Analyzing the methodology for the use of eDNA in detecting Batrachochytrium dendrobatidis

MSc (Environmental Sciences), North-West University, Potchefstroom Campus

Batrachochytrium dendrobatidis (Bd) is an aquatic fungal pathogen that has caused mass mortalities in amphibian populations globally. Conventional diagnostic assays for the pathogen are invasive, and involve the collection of toe clips, ventral swabs or mouth parts from tadpoles. Environmental DNA (eDNA) offers a non-invasive and time-efficient alternative to diagnosing the pathogen in aquatic environments. This project aimed to develop a Standard Operating Procedure (SOP) for the detection of Bd using eDNA. The protocol was developed under laboratory conditions, and a preliminary method was applied in field application concurrently with laboratory experiments to aid in the further optimisation of the protocol. Four DNA extraction methods were compared in a variety of matrix types, namely sterile, non-sterile and heat-treated. Two crude extraction methods, namely the Heat Lysis and CTAB method, and two commercial DNA extraction kit methods, namely the DNeasy PowerSoil Kit and Zymo Research Bacterial and Fungal Kit, were selected for comparison using the matrix types. One of each method type was selected for the subsequent filtering test phase. The DNeasy PowerSoil Kit and Zymo Research Bacterial and Fungal Kit delivered very similar results, but the DNeasy PowerSoil Kit was selected as the method of choice due to its better quality DNA and greater consistency in terms of Ct-values overall. The crude extraction methods also delivered similar results, but the Heat Lysis method was selected due to its more consistent results with the lowest standard deviations, while being the more cost-effective, as well as less time consuming method. Overall, the kit methods delivered better Ct-values in the non-sterile matrix, but weaker Ct-values in the sterile matrix compared to the crude methods. In the filtering phase, three filteration methods, namely the drill filter, syphon pump, and Continuous Low-level Aquatic Monitoring (C.L.A.M) were tested in the field for their practicality in application and possible filter volume limitations. The syphon pump was chosen due to not requiring any electricity, being capable of sampling in remote regions due to its compact size and light weight, as well as performing similarly in terms of filtering capacity when compared to the other methods. Two filter material preservation mediums, 70% ethanol and Lysis Buffer, were tested over three time periods for the two selected DNA extraction methods. The Dneasy PowerSoil Kit coupled with Lysis Buffer preservation medium performed significantly better compared to the other method and preservation medium combinations. Standards for the qPCR assay were set using a G-Block dilution series as well as a filtered zoospore dilution series to determine the efficiency, lowest limit of detection (LOD) and lowest limit of quantification (LOQ) of the Bd primers. The primers delivered an acceptable efficiency of 93.5%, with a LOD of 100 copies/μl and a LOQ of 1000 copies/μl in the G-block series. A preliminary eDNA assay was applied concurrently with lab experiments to aid in optimising the protocol. This not only gave insight to the limitations of the developed assay, but also provided guidance for future field sampling. A fellow MSc student, Mr. Jacques Potgieter, from the Herpetological Health Lab (HHL) was conducting a study at the Ncandu Nature Reserve, uMsonti Nature Reserve, and Ncandu Private Forest and Grassland Reserve on the distribution of Bd using conventional diagnostic measures. To reduce the impact and stress on the amphibian population, the projects were planned in parallel. This collaboration allowed for a direct comparison between the preliminary eDNA assay and conventional diagnostic methods. To test the validity of the assay, an internal control assay, using Amietia spp. primers, was run alongside the extracted eDNA samples. This primer set was selected due to the high numbers of Amietia delelandii (Delalandi’s river frog) present at the sampling sites. The LOD and LOQ of the primers as well as the specificity was tested prior to field sampling. The LOD of the Amietia spp. was determined to be 10 copies/μl, the LOQ 100 copies/μl, and the assay only amplified Amietia spp. DNA based during the specificity test. When the preliminary assay was applied in the field, only 70% ethanol had been tested for both the selected extraction methods under laboratory conditions. The DNeasy PowerSoil kit demonstrated a high level of inhibition in downstream application when applying the ethanol and was deemed insufficient for field application. Thus, the Heat-Lysis method, coupled with 70% ethanol was selected as the preliminary method. Three of the four sites tested positive for Bd using conventional diagnostic measures, but none for Bd eDNA. However, all four of the sites tested positive for the Amietia spp. eDNA. Thus, indicating that the protocol is sufficient for eDNA application, but the Bd DNA concentrations present in the environment may have been below the detection threshold. After field application of the preliminary method, the Lysis Buffer was tested as a preservation medium and the DNeasy PowerSoil Kit – Lysis Buffer combination delivered significantly better Ct-values compared to the Heat Lysis – Ethanol combination. The sensitivity was significantly increased and the LOD of the final eDNA protocol was determined to be 10 zoospores according to the zoospore filter dilution series. Based on the Amietia spp. results from the field application, much could be gathered from the data regarding which environments present the best conditions for sampling eDNA. It was determined that lentic water bodies with a high level of turbidity and lower water columns present more positive samples and lower Ct-values. It would be recommended to sample eDNA during spring and early summer when flow rates are still low prior to summer rainfall and during the breeding season of the amphibian hosts.

Masters