1 INTRODUCTION

Freshwater aquatic ecosystems cover only around 1% of the Earth's surface, yet they contain ~10% of all species recorded (Strayer & Dudgeon, 2010). Unfortunately, these systems are particularly prone to threatening processes resulting from human activities, which has led to the loss of about three-quarters of all wetlands and ~28,000 freshwater species facing the threat of extinction (Darwall et al., 2018). Modification and loss of wetlands result from a wide variety of causes, including alterations in water quality, degradation or destruction of habitat, exploitation of species and resources, introduction of non-native species, pollution, and climate change (Kingsford et al., 2016).

Examples of the threat faced by inland aquatic biodiversity can be seen in south-western Australia, where significant and increasing survival pressure due to large-scale human endeavors is having a profound impact (Horwitz et al., 2008). South-western Australia has been acknowledged as housing exceptional biological richness, but it has also been recognized as being under exceptional threat, thus qualifying it as one of the 25 global biodiversity hotspots of Myers et al. (2000). Activities such as agriculture, urbanization, groundwater extraction, and mining have altered the hydrological nature of this region (Burnham et al., 2012), and climate change is exacerbating these issues, further threatening aquatic taxa (Hoffmann et al., 2018). An example of this can be seen in the freshwater crayfish genus Engaewa Riek, a group of strongly burrowing crayfish that reach a maximum of ~20-mm occipital carapace length (OCL) in size (Horwitz & Adams, 2000). These crayfish are endemic to the region, and currently 60% of species within the genus are listed as Critically Endangered or Endangered under the Australian Federal EPBC Act (Department of the Environment, 2023).

Engaewa are experiencing a wide range of threatening processes, including removal of vegetation, alteration of hydrology, and compaction of soil, all of which reduce the quantity and quality of habitat and increase habitat fragmentation (Horwitz & Adams, 2000). While all species in the genus show relatively narrow geographical distributions, Engaewa pseudoreducta Horwitz and Adams is the most restricted of the described species (Burnham, 2014). At the time of its listing as Critically Endangered under the State and Federal Acts its known range constituted significantly less than 3 km², it had been eradicated from its type locality, and the species had been found at only one other location, which was experiencing habitat degradation (Horwitz & Adams, 2000). As it is confined to highly restricted and disjunct habitat and possesses low dispersal ability, there are serious concerns regarding the viability of this species (Burnham et al., 2012). There is likely little suitable habitat still present within its presumed range; however, extensive surveying effort has resulted in the detection of the crayfish at a site ~16 km to the north of the previously known population (Burnham et al., 2012). Efforts to identify further populations have been hampered by the difficulty of detecting these crayfish due to their extreme burrowing habit.

Understanding the distribution of species is the essential first step in assessing and managing threatened species; however, the conservation of species dependent upon freshwater ecosystems is frequently challenging as they often cannot be easily visualized, and the difficulty of detecting them is further exacerbated for species that are rare and/or cryptic. This is true of species within the genus Engaewa as they do not occupy open water, which would allow for visual identification via methods such as trapping, which are often used for freshwater taxa. Instead, they burrow deeply on the margins of water bodies to access the water table within their extensive burrow systems (Burnham, 2014). Freshwater crayfish globally display a wide variety of burrowing strategies, from living in open water with virtually no burrow construction, through to species that burrow deeply and live entirely independently of any standing water (Horwitz & Knott, 1983). This leads to a lack of knowledge regarding the general biology, ecology, and population dynamics of those species that lead a predominantly subterranean existence.

It has been suggested that Engaewa species evolved their extreme digging habit as an adaptation to the highly seasonal water availability of south-western Australia, which has allowed them to persist in a climate that has seen a general drying trend (Burnham, 2014). Most species have never been seen outside of their burrows (that may penetrate many meters into the soil) (Burnham, 2014). These burrows provide not only a home for the crayfish but also essential ecosystem services such as facilitating the movement of oxygen, water, and nutrients through the soil profile (Horwitz & Knott, 1983), and creating habitat that can be occupied by other fauna (primarily, though not exclusively, small invertebrates), which are collectively termed pholeteros (Horwitz, 1995). These crayfish demonstrate the characteristics of a keystone species as their loss has the potential to alter drainage, decomposition, erosion, sedimentation, and primary production. As such, it is imperative that these species are monitored from a conservation as well as an ecosystem function perspective.

Currently, the only method available to survey for the presence of Engaewa species begins with visually locating burrows, where burrow entrances are recognized by a pile of soil (referred to as a “chimney”), which results from the crayfish forming soil pellets as they excavate and maintain their burrow that are expelled through the burrow entrance (Burnham, 2014). The pellets that form the chimney can be up to 1–2 cm in diameter, though they are often much smaller, and the chimney itself can range from less than half a dozen small pellets surrounding a small hole to a conical-shaped chimney up to ~35 cm high, formed from hundreds of individual pellets (Burnham, 2014). Initially, the soil forming the chimney will be distinctly pelleted, though it may eventually appear as a simple mound of soil due to weathering (Burnham, 2014). Even when crayfish burrows are found, identifying whether Engaewa are present is not always straightforward, as crayfish from the genus Cherax Erichson occur in sympatry and create burrows with somewhat similar soil pellets. On occasion, experienced researchers can identify distinctions between burrows formed by these two genera as Cherax burrows are simpler in structure and shallower than those constructed by Engaewa, resulting in smaller piles of pellets. The pellets of Cherax chimneys are often deposited more to one side of the burrow and may form a caldera rather than a chimney, and the pellets are generally larger (corresponding with their larger body size) (Burnham, 2014). However, there are cases where the distinction is not always clear even to an expert, and additionally juvenile Cherax will form relatively smaller pellets further confounding identification (Burnham, 2014). As a result, there will, on occasion, be chimney-type structures that are difficult to differentiate on the surface, especially if they have been weathered.

Even if it is assumed that a crayfish from the genus Engaewa did construct a detected burrow, it is not evident which species has done so without collecting specimens through an invasive, destructive, and time- and effort-intensive method (Burnham, 2014). Currently, confirmation of the species present at a site requires collecting a crayfish by hand-excavation of the burrow. As Engaewa burrows can be >2 m deep and are composed of many bifurcating tunnels that are generally only a centimeter or two wide, excavating them with a shovel destroys the burrow system of the crayfish and much of the immediate surrounding habitat. Additionally, due to the burrow complexity and the small size of the crayfish, even significant excavations do not always result in finding a crayfish. It is obvious that an alternative survey method to the current destructive, time-consuming, and unreliable method of sampling a critically endangered species is needed; therefore, we suggest that environmental DNA (eDNA) techniques could be developed for species identification purposes.

Environmental DNA is the term given to trace the amounts of DNA left within an environment by an organism as they pass through it, and collection of an environmental sample requires a relatively small volume of water and/or soil for analysis. As many endangered freshwater crayfish species are rare and cryptic, highly sensitive and noninvasive methods like eDNA-based monitoring are needed to improve conservation outcomes. An additional benefit of utilizing an eDNA approach for studies such as this is that, once a verified DNA reference sequence is obtained, no taxonomic expertise is required for positive detection of the species. It has been demonstrated that eDNA can be used to detect freshwater crayfish (e.g., Chucholl et al., 2021; King et al., 2022; Rusch et al., 2020; Trujillo-Gonzalez et al., 2021); however, the focus of crayfish eDNA studies has often been the early detection of biological invasions (e.g., Geerts et al., 2018; Mauvisseau et al., 2018). There are also some examples of using an eDNA approach to monitor rare or endangered species, even when at low abundances (e.g., Cowart et al., 2018; Jerde et al., 2011); however, these studies have primarily occurred in lotic environments targeting secondary burrowers (i.e., stream-dwelling species). There is only one study to date that has focused on a primary burrowing crayfish species (see Quebedeaux et al., 2023); however, although this study did not target a stream-dwelling species, its burrows are intermittently connected to surface waters and sampling was able to utilize these ephemeral waters for sampling purposes rather than utilizing the burrow substrates.

This study focused on the burrowing crayfish E. pseudoreducta, considered the most imperiled species within the genus. By sampling the environment into which E. pseudoreducta is most likely to deposit DNA (i.e., soil and burrow water), we propose that the detection of this species' presence may be possible without the need for removing the crayfish from its burrow. To the best of the authors' knowledge, there are no publications that have investigated whether crayfish eDNA can be successfully extracted directly from burrow substrates; thus, it is critical to evaluate substrate selection for eDNA-based surveys. The specific aims of this study were to (i) examine the efficacy of eDNA as a tool for detecting E. pseudoreducta and (ii) compare the detectability of E. pseudoreducta from different eDNA substrates, including soil and water.

2 METHODS

2.6 Field validation: eDNA samples

We field-tested the developed qPCR assay on environmental DNA collected from Payne Road (Western Australia), one of only two localities where E. pseudoreducta is known to occur (Figure 1). All suitable habitat at the Payne Road site was surveyed using walked transects during which we recorded the location of all crayfish burrows, measurements of burrow height, width, and diameter of soil pellets and took photographs. A total of 110 burrows were located, and we sampled 11 that were considered the most likely to be made by E. pseudoreducta based on burrow width (small), shape of the chimney (conical), and soil pellet size (small) as per the characteristics outlined in the introduction.

Three sample substrates were targeted for collection: sediment from chimney pellets, sediment from internal burrow scrapes (~5–10 cm below the burrow entrance), and water from inside the burrow. Sediment samples were collected in a falcon tube, and water samples were filtered onto a filter cartridge (surfactant-free cellulose acetate, 1.2 μM pore size, 28 mm diameter) using a syringe, with all samples stored at −20°C after collection. Nitrile gloves were worn and changed after collection of each sample. Due to the high sedimentation present in the water, we filtered varying volumes, with an average of ~140 mL pushed through the filter cartridge. Samples were collected in September, which represents the end of the wet season (and therefore peak water levels) in the study region.

Sediment and water samples were extracted using a DNeasy PowerSoil Pro kit and DNeasy Blood and Tissue kit (Qiagen), respectively. To extract DNA from the filter cartridge, ATL buffer was injected into the cartridge and left to sit for 3 days. The ATL buffer was then aspirated from the cartridge and placed into a tube for subsequent digestion and extraction. Laboratory extraction controls were run alongside samples, with only the extraction reagents included.

Preliminary tests run on a subset of samples at both neat and 1/10 dilution (DNA template input of 2 μL) indicated late amplification in a small number of samples. As such, we performed all further testing using an input volume of 7 μL DNA template. Reactions were run in quadruplicate, and each PCR plate included no-template and positive (synthetic DNA) samples.

3 RESULTS

3.2 Assay LOD and LOQ

On machine analysis of the standard curve showed good precision (R² = 0.988), a slope of −4.11, and an efficiency of 78.33%. The lowest standard to show amplification was 6.4 copies/reaction, where 3 out of 12 replicates were successfully amplified. Based on the criteria applied, the discrete LOD was calculated to be 64 copies/reaction (where 95% of the replicates showed amplification) and the LOQ was 393 copies/reaction. Using the modeled calculation of Klymus et al. (2019), we calculated the LOD to be 20 copies/reaction (Figure 2).

4 DISCUSSION

The successful development of a species-specific assay to detect Engaewa pseudoreducta demonstrates the efficacy of eDNA biomonitoring of burrowing crayfish, which is consistent with the growing body of research that confirms eDNA is a useful tool for the detection of freshwater crayfish species (e.g., Chucholl et al., 2021; King et al., 2022; Quebedeaux et al., 2023; Trujillo-Gonzalez et al., 2021).

We sampled 11 burrows identified as likely constructed by E. pseudoreducta. Six of these were confirmed through positive amplification in at least one qPCR replicate. While we cannot confirm if the other samples tested were taken from burrows made by E. pseudoreducta (i.e., false negatives) as hand excavation did not occur (due to the species status as Critically Endangered), positive detections through eDNA methods prove that this approach can provide a non-invasive detection technique. Although further testing and development would be beneficial, even at a detection efficiency potentially as low as 50%, this method still provides potential benefits when compared to the traditional excavation method. As previously mentioned, excavating the burrows of these systems is destructive and time-consuming (and therefore costly) and frequently results in false negatives due to the challenge of sighting the organism in a huge volume of soil that is removed while trying to follow their large, complex burrow systems.

The assay was sensitive enough to detect the species in chimney pellets and burrow sediment, but not water sampled from burrows. This is consistent with the findings of previous studies that have shown the importance of selecting appropriate eDNA substrates when conducting surveys in aquatic systems (e.g., Buxton et al., 2017; Koziol et al., 2019) but is surprising given freshwater crayfish (e.g., Chucholl et al., 2021; King et al., 2022; Rusch et al., 2020; Trujillo-Gonzalez et al., 2021), including burrowing crayfish (Quebedeaux et al., 2023), have been detected successfully from water. It is possible that non-detection in this substrate could be due to the low volume of water filtered, as it is generally accepted that the probability of detection increases with the volume of water sampled (see Sepulveda et al., 2019 and references therein). While a bigger pore size to increase the volume of water filtered could be tested, the load of suspended solids in the burrow water would likely obstruct even large pore sizes, making it unlikely to succeed in increasing detectability. Another possibility for nondetection in water samples could be the extraction method used for this sample type, where sufficient PCR inhibitors were not removed and they prevented amplification during PCR. While a dilution series was used during the PCR process that should lessen the impact of inhibitors, the biology of this species means that it is likely that limited DNA has been deposited in the environment; therefore, while diluting out PCR inhibitors, the amount of target DNA would also be reduced, thus leading to no samples providing positive amplification.

The sample substrate that provided the most positive detections was the chimney pellets (6 out of 11), followed by the internal burrow scrapes (3 out of 11). It is possible that the chimney pellets provided more detections due to increased contact with the crayfish, through their handling of the sediment as they roll and manipulate it to remove it from the burrow. Chimney pellets are by far the easiest substrate to sample, particularly if sample collection is to be undertaken by nonexperts. No specialist equipment is needed, with only collection tubes, disposable scrapers, and gloves required. However, it should be noted that as soon as a pellet is expelled from the burrow the DNA will start to degrade from UV exposure, in addition to the bacterial action already occurring in the burrow. It will therefore be important to sample those pellets that have been expelled most recently; freshly expelled pellets will appear moist, spherical, and uniform, but they rapidly become weathered either by drying out and crumbling or conversely by being washed away. Additionally, other sampling factors must be considered for future eDNA studies. For example, the importance of seasonality of sampling for accurate eDNA-based surveys and monitoring is widely acknowledged (De Souza et al., 2016; Hayami et al., 2020). For many taxa, sampling when the amount of DNA being deposited into the environment is at its maximum may be important (for example, when they are spawning or molting), as this is when they are most detectable via an eDNA approach. Specifically in relation to Engaewa, their chimneys of soil are often only apparent in the wetter months of the year and the crayfish can be found closer to the surface at these times as they appear to remain around the level of the water table, as has been noted when undertaking hand excavation of burrows (Burnham, 2014). As a result, sampling at this time results in the highest percentage of successful excavations, and it is also likely that this is when DNA presence will be highest as the crayfish increase their burrow maintenance and are expelling fresh soil pellets (Burnham, 2014).

The results obtained suggest that little DNA is left by E. pseudoreducta in the environment due to their biology. An animal with a hard or keratinized outer layer like crustaceans and reptiles might shed less eDNA than an animal with delicate skin (e.g., some mammals; Adams et al., 2019; Andruszkiewicz Allan et al., 2021). While eDNA is also shed into the environment through feces, urine, epithelial cells, external mucous layers, saliva, gametes, and physical remains, including shed exoskeletons of crustaceans, it is thought that relatively limited eDNA shedding from a keratinized exterior may reduce the detectability of some groups of organisms from environmental samples (Adams et al., 2019; Lacoursière-Roussel et al., 2016). Mesocosm trials have been undertaken with stream-dwelling crayfish to better understand this issue (e.g., Baudry et al., 2021; Troth et al., 2020), but as yet there is insufficient data for strongly burrowing species that may live in a semi- or near fully terrestrial environment. As such, it is not surprising that initial PCR tests showed that when using an input of 2 μL of extracted DNA, the target DNA concentration is likely too low to show conclusive amplification. Subsequent tests indicated that increasing this volume to 7 μL is likely to improve testing results; as the environmental tests showed no inhibition in sediment samples, it is unlikely that an increase in template volume would have any associated inhibition issues.

We recognize that the efficiency of the assay developed (78.33%) does not meet what would typically be desired; however, the genetic variability within E. pseudoreducta reference sequences that are available to design primers and probes against is large, so the developed assay was therefore determined to be the most efficient design possible at present. The short barcoding regions of the mitochondrial genome (mitogenome) and marker genes used for assay design often result in suboptimal assay performance. However, the increased accessibility of full mitogenome assemblies may enable the development of more robust eDNA assays (Allison et al., 2023). Therefore, future studies should endeavor to construct mitogenomes for the target as well as other Engaewa species to allow new assays to be developed to improve efficiency and increase the likelihood of detecting the target species. Additionally, an increased replication at both the sampling (biological replicates) and PCR (technical replicates) stage will increase the likelihood of detection.

The development of an eDNA method for the detection of E. pseudoreducta offers an exciting opportunity to expand our understanding of the distribution of the species. The implications of being able to survey a species such as those from the genus Engaewa with less effort, greater confidence, and no impact on the species itself cannot be overstated. eDNA approaches offer the potential to revolutionize biodiversity monitoring and may be most effective in situations where traditional approaches have undesired impacts. While still far from a conservation magic bullet, studies such as this suggest that eDNA, at the very least, has a significant role to play in ongoing efforts to protect biodiversity. This study will improve current conservation efforts for critically endangered freshwater crayfish and be used to inform a species recovery plan.

AUTHOR CONTRIBUTIONS

This study was designed by KLD, QFB, PN, BC, and SH; samples were collected by KLD and BC; samples were processed and molecular research was conducted by KLD; assay design and testing were completed by KLD; statistical analysis was conducted by KLD; all authors contributed to the writing of this paper.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflict of interest.