The African clawed frog Xenopus laevis is invasive on four continents, and is recognized as one of the invasive amphibians that generates the greatest impacts in the ecosystems it invades. Although its diet has been studied in its native habitat and invaded areas, its trophic role is still unclear, especially in the communities it invades. We studied the diet of X. laevis, and looked at its stable isotope signatures and its bioaccumulation of heavy metals, to gain a better understanding of its trophic role. The diet was found to consist mainly of aquatic invertebrates, with some consumption of the native fish Cheirodon pisciculus. The isotope analysis revealed that the assimilation of prey by X. laevis is unrelated to the most-consumed item. Xenopus laevis occupied a high trophic position in its own stream and was segregated from fish in by its use of trophic resources. Despite its high trophic position, only biomagnification of copper and zinc was found in relation to some prey, but not manganese or arsenic.

Bioaccumulation, biomagnification, diet, ¹³C/¹²C, ¹⁵N/¹⁴N

Invasive herpetofauna can cause significant trophic disruptions in the ecosystems they invade, and predation is one of their main mechanisms, as it reduces the abundance of native species, causing population extinctions or the change in nutrient cycles (Kraus 2015). Understanding the trophic position of a species is important in such circumstances since the introduction of top predators can have major ecosystem impacts (Fritts and Rodda 1998). Organisms at intermediate trophic levels, like amphibians, can also have an effect through top-down or bottom-up mechanisms (Kraus 2015). For instance, on a global scale, freshwater fish invasions significantly disturb the trophic structure of invaded ecosystems. In this context, Sagouis et al. (2015), in a global evaluation, reported that in lotic systems, the main changes are driven by the introduction of non-native top predators at the top of the food chain, which can expand the total isotopic niche without altering the trophic niche of native species. Conversely, in lentic systems, changes are primarily associated with introducing mesopredators, which reduce the size of the trophic niche, the range of consumed resources, and the trophic niche breadth of the native fish community, thereby increasing direct and indirect competition.

The African clawed frog Xenopus laevis (Daudin, 1802) has invaded an extensive geographic area on several continents, with established populations in the USA (McCoid and Fritts 1980), France (Fouquet and Measey 2006), Italy (Lillo et al. 2011), Portugal (Rebelo et al. 2010), Japan (Kokuryo 2009), China (Wang et al. 2019), Mexico (Peralta-García et al. 2014), and Chile (Lobos and Measey 2002; Lobos and Jaksic 2005). Xenopus laevis has demonstrated great success in aquatic systems with a Mediterranean climate (Measey et al. 2012), except in Australia where the species has not yet invaded, despite ecological niche models which predict favorable environmental conditions for the species (Rödder et al. 2017). Records of X. laevis have been reported in Chile since the 1970s, with a considerable expansion in the country’s Mediterranean region (Mora et al. 2019). From a trophic perspective, X. laevis has been characterized mainly as a predator of zooplankton invertebrates in lentic environments and of zoobenthos in lotic environments in both its native and invasive populations (see Courant et al. 2017). The same review shows that predation on fish (USA, France, and Portugal) is generally low, similar to that reported by Lafferty and Page (1997) in California. Predation on amphibian eggs and larvae has been observed in South Africa, France, and Portugal. Additionally, in France, cannibalism accounts for 19% of the total predation on eggs (Courant et al. 2017), with lower rates reported in South Africa, Wales, Portugal, and the USA.

Although there are extensive data on the diet of X. laevis in the literature (e.g., Courant et al. 2017), its trophic role (trophic level, range of trophic resources used, nutrient assimilation, and niche overlap with other vertebrates) remains unclear. In this context, stable isotope analysis (SIA) is a tool to determine the trophic positions of organisms (e.g., Davis et al. 2012). Isotopic signatures of carbon (δ¹³C) have been used to identify sources of primary energy production of consumers, and nitrogen (δ¹⁵N) is used to estimate the trophic levels of the consumers (Baeta 2019). SIA studies permit the evaluation of trophic dynamics and help quantify the use of resources among native and invasive species (McCue et al. 2019). Estimating the isotope niche (breadth of δ¹³C and the δ¹⁵N) is widely used to represent the range of resources a consumer uses and the overlap between species’ ecological niches (Davis et al. 2012). In biological invasions, the use of SIA has made remarkable progress in recent years (McCue et al. 2019), enabling in-depth analyses of trophic relationships (Pujol-Buxó et al. 2018), effects on ecosystem functioning (Sagouis et al. 2015), competition among invasive species (Jackson et al. 2012), plasticity and adaptation (Pérez-Diz et al. 2023), and interactions among invasive species (Yelenik and D’Antonio 2013).

Heavy metals are typically found with minerals (such as carbonates and sulfates) and organic substances. They accumulate in aquatic sediments, which serves as their primary reservoir (Hua et al. 2016). Metals can remain stored for long periods, particularly in Mediterranean watercourses, where the climate is characterized by long periods of drought. These metals can bioaccumulate (reaching concentrations greater than in their environment) and/or biomagnify through trophic chains (Barra et al. 2021), generating lethal or sub-lethal effects on local populations (Shahjahan et al. 2022). Chile has important metal reserves and is one of the global producers of copper, which has been linked to surface water pollution issues (Copaja et al. 2016; Copaja et al. 2017; Ali et al. 2020; Barra et al. 2021). However, due to a volcanic history and general plate tectonics, Chilean soils are also naturally high in heavy metals (Oyarzún and Oyarzún 2011). The use of SIA combined with bioaccumulation studies may provide insight into how these heavy metals move through the trophic levels in relation to a species’ diet (Marambio-Alfaro et al. 2021).

Xenopus laevis has been categorized as an aquatic top predator due to its large size, generalist diet, and fully aquatic life history (Courant et al. 2017; Secondi and Raux 2020). If these assumptions are correct, X. laevis will occupy the highest trophic level (δ¹⁵N) in the food web of an aquatic community in a stream impacted by mining activities, a predominant land use across much of Chile. Furthermore, these mining activities are expected to result in metal bioaccumulation and/or biomagnification of heavy metals within the frog’s tissues. Consequently, this study aimed to investigate the trophic role of X. laevis by analyzing its diet, stable isotope signatures, and concentrations of heavy metals (Mn, As, Zn, and Cu) in a Chilean stream associated with mining activity.

Invasive species may affect ecosystem functioning by modifying the trophic structure and dynamics (Sagouis et al. 2015). Understanding the trophic roles of species is necessary to evaluate their impact. This is highly relevant for X. laevis given that it is the amphibian that exerts the second highest environmental and socioeconomic impact on the ecosystems it invades after Rhinella marina (Measey et al. 2016). In previous surveys, the presence of X. laevis was sporadic (3/15 occasions) in the El Cobre stream, likely due to the stream drying out during the driest periods. From the stomach contents investigated here, the diet of X. laevis was mainly composed of benthic invertebrates (Physa 44.8% of the RII, larvae of Coenagrionidae 19.06%) and nektonic invertebrates (Dytiscidae 17.16%). Although Chironomidae larvae were the most abundant in the Surber sampling, they were poorly represented in the diet (RII = 0.32%), where frogs preferred larger prey. Xenopus laevis shares its habitat with native and invasive fish species in the El Cobre stream; there was some predation (8.8% of RII) on the native and endangered C. pisciculus (Habit et al. 2006; Manosalva 2023). This indicates X. laevis functions as a generalist consumer, primarily feeding on aquatic invertebrates in its native and invasive habitats (Courant et al. 2017).

Xenopus laevis was positioned high in the foodweb (δ ¹⁵N = 11.8±1.2), along with the native and invasive fish. However, despite the presence of fish in the stomachs of X. laevis, the frogs occupied a similar trophic position to the fish. The fish present in the stream have been primarily described as consumers of aquatic invertebrates (see Aldunate and De la Hoz 1993 for C. pisciculus, Bonifacio et al. 2019 for C. decenmaculatus, Urzúa et al. 1977 for Basilichthys). The occasional presence of X. laevis in the stream (recorded on 3/15 occasions) may explain itslower trophic position relative to the fish, as frogs only consume fish opportunistically. During the dry and wet seasons, X. laevis can undertake terrestrial migrations over several kilometers, potentially dispersing from irrigation ponds and their associated canal networks (present in agricultural areas near the study site), which may act as reservoirs for their populations (Measey 2016; De Villiers and Measey 2017; Courant et al. 2019).

The δ¹⁵N values are higher than those reported for other native amphibians such as Eusophus insularis (δ ¹⁵N = 6.6±0.8, Lobos et al. 2022), Rhinoderma darwini (3.6±0.7, Molina-Burgos et al. 2018), Pleurodema thaul (7.2±1.03), and Rhinella arunco (7.3±0.08). These last ones are from the El Sauce stream. This is relevant because anurans are generally considered to have intermediate positions in trophic webs (Kraus 2015). Another important aspect was the total segregation in isotopic space (use of trophic resources) of all fishes and X. laevis (Fig. 2), despite sharing the same plant habitat on stream banks, with the exception of some small overlap between C. pisciculus and C. decenmaculatus. Isotopic analysis is a robust methodology to evaluate food assimilation (Davis et al. 2012). Interestingly, detritivores contributed the most to the diet of X. laevis (62.4%), followed by carnivores (36.5%), and finally herbivores (1.1%) (Fig. 3), even though a herbivore snail (Physa) was the main prey item found in frog stomachs (RII 44.8%). Although snails were the most consumed prey (and one of the most abundant in the Surber samples), they did not appear to contribute significantly to assimilation, perhaps due to their hard-to-digest calcareous shells.

The concentrations of heavy metals present in the stream sediments exceeded the international ISQG and PEL standards, aligning with significant contamination reported for some Chilean rivers such as the Aconcagua, attributed to industrial activities, with high levels of Cu, Cd, Zn, Hg, and As (Schalscha and Ahumada 1998). In our study, only the Ephemeroptera (Baetidae) showed high concentrations of all the metals analyzed. These insects were poorly represented in our study area, comprising only 0.48% of the total collected, and this is likely due to their requirement for oxygenated and pollutant-free waters (Figueroa et al. 2007). Consequently, these organisms are often used as bioindicators of water quality (Figueroa et al. 2007). In the rest of the community, only Zn tended to increase in concentration and trophic position. Although X. laevis and the fish have the highest trophic position in the foodweb, biomagnification was only found for Cu (compared to C. pisciculus) and Zn (compared to Physidae and Coenagrionidae). The biomagnification of a metal depends not only on its concentration, but also on its ecology, physiology, the complex structure of the trophic web of which it is part, and environmental factors (latitude, physicochemical factors of the water) (Córdoba-Tovar et al. 2022). In addition, metals that accumulate in an invader like X. laevis pose a new threat from this species, since it can mobilize these contaminants to other sites (due to its migratory capacity) and transfer them to its predators (such as herons and gulls in Chile). Baetidae was the only taxon that approached the threshold value of bioaccumulation. High concentrations of heavy metals in three species of anurans (eggs and larvae) were reported in industrialized areas of Austria, which increased as their development advanced (Grillitsch and Chovanec 1995). In laboratory tests, embryos of X. laevis proved to be sensitive to Cu (hatching success). At the same time, larvae showed lower survival after exposure to Zn, Cu, Pb, and Cd (Haywood et al. 2004). In Chile, few studies have assessed the exposure of aquatic vertebrates to heavy metals, and these have only been conducted on fish. These studies have successfully established the expression of certain genes in response to metal exposure, oxidative stress, and endocrine disruptors (Ali et al. 2020; Barra et al. 2021). Other studies have shown that not all metals have the same accumulation pattern, which depends on the species’ habits (pelagic vs. benthic) and the tissues analyzed (Copaja et al. 2016; Copaja et al. 2017). In this case, the high concentration of metals in sediments and the low levels of bioaccumulation in top predators may be related to the sequestration of metals in the sediment since their concentrations in the water samples were low. Indeed, tissue values for Cu and Zn were lower than those reported for two fish native to central Chile (Copaja et al. 2016). This is relevant for Mediterranean streams in geographic areas subjected to mega-droughts, which, in the current climate change scenario, experience major floods only in rainy years, when a significant amount of sediment is mobilized (Garreaud et al. 2017). In conclusion, due to the nature of Chilean soils and the economic importance of mining, it is imperative to continue studies to advance in evaluating the potential long-term effects of heavy metals.

We thank to the national fishery and aquaculture service of Chile for the capture permits and anonymous reviewers of this manuscript. CG Acknowledges the funding through ANID/FONDECYT/Regular 1210946 and ANID-PIA-INACH-ACT192057.