History, applications, methodological issues and perspectives for the use of environmental DNA (eDNA) in marine and freshwater environments
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
INVITED REVIEW
History, applications, methodological issues and perspectives for the use
of environmental DNA (eDNA) in marine and freshwater environments
Edgardo E. Díaz-Ferguson1,2 & Gregory R. Moyer2
1. Department of Fisheries and Allied Aquacultures, Auburn University, 203 Swingle Hall, Auburn, Alabama;
edgardo_diaz-ferguson@fws.gov
2. U.S. Fish and Wildlife Service, Fish Technology Center, Conservation Genetics Lab, 5151 Spring Street, Warm
Springs, Georgia 31830; greg_moyer@fws.gov
Received 30-I-2014. Corrected 08-v-2014. Accepted 10-vi-2014.
Abstract: Genetic material (short DNA fragments) left behind by species in nonliving components of the envi-
ronment (e.g. soil, sediment, or water) is defined as environmental DNA (eDNA). This DNA has been previously
described as particulate DNA and has been used to detect and describe microbial communities in marine sedi-
ments since the mid-1980’s and phytoplankton communities in the water column since the early-1990’s. More
recently, eDNA has been used to monitor invasive or endangered vertebrate and invertebrate species. While there
is a steady increase in the applicability of eDNA as a monitoring tool, a variety of eDNA applications are emerg-
ing in fields such as forensics, population and community ecology, and taxonomy. This review provides scientist
with an understanding of the methods underlying eDNA detection as well as applications, key methodological
considerations, and emerging areas of interest for its use in ecology and conservation of freshwater and marine
environments. Rev. Biol. Trop. 62 (4): 1273-1284. Epub 2014 December 01.
Key words: environmental DNA (eDNA), detection probability, occupancy models, persistence, metabarcode,
minibarcode.
In order to understand distributions, pat- and in some cases, harmful to the environment
terns, and abundances for populations or (e.g., marine bottom trawls, electrofishing and
species, the collection (detection) and identifi- rotenone poisoning) (Thomsen et al., 2012b).
cation of individuals from their physical origins The advent of novel molecular and forensic
must be undertaken. Thus, species detection methods have provided innovative tools for
is fundamental to scientific disciplines such detecting marine and aquatic organisms that
as phylogenetics, conservation biology, and may circumvent the aforementioned limitations
ecology. However, species detection is some- (Darling & Blum, 2007; Valentini, Pompano, &
times extremely difficult especially in marine Taberlet, 2009; Lodge et al., 2012).
and aquatic environments where organisms One such tool is the detection of an organ-
have complex life cycles, and direct observa- ism’s environmental DNA (eDNA). Defined
tion of early development stages is almost as short DNA fragments that an organism
impossible (Ficetola, Miaud, Pompanon, & leaves behind in non-living components of
Taberlet, 2008). Species detection in these the ecosystem (i.e., water, air or sediments),
environments has been conducted using tra- eDNA is derived from either cellular DNA
ditional direct observation methods (visual or present in epithelial cells released by organ-
acoustic) (Thomsen et al., 2012a). Nonetheless, isms to the environment through skin, urine,
traditional detection methods can have logistic feces or mucus or extracellular DNA that is the
limitations, be time consuming, expensive, DNA in the environment resulting from cell
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1273death and subsequent destruction of cell struc- vertebrates and invertebrates) with the sole dif-
ture (Foote, Thomsen, Sveegaard, Wahlberg, ference being that eDNA fragments are usually
Kielgast, Kyhn, Salling, Galatius, Orlando, & smaller (100bp or less) and other genes besides
Gilbert, 2012; Taberlet, Coissac, Hajibabaei, COI are used. Because of these similarities,
& Rieseberg, 2012a). Methodologically, eDNA some authors refer to eDNA detection as mini-
detection requires the development of genetic barcode detection (Hajibabaei, Singer, Clare,
markers specific to that target taxon or taxa. Hebert, 2007; Baird & Hajibabaei, 2012). In
Once genetic markers are developed, target contrast, microbiologists refer to the analysis
eDNA fragments can be detected using differ- of the whole bacterial communities in an eco-
ent molecular methods including traditional or system obtained from water, sediments or soil
End-Point PCR, and visualization of the PCR samples as metagenomics (Handelsman, 2004).
product through electrophoresis, quantitative More recently, the use of eDNA has gained
or Real Time PCR (qPCR), Sanger sequencing attention for eukaryote detection to assess
or more recent Next-generation DNA sequenc- sources of fecal contamination in aquatic sys-
ing (Taberlet, Coissac, Pompanon, Brochman, tems (Layton, 2006). Yet, it was not until 2008,
& Willerslev, 2012b; Yoccoz, 2012). Method that French researchers first applied eDNA
selection is related to the research or manage- methods to confirm the presence of an aquatic
ment question, sampling logistics, life history invasive species (Rana catesbiana) from water
of the species, and availability of funds. samples in a natural lotic system (Ficetola et
This review provides a summary of the al., 2008). In North America, Jerde, Mahon,
methods, history, current applications, and Chadderton & Lodge (2011) demonstrated the
future areas of interest for eDNA as well as efficacy of eDNA as a detection tool for inva-
areas of concern and uncertainty for its use in sive species in freshwater systems. This study
marine and freshwater ecosystems. focused on the detection of silver and bighead
Asian carp (Hypophthalmichthys molitrix and
History of eDNA H. nobilis). The first study to apply eDNA as a
detection tool for federally endangered organ-
Although perceived as a modern method, isms was also published in 2011 (Goldberg,
eDNA has been utilized since the mid-1980’s Pilliod, Arkle, & Waits, 2011). In this study
for the detection of bacterial communities in researchers explored the effect of seasonal-
marine sediments (Ogram, Sayler, & Barkay, ity on eDNA detection of amphibians. Also in
1987). In the 1990’s, eDNA methods were 2011, the first study on eDNA persistence of
employed to monitor phytoplankton blooms amphibian species inhabiting freshwater eco-
and assess changes in biomass of bacterial systems was published by Dejean, Valentini,
communities (Bailiff & Karl 1991; Weirbauer, Duparc, Pellier-Cuit, Pompanon, Taberlet, &
Fucks, & Peduzzi, 1993; Paul, Kellog, & Jiang, Miaud, (2011).
1996). During this time period eDNA classifi- The first uses of eDNA detection in the
cation was dependent on particulate size (Paul marine environment was conducted by Foote
et al., 1996). Thus, eDNA found in aggregates et al. (2012a) for genetic monitoring of marine
greater than 0.2μm associated with cells (e.g., mammals and by Thomsen et al., (2012b) to
microbial eDNA) was termed particulate DNA estimate marine fish biodiverstiy. The first
or P-DNA while eDNA less than this size (e.g., reviews of eDNA also came in 2012 (Taberlet
dissolved viral DNA) was considered dissolved et al., 2012a; Yaccoz, 2012). Methodological
DNA or D-DNA (Paul et al., 1996). papers trying to make approximations between
Environmental DNA is considered a simi- fish biomass, abundance and eDNA detection
lar molecular identification tool as DNA bar- probability were also published during the
coding (i.e., a 650bp sequence of the mtDNA same time period by Dejean et al. (2011) and
COI gene used for the identification of Takahara, Minamoto, Yamanaka, Hideyuki &
1274 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014Kawabata (2012). Recently, studies have been appears promising (Goldberg et al., 2011;
focused on modeling persistence and detection Takahara et al., 2012; Takahara, Minamoto
using multiple biotic and abiotic factors such as & Doi, 2013). For example, the chucky
vectors, system volume, sample volume, eDNA madtom (Noturus crypticus), a United
dynamics, stream flow, discharge and particle States federally endangered species, was
size (Piaggio, Engeman, Hopken, Humphrey, last observed in 2004 despite intensive,
Keacher, Bruce, & Michael, 2013; Schmidt, survey efforts using traditional sampling
Kery, Ursenbasher, Hyman, & Collins, 2013; methods (i.e., seines and snorkel surveys).
Barnes, Turner, Jerde, Renshaw, Chadderton, The increased sensitivity of eDNA over
& Lodge, 2014; Pilliod, Goldberg, Arkle, & traditional sampling methods may assist in
Waits, 2014; Turner, Barnes, Charles, Jones, the detection of this species; alternatively,
Xu, Jerde, & Lodge, 2014). the lack of detection could prompt a fede-
ral status determination for this species
eDNA applications in ecology as presumed extinct. However, caution
and conservation should be taken when using the lack of
eDNA detection as a means to determine
Multiple applications of eDNA exist for a taxon’s conservation status because the
use in the fields of marine ecology and con- probability of eDNA detection will vary
servation biology, and include wildlife DNA depending on the volume of water sampled
forensics, detection of cryptic and endangered and the presumed (but usually unknown)
species/populations, detection of aquatic inva- density of the target organism in the field.
sive species (AIS), biodiversity and community The development of sampling protocols
assessment, population dynamics, ecosystem that address eDNA detection probabilities
health, trophic interactions, dietary studies, and will be critical for determining the conser-
species historical patterns of distribution. Below vation status of a species in the future.
is a brief synopsis of each of these applications. 3. Detection of aquatic invasive species
(AIS): The use of molecular methods to
1. Wildlife DNA forensics: The field of wild- detect AIS has proliferated in recent years.
life DNA forensics, which is a synthe- The development of molecular markers
sis of conservation genetics and forensic specific to the target species provides a new
genetics, was developed to address the tool for conservation managers that seek
increasing need of DNA forensic tools in to monitor AIS. Thus, the use of eDNA
wildlife law enforcement (Ogden, 2008, provides the possibility of confirming AIS
2009). The applicability of eDNA as a detection in hours or days instead of weeks
forensic tool appears promising to address or months, allowing managers to act quic-
basic forensic identification issues at the kly to minimize dispersal and settlement of
level of individual (e.g., hatchery vs wild the invader (Darling & Mahon, 2011). In
origin, location of origin, species introduc- addition, eDNA detection of AIS provides
tion), population, or species. Detection of clues to determine origin of the introduc-
eDNA fragments can also provide eviden- tion and possible routes of invasion.
ce for illegal wildlife trade and traceability 4. Biodiversity and community structure:
of illegal fishing products (e.g., shark or The term DNA metabarcoding is used to
rayfins). designate multispecies identification using
2. Detection of low density populations: The eDNA samples (Taberlet et al., 2012b).
detection of eDNA for species or popu- The approach relies on Next-generation
lations that are at low densities (e.g., sequencing (permitting the sequencing of
threatened or endangered taxa) or are billions of 100 base pair reads) and the
visually evasive (e.g., madtom catfishes) creation of taxonomic reference libraries
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1275(e.g., the Barcode of Life). Thus instead of diversity can also serve as a proxy for
identifying one species from a single water assessing ecosystem health. More speci-
sample, an eDNA metabarcoding approach fically, changes in species diversity can
has the potential to identify the eDNA influence overall ecosystem dynamics
of any taxon collected in a water sample directly or indirectly, by reducing water
given that the DNA sequences are already quality (Strayer, 2010), changing nutrient
deposited in a repository. Thus, metabar- dynamics (Didham, Tylianakis, Hutchin-
coding can be used as a tool to produce son, Ewers, & Gemmel, 2005) or affecting
biodiversity estimates that are taxonomi- the distribution of submerged machro-
cally comprehensive, quicker to produce, phytes (Strayer, 2010). Therefore, eDNA
and less reliant on taxonomic expertise (Ji could be a useful tool for future risk-based
et al., 2013). decision making of natural resources (Wil-
5. Population dynamics: Detection and quan- son & Wright, 2013) or environmental
tification of eDNA can be used as a referen- impact assessments (Veldhoen, Ikonomou,
ce or indirect measurement of population & Helbing, 2012).
attributes such as abundance, distribution, 7. Trophic interactions and dietary studies:
and biomass. Distributions of organisms Traditional studies allow quantifying and
and biomass in freshwater systems have estimating the relationship between pre-
been correlated with eDNA concentra- dator and prey as well as herbivore and
tion (Takahara et al., 2012, 2013). Thus
plant relationships using stomach con-
eDNA concentration has been used as a
tents, feces, or fecal pellets. The direct
proxy for population distribution in amphi-
observation or identification of the prey
bians (Ficetola et al., 2008; Goldberg et
in the stomach or in the fecal material
al., 2011), fishes (Mahon, Jerde, Galaska,
often is difficult and can reduce the taxo-
Bergner, Chadderton, Lodge, Hunter, &
nomic resolution or introduce bias (Bra-
Nico, 2012; Minamoto, Yamanaka, Taka-
ley, Goldsworthy, Page, Steer, & Austin,
hara, Honjo, & Kawabata, 2012) and repti-
2010). Thus, the use of eDNA fragments
les species (Piaggio et al., 2013). In marine
species no studies have been conducted to or metabarcoding approach using stomach
determine the relationship between eDNA contents as target DNA can be used for
concentration and species distribution, dietary and trophic studies without the
abundance and biomass. However, spa- observation or identification of the prey
tial and temporal oscillations of bacte- in the stomach or feces (Zarzoso-Lacoste,
rial and phytoplankton communities have Corse, & Vidal, 2013).
been correlated with DNA concentration 8. Species historical patterns of distribution:
in coastal waters during blooming events Short DNA sequences can persist for long
(Bailiff & Karl, 1991). time periods mainly in cold environments
6. Ecosystem health: Presence of AIS as with reduced exposure or absence of light
well as introduced pathogens (viral or (Willerslev et al. 2003). This has been
fungal) can have severe demographic and observed in studies conducted in old sedi-
genetic impacts to existing native popula- ments and ice cores (Hofreiter, Mead,
tions (Blanc, 2001). By using eDNA to Martin, & Poinar, 2003; Willerslev, Cape-
monitor virus concentration (Minamoto, llini, Boomsman, & Nielsen, 2007). Thus,
Honjo, & Kawabata, 2009) or invasive genetic information (ancient eDNA) in
species, managers can indirectly monitor sediments, ice cores and other environ-
for ecosystem health. In addition, the use mental sources could allow scientists to
of eDNA to monitor changes in commu- reconstruct community structure and his-
nity composition and reductions in species torical ecological process.
1276 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014eDNA persistence Golberg et al., 2011; Barnes et al., 2014). For
example, eDNA persistence in freshwater len-
DNA persistence can be defined as the tic systems has been shown to be as great as
continuance of DNA fragments once the source 30 days (Ficetola et al., 2008) while for marine
of this DNA is removed from the system systems (open and highly dynamic systems) it
(Dejean et al. 2011). Once extracellular DNA only averages approximately seven days (Foote
is released in the environment, it may persist, et al., 2012; Thomsen et al., 2012b).
be absorbed in organic and inorganic particles, Fragment size of target eDNA is also
be degraded, or transformed by microorgan- important when persistence is analyzed. It has
isms. Persistence is strongly correlated with been shown that fragments within a range of
species density, body size and the ratio between 300-400bp can persist in the aquatic environ-
amounts of released/degraded DNA by species ment for at least a week in controlled condi-
into the environment. In control conditions, tions (Alvarez, Yumet, Santiago, & Torantos,
DNA detection decreases with time once the 1996; Zhu, 2006). In contrast, shorter frag-
source of DNA is removed from the environ- ments of DNA (100bp or less) have been
ment (Dejean et al., 2011). Persistence values known to remain stable after several weeks and
vary across taxa and life history i.e., changes even years depending on environmental condi-
in size, behavior and stage of development can tions (Willlerslev et al., 2003, 2007; Taberlet et
affect the amount of DNA available at a local al., 2012a).
scale (Thomsen et al. 2012a).
Persistence estimates have been deter- Freshwater vs marine systems
mined for different taxonomic groups: from 15
to 30 days for fresh water fishes (Dejean et al., Freshwater and marine ecosystems consti-
2011 & Takahara et al., 2012a), 15 to 30 days tute great reservoirs of eDNA. For both, eDNA
for amphibians (Ficetola et al., 2008; Goldberg detection is correlated with abundance of the
et al., 2011), 0.9 to 7 days for marine mammals target species and the rate by which DNA is
(Foote et al., 2012), 21 days for mudsnails released and degraded by biotic and abiotic
(Goldberg, Sepulveda, Ray, Baumgardt, & factors (Thomsen et al., 2012a, b). In freshwa-
Waits, 2013), and 14 days for reptiles (Piaggio ter systems, eDNA is usually homogeneously
et al., 2013). In addition to density, size and distributed and the successful use of this tech-
life history features of the target taxa, eDNA nology has been proven in a number of studies
persistence can be influenced by other biotic comprising a range of taxa (Lodge et al., 2012;
factors such as bacterial and fungal concentra- Thomsen et al., 2012a). For marine systems,
tions (Dejean et al., 2011). The role of abi- the detection of eDNA is possible but less reli-
otic factors on eDNA persistence has also been able than in freshwater systems (Foote et al.
reported (e.g., oxygen concentration, nuclease 2012; Jones, 2013). Despite the high density
activity, pH, conductivity, UV radiation, pH of marine biota, the vast volume of sea water
and temperature) (Shapiro, 2008; Dejean et in relation to biomass promotes eDNA dilu-
al. 2011; Barnes et al., 2014). Nonetheless, tion and dispersion. This compounded by the
the effect of these factors has been studied dynamic nature of marine environments (e.g.,
independently and only one of these studies strong tides, current system and oceanographic
had measured how covariation of these fac- events) and the increased inhibition of subse-
tors (specific environmental conditions) affects quent molecular procedures due to the high
persistence (Barnes et al., 2014). Other abiotic salinity environments, make eDNA detection
factors related to the system that can also affect more challenging when compared to freshwater
persistence include: stream flow, currents, tidal systems. Currently, eDNA studies conducted in
oscillations, type of sediment, and salinity the marine environment are limited to detec-
(Corinaldesi, Beolchini, & Dell’Anno, 2008; tion of fishes, marine mammals and microbial
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1277communities in temperate areas (Venter et al. eDNA fragments; therefore, depending on the
2004; Foote et al. 2012; Thomsen et al. 2012b; research question, it may be necessary to incor-
Kelly, Port, Yamahara, & Crowder, 2013). porate information regarding the life history of
Despite of the success on eDNA detection of the organism in to the experimental design. For
these taxa in the marine systems; scientists example, during spawning events, there is a
have to consider that eDNA degrades rapidly release of multiple sources of genetic material
in seawater (less than seven days) and detec- into the environment including blood, urine,
tion of marine species is considered to be local fecal material, epithelial cells and gametes.
(Thomsen et al. 2012b). Therefore, the use of Taking advantage of this information may
eDNA in marine systems is still a challenge, allow a better detection of the species; alterna-
especially in applications such as the establish- tively, it could also overestimate the presence
ment of relationships between distribution, of a species in a particular period.
abundance, or biomass of species with eDNA Demographic connectivity patterns:
detection probabilities, since dilution and dis- Movement of organisms is a measurement of
persion of eDNA is greater. marine and freshwater connectivity and implies
In addition to the aforementioned issues, movement of genetic material. Therefore, the
marine and freshwater tropical environments presence of eDNA from a particular species
have high surface temperatures (sometimes between geographically or oceanographically
above 30°C) and elevated UV radiation at sea connected areas could be a sign of population
level that may increase eDNA degradation rate connectivity. Understanding connectivity is
and reduce the persistence period in the water, essential for the establishment of population
decreasing the detection probability (Barnes boundaries, marine corridors and common pro-
et al., 2014). For this reason the selection of tected areas (Díaz-Ferguson, Haney, Wares, &
the appropriate detection method, preliminary Silliman, 2010; Díaz-Ferguson, 2012).
laboratory and aquarium eDNA assays and Taxa and animal size: Based on freshwater
consideration of species size, vertical distribu- studies (mainly focused in fishes and amphib-
tion, current or water flow, life history and deg- ians) the greater the size of the individual,
radation rates among different environments, the more likely it is to be detected because
are required before the beginning of any field of the increased amount of eDNA input into
detection protocol (Kelly et al., 2013). the system. The eDNA of invertebrates is less
likely to be detected than in vertebrates due to
eDNA considerations for the presence of an exoskeleton (Golberg et al.,
experimental design 2013). However, high detection probabilities
have been found for freshwater tadpole shrimp
Successful eDNA detection of eukaryotes (Lepidurus apus) and the New Zealand mud-
in freshwater and marine ecosystems relies on snail (Potamopyrgus antipodarum) (Goldberg
accuracy of the experimental design in accor- et al., 2013). Note that depending on the taxon,
dance with the environment, life history, and exoskeletons are often shed at known times of
behavior of the target species. Therefore the the year; thus making them a potential source
following topics along with clearly defined of eDNA.
hypotheses should be considered/articulated Vectors: eDNA can be transported by alter-
before the method is applied to a natural system. native pathways or introduced into a system
Life history considerations: Season- not only because the target species inhabits the
al changes in behavior including spawning system but by other means (Darling & Mahon,
migrations, period of larval development, phy- 2011). Vectors are often problematic for many
lopatry, vertical movements and tidal oscil- eDNA applications, because while the eDNA
lations can influence eDNA concentration as is detected in the area, the taxon of interest
well as temporal and spatial distributions of is not actually there, creating a false positive.
1278 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014The following are the most common vectors of Young, Jane, Lowe, Whiteley, & Schwartz,
DNA in freshwater and marine systems: 2013). In doing so, researchers minimize the
detection of false positives. Another method
1. Animals: Introduction of DNA by animal to limit the uncertainty associated with eDNA
vectors is mainly driven by birds in natural false positives, is to confirm the identity of all
ecosystems. However, any predator can positive eDNA via Sanger sequencing (Darling
transport eDNA from other species via & Mahon, 2011).
defecation or prey transport. False negatives are also a concern for
2. Water contamination: Input of water from eDNA detection. Although false negatives will
external sources can also be an alterna- always be a possibility due to the myriad array
tive pathway to introduce eDNA into a of abiotic and biotic factors affecting eDNA
system. Discharge of ballast water from persistence, there are several steps that can be
ships has the potential to harbor eDNA incorporated into an experimental design to
because vessels often take on ballast water minimize eDNA false negatives. The detection
from coastal areas rich in biodiversity and of eDNA is contingent on PCR; thus any inhi-
transport species or DNA to other areas bition of the PCR reaction, if not accounted for,
where these species DNA have not been will lead to false negatives. Leading causes of
reported (Rothlisberger & Lodge, 2013). PCR inhibition include high salinity in marine
In addition, accidental input of water from systems (Foote et al., 2012) and humic/fluvic
the aquaculture industry can also cause acids, tannins and polyphenols in freshwater
introduction of exogenous DNA into a environments (Matheson, Gurney, Esau, &
natural system. Lehto, 2010). Thus, a DNA extraction kit that
removes potential inhibitors is recommended
Perspective and future of eDNA along with appropriate PCR controls (Piaggio
et al. 2013). Specifically for marine environ-
The perspective and future of eDNA as a ments eDNA detection is considered to be local
method species detection method and subse- due to the reduced persistence period (Foote
quent applications will focus on three general et al., 2012; Thomsen et al. 2012b). This fact
areas as follows: resolving eDNA methodologi- makes marine environments prone to high false
cal issues, improving eDNA technologies, and –negative detection rates. Therefore, future
exploring new of eDNA applications. Below eDNA studies focused on marine ecosystems
we discuss upon each of these areas. should consider preliminary laboratory assays
Resolving eDNA methodological issues: that test the effect of the ecosystem dynam-
Future success of eDNA as an effective tool for ics on eDNA distribution (i.e., particle size
species detection will rely on resolving eDNA dynamics, influence of local currents, advec-
issues related to the optimization of molecu- tion transport and salinity gradients) (Turner et
lar assays and confirmation of positive sam- al., 2014), and the impact of this dynamic on
ples. Optimization of eDNA molecular assays false negative rates.
involves testing for sensitivity (i.e., using Furthermore, eDNA, like most if not all
known amounts of DNA from the target species traditional sampling methods, suffers from
in order to determine the minimum amount of imperfect detection (Schmidt et al. 2013).
target DNA required for detection), specificity Therefore, the incorporation of occupancy
(i.e., conducting cross species amplification to models (see below) in an eDNA experimental
make sure that the assay only detect the pres- design will allow researchers to evaluate and
ence of the species of interest), and utilization quantify the false-negative measurement error
of strict field and laboratory standards and in freshwater and marine ecosystems (Morde-
controls (Dejean, Valentini, Miguel, Taberlet, cai, Mattsson, Tzilkowski, & Cooper, 2011).
Belleman, & Miaud, 2012; Wilcox, McKelvey, The same models can also account for eDNA
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1279false positives (Miller, Talley, Lips, & Camp- the conservation status of many threatened and
bell Grant, 2012). endangered organisms.
Improving supporting eDNA technolo- New frontiers and eDNA applications:
gies: The use of eDNA as a detection tool is Emerging areas of interest for eDNA methods
generating an increasing amount of genetic include areas where traditional sampling meth-
information associated to spatial and temporal odologies remain unfeasible such as underwa-
variables (i.e., annual data by season, site and ter plains on the deep ocean floor (i.e., abyssal
region). This fact will be a computational chal- plains especially hadopelagic zones). The com-
lenge for scientists in the near future. Therefore munity composition in these deep ocean envi-
it will be necessary to increase data base size, ronments still remains relatively unknown,
improve data management, and create archival primarily because studies pertaining to hadal
eDNA data bases (Yaccoz, 2012). For instance, ecosystems are rare (Bull, Stach, Ward, &
the collection of water samples and the DNA Goodfellow, 2005). Another emerging area of
extracted from those samples is an important eDNA interest is using the method to assess
reference to assess future spatial distribution specific biological events such as spawning,
of species, and monitor biodiversity and com- settlement, and recruitment using fluctuations
munity structure. on target eDNA concentration when and where
Statistical analyses: Sampling methods a particular life history event has occurred.
that rely on presence/absence data to estimate This may be of importance when predicting the
the number of individuals in an area, often suf- structure and dynamics of marine communi-
fer from imperfect detection (Pollock, Nichols, ties from the environmental conditions they
Simmons, Farnworth, Bailye, & Sauer, 2002); experience, because of abundance, biomass or
and eDNA studies will too (Schmidt et al. productivity changes, and theoretically should
2013). To date, there has been little attention the amount of eDNA that may be detected. For
given to eDNA detection probabilities with example, eDNA detection methods may some-
most studies assuming perfect or near perfect day serve as tools to elucidate the processes that
detection of the taxon of interest (Hyman & structure phytoplankton communities, which is
Collins, 2012; Thomsen et al., 2012a). Mean- of importance to assessing impacts of climate
while biotic and abiotic factors influencing change on ecosystem function (Eggers, Lewan-
eDNA are beginning to be understood (MacK- dowska, Ramos, Blanco-Ameijeiras, Gallo, &
ensie et al., 2002; MacKensie, Nichols, Royle, Mathiesen, 2013), trophic dynamics (Sterner
Pollock, & Hines, 2006). However, how these & Elser, 2002), and water quality (Anderson,
parameters might affect eDNA detection prob- Cembella, & Hallegraeff 1998).
abilities, remains unclear (Pilliod et al., 2014). In summary, the detection of trace amounts
Site occupancy models provide a means to of DNA in environmental samples is well
account for imperfect detection of various sam- established in the field of microbiology, but
pling methods (Pollock et al., 2002; Andrew this technique has only recently garnered appli-
Royle, & Dorazio, 2008; Pilliod, Goldberg, cability in conservation biology and ecology –
Arkle, & Waits, 2013) including eDNA meth- particularly for the detection of aquatic invasive
ods (Schmidt et al., 2013). Occupancy models species or endangered organisms, and inven-
can be used to study the effects of various tory and monitoring. While eDNA applications
abiotic and biotic factors that influence detec- appear promising, the detection of eDNA from
tion probabilities (both in the field and in the freshwater and marine ecosystems will depend
laboratory), and to determine the number of on clearly articulated hypotheses and proper
visits, number of samples and volume of water experimental design, in both the laboratory and
needed to be confident that a species is absent field, to adequately address these hypotheses
from a site (Schmidt et al., 2013). The latter (i.e., laboratory studies should seek minimize
will be particularly important for determining false positives and negatives and field studies
1280 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014should address imperfect detection) Finally, principales consideraciones metodológicas y áreas emer-
the use of eDNA to detect an organism is sim- gentes para su uso en ecología y conservación de ambientes
marinos y de agua dulce.
ply another tool to be used by biologists; thus,
results should be corroborated with other detec- Palabras clave: ADN ambiental (ADNa), probabilidad de
tion techniques, especially when the research detección, modelos de ocupación, persistencia, meta códi-
has conservation and management implications go de barras, mini código de barras.
(i.e., presence/absence of endangered organ-
isms or invasive species). As DNA barcoding REFERENCES
moves to become a global standard for species
identification (Bucklin, Steinke & Blanco- Alvarez, A. J., Yumet, G., Santiago, C., & Torantos, G.
Bernal, 2011; Shen, Chen, & Murphy, 2013), (1996). Stability of manipulated plasmid DNA in
aquatic environments. Environmental Toxicology and
eDNA, as a detection method, it is poised to be
Water Quality, 11, 129-135.
a potentially reliable, cost effective and expedi-
Anderson, D. M., Cembella, A. D., & Hallegraeff, G. M.
tious tool for the inventory and monitoring of (Eds.) (1998). The Physiological Ecology of Harmful
species, and the estimation of biodiversity and Algal Blooms. Heidelberg: Springer-Verlag.
community structure in freshwater ecosystems. Andrew Royle, J. & Dorazio, R. (2008). Hierarchical
In marine ecosystems more research is needed modeling and inference in ecology: the analysis of
in order to avoid false negatives, considering data from populations, metapopulations and commu-
the reduced persistence of eDNA on these nities. Oxford, UK: Elsevier.
environments, that limits detection probability Bailiff, M. & Karl, D. (1991). Dissolved and particu-
to a local scale. late DNA dynamics during a spring bloom in the
Antartic Peninsula region, 1986-1987. Deep Sea
Research. Part 1. Oceanographic Research, 38(8-9),
ACKNOWLEDGMENTS 1077-1095.
Baird, D. & Hajibabaei, M. (2012). Biomonitoring 2.0 a
We want to thank the Fish and Wildlife new paradigm in ecosystem assessment made possi-
Service and the AIS inventory and monitoring ble by next generation DNA sequencing. Molecular
Ecology, 21, 2039-2044.
program for funding our research. We also want
Barnes, M. A., Turner, C. R., Jerde, C. L., Renshaw, M. A.,
to thank Fred Utter and Jorge Cortés for provid-
Lindsay Chadderton, W., & Lodge, D. (2014). Envi-
ing feedback and comments to this review. ronmental conditions influence eDNA persistence in
aquatic systems. Environmental Science and Techno-
logy, 48, 1819-1827.
RESUMEN
Blanc, G. (2001). Introduction of pathogens in European
Historia, aplicaciones, aspectos metodológicos y aquatic ecosystems: attempt of evaluation and reali-
perspectivas para el uso del ADN ambiental (ADNa) ties. In A. Uriarte & B. Basurco (pp. 37-50) (Eds.).
en ecosistemas marinos y de agua dulce. El material Environmental Impact Assessment of Mediterranean
genético que liberan los organismos en los componentes no Aquaculture Farms.
vivos del ecosistema (aire, suelo, agua y sedimentos) reci- Braley, M., Goldsworthy, S., Page, B., Steer, M., & Austin,
be el nombre de ADN ambiental (ADNa) (eDNA, por su J. (2010). Assessing morphological and DNA-based
nombre en inglés). Este ADN previamente definido como diet analysis techniques in a generalist predator, the
ADN particulado ha sido utilizado desde mediados de la arrow squid Nototodarus studies gouldi. Molecular
década de los ochenta y principios de los noventas para Ecology Research, 10(3), 466-474.
describir la composición de las comunidades microbianas
Bucklin, A., Steinke, D., & Blanco-Bernal, L. (2011). DNA
en sedimentos marinos y de comunidades microbianas y
barcoding of marine metazoa. Annuals Reviews in
fitoplanctónicas en la columna de agua. Recientemente
Marine Science, 3, 471-508.
el ADNa es utilizado principalmente para la detección y
monitoreo de especies invasoras y en peligro. No obstante, Bull, A., Stach, J., Ward, A., & Goodfellow, M. (2005).
existen múltiples áreas en las que este método puede ser Marine actinobacteria: perspectives, challenges,
utilizado como por ejemplo en ciencias forenses, ecología future directions. Antonie van Leeuwenhock, 87(3),
de poblaciones y comunidades, y taxonomía. Esta revisión 65-79.
proporciona información sobre esta nueva herramienta Corinaldesi, C., Beolchini, F., & Dell’Anno, A. (2008).
molecular, sus actuales y futuras aplicaciones, historia, Damage and degradation rates of extracellular DNA
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1281in sediments: implications for the preservation of (Potamopyrgus antipodarum). Freshwater Science,
gene sequences. Molecular Ecology, 17, 3939-3951. 32 (3), 792-800.
Darling, J., & Blum, M. (2007). DNA-based methods for Hajibabaei, M., Singer, G., Clare, E., & Hebert, P. (2007).
monitoring invasive species: a review and prospectus. Design and applicability of DNA arrays and DNA
Biological Invasions, 9, 751-765. barcodes in biodiversity monitoring. BMC Biology,
Darling, J. & Mahon, A. (2012). From molecules to mana- 5, 24-30.
gement: Adopting DNA based methods for monito- Handelsman, J. (2004). Metagenomics: applications of
ring biological invasions in aquatic environments. genomics to uncultured microorganisms. Microbiolo-
Environmental Research, 111(7), 978-988. gy and Molecular Biology Reviews, 68, 669.
Dejean, T., Valentini, A., Duparc, A., Pellier-Cuit, S., Pom- Hofreiter, M., Mead, J., Martin, P., & Poinar, H. (2003).
panon, F., Taberlet, P., & Miaud, C. (2011). Persisten- Molecular carving. Current Biology, 13, R693-R695.
ce of environmental DNA in freshwater ecosystems. Hyman, O. & Collins, J. (2012). Evaluation of a filtration-
PLoS ONE, 6 (8), e23398. based method for detecting Batrachochytrium dedro-
Dejean, T., Valentini, A., Miquel, C., Taberlet, P., Belle- batidis in natural bodies of water. Diseases of Aquatic
main, E., & Miaud, C. (2012). Improved detection Organisms, 97 (3), 185-195.
of an alien invasive species through environmental Jerde, C., Mahon, A., Chadderton, W., & Lodge, D. (2011).
DNA barcoding: the example of the American bull- “Sight unseen” detection of rare aquatic species
frog Lithobates cathesbeianus. Journal of Applied using environmental DNA. Conservation Letters, 4,
Ecology, 49, 953-959. 150-157.
Díaz-Ferguson, E., Haney, R., Wares, J., & Silliman, B. Ji, Y., Ashton, L., Pedley, S., Edwards, D., Tang, Y., Naka-
(2010). Populations genetics of a trochid gastropod mura, A., Kitching, R., Dolman, P., Woodcock, P.,
broadens picture of Caribbean Sea connectivity. PLoS Edwards, F., Larsen, T., Hsu, W., Benedick, S., Ham-
ONE, 5(9), e12675. mer, K., Wilcove, D., Bruce, C., Wang, X., Levi, T.,
Díaz-Ferguson, E. (2012). Introducción a la ecología Lott, M., Emerson, B., & Yu, D. W. (2013). Reliable,
molecular marina: aplicaciones y perspectivas. Via verifiable and efficient monitoring of biodiversity via
Argentina, Panamá, República de Panamá. Universal metabarcoding. Ecology Letters, 16, 1245-1257.
Books. 212p. Jones, M. (2013). Environmental DNA: Genetics steps
Didham, R., Tylianakis, J., Hutchinson, M., Ewers, R., & forward when traditional ecological surveys fall
Gemmel, N. (2005). Are invasive species the drivers short. Fisheries, 38(7), 332-333.
of ecological change. Trends in Ecology and Evolu- Kelly, R., Port, J., Yamahara, K., & Crowder, L. 2013.
tion, 20(9), 470-474. Using environmental DNA to census marine fishes in
Eggers, S., Lewandowska, A., Ramos, B., Blanco-Ameijei- a large mesocosm. PLoS ONE, 9(1), e86175.
ras, J., Gallo, F., & Mathiesen, B. (2013). Community Layton, A. (2006). Development of Bactereroides 16S
composition has greater impact on the functioning of rRNA Gene Taqman-base real time PCR assays for
marine phytoplankton communities than ocean aci- estimation of total, human and bovine fecal pollu-
dification. Global Change Biology, doi: 10.1111qcb. tion in rivers. Applied Environmental Microbiology,
12421 72(6), 4214-4224.
Ficetola, F., Miaud, C., Pompanon, F., & Taberlet, P. Lodge, D., Turner, C., Jerde, C., Barnes, M., Chadderton,
(2008). Species detection using environmental DNA L., Egan, S., Feder, J., Mahon, A., & Pfrender, M.
from water samples. Biological Letters, 4, 423-425. (2012). Conservation in a cup of water: estimating
Foote, A., Thomsen, P., Sveegaard, S., Wahlberg, M., Kiel- biodiversity and population abundance from envi-
gast, J., Kyhn, L., Salling, A., Galatius, A., Orlando, ronmental DNA. Molecular Ecology, 11, 2555-2558.
L., Thomas, M., & Gilbert, T. (2012). Investigating MacKenzie, D., Nichols, J., Lachman, G., Droege, S.,
the potential use of environmental DNA (eDNA) for Royle, J., & Langtimm, C. (2002). Estimating site
genetic monitoring of marine mammals. PLoS ONE, occupancy rates when detection probabilities are less
7 (8), e4178. than one. Ecology, 84, 2200-2207.
Goldberg, C., Pillod, D., Arkle, R., & Waits, L. (2011). MacKenzie, D., Nichols, J., Royle, J., Pollock, K., &
Molecular detection of vertebrates in stream water: a Hines, J. (2006). Occupancy Estimation and Mode-
demonstration using rocky mountain tailed frogs and ling: Inferring Patterns and Dynamics of Species
Idaho giant Salamanders. PLoS ONE, 6(7), e22746. Occurrence. San Diego, United States: Elsevier.
Goldberg, C., Sepulveda, A., Ray, A., Baumgardt, J., & Mahon, A., Jerde, C., Galaska, M., Bergner, J., Chadderton,
Waits, L. (2013). Environmental DNA as a new W., Lodge, D., Hunter, M., & Nico, L. (2012). Vali-
method for early detection of New Zealand mudsnails dation of eDNA surveillance sensitivity for detection
1282 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014of Asian Carps in controlled and field experiments. monitoring studies: statistical methods for design and
PLoS ONE, 8(3), e58316. analysis. Environmetrics, 13, 105-119.
Matheson, C., Gurney, C., Esau, N., & Lehto, R. (2010). Rothlisberger, J. & Lodge, D. (2013). The Laurentian
Assessing PCR inhibition from humic substances. Great Lakes as a beachhead and a gathering place for
The Open Enzyme Inhibition Journal, 3, 38-45. biological invasions. Aquatic Invasions, 4, 361-374.
Miller, D., Talley, B., Lips, K., & Campbell Grant, E. Shapiro, B. (2008). Engineered polymerases amplify the
(2012). Estimating patterns and drivers of infection potential of ancient DNA. Trends in Biotechnology,
prevalence and intensity when detection is imperfect 26, 285-287.
and sampling error occurs. Methods in Ecology and Shen, Y., Chen, X., & Murphy, R. (2013). Assessing DNA
Evolution, 3(5), 850-859.
Barcoding as a tool for species identification and data
Minamoto, T., Honjo, M., & Kawabata, Z. (2009). Sea- quality control. PLoS ONE, 8(2), e57125.
sonal distribution of cyprinid herpes virus 3 in Lake
Shmidt, B., Kery, M., Ursenbasher, S., Hyman, O., & Coll-
Biwa Japan. Applied Environmental Microbiology,
ins, J. (2013). Site occupancy models in the analysis
75, 6900-6904.
of environmental DNA presence/absence surveys:
Minamoto, T., Yamanaka, H., Takahara, T., Honjo, M., & A case study of an emergent amphibian pathogen.
Kawabata, Z. (2012). Surveillance of fish species Methods in Ecology and Evolution, 4(7), 646-653.
composition using environmental DNA. Limnology,
Sterner, R. & Elser, J. (2002). Ecological Stoichiometry:
13, 193-197.
The biology of elements from molecules to the bios-
Mordecai, R., Mattsson, B., Tzilkowski, C., & Cooper, R. phere. Princeton: Princeton University Press.
(2011). Addressing challenges when studying mobile
Strayer, D. (2010). Alien species in fresh waters: ecolo-
or episodic species: hierarchical Bayes estimation
gical effects, interactions with other stressors, and
of occupancy and use. Journal of Applied Ecolo-
prospects for the future. Freshwater biology, 55(1),
gy, 48(1), 56-66.
155-174.
Ogden, R. (2008). Fisheries forensics: the use of DNA tools
Taberlet, P., Coissac, E., Hajibabaei, M., & Rieseberg, L.
for improving compliance, traceability and enforce-
ment in the fishing industry. Fish and Fisheries, 9, (2012a). Environmental DNA. Molecular Ecology,
462-472. 21, 1789-1793.
Ogden, R. (2009). Wildlife DNA forensics- Bringing the Taberlet, P., Coissac, E., Pompanon, F., Brochmann, C.,
gap between conservation genetics and law enforce- & Willerslev, L. (2012b). Towards next-generation
ment. Endanger Species Research, 9, 179-195. biodiversity assessment using DNA metabarcoding.
Molecular Ecology, 21, 2045-2050.
Ogram, A., Sayler, G., & Barkay, T. (1987). The extraction
and purification of microbial DNA from sediments. Takahara, T., Minamoto, T., Yamanaka, H., Hideyuki, D.,
Journal of Microbiological Methods, 7, 57-66. & Kawabata, Z. (2012). Estimation of fish biomass
using environmental DNA. PLoS ONE, 7(4), e35868.
Paul, J., Kellogg, C., & Jiang, S. (1996). Viruses and DNA
in marine environments. In R. R. Colwell (pp. 115- Takahara, T., Minamoto, T., & Doi, H. (2013). Using
124) (Eds.). Microbial diversity in time and space. environmental DNA to estimate the distribution of
Plenum Press, New York. an invasive fish species in ponds. PLoS ONE, 8(2),
e56584.
Piaggio, A., Engeman, R., Hopken, M., Humphrey, J., Kea-
cher, K., Bruce, W., & Michael, A. (2013). Detecting Thomsen, P., Kielgast, J., Iversen, L., Wiuf, C., Rasmus-
an elusive invasive species: a diagnostic PCR to detect sen, M., Gilbert, M., Orlando, L., & Willerslev, E.
Burmese python in Florida waters and an assessment (2012a). Monitoring endangered freshwater biodiver-
of persistence of environmental DNA. Molecular sity using environmental DNA. Molecular Ecology,
Ecology Research, doi:101111/1755-0998.12180. 21, 2565-2573.
Pilliod, D., Goldberg, C., Arkle, R., & Waits, L. (2013). Thomsen, P., Kielgast, J., Iversen, L., Moller, P., Ras-
Estimating occupancy and abundance of stream mussen, M., & Willerslev, E. (2012b). Detection of
amphibians using environmental DNA from filtered a diverse marine fauna using eDNA from seawater
water samples. Canadian Journal of Fisheries and samples. PLoS ONE, 7(8), e41732.
Aquatic Sciences, 70, 1123-1130. Turner, C., Barnes, M., Charles, C., Xu, Y., Jones, S., Jerde,
Pilliod, D., Golberg, C., Arkle, R., & Waits, L. (2014). C., & Lodge, D (2014). Particle size distribution and
Factors influencing detection of eDNA from a stream optimal capture of aqueous macrobial eDNA. BioR-
dwelling amphibian. Molecular Ecology Resources, xiv doi: 10.1101/001941.
14(1), 109-116. Valentini, A., Pompano, F., & Taberlet, P. (2009). DNA
Pollock, K., Nichols, J., Simmons, T., Farnworth, G., barcoding for ecologists. Trends in Ecology and Evo-
Bailye, L., & Sauer, J. (2002). Large scale wildlife lution, 24, 110-117.
Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014 1283Veldhoen, N., Ikonomou, M., & Helbing, C. (2012). & Cooper, A. (2003). Diverse plant and animal gene-
Molecular profiling of marine fauna: Integration of tic record from Holocene and Pleistocene sediments.
omics with environmental assessment of the world’s Science, 300, 791-795.
oceans. Ecotoxicology and Environmental Safety, Willerslev, E., Capellini, A., Boomsman, W. R., & Nielsen,
76(1), 23-38.
M. (2007). Ancient biomolecules from deep ice cores
Venter, J., Remington, K., Heidelberg, J., Hallpern, A., reveals a forested Southern Greenland. Science, 317,
Rusch, D., Eisen, J., Wu, Du., Paulsen, I., Nelson, 111-114.
K., Nelson, W., Fouts, D., Levy, S., Knap, A., Lomas,
Wilson, C. & E. Wright. (2013). Using environmental DNA
M., Nealson, K., White, O., Peterson, J., Hoffman,
(eDNA) as a tool in risk-based decision making.
J., Parsons, R., Baden-Tillson, H., Pfannkoch, C.,
Technical Report, Ontario Ministry of Natural Resou-
Rogers, Y., & Smith, H. (2004). Environmental geno-
me shotgun sequencing of the Sargasso Sea. Science, rces, Aquatic and Research Development Section.
304, 66-74. Aquatic Research Series, 2013-01.
Weirbauer, M., Fucks, D., & Peduzzi, P. (1993). Distribu- Yoccoz, N. (2012). The future of environmental DNA in
tion of viruses and dissolved DNA along a coastal ecology. Molecular Ecology, 21, 2031-2038.
trophic gradient in the Northern Adriatic Sea. Applied Zarzoso-Lacoste, D., Corse, E., & Vidal, G. (2013). Impro-
Environmental Microbiology, 59(12), 4074-4082. ving PCR detection of prey in molecular diet studies:
Wilcox, T., McKelvey, K., Young, M., Jane, S., Lowe, importance of group specific primers set selection
W., Whiteley, A., & Schwartz, M. (2013). Robust and extraction protocol performances. Molecular
detection of rare species using environmental DNA: Ecology Resources, 13(1), 117-127.
The importance of primer specificity. PLoS ONE, Zhu, B. (2006). Degradation of plasmid and plant DNA in
8(3), e59520. water microcosms monitored by natural transforma-
Willerslev, E., Hansen, A., Binladen, J., Brand, T., Gilbert, tion of real time polymerase chain reaction (PCR).
M., Shapiro, B., Bunce, M., Wiuf, C., Gilichisky, D., Water Research, 40, 3231-3238.
1284 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 62 (4): 1273-1284, December 2014You can also read
