Research Article: Development and Application of an eDNA Method to Detect the Critically Endangered Trinidad Golden Tree Frog (Phytotriades auratus) in Bromeliad Phytotelmata

Date Published: February 15, 2017

Publisher: Public Library of Science

Author(s): Sarah Brozio, Chloe Manson, Eleanor Gourevitch, Thomas J. Burns, Mark S. Greener, J. Roger Downie, Paul A. Hoskisson, Hideyuki Doi.


The use of environmental DNA (eDNA) to monitor rare and elusive species has great potential for conservation biology. Traditional surveying methods can be time-consuming, labour-intensive, subject to error or can be invasive and potentially damaging to habitat. The Trinidad golden treefrog (Phytotriades auratus) is one such species that would benefit from such an approach. This species inhabits the giant bromeliad (Glomeropitcairnia erectiflora) on two peaks on the Caribbean island of Trinidad. Traditional survey methods for this species have required the destruction of the giant bromeliad, which is the only known habitat of this frog. Here we described the development of an eDNA PCR-based assay that uses water drawn from the water-filled phytotelmata of the giant bromeliad along with the use of a synthetic DNA positive control that can be easily amplified in the bacterium Escherichia coli. The assay can detect to a DNA concentration of 1.4ng. Sampling of 142 bromeliads using this method revealed 9% were positive for P. auratus DNA. These data suggest that eDNA methods also have great potential for revealing the presence of elusive species in arboreal habitats.

Partial Text

The extraction and identification of DNA from environmental samples has recently shown great potential for the monitoring of endangered and elusive species[1,2]. Traditional surveying methods in zoology for the determination of a particular species are generally time consuming, highly labour-intensive and vulnerable to observer error[1]. Complementary approaches to identify elusive or rare species have been sought for a long period to aid in surveying. Environmental DNA (eDNA) is nuclear or mitochondrial DNA that is released from an organism in to the environment. The analysis of eDNA has shown great potential in the field of conservation biology over a range of habitats [1–4]. Sources of eDNA include faeces, mucous, gametes, shed skin and carcasses and as such can report on the presence of a given species in a habitat, thus finding application in a range of ecological studies [1]. In aquatic environments eDNA is dispersed within the water and can persist for between one and 21 days depending on the prevailing environmental conditions and has successfully been used in invasive species monitoring[5], disease monitoring [6] and amphibian conservation monitoring [3].

The golden tree frog (Phytotriades auratus) is a highly elusive species, with a constrained range in Trinidad (by altitude and breeding site availability) which is likely to be significantly affected by climate change [7]. Current survey methods require destructive sampling of the only known breeding sites of P. auratus–the phytotelmata of the giant bromeliad (Glomeropitcairnia erectiflora)[7,13]. Visual sampling for elusive species is time consuming, labour-intensive and subject to observer error [1] and in the case of P. auratus destructive, as G. erectiflora needs to be removed from branches and the leaves separated, thus destroying P. auratus habitat. To avoid damage to habitat we have developed a non-destructive, non-invasive method to utilise eDNA from the water filled phytotelmata of G. erectiflora to survey for this species. Recent work on using eDNA to survey for amphibians has shown the value of such methods as sensitive and efficient tools to detect elusive or difficult to survey species and is becoming the preferred method of surveying given the cost and time effective nature of its application [3,14–16]. The decision to use standard PCR to detect eDNA in this study was based on the need to prove presence or absence of the frogs. It has been suggested that qPCR can give quantitative measurements of population size [2], but these approaches have several potential pitfalls in terms of their application and execution related to the mass of the organism, the quality of the water and the release rate of DNA containing material by the target species[1]. There are advantages to using qPCR approaches for detecting eDNA such as increased sensitivity [2], however the relatively small volumes of water present in phytotelmata and the ability to use this method in laboratories without sophisticated qPCR machines makes standard PCR more useful in this situation.