3 Progress of domestic and international eDNA-based surveys of amphibians

3.1 Species diversity surveys

With the signing of the Convention on Biological Diversity (CBD) and the 15th Conference of the Parties (COP15) to the United Nations Framework Convention on Climate Change (UNFCCC), global biodiversity has received wide attention. eDNA technology combined with Next-generation sequencing offers a more sensitive and efficient method for biodiversity surveys. Svenningsen et al. used simultaneously traditional survey methods and eDNA metabarcoding to detect amphibian diversity in Danish lakes and ponds, and evaluated the detection efficiency of the two methods from the perspectives of species richness, the average number of species detected at various points, and frequency of detection. They found that both eDNA metabarcoding and traditional surveys yielded a large number of observations unique to the respective method, suggesting that eDNA metabarcoding should be combined with traditional methods as a complement to enhance determination of the total composition of species (Svenningsen et al., 2022).
eDNA detection of amphibians has yielded good results in temperate regions (Deiner et al., 2017; Thomsen & Willerslev, 2015; Yates et al., 2019). However, it has been demonstrated that high temperatures and strong UV light favor the growth of microorganisms and promote the degradation of DNA. To validate eDNA metabarcoding surveys of amphibian diversity in the tropics, Sasso et al. analyzed four streams in the Atlantic Forest of southeastern Brazil during a five-year period of traditional field surveys (using visual and acoustic methods) as well as short-term sampling (4 days) of eDNA. A total of 10 species of aquatic amphibians were detected over the 5-year period of the traditional survey, while nine species were detected by eDNA metabarcoding, including one species that had only been detected once during the 5-year traditional survey. There was a high degree of similarity in stream community composition between the results of the two methods (Sasso et al., 2017). Li et al. used eDNA metabarcoding to assess amphibian diversity at 288 sites in 18 regions of Hainan Island and found 15 species, including a number difficult to detect with traditional monitoring methods (Li et al., 2021). These studies demonstrate the effectiveness of the eDNA metabarcoding technique in detecting amphibians in the tropics.

3.2 Detection of invasive species

Alien invasive species are one of the main causes of global biodiversity loss and homogenization (Ehrenfeld, 2010; Ficetola et al., 2007; Pysek & Richardson, 2010; Vitousek et al., 1997). Invasive species may disrupt local ecological balance and disturb the ecosystem, with substantial impacts on the variety and abundance of native species (Ficetola et al., 2008). The early presence of invasive species is difficult to detect in a timely manner using traditional methods, while eDNA technology allows for rapid and sensitive monitoring of invasive organisms. Invasive amphibians were the first vertebrates for which species distribution was successfully assessed by extracting DNA from water samples (Ficetola et al., 2008). Specific primers were used to expand short mitochondrial DNA sequences to trace the presence of invasive bullfrogs (Rana catesbeiana ) in controlled environments and natural wetlands. Similarly, Dejean et al. assessed the distribution of this species in ponds in western France using traditional monitoring (visual and acoustic surveys) and eDNA. Bullfrogs were found in only 14% of sites in the traditional survey, while the eDNA survey detected a much higher presence (78%), demonstrating substantial effectiveness and sensitivity (Dejean et al., 2012).

3.3 Detection of rare and endangered species

eDNA has great potential for detecting rare and endangered species. Pierson et al. investigated the occurrence of the tiny, rare, multi-toothed salamander Urspelerpes brucei by using both the litter bag survey and eDNA methods. The results showed that the detection probability of the eDNA survey was 0.788, much higher than that of the litter bag survey (0.048). The eDNA method also imposes weaker disturbances on aquatic habitats and has lower costs than the litter bag method (Pierson et al., 2016). Goldberg et al. designed specific primers to detect two rare species in the United States,Ascaphus montanus and Dicamptodon aterrimus . In the experiment, water samples were collected from five rivers with different densities of the target species, and PCR amplification was performed to detect whether the species’ DNA existed in the samples.Dicamptodon was detected in all water samples and Ascaphusin four of the five rivers, proving the effectiveness of eDNA detection for surveys of these rare species (Caren S Goldberg et al., 2011).
McKee et al. studied the species Ambystoma cingulatum ,Ambystoma bishopi , Notophthalmus perstriatus andLithobates capito in the southeastern longleaf pine region of the United States using both eDNA and traditional methods (mainly dip-netting, trapping, and visualization). The eDNA method succeeded in detecting the target species at six sites where traditional methods did not show a presence, again indicating that eDNA can be used as a supplement to traditional methods for investigating endangered species (A. M. McKee et al., 2015). Voros et al. used eDNA to detect the presence of the rare cave-dwelling amphibian Proteus anguinus in all of 15 sampled caves in Croatia, five of which constituted first records for the species (Voros et al., 2017). Lopes et al. analyzed water samples from the Atlantic Coast Forest and six mountainous areas of the adjacent Cerrado grasslands in Brazil for traces of DNA of rare and endangered amphibian species. They successfully detected four declining species (Hylodes ornatus , Hylodes regius ,Crossodactylus timbuhy and Vitreorana eurygnatha ), two species considered locally extirpated (Phasmahyla exilis andPhasmahyla guttata ), and a species not seen since 1968 (Megaelosia bocainensis ) (Lopes et al., 2021). These studies successfully confirmed the presence of species not detected by traditional methods, and underline the effectiveness of the eDNA method for biodiversity monitoring at low population densities.

3.4 Species abundance and biomass assessment

eDNA allows for the assessment of the relative abundance and biomass of faunal communities. In 2019, Kelly et al. analyzed eDNA macro-barcodes of three simulated biomes with different biomass ratios. The effects of these ratios on biodiversity estimates were analyzed by assuming that the amount of DNA collected in the environment was proportional to abundance of individuals in the water, and that the DNA shedding rate and amplification efficiency of certain primers of isotaxa was constant. The analysis results showed that in the case of high amplification efficiency, such as primer adaptation, the ratio index of eDNA macro-barcode readings was highly indicative of trends in group biomass (Kelly et al., 2019). This result has been replicated in amphibian studies. For example, Thomsen et al. investigated six different species including two amphibians (Pelobates fuscus and Triturus cristatus ) using eDNA and qPCR methods, finding that the eDNA concentrations of these two species in water samples correlated with density and biomass (P. F. Thomsen, J. Kielgast, L. L. Iversen, C. Wiuf, et al., 2012). Pilliod et al. used eDNA and traditional methods to assess the diversity and abundance of two amphibian species,Ascaphus montanus and Dicamptodon aterrimus , in 13 streams in central Idaho, USA. They reported that eDNA reading proportion was positively correlated with density and biomass as measured by traditional methods, and showed that the accuracy of eDNA-based abundance assessments increased with the amount of eDNA in water and the number of replicate samples collected (Pilliod et al., 2014). Everts et al. used droplet digital PCR (ddPCR) analysis of eDNA to detect the abundance of bullfrogs in the United States, and found that eDNA concentration increased significantly with increasing abundance of bullfrog eggs and tadpoles (Teun Everts et al., 2021).

3.5 Reconstruction of ancient ecosystems

eDNA for amphibian surveys is mainly used to detect the presence of target species in current communities. However, it has been shown that since DNA molecules can exist in the environment for a long time, ancient ecosystems can be assessed by extracting remnant DNA molecules from environmental sources. Ficetola and Taberlet detected eDNA of several amphibian species (Bufo bufo , Rana temporaria andIchthyosaura alpestris ) in 1,000-year-old sediments from a lake in the French Alps. This suggests that eDNA stored in lake sediments can be used to understand changes in species distribution over time. In addition, it can be combined with other paleoecological data to understand species’ responses to environmental changes (such as habitat alterations, climate change, and the introduction of alien species) (Giguet-Covex et al., 2019).

3.6 Amphibian disease surveillance

In addition to factors such as habitat degradation and environmental pollution, viral infections have also emerged as a significant contributor to the decline of amphibian populations (Rachowicz et al., 2006). According to Scheele et al., the amphibian chytrid fungus (Batrachochytrium dendrobatidis ; Bd) has led to the decline of at least 501 amphibian species worldwide, with 90 species having gone extinct, including the Darwin’s frog (Rhinoderma darwinii ) (Scheele et al., 2019). In recent years, eDNA methods have demonstrated remarkable capabilities for detecting Bd due to their convenience, sensitivity, and non-invasive nature.
Researchers like Stephen et al. have successfully determined the presence and quantity of chytrid pathogens by collecting skin swabs from amphibian hosts, extracting DNA from these swabs, and employing qPCR techniques (Boyle et al., 2007). Beyond skin swab collection, pathogenic DNA can also be detected from environmental samples. Walker et al., for instance, successfully detected the presence of the Bd in small-volume (<1 L) water samples by filtering and utilizing qPCR (Walker et al., 2007). Similarly, Julie et al. employed filtration to capture zoospores from water samples and subsequently detected pathogen DNA from filtered particles at concentrations as low as 0.06 zoospores per water sample, highlighting the high analytical sensitivity of the eDNA method (Kirshtein et al., 2007).
The potential of eDNA technology for early detection of Bd in the environment is noteworthy. Colleen et al. collected eDNA from filtered water samples and detected Bd at three sites, all of which a month later turned out to experience Bd-induced mortality, while Bd was not detected in sites without later mortality (Kamoroff & Goldberg, 2017). This underscores the potential of eDNA technology for early detection of Bd presence.