eDNA‐based detection of a vulnerable crocodile newt (Tylototriton uyenoi) to influence government policy and raise public awareness

Although at least five Tylototriton species were recorded in Thailand, only Tylototriton verrucosus is registered as a protected species under the Wildlife Preservation and Protection Act in Thailand. Populations of T. uyenoi are now severely declining, caused by anthropogenic activities. As intense human pressure is having profound effects on the diminishment in T. uyenoi populations, a conservation plan is needed. Information such as the abundance and distribution of a species is necessary. Yet, current established survey methods are either time‐consuming or labour‐intensive. Here, eDNA‐based detection was developed for tracking the presence of the T. uyenoi.


| INTRODUC TI ON
Amphibians are facing massive population declines (Stuart et al., 2004;Wake & Vredenburg, 2008). Amphibians play a major role in ecosystems, and they are critical components of both aquatic and terrestrial communities. Thus, anthropogenic factors that negatively affect amphibians could influence entire ecosystems. Both direct and indirect anthropogenic activities such as habitat modification, overexploitation, exotic species introductions, emerging pathogens, chemical pollutants and climate change are all proposed causes for amphibian declines (Anderson 2019; Wake & Vredenburg, 2008).
As with other amphibians, humans are the main reason behind an unparalleled decline in newt populations. Illegal trading, habitat alteration (loss) and hunting for food or medicine are major causes of local declines of the Tylototriton newts. The genus Tylototriton currently contains at least 19 species/subspecies of newts native to Southern and Southeast Asia (Bernardes et al., 2020;Nishikawa, Khonsue et al., 2013;Phimmachak et al., 2015). Many of them are now recognized as threatened species by the International Union for Conservation of Nature (IUCN) Red List. In Thailand, Tylototriton verrucosus has long been documented as the only newt. Recently, three new Tylototriton were described: Tylototriton uyenoi, Tylototriton panhai and Tylototriton phukhaensis (Nishikawa, Khonsue et al., 2013;Pomchote et al., 2020). Here, we focus on the T. uyenoi or Chiang Mai crocodile newt. Pui and Doi Inthanon National Park, reveal that over the past few years T. uyenoi have been severely declining, with causes ranging from habitat loss to being hunted for food by locals. As only T. verrucosus is protected by law, the risk of T. uyenoi disappearing from the areas is high.
In order to develop effective conservation interventions, information on the abundance and distribution of imperilled species is necessary. Species monitoring is therefore needed. Yet, currently established survey methods are either time-consuming or labourintensive. Detecting a species' occurrence, decline and extinction is considerably challenging due to the high level of effort and cost associated with the assessment and monitoring of biodiversity.
The advancement of molecular genetic methods has dramatically improved the reliability, portability and widespread application of biomonitoring with environmental DNA (eDNA) in the past decade.
The eDNA analysis is based on the retrieval of genetic material naturally released by organisms in their environments such as water, soil and air (Ficetola et al., 2008). Two main applications of eDNA use are (1) to gain an insight of species richness, estimates of relative abundance and quantification of biodiversity using next-generation sequencing and (2) to target surveillance of a single-specific species using conventional PCR, quantitative PCR (qPCR) or digital PCR (ddPCR). The second application, targeted surveillance, is useful for our focal species. Environmental DNA-based species detection has already proven valuable in tracking various organisms including rare, elusive, endemic and endangered species (e.g., Dejean et al., 2012;Fukumoto et al., 2015;Katz et al., 2020;Pilliod et al., 2013;Preißler et al., 2018). As with many other amphibians, life history characteristics of salamanders and newts, including small population sizes and variation in annual reproductive success, can have an effect on detectability. Thus, different sampling techniques are found to have varying effectiveness (Bevelhimer et al., 2008;Farmer et al., 2009;Gunzburger, 2007;Liner et al., 2008;Parris et al., 1999;Smith et al., 2006). In addition, conditions at the sampling sites also affect the choice of sampling method (Smith et al., 2006). Recent studies have demonstrated the potential of eDNA for detecting salamanders and newts, which reduce those limitations of other survey methods. Some work has demonstrated superior sensitivity and effectiveness of the eDNA survey over physical capture or visual observation (Katano et al., 2017;McKee et al., 2015;Rees et al., 2014).
Environmental DNA is useful for detecting not only adults but also larvae (Bevelhimer et al., 2008;Preißler et al., 2018). Here, the eDNA survey and monitoring of Chiang Mai crocodile newt (T. uyenoi) were developed, which can be incorporated into a conservation plan for monitoring of the species.

| qPCR assay
To design primers and a probe specific to Chiang Mai crocodile newt (T. uyenoi), the partial NADH dehydrogenase subunit 2 region (ND2) of the mitochondrial DNA (mtDNA) region of six individuals was sequenced. Tylototriton uyenoi were caught from Doi Suthep (three individuals) and from Doi Inthanon (three individuals).
All the DNA analysed originated from the mucus of the individual K E Y W O R D S eDNA, monitoring, crocodile newt, government policy, conservation T. uyenoi. The animals were released back into nature after mucus collection. Total DNA was extracted from the mucus sample using the DNeasy Blood and Tissue Kit (Qiagen) according to the manufacturer's protocol. Mucus DNA concentrations were determined using the Qubit dsDNA HS Assay (Invitrogen). Extracted DNA was first used as a template for sequencing with the primers Sal_ND2_F1: 5'-AAGCTTTTGGGCCCATACC-3' and Sal_ND2_R2: 5'-GGTTGCATTCAGAAGATGTG-3' both of which were from previously reported primers (Nishikawa, Khonsue et al., 2013).
Species-specific primers and a minor-groove binding (MGB) probe incorporating a 5′ FAM reporter dye and a 3′ non-fluorescent quencher were then designed based on the consensus sequences generated from this study and the public database to amplify a 116-bp amplicon targeting the ND2 region for the T. uyenoi, using Primer Express (V3.0, Life Technologies; Table 1). Probe and primer sequences were matched against the National Centre for Biotechnology Information (NCBI, http://www.ncbi.nlm.nih.gov/) nucleotide database with BLASTn (Basic Local Alignment Search Tool) to confirm the species' specificity for the T. uyenoi in in silico assays.
The qPCR assay was deployed using Environmental Master Mix (Applied Biosystems) on the mucus samples from the T. uyenoi and non-target species including Amolops marmoratus, Limnonectes taylori, Limnonectes kuhlii and Megophrys parva to ensure that the assay only amplified the T. uyenoi. Also, the qPCR assay was carried out using synthetic fragments of T. anguliceps, T. panhai and T.
verrucosus. As we had not yet been able to obtain these vulnerable Tylototriton species, the synthesised target DNA was designed based on sequences retrieved from GenBank, Accession Numbers LC505611, MK097271 and KT304276 for T. anguliceps, T. panhai and T. verrucosus, respectively. In addition, water containing DNA extracted from mucus (concentration 0.5 ng/μl) was included as a positive control for the presence of amplifiable eDNA in water samples.
All eDNA qPCR amplifications were performed using a Rotor-Gene Q (Qiagen) in three replicates in a final volume of 20 µl, using 10.0 µl of 2× TaqMan Environmental Master Mix 2.0 (Thermo Fisher Scientific), 2.0 µl of DNA template, 900 nM each of the F/R primers and 125 nM of the probe. Samples were run under the following conditions: an initial 10 min of incubation at 95°C followed by 50 cycles of denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min. Negative controls with all PCR reagents but no template (three replicates) were run in parallel to assess potential contamination. The quantification cycle (Cq) was converted to quantities per unit volume using the linear regression obtained from the synthesised target gene standard curve (Integrated DNA Technologies Pte. Ltd.). The T. uyenoi eDNA concentrations were then reported as copies/ml. The limit of detection (LOD) and the limit of quantification (LOQ) were also measured using the standard dilution series of synthesised target gene fragment with known copy numbers. The concentrations of the standards were adjusted to 15,000, 1,500, 150, 15 and 1.5 copies per reaction with 13 technical replicates used for each of the dilution steps. The calculation of LOD and LOQ was done using published R script by Klymus et al. (2020).

| eDNA field collection
Water samples were collected in 2017 and 2018, once a month from August to January at three sites in Doi Suthep including site (1) artificial pond, site, (2) artificial pond and site and (3) natural pond. In addition, water from three extra sites with an unknown occupancy of T. uyenoi but located within the species' range was collected (August to October). Each site was sampled in triplicate, and 300 ml samples of water were collected and filtered on GF/F filter (0.7 μm; Whatman International Ltd.). Filtering control (300 ml of distilled water) was filtered as a last sample on each day of sample filtration.
All filters were folded inwards, placed in 2-ml tubes and stored at −20°C until DNA extraction, which took place within 48 hr. To avoid contamination, all field equipment was sterilized using 10% bleach, UV crosslinker or autoclaved and sealed prior to transport to the study site, and a separate pair of nitrile disposable gloves was used for each sample.

| DNA extraction from the filters
DNA trapped on the filters obtained from field collections was extracted using Qiagen DNeasy Blood and Tissue Kit (Qiagen) using a protocol modified from the manufacturer's protocol with the following changes: the DNA from all samples was eluted twice with 25 µl AE buffer, in a total volume of 50 µl to obtain a more concentrated eDNA solution. The volumes of ATL buffer (360 µl), proteinase K (40 µl), AL buffer (400 µl) and ethanol (400 µl) were doubled.

| Statistical analysis
An ANOVA was used to determine whether the collection site or sampling month had had any statistically significant effects on DNA concentration. Statistical analyses were conducted in R version 3.3.1 (R Core Team 2017).

| Species-specific primer design
The partial ND2 region sequences of the T. uyenoi from two locations in Chiang Mai were generated and deposited in GenBank (Accession Numbers MG018993 and MG018994). The sequences obtained from this study and GenBank were aligned and analysed (Table 2).
There are seven variable sites found within 497 bp analysed ND2 fragments of T. uyenoi. Nucleotide variations were also found be- with at least one mismatch on the forward and four mismatches on the reverse primers (Table 3). When comparing with other amphibian species found in Doi Suthep, the homology was found to be lower (51%-62%). The primers designed were found to be specific to T. uyenoi after testing with other Tylototriton found in Thailand (T. anguliceps, T. panhai and T. verrucosus) and other non-target species (Table 4). The same result was achieved when the probe was added for use in qPCR (Table 4) and resulted in the single detection of the target species, T. uyenoi.

| qPCR assay
The assay designed in this study was found to be species-specific to T. uyenoi. Both PCR and qPCR did not result in any positive detection of the non-target species. The LOD and the LOQ of the assay were determined by an analysis of the standard curve (slope = −3.568, Y inter = 42.376, R 2 = 0.988, Eff% = 90.664) generated from a dilution experiment under laboratory conditions. The LOD was 7.9 copies per reaction, and the LOQ was 9.0 copies per reaction.

| eDNA survey
As eDNA is thought to rapidly degrade into short fragments, PCR primer sets for eDNA studies should be designed to yield short amplicons (100-180 bp) (Dejean et al., 2011). Also, it is noteworthy that short-amplicon PCR assays have high detection sensitivity (Huver et al., 2015). The specific primers for the T. uyenoi generated in this study were a desirable length of 116 bp and thus suitable for an eDNA assay. Also, as intraspecific nucleotide variation was found, the primers and probe were designed outside those variable regions.
This would make sure that the eDNA of T. uyenoi will be amplified across its range.
Tylototriton uyenoi eDNA was found to be varied between sampling months. Highest eDNA concentrations were found in August with a reduction throughout the other sampling months (from August to January), being lowest in January. Breeding season and larvae abundance may also affect on the eDNA concentration (Buxton et al., 2017). Increased eDNA levels of great crested newts (Triturus cristatus) were found at the end of breeding season (early June). Similarly, in our study, eDNA was also detected at high concentration in July-August when the sampling ponds have high occupancy by larvae.
Previously published work on T. uyenoi in Thailand is limited to systematic and diet studies. Because amphibian detectability can vary among conventional, capture-based survey methods due to life history and population size, alternative survey methods are needed (Bevelhimer et al., 2008;Liner et al., 2008;Smith et al., 2006;Strain et al., 2008). The findings here show that eDNA-based survey is useful for tracking the presence of the Chiang Mai crocodile newt.
To our knowledge, this was the first study, which had been carried out in Thailand. In this study, eDNA was proved to be an alternative method for rapid and sensitive surveys.

| Where could the eDNA-based survey stand in the conservation action plan?
Responding to the Amphibian Conservation Summit's response to the amphibian crisis worldwide, the Amphibian Conservation Action Plan (ACAP) was designed and first announced in 2007 (Gascon et al., 2007). Eleven relevant actions for global amphibian conservation were included in ACAP. Since then, although updated in 2015, questions and gaps remain that should be taken into consideration for effective conservation (Wren et al., 2015). For example, for reintroduction, one of the eleven proposed actions to take place, it is essential that species are carefully appraised for their suitability for reintroduction so that status and distribution of the species are required. However, due to highly diverse range of habitat requirements and life history traits displayed by amphibians, conventional monitoring methods such visual TA B L E 3 Homology of the query to the forward and reverse primers and probe, and percentage identity as a function of the number of matching base sites divided by 61 (total number of base sites across the primer pair and probe). Base site homology between the query and the primer is shown as a dot

Identity (%) GenBank
Tylototriton uyenoi TA B L E 4 Results of PCR and qPCR using the primers and probes targeting the ND2 region of Tylototriton uyenoi on seven different species

PCR qPCR
Tylototriton species Non-target species

Amolops marmoratus x x
Limnonectes taylori x x

Limnonectes kuhlii x x
Megophrys parva x x observation are commonly ineffective. In addition, to prevent further habitat loss, the effects of anthropogenic perturbations on amphibian populations are another big question that needs to be addressed. Traditional field survey cannot provide up-to-date information, which would normally help to answer the question.
Environmental DNA poses potential in filling these gaps and also could provide answers to these questions. In

ACK N OWLED G EM ENTS
This work was supported by Chiang Mai University. We would like to thank Dr Santhiti Vadthanara and Dr Chatmongkon Suwannapoom for their involvement in the collection of the samples. We are thankful to our colleagues and students, for every little help from them and also Dr Lauren R. Clark and Mr Leslie C. Cunningham for English editing.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study were derived from the following resource available in the public domain: GenBank  eDNA-based detection of a vulnerable crocodile newt (Tylototriton uyenoi) to influence government policy and raise public awareness. Divers Distrib. 2021;27:1958-1965. https:// doi.org/10.1111