Parotoid gland extract preparation
Adult cane toads (obtained in south-eastern Queensland, December 2018) were killed humanely using the cool/freeze method8 and stored at − 20 °C. Parotoid glands (54 g) excised from 23 thawed toads were macerated in H2O (250 mL) with a commercial blender, and filtered through a bed of Celite 545. The filtrate was concentrated in vacuo at 40 °C, and was partitioned into ethyl acetate (EtOAc) and H2O solubles. The EtOAc extract (750 mg) containing mostly bufagenins (Fig. 1a) was used in the attractant assay without further purification.
Analytical HPLC (298 nm) chromatogram of extracts obtained from (a) frozen parotoid gland, (b) eggs, (c) early-development tadpole, (d) late-development tadpole and (e) fresh parotoid secretion of cane toads, Rhinella marina [HPLC condition: Agilent Zorbax C8 column, 5 μm, 4.6 × 150 mm, 1 mL/min flow rate 15 min gradient elution from 90% H2O in MeCN, to 100% MeCN, with a constant 0.01% TFA in MeCN modifier]. Highlights: light blue = unspecified bufagenins (MW 400–432); light pink = unspecified bufolipins (MW 630–700); blue = bufagenins 1–5; red = bufotoxins 6–10; pink = bufolipin 11. Structures for 1–11 are shown in Fig. 2, and were assigned on the basis of spectroscopic analysis and comparisons with authentic standards.
Egg extract preparation
Cane toad eggs obtained from two laboratory-laid clutches (see method below, Northern Territory, October 2010) were stored at − 20 °C until extraction. Frozen eggs were freeze-dried to yield dry material (1.5 g) that was extracted overnight at room temperature with 90:10 MeOH:H2O (100 mL). The resulting solvent extract was concentrated in vacuo at 40 °C to give a crude material (251 mg) which was partitioned into EtOAc and H2O solubles. The EtOAc (160 mg) extract containing mostly bufagenins and bufolipins (Fig. 1b) was used in the attractant assay without further purification.
Early-development tadpole extract preparation
Early developmental stage cane toad tadpoles were collected live from the wild (Northern Territory, March 2010), and stored at − 18 °C until extraction. Frozen tadpoles were freeze-dried to yield dry material (2.6 g) that was extracted overnight at room temperature with 90:10 MeOH:H2O (100 mL). The resulting solvent extract was concentrated in vacuo at 40 °C to give a crude material (1062 mg), which was partitioned into n-BuOH and H2O solubles. The BuOH extract (582 mg) containing mostly bufagenins and bufolipins (Fig. 1c) was used in the attractant assay without further purification.
Late-development tadpole extract preparation
Mid to late developmental stage tadpoles were collected live from the wild (Northern Territory, December 2010), and stored at − 18 °C until extraction. Frozen tadpoles were freeze-dried to give dry material (13.4 g) that was extracted overnight at room temperature with 90:10 MeOH:H2O (100 mL). The resulting solvent extract was concentrated in vacuo at 40 °C to give a crude material (6899 mg), which was partitioned into n-BuOH and H2O solubles. The BuOH extract (3885 mg) containing mostly bufagenins (Fig. 1d) was used in the attractant assay without further purification.
Parotoid secretion extract preparation
Parotoid secretion was obtained from a live adult toad (Northern Territory, August 2011) by mechanical compression of the parotoid gland directly into MeOH, which following concentration in vacuo yielded a crude MeOH extract (26.2 mg). The crude MeOH extract containing mostly bufotoxins (Fig. 1e) was used in the attractant assay without further purification.
Pure compounds preparation
Marinobufagin (1), marinobufotoxin (6) and suberoyl-l-arginine (13) were obtained from our in-house pure compound library, and their purities were confirmed by LCMS, HRMS and NMR (see Supporting Information for 1H NMR spectra of the pure compounds). Plant cardenolides: digitoxigenin (14), ouabain (15) and digoxin (16) (Fig. 2) were purchased from Sigma Aldrich and were used in the attractant assay without further purification.
Compounds identified in different stages of cane toad (Rhinella marina) (1–13) and plant derived cardenolides (14–16).
Chemical analyses
Analytical HPLC was performed using an Agilent 1100 series module equipped with a diode array detector on an Agilent Zorbax Stable Bond C8 column (4.6 × 150 mm, 5 μm), 1 mL/min flow rate, 15 min gradient elution from 90% H2O in MeCN to 100% MeCN with a constant 0.01% TFA in MeCN modifier. All analytes were prepared in MeOH stock solutions (1 mg/mL) and an aliquot (10 μL) used for each analysis. HPLC chromatograms were monitored at 298 nm (the α-pyrone chromophore common to all bufadienolides). Compounds 1–12 (Fig. 2) present in the extracts were identified by LC-DAD-ESIMS and comparison with authentic standards (see Supporting Information Table S1). LC-DAD-ESIMS (Liquid Chromatography coupled to Diode Array Detector and Electrospray Ionization Mass Spectra) was acquired using an Agilent 1100 Series LC/MSD mass detector in both positive and negative modes using Agilent Zorbax Stable Bond C8 column (4.6 × 150 mm, 5 μm) with 1 mL/min flow rate, 15 min gradient elution from 90% H2O in MeCN to 100% MeCN with a constant 0.05% formic acid in MeCN modifier.
Bait preparations
Stock solutions of all attractant extracts were prepared in MeOH (20, 2.0 and 0.20 mg/mL concentrations), with a fixed volume (0.5 mL) of each loaded onto porous ceramic rings (Majestic Aquariums, Sydney, NSW) to give a series of loadings per ceramic ring (10, 1.0 and 0.1 mg) per attractant extract preparation. Stock solutions were also prepared for all pure compound attractants in MeOH (5.0 and 0.5 mM) with a fixed volume (0.5 mL) of each loaded onto porous ceramic rings to give a series of loadings per ceramic ring (2.5 and 0.25 µmoles) per attractant pure compound preparation (marinobufagin, 1.00 and 0.10 mg; digitoxigenin, 0.94 and 0.094 mg; marinobufotoxin, 1.78 and 0.178 mg; ouabain octahydrate, 1.82 and 0.182 mg; digoxin, 1.95 and 0.195 mg; suberoyl-l-arginine, 0.825 and 0.0825 mg per ceramic ring, respectively). Negative controls were ceramic rings loaded with MeOH (0.5 mL/ring) only. All impregnated ceramic rings were left in the fume-hood overnight at room temperature to allow the MeOH to evaporate, and to fix the attractants to the ceramic matrix.
Toad breeding
Adult toads were collected from the Adelaide River floodplain, near the city of Darwin in tropical Australia, and the animals were held in outdoor enclosures at The University of Sydney Tropical Ecology Research Facility at Middle Point, Northern Territory (12°34.73′S, 131°18.85″E). Breeding was induced by injection of the synthetic gonadotrophin leuprorelin acetate (Lucrin, Abbott Australasia). Females were injected with 0.75 mL doses of 0.25 mg/mL, while males were injected with doses of 0.25 mL4,9. Toads were injected just prior to sunset, and the pairs were placed in 70 L plastic tubs set on an angle with 8 L water. The following morning, eggs were collected and placed in 18 L tanks holding 9 L aerated water. When eggs developed into free-swimming tadpoles (Gosner10 stage 25), tadpoles were transferred to outdoor 750 L mesocosms located in a shaded area. Tadpoles were fed algae wafers (Kyorin, Japan) ad libitum daily, with 50% of the water in mesocosms changed every 3 days. Tadpoles (stage 30–39) were haphazardly selected from mesocosms for use in attraction trials as required.
Attraction trials
Attraction trials were conducted in a covered outdoor enclosure exposed to ambient temperature between 0930 and 1700 hours (maximum daily water temperature range over all trials: 26–32 °C). Each trial used plastic pools (1 m diameter) filled with 90 L of well water. Within each pool we placed two plastic traps (175 mm × 120 mm × 70 mm), each of which had a funnel (1 cm diameter) attached to one side. The traps were positioned in the centre of the pool 5 cm apart, with the funnels facing outward. Each pool was stocked with 50 tadpoles from a single clutch. Tadpoles were allowed to settle for 2 h, after which we randomly allocated treatments to traps (i.e., control or chemical). A single bait was added to each trap, and the number of tadpoles within each trap was counted hourly for 6 h. Water temperature was measured at hourly intervals using a hand-held thermometer.
Attraction responses to 26 combinations of chemical/concentration were tested, using a total of nine tadpole clutches. Each concentration of each chemical was tested using 4–7 tadpole clutches. The tadpole clutches used for each trial were chosen randomly, with the proviso that they had not been previously tested with the same chemical concentration. Individual tadpoles and baits were used only once in trials.
Statistical analysis
We analysed tadpole attraction as a binomial response (trap preference: chemical trap vs control trap) using logistic regression11 in R12, package MASS:glmmPQL). Models were based on the quasibinomial distribution to account for overdispersion of data, with Treatment (control vs. chemical) and Time (hourly intervals) as fixed effects. Random effects were accounted for by nesting trap within pool and responding tadpole clutch. We did not apply Bonferroni corrections to treatment p values due to the highly subjective nature of deciding when to apply such corrections13,14, see both papers for further problems with use of Bonferroni corrections). Rather, we provide unadjusted treatment p values in association with effect sizes (i.e., odds ratio of trap preference; this being a more meaningful indicator of biological significance) to interpret our attraction results13,14.
Ethics approval
This research was approved under permit 6033 from the University of Sydney Animal Care Committee. All methods were performed in accordance with the relevant guidelines and regulations, including ARRIVE guidelines.
Consent for publication
All authors agree to publication of this work.
Source: Ecology - nature.com
