FIRST RECORD OF BATRACHOCHYTRIUM DENDROBATIDIS IN THE NORTHERN NETHERLANDS

Amphibian populations worldwide are declining because of the chytridiomycosis pandemic, caused by the fungal pathogens Batrachochytrium dendrobatidis (Bd) and Batrachochytrium salamandrivorans (Berger et al., 1998; Fisher and Garner, 2007; Spitzen-van der Sluijs et al., 2013, 2014; Scheele et al., 2019). Batrachochytrium dendrobatidis was first discovered in the 1990s, after dramatic population declines of anuran amphibians in Queensland (Australia) and Panama (Berger et al., 1998). In 2013, B. salamandrivorans was described, following the discovery of massive crashes of salamander populations in northwestern Europe (Martel et al., 2013). Due to its global widespread, the fungus Bd has since received much research attention. Ongoing research aims to identify its global distribution pattern, its impact on the species it infects, how to protect biodiversity, and how to mitigate the disease (Fisher et al., 2009; Olson et al., 2021). Because Bd is capable of hybridizing across lineages and hybrid lineages can be more virulent, it is unknown how amphibian host populations will react to the introduction of a new or hybrid Bd lineage (Voyles et al., 2018). Additionally, small amphibian host communities that are recovering from Bd outbreaks are especially vulnerable to new disease outbreaks (Voyles et al., 2018). Screening of populations that are facing Bd-related die-offs and declines, as well as less affected populations that are not or seem not to be as threatened by Bd-related outcomes, is therefore important for mitigation planning (Byrne et al., 2019).

Bd infects a broad range of species, including 520 anuran amphibians (frogs and toads), urodeles (salamanders and newts), and caecilians (Gower et al., 2013; Olson et al., 2013; Scheele et al., 2019). Most anurans are susceptible to Bd infection during all life stages (excluding eggs), although morbidity and mortality vary between species, life stages, and environmental conditions (Gower et al., 2013). The microhabitat and environment that a host species inhabits are key determinants for infection and disease outcome, as, for instance, virulence is reduced at warmer temperatures (>28 C) (Daskin et al., 2011; Van Rooij et al., 2015). Some tolerant amphibian species maintain infection below lethal levels and function as carrier species (Reeder et al., 2012); this makes them potential pathogen vectors as well as environmental reservoirs (Reeder et al., 2012; Kolby et al., 2014).

In Europe, Bd is considered to be widely distributed, infecting a broad range of amphibian species (Garner et al., 2005). Although in northern, southern, and central Europe Bd is known to be widespread (Martel et al., 2012; Allain and Duffus, 2019; Meurling et al., 2020), limited data exist on its distribution in western Europe. Batrachochytrium dendrobatidis has been present in western Europe (including Belgium and the Netherlands) since the late 1990s, although mass mortalities or steep declines in host populations have not been observed (Spitzen-van der Sluijs et al., 2014). In the northern part of the Netherlands, Bd has been reported at very low incidence and prevalence and was not found in all provinces (Spitzen-van der Sluijs et al., 2014); however, large-scale surveillance has not been conducted. The Netherlands hosts 16 native amphibian species: 5 urodeles and 11 anurans. Additionally, there are some non-native species such as the Italian crested newt (Triturus carnifex). Several anuran and urodele species have tested positive for Bd in the Netherlands and can act as Bd carriers (Spitzen-van der Sluijs et al., 2014). However, at the present time research efforts in the Netherlands mostly focus on B. salamandrivorans (Bd’s sister species), which is responsible for massive die-offs of the fire salamander (Salamandra salamandra) in northwestern Europe (Spitzen-van der Sluijs et al., 2013).

We aimed to screen the province of Groningen for Bd to verify the status of Bd presence and to broaden the existing data on Bd distribution in the Netherlands. We sampled 12 locations across the region and performed molecular diagnostic tests to identify the presence of Bd. Our study expands our knowledge of Bd distribution and supports the value of larger sample sizes in pathogen-monitoring studies.

The province of Groningen is situated in the northeast part of the Netherlands (Fig. 1). The landscape is flat, and a large part of the province is below sea level. About 60.9% of the land is used for agriculture, 9% is built-up or semi-built-up area, 6% is natural vegetation, and 9.7% is inland water (Statistics Netherlands, 2018). The province has an oceanic climate, with a year-round daily mean temperature of 10 C (Royal Netherlands Meteorological Institute, 2019–2021).

We surveyed 12 locations around the province of Groningen in July 2020 and May–June 2021, aiming to sample during the amphibian reproductive period to ensure sufficient sample size (Fig. 1). To achieve that, in 2021 sites in Groningen were checked for active amphibian presence (e.g., active calling, newts in aquatic-reproductive form) twice a week since the beginning of spring (March) to determine when sampling could start. Amphibians were found to aggregate only in late May–June, which is later in the year than what was observed the previous year (observations by authors). Sampling sites included different types of inland water (i.e., canals, streams, natural ponds, and garden ponds). Sampling locations were selected based on the observed amphibian species richness and abundance according to RAVON’s (Reptile, Amphibian and Fish Conservation Netherlands, Nijmegen, the Netherlands) distribution data and public access to the sites. Each site was screened for 40 min by 2 persons, and amphibians were captured with a dip net or by hand (amphibians were always handled with sterile nitrile gloves). Our goal was to sample at least 20 amphibian individuals per site, based on the prevalence estimates for Bd in the Netherlands, as that was 4.8% (Spitzen-Van Der Sluijs et al., 2014) and at least 20 individuals need to be sampled for obtaining an objective subset of the population to estimate Bd prevalence. Captured adults and juveniles were kept in individual bags, and tadpoles were kept in water-filled buckets until Bd sampling. All collected amphibians were handled according to safety protocols to prevent the spreading of Bd, and all equipment was disinfected with 70% ethanol between sites (Van Rooij et al., 2017). Each specimen was handled by 1 person, while another person collected skin samples by rubbing the abdomen (adults and juveniles) or the oral disc (tadpoles) with rayon swabs (Hyatt et al., 2007; Image DELTALAB, Barcelona, Spain). Two swab replicates were collected simultaneously (for tadpoles, samples were collected sequentially because of the impossibility of placing 2 swabs in a tadpole disc at the same time) per specimen and were placed in aseptic 1.5 ml microcentrifuge tubes. Pictures of all individuals were taken, and identification of species, sex, and life stage was performed on-site. Individuals were categorized as tadpoles, juveniles, or adults. Individuals that had not reached sexual maturity (<4.5 cm snout-vent length and no clear nuptial pads) were classified as juveniles, and all sexually mature individuals were classified as adults and sexed afterward. If metamorphosis was not complete, individuals were grouped into tadpoles. Individuals were released immediately after sampling. One duplicate of each swab was kept at −20 C until DNA extraction (University of Groningen); the other was kept at −80 C until all extractions, and qPCRs of the first duplicate were performed, to be sent to an independent external laboratory (University of Leipzig) for validation.

At the University of Groningen, DNA extraction of skin swabs was performed with the Biokè NucleoSpin tissue kit (Cat. No./ID 740952.250, Biokè, Leiden, the Netherlands) with an additional step of enzymatic lysis to enhance the lysis of the fungal cell (Belden et al., 2015). DNA extraction protocols followed the single-swab method described in Mantzana-Oikonomaki et al. (2021). During each round of extractions, a blank sample was extracted (no swab) as an extraction negative control (extraction negative controls were all negative). Eluted DNA was stored at −20 C until qPCR runs.

At the University of Leipzig, DNA extraction of skin swabs was performed with the Qiagen Blood and Tissue kit (Qiagen, Hilden, Germany) with the same enzymatic lysis step described above (Belden et al., 2015). Eluted DNA was stored at −20 C until qPCR runs.

After DNA extractions, qPCR was performed using established protocols and Bd-specific primers (Bd [ITS] 5.8S region) (Boyle et al., 2004; Hyatt et al., 2007). In Groningen qPCRs were performed on a CFX96 Real-Time System (Bio-Rad Laboratories Inc., Hercules, California); in Leipzig, qPCRs were performed on a qTower³ (Analytik, Jena, Germany). Each qPCR plate included a series of 5 plasmid-based Bd (Standish et al., 2018) standard dilutions (10, 100, 1,000, 10,000, and 100,000 ITS copies in University of Groningen and 100, 1,000, 10,000, 100,000, and 1,000,000 ITS copies in University of Leipzig) and a negative control containing deionized water. In each qPCR run, the samples, the negative control, and standard dilutions were run in duplicates. In case the replicates showed contradictory results, a third replicate was run.

For both laboratories, a sample was considered Bd positive when 2 qPCR replicates provided a Cq (quantification cycle) value lying between the amplification signals estimated for the lowest and highest standard. In addition, for a sample to be considered positive, the amplification curve had to be sigmoidal, and the standard error had to be smaller than the mean of the 2 replicates. The quantification of zoospore equivalents (i.e., 1 zoospore equivalent [ze] = 10 ITS copies) was calculated as the average of the replicates for that sample. An individual was considered Bd infected if both sample duplicates (one processed in Groningen, the other in Leipzig) were positive.

A total of 359 individuals (adult, juveniles, and tadpoles or larvae) from 8 different amphibian species were sampled in years 2020 and 2021 (110 collected in 2020 and 249 in 2021): Pelophylax lessonae (n₂₀₂₀ = 17, n₂₀₂₁ = 31), Pelophylax esculentus (n₂₀₂₀ = 1, n₂₀₂₁ = 4), Pelophylax ridibundus (n₂₀₂₀ = 4), unidentified juvenile/tadpole Pelophylax sp. (n₂₀₂₀ = 62, n₂₀₂₁ = 82), Rana temporaria (n₂₀₂₀ = 11, n₂₀₂₁ = 2), Bufo bufo (n₂₀₂₀ = 2, n₂₀₂₁ = 57), Epidalea calamita (n₂₀₂₀ = 1), Lissotriton vulgaris (n₂₀₂₁ = 73), and Triturus cristatus (n₂₀₂₀ = 12) (Suppl. Data, Table S1).

In total, 11 out of the 249 samples collected in 2021 and 2 out of the 110 samples collected in 2020 tested positive for Bd. Prevalence of Bd was estimated at 4.4% for 2021 and 0.02% for 2020. For both sampling periods, positive individuals were found in 4 out of 12 sites (∼33%) with 2 sites being infected in 2020 and 2 new sites being infected in 2021. Overall, for both years pooled together, across Bd-positive sites, prevalence was between 1% and 15%. There was an 18% prevalence of Bd in all anuran juvenile individuals captured (no Caudata juveniles were captured) and a 4% prevalence of Bd in all captured adults (Fig. 2; Tables I, S2). No Bd-infected tadpoles or newts were found. Average Bd load for positive samples ranged from 11.07 to 681.07 ze (Table II). Of the 13 Bd-positive samples, 8 samples came from unidentified Pelophylax juveniles.

This study provides the first documentation of the chytrid pathogen Bd in the province of Groningen in the Netherlands (Fig. 2). A previous survey, between March and September 2009, in this province, did not detect Bd (Spitzen-van der Sluijs et al., 2010, 2014). However, the small sample size of total individuals sampled in Groningen in that study (12 individuals across 3 sites; Spitzen-van der Sluijs et al., 2010) does not allow us to conclude that the pathogen was only recently introduced into the province of Groningen. Instead, and in line with what was suggested by Spitzen-van der Sluijs et al. (2010), it is more likely that the pathogen was already present, but not detected, due to very low prevalence and small sample size. In the present study, a large enough proportion of the population was sampled to detect Bd and conclude that Bd is present in the province of Groningen.

We encountered Bd infected individuals of the species P. lessonae, P. esculentus, and unidentified Pelophylax sp. juveniles. In contrast, sampled B. bufo, R. temporaria, P. ridibundus, E. calamita, L. vulgaris, and T. cristatus were not found to be infected in any of the sampling sites. The fact that no Caudata species (0 infected individuals out of 85 total Caudata individuals sampled) were found infected agrees with the overall trend for a higher tendency of anuran species getting infected with Bd rather than Caudata, as shown also in earlier surveys (Spitzen-van der Sluijs et al., 2014). In each of the sampling sites, only a very small percentage of the samples tested positive for Bd (1–15%), which is in line with the prevalence of Bd across the Netherlands and Belgium (4.8%; Spitzen-van der Sluijs et al., 2014). A large majority of the total infected individuals were juveniles (61.5%), showing a higher proportion of infected individuals in this life stage. This is not surprising as juvenile anuran amphibians have higher Bd-related mortality rates than adults (Russell et al., 2010), and in Europe chytridiomycosis has the tendency to be expressed in newly metamorphosed animals (Bakar et al., 2016). Interestingly, the average number of zoospores found in infected individuals was much higher in the province of Groningen (this study), in comparison to other areas in the Netherlands and Belgium (Martel et al., 2012; Spitzen-van der Sluijs et al., 2014). This pattern could be related to the fact that during our sampling period (17 C, May 2020 and 17–20 C, May–June 2021), temperatures were slightly lower than average temperatures during the sampling periods in Spitzen-van der Sluijs et al. (2014) (18–20 C, May–June 2009) (Royal Netherlands Meteorological Institute, 2019–2021), and therefore conducive to higher Bd zoospore production (Woodhams et al., 2011).

Previous reports indicate that the species found in this study, like P. esculentus and L. vulgaris, have a high tolerance against Bd and are disease-resistant (Cheatsazan et al., 2013; Ujszegi et al., 2021). Tolerant species, however, can act as reservoirs of the disease, both to sympatric species and to their habitat, and can further spread the pathogen or lead to new disease variants emerging. While increasing the sample size of disease screenings leads to higher confidence when asserting that a site is uninfected, the required sample size is only rarely met (DiGiacomo and Koepsell, 1986). Absolute certainty that a site is Bd-negative is impossible to achieve, and therefore it is safest to assume that all sites are potentially Bd-positive. We therefore recommend that all protection measures (e.g., disinfecting equipment, using gloves) are put into place. Additionally, strict disease mitigation practices are needed in infected areas, independently of the potential hosts’ susceptibility to the disease, to protect other areas, populations, and species, and to limit the possibility of new variants emerging.

In this study we report the presence of the Bd pathogen in Groningen for the first time. We observe differences in infection prevalence between species and between sampling years, highlighting the importance of regular screenings and continuous efforts to maintain basic hygiene procedures. Finally, since the Bd pathogen is now known to have spread over almost all of the Netherlands apart from Zuid-Holland and Zeeland (Spitzen-van der Sluijs et al., 2010), conservation efforts should focus on preventing introductions of new and more highly virulent disease strains that can hybridize and could shift the disease outcome and eventually lead to an increase in mortality rate or to a new variant with higher virulence emerging.

We thank Mees van der Donk and Henrieke te Brake for their assistance during fieldwork. We are grateful to Ellis Mulder for laboratory assistance. This project was supported by the British Ecological Society, London, U.K. (SR20/1224 to J.S.P.) and the Deutsche Forschungsgemeinschaft, Bonn, Germany (SA3786 to J.S.P.). All sampling permits were acquired under RAVON (FF/75A/2016/022).