morphologically
described species exhibited low levels of interspecific divergence or shared haplotypes.
Specifically, 18 currently recognized species were indistinguishable using Fcox1
barcodes (Table 2-3; Table 2-4) and were recovered as non-monophyletic in our NJ
tree. Two species (Fusconaia burkei and Fusconaia escambia) were characterized by
only 0.78% sequence divergence at F-cox1 (Table 2-3) and shared a BIN (Table 2-5;
Figure 2-3). However, these two species are clearly separable morphologically and
geographically (Figure 2-6), and are genetically diagnosable by 5 nucleotides. These
results align with previous phylogenetic studies that have included these taxa and
suggest a recent evolutionary origin (e.g. Campbell and Lydeard 2012; Pfeiffer et al.
2016). Similarly, Quadrula infucata and Quadrula kleiniana shared a BIN with a 2.04%
sequence divergence. These taxa are morphologically similar, but allopatric and
molecularly diagnosable by 11 nucleotides, and likely represent valid species.
Individuals within the genus Elliptio showed the greatest inconsistency between
morphology and genetic signature, with 12 currently recognized species exhibiting
overlap between intra- and inter-specific divergences (Table 2-3; Table 2-4). These
36
Elliptio BINs were scored as MERGE (10) or MIXTURE (2). Elliptio has been considered
the most diverse freshwater mussel genus in North America, with 38 currently
recognized species occurring in Atlantic and Gulf of Mexico drainages (Turgeon et al.
1998; Williams et al. 2014). However, the genus has a tangled history of taxonomic
treatments due to the subjectivity of using morphological characters to separate
intraspecific variability from interspecific similarity (Simpson 1892; Clench and Turner
1956; Johnson1970; 1972; Williams et al. 2008; 2014). Most malacologists recognize 3
or 4 morphological groups in the genus (Johnson 1970; Johnson 1972; Williams et al.
2008; Williams et al. 2014). Our analyses of F-cox1 barcodes assigned members of the
genus Elliptio to 4 BINs; however, BIN membership did not follow current taxonomy or
morphological groups (Table 2-3; Table 2-4). The Elliptio BIN (ACY9226) included two
imperiled taxa, E. chipolaensis (N=19) and E. nigella (N=21) (Figure 2-5), both endemic
to the ACF basin, but considered allopatric (Williams et al. 2014). These taxa also
shared M-cox1 barcodes (Figure 2-4) and no diagnostic nucleotides were identified to
separate these two taxa at either locus. Although the two taxa superficially resemble
each other morphologically and occur in the same Gulf drainage (Figure 2-6), we
suggest a more comprehensive investigation to validate the relationships between these
two imperiled species.
Overall, barcode sequences did not provide a strong signal for biogeographic
patterns and allopatric taxa within the genus Elliptio. In fact, only one Elliptio BIN
(AAX7569) formed a geographically isolated group, which contained specimens
collected from the Alabama River identified as E. arctata (Figure 2-4; Figure 2-7). The
remaining E. arctata shared BIN AAB3266 with nine other species of Elliptio (E. arctata,
37
E. crassidens, E fraterna, E. fumata, E. jayensis, E. mcmichaeli, E. monroensis, E.
occulta, E. pullata, and E. purpurella), which included 206 specimens collected from
throughout the study area (Figure 2-4; Figure 2-7). A fourth BIN (ACZ0072) was
comprised of specimens identified as Elliptio jayensis collected from the Withlacoochee
(Tampa Bay drainage) (N=4) and St. Johns (N=2) rivers and E. pullata from Spring
Creek in the Chipola River Basin (N=2) (Figure 2-4; Figure 2-7). A congruent signal was
observed at Mcox1, indicating that these Elliptio BINs do not recover currently
recognized morphospecies as monophyletic clades (Figure 2-4; Figure 2-7), highlighting
the need for future investigations to elucidate the taxonomic validity of morphospecies in
the genus Elliptio.
Cases of Deep Intraspecific Divergence and Putative Cryptic Diversity
At the other end of the spectrum were species that showed deep intraspecific
divergence. Included were 15 currently recognized species in 10 genera that were
divided into 39 BINs (Table 2-5). In this section, we provide details for two classes of
potential cryptic diversity organized by tribe. The first includes cases of high intraspecific
divergence between conspecifics that resulted in monophyletic clades that were found
to be allopatric. These BINs were both genetically and geographically diagnosable. We
also describe cases in which multiple BINs were identified for a given species, but the
sequence clusters revealed sharing among river drainages.
Anodontini
Anodontoides radiatus occurs in Gulf of Mexico drainages from western Florida
to Louisiana. Among the three drainages sampled during this study, two allopatric BINs
were delineated; one for the Apalachicola-Chattahoochee-Flint (ACF) Basin and
another for the Choctawhatchee and Escambia river basins (Figure 2-3; Figure 2-8) with
38
3.52% genetic distance separating the two BINs. The fact that high intraspecific
divergence was observed within only a portion of the species’ range suggests additional
cryptic diversity may exist. These findings are particularly important given A. radiatus is
currently under review for ESA protection (USFWS 2011).
Pyganodon grandis has been considered the most widespread freshwater
mussel in North America, occurring throughout the Gulf of Mexico drainages from
Florida to Mexico and throughout the Interior Basin and Great Lakes drainage (Williams
et al. 2008). We included samples from four eastern Gulf of Mexico drainages and
Table 2-3. Sample sizes (n), mean and maximum uncorrected p-distance (%), and the distance to the nearest neighbor (NN) species based on 1551 Fcox1 sequences for 57 currently recognized freshwater mussel species in the southeastern United States. Dashes (-) for p-distances represent cases for which the species was only represented by a single specimen
Table 2-4. Sample sizes (n), mean and maximum uncorrected p-distance (%), and the distance to the nearest neighbor (NN) species based on 377 M-cox1 sequences representing 37 currently recognized freshwater mussel species in the southeastern United States. Dashes (-) for p-distances represent cases for which the species is a singleton.
Table 2-5. BIN assignments based on 1551 F-cox1 DNA barcodes representing 57 freshwater mussel species in southeastern United States. Each taxon was assigned one of four scores (match, merge, split, or mixture). Sample sizes (n), mean and maximum pairwise uncorrected p-distances, and distance to nearest neighbor for each BIN are provided.
Figure 2-1. Collection sites in the southeastern United States for the 1571 specimens
analyzed in this study. Each point may represent several collection sites and multiple taxa collected from the same locality.
54
Figure 2-2. Circular phylogram based on 1551 F-cox1 gene sequences representing 23
genera and 57 species of freshwater mussels collected from the southeastern United States.
55
Figure 2-3. Neighbor-joining tree based on 1551 F-cox1 sequences. Taxonomy, specimen identifiers, and Barcode Index Numbers (BIN) assigned to each taxon are given as BOLD:XXXXXXX. Drainages of collection are abbreviated as follows: Apalachicola (Apa), Aucilla (Auc), Chattahoochee (Cha), Chipola (Chi), Choctawhatchee (Cho), Econfina (Eco), Escambia (Esc), Everglades (Eve), Fenhalloway (Fen), Flint (Fli), Mississippi (Mis), Mobile (Mob), Mayakka (Mya), Ochlockonee (Och), Peace (Pea), St. Johns (StJ), St. Marys (StM), Steinhatchee (Ste), Suwannee (Suw), Tampa Bay (Tam), Tomoka (Tom), Withlacoochee (Wit), and Yellow (Yel).
56
Figure 2-3. Continued
57
Figure 2-3. Continued
58
Figure 2-3. Continued
59
Figure 2-3. Continued
60
Figure 2-3. Continued
61
Figure 2-3. Continued
62
Figure 2-3. Continued
63
Figure 2-3. Continued
64
Figure 2-3. Continued
65
Figure 2-3. Continued
66
Figure 2-3. Continued
67
Figure 2-3. Continued
68
Figure 2-3. Continued
69
Figure 2-3. Continued
70
Figure 2-3. Continued
71
Figure 2-3. Continued
72
Figure 2-4. Neighbor-joining tree based on 373 M-cox1 sequences. Taxonomy, specimen identifiers, and Barcode Index Numbers (BIN) assigned to each taxon are given as BOLD:XXXXXXX. Drainages of collection are abbreviated as follows: Apalachicola (Apa), Aucilla (Auc), Chattahoochee (Cha), Choctawhatchee (Cho), Econfina (Eco), Escambia (Esc), Everglades (Eve), Fenhalloway (Fen), Mississippi (Mis), Mobile (Mob), Mayakka (Mya), Ochlockonee (Och), Peace (Pea), St. Johns (StJ), St. Marys (StM), Steinhatchee (Ste), Suwannee (Suw), Tampa Bay (Tam), Tomoka (Tom), Withlacoochee (Wit), and Yellow (Yel).
73
Figure 2-4. Continued
74
Figure 2-4. Continued
75
Figure 2-4. Continued
76
Figure 2-4. Continued
77
Figure 2-5. Frequency distribution histogram of uncorrected pairwise genetic distance
based on 1551 Fcox1 sequences assigned to 57 currently recognized freshwater mussel species in the southeastern United States. The intraspecific and interspecific distances are displayed using black and gray columns, respectively.
78
Figure 2-6. Neighbor-joining subtrees illustrating BIN sharing between Fusconaia burkei
and Fusconaia escambia (left) and Elliptio chipolaensis and Elliptio nigella (right).
79
Figure 2-7. Neighbor-joining subtrees illustrating examples of shallow interspecific
divergence for which two or more species formed a single genetic cluster.
80
Figure 2-8. Neighbor-joining subtrees illustrating examples of deep intraspecific
divergence for which conspecifics were split into two or more BINS, indicating possible cases of cryptic diversity.
81
Figure 2-9. Neighbor-joining subtrees illustrating examples of high intraspecific
divergence in which conspecific individuals were assigned to two or more BINS, indicating possible cases of cryptic diversity.
82
CHAPTER 3 APPLYING DNA BARCODES TO INVESTIGATE ECOLOGICAL HOST
ASSOCIATIONS AND SPECIES BOUNDARIES FOR FRESHWATER MUSSELS
North America is the epicenter of freshwater mussel diversity with nearly 300
species from 55 genera in the United States and Canada (Williams et al. 2017).
Anthropogenic disturbances, including broad-scale habitat modification and degradation
of water quality, have had detrimental impacts on mussel diversity; recent assessments
are that at least 70% of species in North America are imperiled or extinct (Williams et al.
1993, Turgeon et al. 1998, Lydeard et al. 1999, Haag 2012; Haag and Williams 2014).
Much of this diversity resides within isolated river basins that drain into the Gulf of
Mexico where high numbers of endemic taxa have evolved (Haag 2010). Taxonomic
uncertainties and incomplete understanding of basic ecological and life history
requirements for many mussel species of conservation concern are impediments to the
protection and recovery of remaining populations (Haag and Williams 2014; Johnson et
al. 2016; Pfeiffer et al. 2016; McLeod et al. 2017).
Freshwater mussel taxa differ in many aspects of their biology and life history
traits, but nearly all have a highly specialized larval stage (glochidia) that must
parasitize the gills or fins of freshwater fishes to complete metamorphosis to the juvenile
stage. Host use varies widely among species, from ‘host specialist’ that rely on one
species of fish to ‘host generalists’ with larvae that may complete metamorphosis on a
wide variety of fishes (Barnhart et al. 2008; Haag 2012). This difference in host fish
specificity has a profound effect on several important population processes including
recruitment (e.g. availability of suitable hosts), dispersal (e.g. vagility of host), and
resiliency (e.g. host generalist vs. specialist) (Berg et al. 2008; Strayer 2008).
Information on mussel-host relationships becomes especially critical for managers
83
working to protect or recover remaining mussel populations using options that include
captive propagation, augmentation, and reestablishment (Neves 1997; Jones et al.
2006; Haag and Williams 2014; McMurray and Roe 2017).
Over the past century, the study of unionid-host associations has been extensive
(Lefevre and Curtis 1910; Barnhart et al. 2008; Johnson et al. 2016), yet remains
incomplete, as detailed information is lacking or potentially erroneous for many species.
What information is available has been collated into a comprehensive on-line database
(Freshwater Mussel Host Database 2017). Most studies characterized mussel hosts by
ex situ infestations using laboratory inoculation trials that immerse a suite of host fishes
in a glochidia-bath under controlled conditions (Neves et al. 1985; Johnson et al. 2016).
Inoculation trials can be replicated and enable distinction between glochidia that are
rejected from non-suitable hosts and glochidia that complete metamorphosis to the
juvenile life stage. This approach has led to the identification of so called ‘physiological
hosts’ and may not represent host-use under natural conditions (Berg et al. 2008;
Levine et al. 2012). For example, many mussel species have host attraction strategies
that mimic prey items to aid infection of host fishes within specific feeding guilds (e.g.
piscivore vs. insectivore) (Barnhart et al. 2008). During inoculation trials, exposure of
glochidia is human-mediated and does not test ecological compatibility in situ (Hoggarth
1992). Other sources are based on in situ infestations of host fishes to identify
‘ecological hosts’ (Berg et al 2008; Levine et al. 2012). The majority of these sources
rely on observing encysted glochidia on the gills, fins, or skin of fishes infected with
mussel larvae in the wild (Wilson 1916; Coker et al. 1921; Boyer et al. 2011; Levine et
al. 2012). Results from host experiments are routinely categorized by evidence type
84
following Hoggarth (1992). Laboratory-based observations that lack confirmation of
metamorphosis are scored as laboratory infestation (LI), whereas laboratory
transformation (LT) indicates observations of metamorphosis from glochidia to juveniles
under controlled conditions. Similarly, field based observations are designed as natural
infestation (NI) or natural transformation (NT), with the latter based on evidence of
metamorphosis following in situ infestations of glochidia on host fishes.
The laboratory inoculation and in situ approaches have inherent shortcomings
and observations from any study (laboratory or field-based) in which metamorphosis
was not documented should be considered tenuous because glochidia will attach to
nonanimal objects and non-hosts without completing metamorphosis (Lefevre and
Curtis 1910; Haag and Warren 2003; Lellis et al. 2013). Additionally, observations
based on natural infestations are problematic, given difficulties with species-level
identification of encysted glochidia or metamorphosed juveniles (Haag and Warren
2003; Kennedy and Haag 2005). However, recent advancements in molecular
techniques, such as DNA barcoding, have improved our ability to identify mussels of all
life stages (White et al. 1994; Gerke and Tiedemann 2001; Kneeland and Rhymer 2008;
Boyer et al. 2011; Ziertiz et al. 2012; Chapter 2).
Gulf Coast Rivers of the southwestern United States harbor a considerable
portion of the overall mussel diversity in North America and at least 52 species have
been reported from the inland waters of Texas (Howells et al. 1996; Williams et al.
2017). Rivers of the Western Gulf Province (Rio Grande to Brazos River) historically
supported at least 31 species, including 11 endemics (Haag 2010). Nine of these
endemic species, however, are considered threatened by the State of Texas and also
85
are being considered for listing under the US Endangered Species Act (ESA) as
threatened or endangered with critical habitat designation. Taxonomic uncertainties
coupled with limited understanding of ecological hosts, however, complicate listing
decisions and conservation actions. To help guide management efforts for the
candidate species, we evaluated current taxonomy and characterized ecological fish
hosts using DNA barcodes. Specifically, we developed a comprehensive library of DNA
barcodes derived from expert-identified reference material and used it to assign
species-level identifications of metamorphosed juvenile mussels recovered from
naturally parasitized fishes. The goal was to develop a reliable approach for
understanding ecological host-fish requirements for freshwater mussels, using a central
Texas example in which six candidate species were being considered for listing under
the ESA (False Spike, Fusconaia mitchelli; Texas Fatmucket, Lampsilis bracteata;
Texas Fawnsfoot, Truncilla macrodon; Golden Orb, Quadrula aurea; Smooth
Pimpleback, Quadrula houstonensis; and Texas Pimpleback, Quadrula petrina). Finally,
we used DNA barcodes to explore current classification based largely on shell
morphology and biogeography and highlight taxa that warrant further investigation.
Methods
Specimen Collection
Adult freshwater mussels were collected July 2008 – August 2013 using a
combination of tactile searches, hand rakes, and visual underwater searches (e.g.
SCUBA). We aimed to collect ≥ 3 individuals for each of the 24 species known to occur
in the Colorado and Guadalupe River drainages of central Texas (Figure 3-1). We also
sampled outside the drainages to obtain hard to find species that have been reported in
these two drainages. Specimens were identified to the lowest taxonomic unit possible
86
by regional freshwater malacologists using a combination of morphological characters
and geographic location consistent with original species descriptions and primary
literature. All specimens were photographed, preserved in 95% ethanol, assigned
catalog numbers, and deposited at the Florida Museum of Natural History (FLMNH).
Potential host fish were collected from four sites in the Colorado River drainage
and from two sites in the Guadalupe River drainage, in areas where the six ESA
candidate species were known to occur (Table 3-1; Figure 3-1). Fishes were sampled
from all available habitat types using seines or electrofishing by boat or barge and
captured in close proximity to mussel beds identified during previous surveys (Braun et
al. 2015). A sample of fishes representing members of different families and species
were selected to provide a diverse array of potential host fishes. These fish were
transported in aerated coolers to the US Fish and Wildlife Service hatchery in San
Marcos, Texas and separated by species into holding tanks held at 21-24°C. The
bottom of each tank was siphoned daily for 28 days to recover metamorphosed
juveniles or sloughed glochidia. The siphonate was filtered through individual 100 um
mesh sieves and contents were examined using a dissecting microscope.
Metamorphosed juveniles were distinguished from sloughed glochidia by the presence
of soft parts between the valves or observation of individuals actively pedal feeding.
Each metamorphosed juvenile was photographed, measured, and preserved in
individual wells on a 96-well plate containing 95% ethanol.
DNA Sequencing and Data Analyses
Tissue samples and whole juveniles were extracted using a modified plate
extraction protocol (Ivanova et al. 2006). Primers for amplification and bidirectional
sequencing of the maternal copy of cytochrome c oxidase subunit I (F-cox1) were: COI
87
dgLCO-1490 - GGTCAACAAATCATAAAGAYATYGG and COI dgHCO-2198 –
TAAACTTCAGGGTGACCAAARAAYCA (Meyer 2003). The PCR amplifications were
conducted in 27-µl reactions with the following reagents and volumes: H20 (14.74 µl),
We ranked fish hosts depending on the number of transformed juveniles recovered for a
specific mussel species (Figure 3-3). For example, a total of 30 A. plicata, 13 L.
hydiana, and 2 U. imbecillis juveniles were recovered after completing metamorphosis
and naturally releasing from M. salmoides.
The number of fishes identified as hosts for each mussel species differed
considerably and host relationships for most species were concomitant with previous
observations. Similar to previous studies, we categorized each mussel species as a
‘host specialists’ or ‘host generalists’ depending on the taxonomic breadth of fish host
diversity (Barnhart et al. 2008; Haag 2012). For host specialists in our study, we
92
documented I. punctatus as an ecological host for Q. aurea, Q. houstonensis, and Q.
petrina, which is consistent with existing NI and LT observations on closely related
species (Q. pustulosa: Coker et al. 1921). In addition, we identified M. punctulatus as an
ecological host for Q. aurea. This fish had not been previously reported as a host for
any congeners. In general, Quadrula species in the Pustulosa group (Simpson 1900;
Serb et al. 2003; Chapter 4) have been considered host specialists that primarily infect
ictalurid catfishes via reflexive release of glochidia (Barnhart et al. 2008; Sietman et al.
2012). Hosts identified for L. bracteata and L. hydiana were primarily centrarchids,
consistent with previous studies on congeners (Howells 1997; Johnson et al. 2012).
Females of the genus Lampsilis typically have modified mantle tissue, which serve as
mimetic lures to aid in host attraction (Barnhart et al. 2008). Our study revealed that L.
hydiana also utilize several small-bodied fishes of the Cyprinidae (M. marconis) and
Percidae (E. spectabile and P. carbonaria), albeit each of these relationships was based
on recovery of a single metamorphosed juvenile. This finding contradicts the
conventional hypothesis that mussels with large mantle-flap displays utilize only large-
bodied, predatory fishes as ecological hosts (e.g. M. salmoides) (Kraemer 1970;
Barnhart et al. 2008). For T. texasiense, we identified three novel hosts from two
families (Centrarchidae: L. auritus and L. megalotis; Percidae: E. gracile). Prior to our
study, only two hosts were identified for T. texasiense, both centrarchids based on NI
evidence (M. macrochirus and L. gulosus: Stern and Felder 1978).
Our study included two host generalists, A. plicata and U. imbecillis, with at least
29 species (8 families) and 20 species (6 families) of known host fishes, respectively
(Freshwater Mussel Host Database 2017). For A. plicata, we identified six host fishes,
93
three of which were not previously reported (E. gracile, L. auritus, and M. punctulatus).
We confirmed three previously reported host fishes and identified two novel hosts (L.
auritus and P. vigilax) for U. imbecillis.
Discussion
The primary objectives of our project were to evaluate current taxonomy and
investigate ecological host fishes for freshwater mussels in central Texas using our
reference DNA barcode library. We demonstrated that results from DNA barcode
analyses align with current taxonomy for most taxa and highlight cases of incongruence.
We provide information on the ecological hosts for eight freshwater mussel species,
including four of the six federal candidates known from the study area. Our recalibration
of taxonomy and characterization of ecological fish hosts provides information critical for
managers who are working to protect or recover remaining freshwater mussel
populations.
DNA Barcode Reference Library
Our study represents the first comprehensive molecular survey of freshwater
mussel diversity within two river basins in central Texas and joins a growing body of
literature using DNA barcoding to evaluate taxonomy and answer ecological questions
(e.g. Chapter 2). All 24 species of mussels currently recognized from this region were
assessed using F-cox1 barcodes. Our study includes the first BOLD entries for 17 of the
24 species and the first F-cox1 sequences reported for 2 species (L. bracteata and Q.
aurea). The observed distribution of intra- and interspecific genetic distances was within
the range of previously reported values for unionids (Campbell et al. 2008; Boyer et al.
2011; Pfeiffer et al. 2016; Perkins et al. 2017; Chapter 2). DNA barcode clusters were
generally congruent with current taxonomy, indicating agreement with existing
94
approaches used to delineate freshwater mussel taxa in this region. The lack of a
barcode gap (Meyer and Paulay 2005) for some taxa and a few misidentifications,
however, highlight cases that require additional scrutiny.
Importance of DNA Reference Libraries
To ensure reproducibility of our findings and facilitate future research, we
followed DNA barcode data standards (Ratnasingham and Hebert 2007; Benson et al.
2012; Chapter 2) when generating our DNA reference library by including the required
metadata for sequences linked to voucher specimens cataloged in public museums.
These steps are essential to provide a reliable molecular framework for future
ecological, systematic, and conservation studies in light of ongoing taxonomic issues
(Williams et al. 2017), high occurrences of misidentification (Shea et al. 2011; Chapter
2), and inaccuracies of existing mussel sequence data on GenBank (Boyer et al. 2011;
Campbell and Lydeard 2012).
Shallow Interspecific Divergence
Our F-cox1 sequences enabled clear assignment of currently recognized species
to barcode clusters except for Q. aurea and Q. houstonensis (Figure 3-2; Figure 3-4;
Figure 3-5), which are two of six species in the study area currently being considered for
listing under the ESA. The observed overlap between intra- and interspecific genetic
distance values limits our ability to rely on F-cox1 barcodes to distinguish between
these two taxa. As a result, Q. aurea and Q. houstonensis were considered members of
the same OTU and merged into a single BIN (Table 3-4). Current classification is largely
based on the allopatric distribution of these two taxa, with Q. aurea considered endemic
to the Guadalupe and Nueces river basins and Q. houstonensis endemic to the Brazos
and Colorado river basins (Howells et al. 1996). These two taxa are similar
95
morphologically and may represent the same species. Further research is needed to
assess species boundaries among these two imperiled taxa (see Chapter 4).
Deep Intraspecific Divergence
Results obtained in this work and previous studies (Chapter 2) reveal the
existence of cryptic diversity in several species, suggesting freshwater mussel diversity
is still underestimated in some lineages. For example, we observed high intraspecific p-
distances separating allopatric populations of Q. petrina in the Guadalupe and Colorado
rivers (Table 3-3), with each allopatric population representing a distinct OTU BIN
(Table 3-4). In addition, high intraspecific F-cox1 divergence (2.29%) was observed
among U. imbecillis specimens from the Pascagoula River (Table 3-4). These cases
represent the potential for cryptic diversity and warrant further investigation using
additional molecular markers and other independent datasets before formal recognition
and description.
Misidentifications
Misidentification rates of freshwater mussels have been documented in previous
studies (Shea et al. 2011; Chapter 2) and we revealed two cases of misidentification in
our dataset using DNA barcodes. Misidentification of congeners L. hydiana and L.
bracteata was not surprising because these taxa are morphologically similar with
overlapping geographic distributions. For example, both taxa exhibit sexually dimorphic
shell shapes and typically have dark rays that radiate from the umbo to the ventral
margin of the shell (Howells et al. 1996). Similarly, Q. couchiana and Q. petrina are
morphologically similar congeners. The current and historical distribution of Q.
couchiana, however, remains unclear and it is likely that this species is extinct (Williams
et al. 1993; 2017; Howells et al. 1996; Turgeon et al. 1998; Serb et al. 2003).
96
Lack of morphological distinctiveness increases risks of misidentification,
especially for inexperienced malacologist (Shea et al. 2011). This is cause for concern
for imperiled taxa like L. bracteata and Q. petrina, which are being considered for listing
under the ESA. False positive and false negative errors represent sources of error in
survey data and may lead to erroneous abundance estimates, distributional information,
and conservation assessments (Royle and Link 2006; Shea et al. 2011). For example,
instances in which L. bracteata are mistaken for L. hydiana represent false negatives
that underestimate the abundance and distribution of L. bracteata. In contrast, false
positives for L. bracteata may inflate demographic and distribution estimates and
mislead conservation efforts. Our DNA barcode library is a valuable tool for identification
of closely related species and can help resolve identification issues that arise from a
reliance on morphology-based taxonomy.
Fish Hosts
We used our DNA barcode library to provide reliable evidence about ecological
hosts for eight freshwater mussel species, including four of the six federal candidates
known from the study area. Our results were categorized as NT, based on observations
of metamorphosis from glochidia to juveniles on host fishes that were collected after in
situ infestation. Prior to our study, host relationships based on NT observations were
limited to only 11 unionids and 4 host fishes (Howard, 1914; Boyer et al. 2011; Hove et
al. 2012; Freshwater Mussel host Database 2017). Existing host information was
available for five of the eight mussel species, including one of the federal candidates (L.
bracteata). Using our approach, we identified novel ecological hosts for seven unionids,
including the first reported hosts for three federal candidates (Q. aurea, Q.
houstonensis, and Q. petrina).
97
Our approach to characterizing host fish requirements for freshwater mussels
using naturally infested fishes and DNA-based identification of metamorphosed
juveniles circumvents deficiencies inherent with previous methods. Regardless of the
method used, it is critical that host relationships be based on observations of encysted
glochidia that complete metamorphosis and are naturally released from the host, not
merely on observations of encysted glochidia. A replicated inoculation host trial
revealed that reported host-mussel relationships based on the occurrence of encysted
glochidia on naturally infected fishes were erroneous (Fritts et al. 2012). We agree with
Fritts et al. (2012) and urge caution with respect to observations that lack confirmation
of metamorphosis (i.e. LI and NI evidence types), given that glochidia attach to
nonanimal objects and non-hosts without completing metamorphosis (Lefevre and
Curtis 1910; Lellis et al. 2013; Johnson et al. 2016; McLeod et al. 2017). In contrast, a
field study based on observations of encysted glochidia showed that laboratory trials
may overestimate the number of ecological hosts (Levine et al. 2012). Field-based
studies that rely on morphological characters to identify encysted glochidia or
metamorphosed juveniles however, are problematic, given difficulties with identification
to the species-level (Haag and Warren 2003; Kennedy and Haag 2005), except in cases
of extremely low diversity (Levine et al. 2012). These conflicts raise questions about the
reliability of both field and laboratory-based studies and support the use of replicated LT
experiments to confirm the results of NT studies, especially those that lack molecular-
based identifications of metamorphosed juveniles (Boyer et al. 2011; Fritts et al. 2012).
As with many other metrics used to assess conservation status and extinction
risks, there is often a lack of detailed information on the life histories of species that are
98
of conservation concern. Intuitively, one may hypothesize that mussels relying on rare,
narrowly distributed fishes are more susceptible to imperilment than those that depend
on common, widespread fishes. However, our findings, along with previous studies
(Haag and Warren 2003; Haag 2012; Haag and Williams 2014), contradict this
assumption, in that four of the five candidate species use common, widespread fishes
as hosts (e.g. I. punctatus and L. macrochirus). These data provide insights on
important population processes including recruitment (e.g. availability of suitable hosts),
dispersal (e.g. vagility of host), and resiliency (e.g. host generalist vs. specialist),
making this information critical for managers who are working to protect or recover
remaining freshwater mussel populations. Furthermore, the greater understanding of
reproductive traits and requirements of these animals provides an opportunity for future
recovery efforts that may include propagation and culture, which is a common method
used to recover mussel species and protect populations from extinction (Neves 1997;
Jones et al. 2006; Haag and Williams 2014; McMurray and Roe 2017).
99
Table 3-1. Collection sites (site abbreviations) and sampling dates for the five fish surveys where metamorphosed juveniles were recovered and identified using DNA barcodes.
Site Location Latitude Longitude Drainage Collection Dates
Juvenile identifications
1 San Saba River Bois D'Arc Road Menard, Texas
30.8995 -99.9107 Colorado 22 May 2012 4 April 2013
Table 3-3. Sample sizes (n), mean and maximum intraspecific p-distances (d), and distance to nearest neighbor species (NN) shown as percentages for taxa included in the F-cox1 barcode library.
Table 3-4. Barcode index number (BIN) assignments based on 124 F-cox1 DNA barcode sequences representing 24 freshwater mussel species known from central Texas. Each taxon was assigned to one of four categories (match, merge, split, or mixture). Sample sizes (n), mean and maximum pairwise uncorrected p-distances, and distance to nearest neighbor species (NN) for each BIN are provided.
Taxa BIN Score n Mean
Distance Max
Distance NN
Distance
Amblema plicata BOLD:AAA8507 Match 9 0.43 1.32 10.03 Arcidens confragosus BOLD:ACQ0671 Match 1 0.00 0.00 12.23 Cyrtonaias tampicoensis BOLD:AAE7911 Match 6 0.47 0.93 10.24 Fusconaia mitchelli BOLD:ATZ1101 Match 9 0.85 2.23 10.80 Glebula rotundata BOLD:AAF5442 Match 19 0.05 0.31 10.34 Lampsilis bracteata BOLD:ATZ1105 Match 6 0.09 0.19 8.57 Lampsilis hydiana BOLD:ATZ1118 Match 2 2.14 2.14 7.72 Lampsilis teres BOLD:AAF4542 Match 8 0.34 0.62 7.72 Leptodea fragilis BOLD:ADC8698 Match 1 0.00 0.00 6.88 Ligumia subrostrata BOLD:ACZ0248 Match 5 0.06 0.15 8.98 Megalonaias nervosa BOLD:AAX3278 Match 2 0.15 0.15 11.36 Potamilus purpuratus BOLD:AAE3348 Match 1 0.00 0.00 6.88 Pyganodon grandis BOLD:ACY9847 Match 3 0.31 0.45 12.69 Quadrula apiculata BOLD:ACH7152 Match 1 0.00 0.00 8.45 Quadrula aurea BOLD:AAI0636 Merge 13 0.92 2.01 4.46 Quadrula houstonensis BOLD:AAI0636 Merge 6 0.92 2.01 4.46 Quadrula petrina BOLD:ATZ1107 Split 3 0.55 0.86 3.40
BOLD:ATZ1115
6 0.64 1.07 3.40
Strophitus undulatus BOLD:AAI0012 Match 2 0.00 0.00 11.31 Toxolasma texasiense BOLD:AAK1891 Match 2 0.15 0.15 10.86 Tritigonia verrucosa BOLD:AAW7935 Match 2 0.31 0.31 9.41 Truncilla macrodon BOLD:ACH2404 Match 3 0.10 0.16 9.17 Uniomerus declivis BOLD:AAW8486 Match 2 0.00 0.00 4.67 Uniomerus tetralasmus BOLD:ACY9741 Match 6 0.05 0.16 4.67 Utterbackia imbecillis BOLD:ATZ1123 Split 2 0.00 0.00 2.29
BOLD:ACD2688
4 0.00 0.00 2.29
106
Table 3-5. Naturally infested host fishes that produced juvenile freshwater mussels. Site numbers follow Table 3-1. The total number of juveniles for each mussel species is shown in parenthesis.
Family and species Common Name Site Collection Date Number of Juveniles
Centrarchidae Lepomis auritus
Redbreast Sunfish
1 6
22 May 2012 29 May 2012
U. imbecillis (2) A. plicata (1) T. texasiense (1)
Lepomis cyanellus Green Sunfish 1 22 May 2012 U. imbecillis (3) Lepomis macrochirus Bluegill 1
3 6
22 May 2012 29 April 2013 29 May 2012
L. bracteata (1) U. imbecillis (3) U. imbecillis (1) A. plicata (1)
Lepomis megalotis Longear Sunfish 4 5 6
11 June 2013 30 May 2012 29 May 2012
A. plicata (3) Q. houstonensis (2) A. plicata (9) T. texasiense (5) A. plicata (21) T. texasiense (1)
Micropterus punctulatus Spotted Bass 5 30 May 2012 A. plicata (4) L. hydiana (2) Q. aurea (1)
Micropterus salmoides Largemouth Bass 2 3 6
21 Aug 2012 29 April 2013 29 May 2012
L. hydiana (1) U. imbecillis (2) A. plicata (30) L. hydiana (12)
Cyprinidae Macrhybosis marconis
Burrhead Chub
5
30 May 2012
L. hydiana (1)
Pimephales vigilax Bullhead Minnow 3 29 April 2013 U. imbecillis (1) Ictaluridae Ictalurus punctatus
Channel Catfish
3 5 6
29 April 2013 30 May 2012 29 May 2012
Q. petrina (6) Q. aurea (7) Q. aurea (7)
Percidae Etheostoma gracile
Slough Darter
5 6
30 May 2012 29 May 2012
A. plicata (2) T. texasiense (1) A. plicata (4)
Etheostoma spectabile Orangethroat Darter 6 29 May 2012 L. hydiana (1) Percina carbonaria Texas Logperch 3 29 April 2013 L. hydiana (1)
107
Figure 3-1. Sampling locations in central Texas for fishes that produced juvenile
freshwater mussels identified using DNA barcodes.
108
Figure 3-2. Neighbor-joining tree based on 124 F-cox1 sequences. Taxonomy, specimen identifiers, and Barcode Index Numbers (BIN) assigned to each taxon are given as BOLD:XXXXXXX. Drainages of collection are abbreviated as follows: Colorado (Col), Galveston Bay (Gav), Guadalupe (Gua), Pascagoula (Pas), and Sabine (Sab).
109
Figure 3-2. Continued.
110
Figure 3-2. Continued.
111
Figure 3-3. Most likely topology generated in the BI analysis with indications of clades
containing juvenile mussels recovered from naturally infested fishes. Values above and below the branch lengths equal BI posterior probability and ML bootstrap support, respectively. Colors indicate the number of juvenile mussels recovered and identified from each fish species.
112
Figure 3-4. Intraspecific and interspecific uncorrected p-distances with cases of high
intraspecific variation and low interspecific divergence indicated.
113
Figure 3-5. Scatterplot illustrating the overlap of maximum intraspecific p-distances with
the nearest neighbor distances. Points above the diagonal line indicate species with a barcode gap.
114
CHAPTER 4 INTEGRATIVE TAXONOMY RESOLVES GENERIC PLACEMENT AND SPECIES
BOUNDARIES FOR IMPERILED FRESHWATER MUSSELS
Accurate taxonomy is critical to accurately define and classify biodiversity. In
addition, good taxonomy has profound implications for biological inferences regarding
biological characteristics, ecological responses, and conservation priorities, among
other pursuits (Barraclough and Nee 2001; Mace 2004). Methods used to delineate
genera and species continue to evolve and conflicts often reflect different interpretations
of available data types (e.g. morphological vs. molecular). Model-based approaches
such as multispecies coalescent models (MSC) are also powerful methods (Rannala
and Yang 2003) that have been utilized to delimit species boundaries (Leaché and
Fujita 2010; Fujita and Leaché 2011). Recent empirical studies, however, have criticized
MSC models for identifying population structure rather than species boundaries (Hedin
2015; Pfeiffer et al. 2016; Sukumaran and Knowles 2017; Willis 2017), which suggests
caution is prudent when basing species hypotheses solely on MSC models. Integrative
approaches that draw inference from multiple independent lines of evidence have been
called for increasingly (Dayrat 2005; Leache et al. 2009; Knowles and Carstens 2007;
Padial et al. 2010; Schlick-Steiner et al. 2010; Carstens et al. 2013) and reveal that
morphological characters or geographic distributions alone are not necessarily
diagnostic at the generic and species levels (Jones et al. 2006; Huang and Knowles
2016; Pfeiffer et al. 2016; Perkins et al. 2017).
An example of a high-diversity taxonomic group that has been characterized
historically based on distributional patterns and phenotypic diagnostics is freshwater
mussels (Bivalvia: Unionidae), which is among the most critically endangered faunas on
Earth. At least 10% of the unionid taxa in the United States are extinct, and 65% of the
115
remaining species are considered imperiled (Williams et al. 1993; Haag 2012; Haag and
Williams 2014). Conservation efforts focused on freshwater mussels are complicated by
taxonomic uncertainty that stems from limited discrete morphological characteristics that
would enable species diagnosis or determination of evolutionary lineages. Molecular
phylogenetics has changed the delimitation of freshwater mussel species boundaries
dramatically by revealing morphologically cryptic diversity (Roe and Lydeard 1998; King
et al. 1999; Lydeard et al. 2000; Jones et al. 2006; Pfeiffer et al. 2016) and
demonstrating that morphology-based assessments alone can misguide classification
and conservation (Mulvey et al. 1997; Serb et al. 2003; Pfeiffer et al. 2016; Perkins et al.
2017).
New molecular tools, analytical methods, and studies diagnosing other intrinsic
traits offer opportunities to further recalibrate unionid taxonomy. This work is urgent
when taxonomic uncertainties complicate the development and implementation of
conservation and recovery programs. North American unionids in the tribe Quadrulini
have been the focus of several generic, species, and population level genetic studies
(Berg et al. 1998; Serb et al. 2003; Roe and Boyer 2015) but a comprehensive sampling
using multiple, independently evolving molecular markers is lacking. Recent efforts to
compile and expand life history information for several members of Quadrulini (Hove et
al. 2012; Sietman et al. 2012; Harriger et al. 2015) coupled with several members of the
genus Quadrula being considered for protection under the Endangered Species Act
(ESA) has stimulated interest in revisiting the systematics of the Quadrulini.
Systematics of the Quadrulini has long been a source of taxonomic debate and
confusion and remains in a chaotic state at the generic and species level (Simpson
116
1900; 1914; Ortmann 1912; Frierson 1927; Vidrine 1993; Howells et al. 1996; Serb et al.
2003; Graf and Cummings 2007; Campbell and Lydeard 2012). Recent taxonomic
treatments recognized between five and nine genera in Quadrulini based on
authoritative interpretation of morphological and life history characters (Williams et al.
1993; 2008; 2014; 2017; Turgeon et al. 1988; 1998; Serb et al. 2003; Graf and
Cummings 2007; Campbell and Lydeard 2012; Lopez-Lima et al. 2017). Species
boundaries are complicated by a variety of morphological and geographic forms that
have perplexed systematists for decades, a consequence of high intraspecific variation
in shell morphology that often overlaps between species (Valentine and Stansbery,
1971; Neck, 1982; Vidrine et al. 1993; Howells et al. 2002; Serb et al. 2003; Williams et
al. 2008). Several taxa that occupy Gulf of Mexico drainages and lower sections of the
Interior Basin of North America were recognized as either distinct species or subspecies
of Q. pustulosa by recent taxonomic treatments (Turgeon et al. 1988; 1998; Vidrine
1993; Williams et al. 1993; 2008; 2014; Howells et al. 1996; Graf and Cummings 2007).
Figure 4-4). Both BI and ML analyses resolved Q. petrina as paraphyletic with regard to
125
Q. nodulata. The two divergent Q. petrina clades correspond to individuals sampled
from the Colorado and Guadalupe rivers, with the Colorado River clade being sister to
Q. nodulata. This provides credible evidence that species-level diversity is
underestimated in this complex. mtDNA sequence divergence exhibited a clear gap
between intraspecific variation and interspecific divergence among the three
geographically isolated clades (Figure 4-5), indicative of species-level divergence and
similar to values reported for several other freshwater mussel species (Roe and
Lydeard 1998; Serb et al. 2003; Jones et al. 2006; Campbell et al. 2008; Inoue et al.
2014; Pfeiffer et al. 2016; Perkins et al. 2017). Sequence divergence at ITS1 was lower
relative to both mtDNA loci but consistent with patterns observed in previous studies
utilizing these genes (Pfeiffer et al. 2016; Perkins et al. 2017). Morphometric analyses
also suggest clear separation of Q. nodulata and the Colorado and Guadalupe Q.
petrina clades (Figure 4-3).
Prior to our study, little information was available regarding phylogenetic
relationships between members of the Q. pustulosa complex. Previous researchers
questioned the validity of taxa in the Q. pustulosa complex because of difficulties
distinguishing between morphological forms, geographic variants, and distinct species
(Strecker 1931; Turgeon et al. 1988; 1998; Vidrine 1993; Williams et al. 1993; 2008;
2014; Howells et. al 1996; Graf and Cummings 2007; Watters et al. 2009). For our
assessment, we allowed geographic distributions based on current taxonomy to
represent the null species hypotheses. Our molecular and morphometric data indicate
that current taxonomy overestimates species-level diversity in the Q. pustulosa
complex. In fact, our data show greater genetic divergence and morphological
126
distinctiveness between the two geographically isolated populations of Q. petrina than
between all Q. aurea, Q. houstonenis, Q. mortoni, Q. pustulosa, and Q. refulgens
sampled. All five taxa previously recognized as species or subspecies in the Q.
pustulosa species complex exhibited extensive paraphyly (Figure 4-4), with no clear
distinction between intraspecific variation and interspecific divergence at mtDNA loci or
clear signals for diagnosis using morphological characters (Figure 4-4). With the
exception of Q. succissa, relationships among mtDNA haplotypes show weak
associations with currently recognized taxonomy and several nominal taxa share ITS1
haplotypes (Figure 4-5). Additionally, morphometric analyses illustrated limited ability to
distinguish between members of the Q. pustulosa complex using shell measurements.
Specifically, Q. houstonensis, Q. mortoni, Q. pustulosa, and Q. refulgens were
indistinguishable. Both Q. aurea and Q. succissa were found to be significantly more
compressed than other members of the complex yet only 74% of individuals identified
morphologically as Q. aurea were binned correctly, with 25% assigned to Q. succissa.
Our molecular-based analyses, however, do not support the recognition of Q. aurea as
a distinct species and we suspect that the observed morphological differences in Q.
aurea may be a product of ecophenotypic variation, a common phenomenon in
freshwater mussels (Ortmann 1920; Eagar 1954; Zieritz et al. 2010; Inoue et al. 2013;
Bourdeau et al. 2015; Fassatoui et al. 2015; Zajac et al. 2017).
Implications for Taxonomy and Conservation
Our study is the first to analyze extensive phylogeographic and morphometric
variation in the Q. pustulosa and Q. petrina species complexes, joining a growing
number of empirical studies that show patterns of diversity in freshwater mussels are
complex and do not always match expectations based on morphological characters
127
(Inoue et al. 2013; 2014; Pfeiffer et al. 2016; Perkins et al. 2017). Considering the lack
of diagnosable units across multiple independent lines of evidence, we suggest that Q.
aurea, Q. houstonensis, Q. mortoni, and Q. refulgens be designated as synonyms of Q.
pustulosa. This expands the distribution of Q. pustulosa from the Pascagoula River
drainage west to the Nueces River drainage in south Texas (Figure 4-1). Our
phylogeographic assessment shows geographic structuring of populations within Q.
pustulosa sensu lato, which provides resource managers with valuable information for
future recovery efforts, especially those involving propagation, augmentation,
translocation, and reintroduction (Jones et al. 2006; McMurray and Roe 2017).
Additionally, our findings provide compelling evidence for recognition of an undescribed
species in the Q. petrina species complex that is endemic to the Guadalupe River.
These taxonomic treatments may impact listing decisions by resource management
agencies considering our findings suggest that two species (Q. aurea and Q.
houstonensis) may be synonyms of Q. pustulosa, and another species (Q. petrina)
contains a cryptic lineage that may represent an undescribed species.
Discussion of Generic-level Relationships
Several recent molecular phylogenies have helped resolve the supraspecific
relationships of the Quadrulini (Serb et al. 2003; Campbell and Lydeard 2012), but
interpretations of these relationships have led to several incongruent generic-level
classifications (Serb et al. 2003; Graf and Cummings 2007; Williams et al. 2008; 2014;
2017; Campbell and Lydeard 2012; Lopez-Lima et al. 2017; Williams et al. 2017). Our
phylogenetic analyses support the recognition of six genera within Quadrulini:
Cyclonaias, Megalonaias, Quadrula, Theliderma, Tritogonia, and Uniomerus (Table 4-1;
Figure 4-2). Similar to previous molecular studies, Theliderma was recovered as sister
128
to Tritogonia verrucosa and Quadrula s.s. (Serb et al. 2003; Campbell and Lydeard
2012). Tritogonia has been treated either as a synonym of Quadrula s.s. (Serb et al.
2003; Williams et al. 2008) or as a monotypic genus (Graf and Cummings 2007;
Watters et al. 2009; Lopez-Lima et al. 2017; Williams et al. 2017). Our results support
the two most recent assessments, which recognize Tritogonia as a monotypic genus
distinct from Quadrula s.s. (Watters et al. 2009; Lopez-Lima et al. 2017; Williams et al.
2017).
The genus Cyclonaias has long been considered monotypic and distinguished
from Quadrula, Theliderma, and Tritogonia by only brooding larvae in the outer two gills
(Simpson 1900; 1914; Ortmann 1912; 1919; Walker 1918; Williams et al. 2008; Watters
et al. 2009). However, Frierson (1927) reported C. tuberculata to brood larvae in all four
gills and subsequently described the genus Cyclonaias as “recalcitrant” and playing
“havoc with classification” because of variability in brooding morphology. Furthermore,
at least three other species, Q. apiculata, Theliderma cylindrica, and T. verrucosa, have
been reported to brood larvae in two or four gills (Simpson 1914; Yeager and Neves
1986; Williams et al. 2008). Phylogenetic relationships do not support previous
classifications based on larval brooding morphology indicating that the number of gills
involved in larval brooding can vary and may not represent shared ancestral states
among genera and species of the Quadrulini (Figure 4-2; Campbell and Lydeard 2012).
We recovered C. tuberculata within a well-supported clade (BS/PP=97/94) that
included taxa previously assigned to Amphinaias (Graf and Cummings 2007), Quadrula
(Simpson 1914; Williams et al. 1993; Turgeon et al. 1988; 1998; Serb et al. 2003;
Williams et al. 2008; 2014), Quincuncina (Graf and Cummings 2007), and Rotundaria
129
(Campbell and Lydeard 2012) (Figure 4-2). Several genus- or subgenus-level names
have been used recently for this group but no consensus has been reached. Graf and
Cummings (2007) resurrected Amphinaias based on the molecular phylogeny of Serb et
al. (2003) and the morphological groups of Simpson (1900). Campbell and Lydeard
(2012) resolved C. tuberculata nested within a paraphyletic Amphinaias, and
subsequently resurrected the epithet Rotundaria (Agassiz 1852) to represent this clade
based on statements in Valenciennes (1827). However, Valenciennes (1827) did not
explicitly state that C. tuberculata was the type of Rotundaria. Ortmann and Walker
(1922) clarified this issue pointing out that Herrmannsen (1848) designated Obovaria
subrotunda as the type of Rotundaria, relegating Rotundaria to a junior synonym of
Obovaria and recognized Cyclonaias tuberculata. Therefore, treatment of Rafinesque’s
type of Unio tuberculata as the type species of Rotundaria is invalid.
The type species of Amphinaias, A. couchiana (Lea 1860), could not be included
in this analysis and is thought to be extinct (Williams et al. 1993; Howells et al. 1996;
Turgeon et al. 1998; Serb et al. 2003; Williams et al. 2017). Morphologically, A.
couchiana most closely resembles members of Quadrula s.s. and has been allied with
this group in previous assessments (Simpson 1900; 1914; Strecker 1931) and we follow
Williams et al. (2017) by supporting the combination Quadrula couchiana. Regardless of
the generic placement of Unio couchiana, the inclusion of C. tuberculata in the clade
representing 12 taxa, including Q. pustulosa, makes Cyclonaias the oldest name
available. The priority of Cyclonaias applies to the generic epithet Pustulosa (Frierson,
1927) as well. Accordingly, we propose that the following 12 species included in this
study be assigned to the genus Cyclonaias: C. aurea, C. asperata, C. houstonensis, C.
130
infucata, C. kleiniana, C. mortoni, C. nodulata, C. petrina, C. pustulosa, C. refulgens, C.
succissa, and C. tuberculata (Table 4-1). This finding follows closely the most recent
taxonomic revision of North American unionids (Williams et al. 2017).
Conclusions
In this study, we used an integrative approach that considered molecular,
distribution, and morphology data to evaluate relationships within and among several
genera of the Quadrulini. Our phylogenetic analyses revealed that morphological and
anatomical characters considered to be synapomorphic at the genus-level may have
misled prior taxonomy. We used our findings to revise generic-level classifications
(Table 4-1; Figure 4-2). At the species level, congruence across all lines of evidence
indicates that current taxonomy overestimates diversity in the Cyclonaias (Quadrula)
pustulosa species complex, while underestimating diversity in the Cyclonaias
(Quadrula) petrina species complex. These findings may affect future conservation and
management efforts, especially for the three species (C. aurea, C. houstonensis, and C.
petrina) under consideration for listing by the US Fish and Wildlife Service (USFWS
2011).
131
Table 4-1. Taxa sampled, drainage of collection, and number of sequences for all individuals included in molecular analyses. * denotes taxa within the Cyclonaias (Quadrula) petrina complex and ** denotes taxa within the Cyclonaias (Quadrula) pustulosa complex.
Table 4-2. Analysis of molecular variance (AMOVA) among members of the Cyclonaias (Quadrula) petrina and Cyclonaias (Quadrula) pustulosa species complexes. Samples were grouped according to current taxonomy. All values were significant (P < 0.0001).
Percentage of variance Source of variation CO1 ND1
Cyclonaias (Quadrula) petrina complex Among groups 87.45 88.98 Within groups 12.55 11.02 Cyclonaias (Quadrula) pustulosa complex Among groups 47.58 48.68 Within groups 52.42 51.32
133
Figure 4-1. Map showing sampled localities (dots) for members of the Cyclonaias
(Quadrula) pustulosa species complex (left) and Cyclonaias (Quadrula) petrina species complex (right). Colors correspond to groups within each complex: C. pustulosa complex - red (C.aurea), green (C. houstonensis), purple (C. mortoni), yellow (C. pustulosa), blue (C. refulgens), and cyan (C. succissa); C. petrina complex - red (C. nodulata), green (C. petrina), and blue (C. sp. cf. petrina).
134
Figure 4-2. Maximum likelihood (ML) phylogeny based on concatenated mtDNA and
nDNA datasets for Quadrulini. Nodes are collapsed into single species-level clades with revised taxonomy (old generic names in parentheses). Asterisks above and below nodes represent ≥ 99% bootstrap and 0.99 posterior probability support, respectively. Number in parentheses after taxon name indicates sample size.
135
Figure 4-3. Comparison of results for members of the Cyclonaias (Quadrula) petrina
species complex. Clockwise from the top-left panel: Fully-resolved and expanded phylogeny based on COI, NDI, and ITS1 sequences; COI+ND1 haplotype network; PCA plots with 95% CI ellipses and arrows for biplot variables (HL, height/length; WL, width/length; WH, width/height); and ITS1 haplotype network. Colors indicate the following taxa: Red (Cyclonaias nodulata); Green (Cyclonaias petrina); Blue (Cyclonaias sp. cf. petrina). Black dots on the networks represent missing or unsampled haplotypes.
136
Figure 4-4. Comparison of results for members of the Cyclonaias (Quadrula) pustulosa
species complex. Clockwise from top-left panel: Fully resolved and expanded phylogeny based on COI, NDI, and ITS1 sequences; COI+ND1 haplotype network; PCA plots with 95% CI ellipses and arrows for biplot variables (HL, height/length; WL, width/length; WH, width/height); and ITS1 haplotype network. Colors indicate the following taxa: Red (Cyclonaias aurea); Green (Cyclonaias houstonensis); Purple (Cyclonaias mortoni); Yellow (Cyclonaias pustulosa); Blue (Cyclonaias refulgens); Cyan (Cyclonaias succissa). Black dots on the networks represent missing or unsampled haplotypes.
137
Figure 4-5. Histograms illustrating the distribution of all intraspecific and interspecific
pairwise uncorrected-p distances for Cyclonaias (Quadrula) petrina complex (top) and Cyclonaias (Quadrula) pustulosa complex (bottom) based on COI (left) and ND1 (right).
138
CHAPTER 5 CONCLUSIONS
In this dissertation, I aspired to initiate a new epoch of biodiversity assessments
for freshwater mussels. By utilizing DNA barcoding as a method for i) testing current
taxonomy, ii) identifying cases of cryptic diversity, and iii) accounting for
misidentifications, I have developed a viable model for modern taxonomic treatment of
freshwater mussels that overcomes some of the historical issues associated with
dependency on difficult morphological traits. My utilization of the DNA barcoding
approach to answer important ecological questions related to the complex life history of
freshwater mussels not only expands our knowledge with respect to host fish
requirements, but also provides timely information needed to make conservation
management decisions. Finally, I demonstrated how DNA barcodes represent an
important taxonomic first-step and that multiple independent lines of evidence (e.g.
morphology, genetics, behavior, geography) can be integrated to make holistic
decisions regarding evolutionary relationships. My results have important implications
for the systematics and conservation of a highly imperiled group of freshwater mussels,
and have the potential to catalyze a unified effort toward taxonomy and conservation at
regional, national, and global scales.
This project began as my Ph.D. dissertation and has evolved into a federal
research program focused on understanding the diversity and distributions of unionids.
My efforts to advance science and conservation of freshwater mussels provide more
than only publications and scientific reports. For example, beyond publishing my
findings in peer-reviewed journals, results of my research have been disseminated
through professional workshops, guest lectures, webinars, presentations at professional
139
meetings, and online resources (e.g. Barcode of Life Database, US Geological Survey
homepage, and ResearchGate). Such outlets have allowed me to connect with a
diverse audience of scientists, resource managers, consultants, students, and other
members of the general public. This is an important contribution given that many
resource managers and non-scientists have difficulties understating the value of
characterizing patterns of speciation or quantifying biodiversity within a highly imperiled
group of organisms.
Until recently, taxonomic assessments that address unionids relied primarily on
morphological characters and biogeographic patterns to delineate species boundaries
(Turgeon et al. 1988; 1998; Williams et al. 1993). Since 1998, modern taxonomy and
systematics have routinely utilized molecular characters along with other lines of
evidence (e.g. soft anatomy, behavior, geography), which has reopened the question of
how many species of freshwater mussels currently exist. In fact, a recent account of
North American freshwater mussels identified > 100 taxonomic changes for 298 species
(Williams et al. 2017), the majority resulting from studies that included DNA evidence.
This underscores the value of taxonomic assessments that incorporate molecular data.
Unfortunately, the vast majority of molecular data for unionids lack associated
voucher specimens and adherence to DNA barcode community standards for
maintenance of records, voucher specimens, and supporting information. This is what
sets the BOLD database apart from GenBank (Ransasingham and Hebert 2007). As of
October 2017, a total of 5,659 COI sequences were available for the Unionidae in
GenBank. The percentage of these with linkable voucher specimens is unknown, but
most seem to lack a definitive association with any voucher information. Ambiguous
140
DNA sequences linked to vouchers specimens can be evaluated easily to determine
whether issues are based on misidentification, laboratory contamination, or taxonomic
problems. Without vouchers, the issue remains dubious and can mislead future
research. For example, a study that used DNA barcodes to identify juvenile mussels
encountered issues with identifications using GenBank data (Boyer et al. 2011). Another
case is based on a contaminated sequence has impacted a variety of studies and
misguided generic assignments of several taxa within the tribe Quadrulini (Serb et al.
2003; Graf and Cumming 2007; Campbell and Lydeard 2012; Williams et al. 2017;
Chapter 4).
My combined efforts in this dissertation provided the first DNA-compliant barcodes
for nearly 1700 specimens representing 81 currently recognized species. This is nearly
one-third of the extant unionid diversity known from the United States and Canada.
Moving forward, my goal is to coordinate the creation of a comprehensive DNA barcode
library for all North American unionids known as UNIO-BARCODE. The completion of
this task will offer powerful tools for expanding our understanding of the diversity,
distribution, and ecology of freshwater mussels. This DNA barcode initiative will
facilitate fast, reliable species-level identification by specialists and non-specialists alike,
and remove a major impediment to the conservation and recovery of remaining
freshwater mussel diversity.
Organizing the effort through BOLD provides online access to large-scale
collections with high-resolution digital images and comprehensive DNA libraries. It
provides an online identification guide that puts results of accurate taxonomy and
systematics in the hands of scientists and non-scientists, including the public and
141
conservation managers. The online framework provides a resource for teaching
workshops and guiding future taxonomic assessments. It also facilitates access to
reliable, up-to-date information by allowing collections to be easily curated following
changes in taxonomy. Finally, it facilitates collaboration among experts in multiple
and identify all specimens to the species level at any life stage. Given the taxonomic
breadth of this endeavor, no single researcher can be expected to carry out this project.
The task, however, can be completed through a network of regional collaborations. The
goal is that all researchers involved with freshwater mussel research and conservation
management will benefit from this project.
142
LIST OF REFERENCES
April, J., Mayden, R.L., Hanner, R.H., Bernatchez, L., 2011. Genetic calibration of species diversity among North America's freshwater fishes. Proceedings of the National Acadamy of Sciences. 108, 10602–10607. doi:10.1073/pnas.1016437108/-/DCSupplemental/pnas.201016437SI.pdf
Barnhart, M.C., Haag, W.R., Roston, W.N., 2008. Adaptations to host infection and
larval parasitism in Unionoida. Journal of the North American Benthological Society 27, 370–394. doi:10.1899/07-093.1
Barraclough T.G., and S. Nee. 2001. Phylogenetics and speciation. Trends in Ecology
and Evolution 16:391–399. Benson, D.A., Karsch-Mizrachi, I., Clark, K., Lipman, D.J., Ostell, J., Sayers, E.W.,
2011. GenBank. Nucleic Acids Research 40, D48–D53. doi:10.1093/nar/gkr1202 Berg, D.J., Levine, T.D., Stoeckel, J.A., Lang, B.K., 2008. A conceptual model linking
demography and population genetics of freshwater mussels. Journal of the North American Benthological Society 27, 395–408. doi:10.1899/07-091.1
Blaxter, M., 2004. The promise of a DNA taxonomy. Philosophical Transactions:
Biological Sciences 359, 669–679. Bouckaert, R., Heled, J., Kühnert, D., Vaughan, T., Wu, C.H., Xie, D., Suchard, M.A.,
Rambaut, A., Drummond, A.J., 2014. BEAST 2: A Software Platform for Bayesian Evolutionary Analysis. PLoS Computational Biology 10, e1003537. doi:10.1371/journal.pcbi.1003537.
Hollander, J., 2015. What can aquatic gastropods tell us about phenotypic plasticity? A review and meta-analysis 115, 312–321. doi:10.1038/hdy.2015.58
Boyer, S., Howe, A., Juergens, N., Hove, M., 2011. A DNA-barcoding approach to
identifying juvenile freshwater mussels (Bivalvia: Unionidae) recovered from naturally infested fishes. Journal of the North American Benthological Society 30, 182–194.
Braun, C. L., C. L. Stevens, P. D. Echo-Hawk, N. A. Johnson, and J. B. Moring. 2014.
Abundance of Host Fish and Frequency of Glochidial Parasitism in Fish Assessed in Field and Laboratory Settings and Frequency of Juvenile Mussels or Glochidia Recovered from Hatchery-Held Fish, Central and Southeastern Texas, 2012–13. Scientific Investigations Report 2014–5217
143
Breton, S., Beaupré, H., Stewart, D., Hoeh, W., Blier, P., 2007. The unusual system of doubly uniparental inheritance of mtDNA: isn’t one enough? Trends in Genetics 23, 465–474.
S., Stewart, D.T., Guerra, D., 2017. The extremely divergent maternally- and paternally-transmitted mitochondrial genomes are co-expressed in somatic tissues of two freshwater mussel species with doubly uniparental inheritance of mtDNA. PLoS ONE 12, e0183529–13. doi:10.1371/journal.pone.0183529
Burkhead, N.M., 2012. Extinction Rates in North American Freshwater Fishes, 1900–
2010. BioScience 62, 798–808. doi:10.1525/bio.2012.62.9.5 Campbell, D., Serb, J., Buhay, J., Roe, K., Minton, R., Lydeard, C., 2005. Phylogeny of
North American amblemines (Bivalvia, Unionoida): prodigious polyphyly proves pervasive across genera. Invertebrate Biology 124, 131–164.
Lydeard, C., 2008. Identification of “extinct” freshwater mussel species using DNA barcoding. Molecular Ecology Resources 8, 711–724. doi:10.1111/j.1755-0998.2008.02108.x
Campbell, D.C., Lydeard, C., 2012. The Genera of Pleurobemini (Bivalvia: Unionidae:
Ambleminae). American Malacological Bulletin 30, 19–38. Casiraghi, M., Labra, M., Ferri, E., Galimberti, A., De Mattia, F., 2010. DNA barcoding: a
six-question tour to improve users' awareness about the method. Briefings in Bioinformatics 1–14. doi:10.1093/bib/bbq003
Chambers, E.A., Hebert, P.D.N., 2016. Assessing DNA Barcodes for Species
Identification in North American Reptiles and Amphibians in Natural History Collections. PLoS ONE 11, e0154363. doi:10.1371/journal.pone.0154363
Clement M, Snell Q, Walker P, Posada D, Crandall K (2002). TCS: Estimating gene
genealogies. Molecular Ecology 9, 1657–1659. Coker, R.E., A.F. Shira, H.W. Clark, and A.D. Howard. 1921. Natural history and
propagation of fresh-water mussels. Bulletin of the Bureau of Fisheries. 37(1919-20):75-181 + 17 plates.
Collins, R.A., Cruickshank, R.H., 2012. The seven deadly sins of DNA barcoding.
Molecular Ecology Resources n/a–n/a. doi:10.1111/1755-0998.12046 Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. jModelTest 2: more models,
new heuristics and parallel computing. Nature Methods 9, 772–772. doi:10.1038/nmeth.2109
144
Dayrat, B., 2005. Towards integrative taxonomy. Biological Journal of the Linnean Society 85, 407–415
de Carvalho, M.R., Bockmann, F.A., Amorim, D.S., Brandão, C.R.F., de Vivo, M., de
Figueiredo, J.L., Britski, H.A., de Pinna, M.C.C., Menezes, N.A., Marques, F.P.L., Papavero, N., Cancello, E.M., Crisci, J.V., McEachran, J.D., Schelly, R.C., Lundberg, J.G., Gill, A.C., Britz, R., Wheeler, Q.D., Stiassny, M.L.J., Parenti, L.R., Page, L.M., Wheeler, W.C., Faivovich, J., Vari, R.P., Grande, L., Humphries, C.J., DESALLE, R., Ebach, M.C., Nelson, G.J., 2007. Taxonomic Impediment or Impediment to Taxonomy? A Commentary on Systematics and the Cybertaxonomic-Automation Paradigm. Evolutionary Biology 34, 140–143. doi:10.1007/s11692-007-9011-6
Dexter, K.G., Pennington, T.D., Cunningham, C.W., 2010. Using DNA to assess errors
in tropical tree identifications: How often are ecologists wrong and when does it matter? Ecological monographs 80, 267–286.
Sietman, B.E., Mulcrone, R.S., Hoeh, W.R., 2012. Pyganodon (Bivalvia: Unionoida: Unionidae) phylogenetics: A male- and female-transmitted mitochondrial DNA perspective. Molecular Phylogenetics and Evolution 63, 430–444. doi:10.1016/j.ympev.2012.01.017
Drummond, A.J., Bouckaert, R.R., 2015. Bayesian Evolutionary Analysis with BEAST.
Cambridge University Press, United Kingdom. 249 pp. Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high
throughput. Nucleic Acids Research 32, 1792–1797. doi:10.1093/nar/gkh340 Elderkin, C.L., Christian, A.D., Metcalfe-Smith, J.L., Berg, D.J., 2008. Population
genetics and phylogeography of freshwater mussels in North America, Elliptio dilatata and Actinonaias ligamentina (Bivalvia: Unionidae). Molecular Ecology 17, 2149–2163. doi:10.1111/j.1365-294X.2008.03745.x
Evans, N., Paulay, G., 2012. Methods in Molecular Biology™. Humana Press, Totowa,
NJ. doi:10.1007/978-1-61779-591-6_4 Fassatoui, C., Ben Rejeb Jenhani, A., Romdhane, M.S., 2015. Geographic pattern of
shell morphology in the endemic freshwater mussel Unio ravoisieri (Bivalvia: Unionidae) from northern Tunisia. Journal of Molluscan Studies 81, 152–160. doi:10.1093/mollus/eyu069
Natural History Survey & Ohio State University Museum of Biological Diversity, 2017. http://wwx.inhs.illinois.edu/collections/mollusk/data/freshwater-mussel-host-database. (Last accessed October 2017).
145
Frierson, L.S., 1927. A classified and annotated check list of the North American naiades.
Early life history and conservation status of the monkeyface, Theliderma metanevra (Mollusca: Bivalvia) in Minnesota and Wisconsin. Walkerana 15, 99–112.
Fujita, M.K., Leaché, A.D., 2011. A coalescent perspective on delimiting and naming
species: a reply to Bauer et al. Proceedings of the Royal Society B 278, 493–495.
based species delimitation in an integrative taxonomy. Trends in Ecology & Evolution 27, 480–488. doi:10.1016/j.tree.2012.04.012
Funk, D.J., Omland, K.E., 2003. Species-Level Paraphyly and Polyphyly: Frequency,
Causes, and Consequences, with Insights from Animal Mitochondrial DNA. Annual Reviews in Ecology, Evolution, and Systematics 34, 397–423. doi:10.1146/annurev.ecolsys.34.011802.132421
Gangloff, M., Williams, J., Feminella, J., 2006. A new species of freshwater mussel
(Bivalvia: Unionidae), Pleurobema athearni, from the Coosa River drainage of Alabama, USA. Zootaxa 1118, 43–56.
Gerke, N., Tiedemann, R., 2001. A PCR-based molecular identification key to the
glochidia of European freshwater mussels (Unionidae). Conservation Genetics 2, 285–287.
Graf, D., Cummings, K., 2007. Review of the systematics and global diversity of
freshwater mussel species (Bivalvia: Unionoida). Journal of Molluscan Studies 73, 291–314.
Homogeneity at Nuclear Microsatellite Loci Masks Mitochondrial Haplotype Diversity in the Endangered Fanshell Pearlymussel (Cyprogenia stegaria). Journal of Heredity 102, 196–206. doi:10.1093/jhered/esq120
Hallerman, E.M., 2006. Patterns of genetic differentiation and conservation of the slabside pearlymussel, Lexingtonia dolabelloides (Lea, 1840) in the Tennessee River drainage. Journal of Molluscan Studies 72, 65–75. doi:10.1093/mollus/eyi055
Haag, W., 2010. A hierarchical classification of freshwater mussel diversity in North
America. Journal of Biogeography 37, 12–26.
146
Haag, W., Warren, M., Jr, 2003. Host fishes and infection strategies of freshwater mussels in large Mobile Basin streams, USA. Journal of the North American Benthological Society 22, 78–91.
Haag, W.R., 2012. North American freshwater mussels: natural history, ecology, and
conservation. Cambridge University Press, Cambridge. 505 pp. Haag, W.R., Williams, J.D., 2014. Biodiversity on the brink: An assessment of
conservation strategies for North American freshwater mussels. Hydrobiologia 735, 45–60. doi:10.1007/s10750-013-1524-7
Haas, F., 1969. Superfamilia Unionacea. Das Tierreich, Leif. 88. Walter de Gruyter and
Co., Berlin. 663 pp. Harriger, K.M., Knight, J.R., Wegener, M.G., 2015. Host-Fish Identifications for Two
Balke, M., 2015. A comprehensive DNA barcode database for Central European beetles with a focus on Germany: adding more than 3500 identified species to BOLD. Molecular Ecology Resources 15, 795–818. doi:10.1111/1755-0998.12354
Henley, W., Grobler, P., Neves, R., 2006. Non-invasive method to obtain DNA from
freshwater mussels (Bivalvia: Unionidae). Journal of Shellfish Research 25, 975–977.
Hickerson, M., Meyer, C., Moritz, C., 2006. DNA barcoding will often fail to discover new
animal species over broad parameter space. Systematic Biol. 55, 729–739. Hoggarth, M.A., 1992. An examination of the glochidia-host relationships reported in the
literature for North American species of Unionacea (Mollusca: Bivalvia). Malacology Data Net 3, 1-30.
147
Holcomb, J., Rowe, M., Williams, J., Pursifull, S., 2015. Discovery of the Ochlockonee Moccasinshell, Medionidus simpsonianus, in the Lower Ochlockonee River, Florida. Southeastern Naturalist 14, 714–720. doi:10.1656/058.014.0415
Kurth, J.E., Bartsch, M.R., Thorpe, W.S., McGill, M.R., Hornbach, D.J., 2012. Early Life History of the Winged Mapleleaf Mussel (Quadrula fragosa). American Malacological Bulletin 30, 47–57. doi:10.4003/006.030.0104
Howard A.D., 1914. Some cases of narrowly restricted parasitism among commercial
species of fresh water mussels. Transactions of the American Fisheries Society, 44, 41–44.
Howells, R.G., 1997. New fish hosts for nine freshwater mussels (Bivalvia: Unionidae) in
Texas. Texas Journal of Science 49, 255-258. Howells, R.G., Neck, R.W., Murray, H.D., 1996. Freshwater Mussels of Texas. Texas
Parks and Wildlife Press, Austin, Texas. 218 pp. Huang, J.-P., Knowles, L.L., 2016. The Species versus Subspecies Conundrum:
Quantitative Delimitation from Integrating Multiple Data Types within a Single Bayesian Approach in Hercules Beetles. Systematic Biology 65, 685–699. doi:10.1093/sysbio/syv119
Inoue, K., Hayes, D.M., Harris, J.L., Christian, A.D., 2013. Phylogenetic and
morphometric analyses reveal ecophenotypic plasticity in freshwater mussels Obovaria jacksoniana and Villosa arkansasensis(Bivalvia: Unionidae). Ecology and Evolution 3, 2670–2683. doi:10.1002/ece3.649
study of a freshwater mussel reveals patterns of habitat use and an association between survival and river discharge. Freshwater Biology 59, 1872–1883. doi:10.1111/fwb.12389
Ivanova, N.V., deWaard, J.R., Hebert, P.D.N., 2006. An inexpensive, automation-
Johnson, M.S., Caccavaie, P.D., Randklev, C.R., Gibson, J.R., 2012. New and
confirmed fish hosts for the threatened freshwater mussel Lampsilis bracteata (Gould, 1855), the Texas Fatmucket (Bivalvia: Unionidae). Nautilus 126, 148–149.
148
Johnson, N.A., McLeod, J.M., Holcomb, J., Rowe, M., Williams, J.D., 2016. Early life history and spatiotemporal changes in distribution of the rediscovered Suwannee moccasinshell Medionidus walkeri (Bivalvia: Unionidae). Endangered Species Research 31, 163–175. doi:10.3354/esr00752
Johnson, R.I., 1970. The systematics and zoogeography of the Unionidae (Mollusca:
Bivalvia) of the southern Atlantic Slope Region. Bulletin of the museum of Comparative Zoology 140, 263–449.
Johnson, R.I., 1972. The Unionidae (Mollusca: Bivalvia) of peninsular Florida. Bulletin of
the Florida State Museum, Biological Sciences 16, 181–249. Jolicoeur, P., 1963. Degree of generality of robustness in Martes americana. Growth 27,
1–27. Jones, J., Hallerman, E., Neves, R., 2006. Genetic management guidelines for captive
propagation of freshwater mussels (Unionoidea). Journal of Shellfish Research 25, 527–535.
Jones, J., Neves, R., Ahlstedt, S., Hallerman, E., 2006. A holistic approach to taxonomic
evaluation of two closely related endangered freshwater mussel species, the oyster mussel Epioblasma capsaeformis and tan riffleshell Epioblasma florentina walkeri (Bivalvia: Unionidae). Journal of Molluscan Studies 72, 267–283.
2015. Endangered Rough Pigtoe Pearlymussel: Assessment of Phylogenetic Status and Genetic Differentiation of Two Disjunct Populations. Journal of Fish and Wildlife Management 6, 338–349.
Jones, J.W., Neves, R.J., 2010. Descriptions of a new species and a new subspecies of
freshwater mussels, Epioblasma ahlstedti and Epioblasma florentina aureola (Bivalvia: Unionidae), in the Tennessee River drainage, USA. The Nautilus 124, 77–92.
Kandl, K.L., Liu, H.-P., Butler, R.S., Hoeh, W.R., Mulvey, M., 2001. A genetic approach
to resolving taxonomic ambiguity among Pleurobema (Bivalvia: Unionidae) of the eastern Gulf Coast. Malacologia 43, 87–101.
Katoh, K., Standley, D.M., 2013. MAFFT Multiple Sequence Alignment Software
Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution 30, 772–780.
149
Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., Drummond, A., 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649. doi:10.1093/bioinformatics/bts199
Kekkonen, M., Hebert, P.D.N., 2014. DNA barcode-based delineation of putative
Kennedy, T.B., Haag, W.R., 2005. Using morphometrics to identify glochidia from a
diverse freshwater mussel community. Journal of the North American Benthological Society 24, 880–889. doi:10.1899/05-001.1
Kerr, K., Lijtmaer, D., Barreira, A., 2009. Probing Evolutionary Patterns in Neotropical
Birds through DNA Barcodes. PLoS ONE 4, e4379. doi: 10.1371/journal.pone.0004379
King, T., Eackles, M., Gjetvaj, B., Hoeh, W., 1999. Intraspecific phylogeography of
Lasmigona subviridis (Bivalvia: Unionidae): conservation implications of range discontinuity. Molecular Ecology 8, 65–78.
Kishino, H., and Hasegawa, M., 1989. Evaluation of the maximum likelihood estimate of
the evolutionary tree topologies from DNA sequence data, and the branching order in hominoidea. Molecular Evolution 29, 170–179.
Kneeland, S.C., Rhymer, J.M., 2008. Determination of fish host use by wild populations
of rare freshwater mussels using a molecular identification key to identify glochidia. Journal of the North American Benthological Society 27, 150–160. doi:10.1899/07-036.1
Knowles, L., Carstens, B., 2007. Delimiting species without monophyletic gene trees.
Systematic Biology 56, 887–895. Köhler, F., 2007. From DNA taxonomy to barcoding - how a vague idea evolved into a
biosystematic tool. Zoosystematics and Evolution 83, 44–51. doi:10.1002/mmnz.200600025
Kowalewski, M., Dyreson, E., Marcot, J.D., Vargas, J.A., Flessa, K.W., Hallman, D.P.,
1997. Phenetic discrimination of biometric simpletons: paleobiological implications of morphospecies in the lingulide brachiopod Glottidia. Paleobiology 23, 444–469.
Kraemer, L.R., 1970. The mantle flap in three species of Lampsilis (Pelecypoda:
Unionidae). Malacologia 10, 225–282.
150
Krebs, R., 2004. Combining paternally and maternally inherited mitochondrial DNA for analysis of population structure in mussels. Molecular Ecology 13, 1701–1705.
Krebs, R.A., Borden, W.C., Evans, N.M., Doerder, F.P., 2013. Differences in population
structure estimated within maternally‐and paternally‐inherited forms of mitochondria in Lampsilis siliquoidea (Bivalvia: Unionidae). Biological Journal of the Linnean Society 109, 229–240.
Selection of Partitioning Schemes and Substitution Models for Phylogenetic Analyses. Molecular Biology and Evolution 29, 1695–1701. doi:10.1093/molbev/mss020
H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J., Higgins, D.G., 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23, 2947–2948. doi:10.1093/bioinformatics/btm404
2009. Quantifying ecological, morphological, and genetic variation to delimit species in the coast horned lizard species complex (Phrynosoma). Proceedings of the National Academy of Sciences 106, 12418–12423. doi:10.1073/pnas.0906380106
Lefébure, T., Douady, C.J., Gouy, M., Gibert, J., 2006. Relationship between
morphological taxonomy and molecular divergence within Crustacea: Proposal of a molecular threshold to help species delimitation. Molecular Phylogenetics and Evolution 40, 435–447. doi:10.1016/j.ympev.2006.03.014
Lefevre, G., and W.C. Curtis. 1910. Experiments in the artificial propagation of fresh-
water mussels. Bulletin of the Bureau of Fisheries. 28, 615–626. Lellis, W.A., White, B.S.J., Cole, J.C., Johnson, C.S., Devers, J.L., Gray, E.V.S.,
Galbraith, H.S., 2013. Newly Documented Host Fishes for the Eastern Elliptio Mussel Elliptio complanata. Journal of Fish and Wildlife Management 4, 75–85. doi:10.3996/102012-JFWM-094
151
Levine, T.D., Lang, B.K., Berg, D.J., 2012. Physiological and ecological hosts of Popenaias popeii (Bivalvia: Unionidae): laboratory studies identify more hosts than field studies. Freshwater Biol 57, 1854–1864. doi:10.1111/j.1365-2427.2012.02844.x
Liu, H.P., Mitton, J.B., Wu, S.K., 1996. Paternal mitochondrial DNA differentiation far
exceeds maternal mitochondrial DNA and allozyme differentiation in the freshwater mussel, Anodonta grandis grandis. Evolution 50, 952–957.
Lopes-Lima, M., Froufe, E., Van Tu Do, Ghamizi, M., Mock, K.E., Kebapçi, Ü., Klishko,
O., Kovitvadhi, S., Kovitvadhi, U., Paulo, O.S., Pfeiffer, J.M., III, Raley, M., Riccardi, N., Şereflişan, H., Sousa, R., Teixeira, A., Varandas, S., Wu, X., Zanatta, D.T., Zieritz, A., Bogan, A.E., 2016. Phylogeny of the most species-rich freshwater bivalve family (Bivalvia: Unionida: Unionidae): Defining modern subfamilies and tribes. Molecular Phylogenetics and Evolution 106, 174–191. doi:10.1016/j.ympev.2016.08.021
Lydeard, C., Cowie, R., Ponder, W., Bogan, A., Bouchet, P., Clark, S., Cummings, K.,
Frest, T., Gargominy, O., Herbert, D., 2004. The global decline of nonmarine mollusks. BioScience 54, 321–330.
Lydeard, C., Garner, J.T., Hartfield, P., Williams, J.D., 1999. Freshwater mussels in the
Gulf region: Alabama. Gulf of Mexico Science 17, 125–134. Lydeard, C., R.L. Minton, and J.D. Williams. 2000. Prodigious polyphyly in imperiled
freshwater pearly-mussels (Bivalvia: Unionidae): a phylogenetic test of species and generic designations. pp. 145-158 in E.M. Harper, J.D. Taylor, and J.A. Crame (eds.). The Evolutionary Biology of the Bivalvia. Geological Society Special Publication No. 177. 494 pp.
Lydeard, C., Mulvey, M., DAVIS, G.M., 1996. Molecular Systematics and Evolution of
Reproductive Traits of North American Freshwater Unionacean Mussels (Mollusca: Bivalvia) as Inferred from 16S rRNA Gene Sequences. Philosophical Transactions: Biological Sciences 351, 1593–1603.
Mace G.M. 2004. The role of taxonomy in species conservation. Philosophical
Transactions of the Royal Society B 359:711–719. Maddison, W. P. and D.R. Maddison. 2011. Mesquite: a modular system for
evolutionary analysis. Version 2.75. http://mesquiteproject.org Marko, P.B., 2002. Fossil Calibration of Molecular Clocks and Divergence Times of
Geminate Species Pairs Separated by the Isthmus of Panama. Molecular Biology and Evolution 19, 2005-2021.
Mayr, E., 1942. Systematics and the Origin of Species. Columbia University Press, New York. 376 pp.
McLeod, J.M., Jelks, H.L., Pursifull, S., Johnson, N.A., 2017. Characterizing the early
life history of an imperiled freshwater mussel (Ptychobranchus jonesi) with host-fish determination and fecundity estimation. Freshwater Science 36, 338–350. doi:10.1086/692096
McMurray, S.E., Roe, K.J., 2017. Perspectives on the controlled propagation,
augmentation, and reintroduction of freshwater mussels (Mollusca: Bivalvia: Unionoida). Ecological Studies 20, 1–12.
Evidence from multi-locus nuclear data conflicts with existing taxonomy in the snake tribe Thamnophiini. Molecular Phylogenetics and Evolution 68, 425–431. doi:10.1016/j.ympev.2013.04.028
Meier, R., Shiyang, K., Vaidya, G., Ng, P., 2006. DNA barcoding and taxonomy in
Diptera: a tale of high intraspecific variability and low identification success. Systematic Biol. 55, 715–728.
Meyer, C.P., 2003. Molecular systematics of cowries (Gastropoda: Cypraeidae) and
diversification patterns in the tropics. Biological Journal of the Linnean Society 79, 401–459. doi:10.1046/j.1095-8312.2003.00197.x.
Meyer, C.P., Paulay, G., 2005. DNA Barcoding: Error Rates Based on Comprehensive
Sampling. PLoS Biol 3, e422. doi:10.1371/journal.pbio.0030422.st003 Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway for
inference of large phylogenetic trees. In: Gateway Computing Environments Workshop (GCE), 2010, pp. 1–8. doi:10.1109/GCE.2010.5676129
Moyer, G.R., Díaz-Ferguson, E., 2012. Identification of endangered Alabama
lampmussel (Lampsilis virescens) specimens collected in the Emory river, Tennessee, USA via DNA barcoding. Conservation Genetics 13, 885–889. doi:10.1007/s10592-012-0328-z
Mulvey, M., Lydeard, C., Pyer, D., Hicks, K., Brim-Box, J., Williams, J., Butler, R., 1997.
Conservation genetics of North American freshwater mussels Amblema and Megalonaias. Conservation Biology 868–878.
Nagy, Z.T., 2010. A hands-on overview of tissue preservation methods for molecular
genetic analyses. Organisms Diversity and Evolution 10, 91–105. doi:10.1007/s13127-010-0012-4
153
Neves, R.J., Bogan, A.E., Williams, J.D., Ahlstedt, S., Hartfield, P.D., 1997. Status of aquatic mollusks in the southeastern United States: A downward spiral of diversity, in: Benz, G.W., Collins, D.E. (Eds.), Aquatic Fauna in Peril: the Southeastern Perspective. Lenz Design and Comunications, Decatur, GA, pp. 1–44.
Neves, R.J., Weaver, L.R.,Zale, A.V., 1985. An evaluation of host fish suitability for
glochidia of Villosa vanuxemi and V. nebulosa (Pelecypoda: Unionidae). American Midland Naturalist 113, 13–19.
Nguyen, L.-T., Schmidt, H.A., Haeseler, A.V., Minh, B.Q., 2015. IQ-TREE: A fast and
effective stochastic algorithm for estimating maximum likelihood phylogenies. Molecular Biology and Evolution 32, 268–274. DOI: 10.1093/molbev/msu300.
R.K., 2014. How global biodiversity hotspots may go unrecognized: lessons from the North American Coastal Plain. Diversity and Distributions 21, 236–244. doi:10.1111/ddi.12278
Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R, Legendre, P., McGlinn, D., Minchin,
P.R., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, H.H., Szoecs, E., Wagner, H., 2016. Vegan: community ecology package. Version 2.4-1. (http://cran.r-project.org/)
Ortmann, A.E., 1920. Correlation of shape and station in freshwater mussels (naiades).
Proceedings of the American Philosophical Society 59, 269–312. Ortmann, A.E., Walker, B.A., 1922. On the nomenclature of certain North American
naiades. Occasional papers of the museum of Zoology 1–80. Packer, L., Gibbs, J., Sheffield, C., Hanner, R., 2009. DNA barcoding and the mediocrity
of morphology. Molecular Ecology Resources 9, 42–50. doi:10.1111/j.1755-0998.2009.02631.x
Padial, J.M., Miralles, A., la Riva, De, I., Vences, M., 2010. The integrative future of
taxonomy. Frontiers in Zoology 7, 16. doi:10.1186/1742-9994-7-16 Parker, R., Hackney, C., Vidrine, M., 1984. Ecology and reproductive strategy of a south
Perkins, M.A., Johnson, N.A., Gangloff, M.M., 2017. Molecular systematics of the
critically-endangered North American spinymussels (Unionidae: Elliptio and Pleurobema) and description of Parvaspina gen. nov. Conservation Genetics 18, 745-757. doi:10.1007/s10592-017-0924-z
154
Pfeiffer, J.M., III, Johnson, N.A., Randklev, C.R., Howells, R.G., Williams, J.D., 2016. Generic reclassification and species boundaries in the rediscovered freshwater mussel “Quadrula” mitchelli (Simpson in Dall, 1896). Conservation Genetics 17, 279–292. doi:10.1007/s10592-015-0780-7
Kamoun, S., Sumlin, W.D., Vogler, A.P., Hedin, M., 2006. Sequence-Based Species Delimitation for the DNA Taxonomy of Undescribed Insects. Systematic Biology 55, 595–609. doi:10.1080/10635150600852011
Ratnasingham, S., Hebert, P.D.N., 2013. A DNA-Based Registry for All Animal Species:
The Barcode Index Number (BIN) System. PLoS ONE 8, e66213. doi:10.1371/journal.pone.0066213.s001
Ricciardi, A., Rasmussen, J., 1999. Extinction rates of North American freshwater
fauna. Conservation Biology 5, 1220–1222. Roe, K., Hartfield, P., 2005. Hamiota, a new genus of freshwater mussel (Bivalvia:
Unionidae) from the Gulf of Mexico drainages of the southeastern United States. The Nautilus (Philadelphia, PA) 119, 1–10.
Roe, K.J., Lydeard, C., 1998. Molecular systematics of the freshwater mussel genus
Potamilus (Bivalvia: Unionidae). Malacologia 39, 195–2054. Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget,
B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice Across a Large Model Space. Systematic Biology 61, 539–542. doi:10.1093/sysbio/sys029
Royle, J.A., Link, W.A., 2006. Generalized site occupancy models allowing for false
positive and false negative errors. Ecology 87, 835–841. doi:10.1890/0012-9658(2006)87[835:gsomaf]2.0.co;2
155
Rubinoff, D., Holland, B., 2005. Between two extremes: mitochondrial DNA is neither the panacea nor the nemesis of phylogenetic and taxonomic inference. Systematic Biology 54, 952–961.
Santos, A.M., Branco, M., 2012. The quality of name-based species records in
databases. Trends in Ecology & Evolution 27, 6–7 doi:10.1016/j.tree.2011.09.011 Schlager, S., Jefferis, G., 2016. Morpho: Calculations and Visualizations related to
geometric morphometrics. R package v 2.4.1.1. https://cran.r-project.org/web/packages/Morpho/index.html
Serb, J.M., Buhay, J.E., Lydeard, C., 2003. Molecular systematics of the North
American freshwater bivalve genus Quadrula (Unionidae: Ambleminae) based on mitochondrial ND1 sequences. Molecular Phylogenetics and Evolution 28, 1–11. doi:10.1016/S1055-7903(03)00026-5
freshwater mussel species (Bivalvia: Unionidae): contributing factors, management implications, and potential solutions. Journal of the North American Benthological Society 30, 446–458.
Shimodaira, H., and Goldman, N., 2002. An approximately unbiased test of
phylogenetic tree selection. Systematic Biology 51, 492-508. Shimodaira, H., Hasegawa, M., 1999. Multiple comparisons of log likelihoods with
applications to phylogenetic inference. Molecular Biology and Evolution 16, 1114–1116.
behaviors of five quadruline freshwater mussel species (Bivalvia: Unionidae). American Malacological Bulletin 30, 39–46.
Simpson, C.T., 1892. On a revision of the American Unionidae. Nautilus 6, 78–80. Simpson, C.T., 1900. Synopsis of the naiades, or pearly fresh-water mussels.
Proceedings of the United States National Museum 22, 501–1044. Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-
analysis of large phylogenies. Bioinformatics 30, 1312–1313. doi:10.1093/bioinformatics/btu033
Stern, E., Felder, D., 1978. Identification of host fishes for four species of freshwater
mussels (Bivalvia: Unionidae). American Midland Naturalist 100, 233–236.
156
Strauss, R.E., 1985. Evolutionary allometry and variation in body form in the South American catfish genus Corydoras (Callichthyidae). Systematic Zoology 34, 381–396.
Strayer, D.L., 2008. Freshwater mussel ecology: A multifactor approach to distribution
and abundance. University of California Press, Berkeley, Los Angeles, London 204 pp.
Sukumaran, J., Holder, M.T., 2010. DendroPy: A Python library for phylogenetic
two common freshwater mussel species, Lampsilis cardium and Quadrula pustulosa (Bivalvia: Unionidae), in a large federally protected waterway (St. Croix River, Minnesota/Wisconsin, USA). American Malacological Bulletin 30, 59–72.
Toews, D.P.L., Brelsford, A., 2012. The biogeography of mitochondrial and nuclear
discordance in animals. Molecular Ecology 21, 3907–3930. doi:10.1111/j.1365-294X.2012.05664.x
Roper, C.F.E., Scheltema, A., Thompson, F.G., Williams, J.D., 1988. Common and scientific names of aquatic invertebrates from the United States and Canada: Mollusks. American Fisheries Society, Special Publication 16. 277 pp.
Mikkelsen, P.M., Neves, R.J., Roper, C.F.E., Rosenberg, G., Roth, B., Scheltema, A., Thompson, F.G., Vecchione, M., Williams, J.D., 1998. Common and scientific names of aquatic invertebrates from the United States and Canada: Mollusks. 2nd Edition. American Fisheries Society, Special Publication 26. 526 pp.
USFWS, 2011. Endangered and Threatened Wildlife and Plants: Partial 90-Day Finding
on Petition to List 404 Species with Critical Habitat in Southeastern United States. Federal Registry 76 (187), 59836-59862.
Vidrine, M.F., 1993. The historical distributions of freshwater mussels in Louisiana. Gail
Q. Vidrine Collectibles, Eunice, LA. 225 pp. Vink, C.J., Paquin, P., Cruickshank, R.H., 2012. Taxonomy and Irreproducible Biological
Vu, V.Q., 2011. ggbiplot: A ggplot2 based biplot. R package v 0.55. http://github.com/vqv/ggbiplot.
Wake, D.B., Vredenburg, V.T., 2008. Colloquium paper: are we in the midst of the sixth
mass extinction? A view from the world of amphibians. Proceedings of the National Academy of Sciences 105 Suppl 1, 11466–11473. doi:10.1073/pnas.0801921105
Walker, J., Bogan, A., Bonfiglio, E., Campbell, D., Christian, A., Curole, J., Harris, J.,
Wojtecki, R., Hoeh, W., 2007. Primers for amplifying the hypervariable, male-transmitted COII-COI junction region in amblemine freshwater mussels (Bivalvia: Unionoidea: Ambleminae). Molecular Ecology Notes 7, 489–491.
Ward, R., Hanner, R., Hebert, P., 2009. The campaign to DNA barcode all fishes, FISH-
BOL. Journal of Fish Biology. 74, 329–356. doi:10.1111/j.1095-8649.2008.02080.x
Watters, G.T., Hoggarth, M.A., Stansbery, D.H., 2009. The Freshwater Mussels of Ohio.
The Ohio State University Press, Columbus. 421 pp. Whelan, N.V., Strong, E.E., 2015. Morphology, molecules and taxonomy: extreme
incongruence in pleurocerids (Gastropoda, Cerithioidea, Pleuroceridae). Zoologica Scripta 45, 62–87. doi:10.1111/zsc.12139
Johnson, N.A., Watters,G.T., 2017. A revised list of the freshwater mussels (Mollusca: Bivalvia: Unionida) of the United States and Canada. Freshwater Mollusk Biology and Conservation, 20, 33–58.
Williams, J.D., Bogan, A.E., Garner, J.T., 2008. Freshwater Mussels of Alabama and
the Mobile Basin in Georgia, Mississippi and Tennessee. University of Alabama Press, Tuscaloosa. 908 pp.
Florida. University of Alabama Press, Tuscaloosa. 490 pp.
158
Williams, J.D., Butler, R.S., Wisniewski, J.M., 2011. Annotated synonymy of the recent freshwater mussel taxa of the families Margaritiferidae and Unionidae described from Florida and drainages contiguous with Alabama and Georgia. Bulletin of the Florida Museum of Natural History 51, 1–84.
Willis, S.C., 2017. One species or four? Yes!...and, no. Or, arbitrary assignment of
lineages to species obscures the diversification processes of Neotropical fishes. PLoS ONE 12: e0172349. doi:10.1371/journal.pone.0172349
Wilson, C.B. 1916. Copepod parasites of fresh-water fishes and their economic
relations to mussel glochidia. Bulletin of the Bureau of Fisheries. 34:333-374 + plates 60–74.
Yang, Z., Rannala, B., 2017. Bayesian Species Identification under the Multispecies
Coalescent Provides Significant Improvements to DNA Barcoding Analyses. Molecular Ecology 26, 3028–3036. doi:10.1111/ijlh.12426
Zachos, F.E., 2016. Species Concepts in Biology. Historical Development, Theoretical
Foundations and Practical Relevance. Springer International Publishing, Switzerland. 220 pp.
Zając, K., Zając, T., Ćmiel, A., 2017. What can we infer from the shell dimensions ofthe
thick-shelled river mussel Unio crassus. Hydrobiologia 1–18. doi:10.1007/s10750-017-3098-2
Zanatta, D., Murphy, R., 2008. The phylogeographical and management implications of
genetic population structure in the imperiled snuffbox mussel, Epioblasma triquetra (Bivalvia: Unionidae). Biological Journal of the Linnaean Society 93, 371–384.
Zieritz, A., Gum, B., Kuehn, R., Geist, J., 2012. Identifying freshwater mussels
(Unionoida) and parasitic glochidia larvae from host fish gills: A molecular key to the North and Central European species. Ecology and Evolution 2, 740–750. doi:10.1002/ece3.220
Zieritz, A., Hoffman, J.I., Amos, W., Aldridge, D.C., 2010. Phenotypic plasticity and
genetic isolation-by-distance in the freshwater mussel Unio pictorum (Mollusca: Unionoida). Evolution and Ecology 24, 923–938. doi:10.1007/s10682-009-9350-0
159
BIOGRAPHICAL SKETCH
Nathan (Nate) Johnson was born in Salem Ohio in August 1979, the first of three
children of John Lewis Johnson, Jr. and Karen May True. He was raised in a rural area
in the outskirts of Richmond, Virginia. Nate spent his childhood summers exploring the
shorelines of the Rappahannock River at the river home of his grandparents, John
Lewis Johnson, Sr. and Hester Ann Johnson. This is where Nate developed his
fascination with aquatic organisms, as he learned to swim, fish, trap blue crabs, and find
freshwater mussels. By the age of 16, Nate was spending the majority of his free time
exploring forests, swamps, and waterways in eastern Virginia and volunteering at a
local U.S. Fish and Wildlife National Fish Hatchery in Charles City, Virginia. He
graduated with honors from Highland Springs High School outside Richmond, Virginia in
1997 and then enrolled in the Department of Fisheries and Wildlife Sciences at Virginia
Tech. In 2001, he graduated with a Bachelor of Science degree in fisheries science,
with a minor in biology. From 2002 to 2003, Nate researched under the supervision of
Dr. Richard Neves and Dr. Jess Jones at the Freshwater Mollusk Conservation Center
in Blacksburg, Virginia, conducting freshwater biological surveys and propagating
endangered freshwater mussels for recovery of populations throughout the
Southeastern United States. From 2003-2005, he worked full-time in the Conservation
Genetics Laboratory of Dr. Eric Hallerman at Virginia Tech on a variety of population
genetics projects, focusing on conservation of native freshwater fish and shellfish.
During the summer of 2005, he conducted an internship at the National Center for Cool
and Coldwater Aquaculture (NCCCWA) in Kearneysville, WV under the direction of Dr.
Yniv Palti. Beginning in the fall of 2005, he enrolled at Virginia Tech as a graduate
research assistant, where he pursued a Master of Science degree in fisheries and
160
wildlife sciences. He received his master’s degree in August of 2007 and moved to
Gainesville, Florida to pursue his doctorate degree, studying the systematics,
phylogeography, and conservation genetics of freshwater mussels at the University of
Florida, under the direction of Dr. James Austin and Dr. James Williams. In August of
2010, Nate accepted a Pathways Student Internship with the U.S. Geological Survey in
Gainesville, Florida, where he worked full-time to build his freshwater mussel research
program and continue his graduate studies as time allowed. On April 25, 2015, he
married Antonia Francesca Brewster and they became the proud parents of Bryce Alan
Johnson on December 29, 2016. Nate received his Doctor of Philosophy in fisheries
and aquatic sciences from the University of Florida in fall 2017 and continued his career
with the U.S. Geological Survey Wetland and Aquatic Research Center in Gainesville,