Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
F1000Res
2025 Jan 09;14:53. doi: 10.12688/f1000research.157055.1.
Show Gene links
Show Anatomy links
Validating the underpinnings of water corticosterone measurement for aquatic amphibians.
Smith TE
,
Holmes AM
,
Emmans CJ
,
Coleman R
,
Hosie CA
.
???displayArticle.abstract???
BACKGROUND: Good animal welfare is important ethically but also to ensure animals provide valid scientific models. Despite thousands of amphibians in research laboratories there is minimal quantitative evidence pertaining to their management and welfare. This study validated methods to non-invasively measure corticosterone, the amphibian 'stress' hormone, from tank water to provide a robust and reliable welfare assessment tool.
METHODS: We report experiments (A) that evaluate parameters linked to the performance of our biochemical extraction methods for waterborne corticosterone and, importantly, associated sampling procedures. We evaluate appropriate sampling water type, sampling vessel, filtration methods, potential degradation of waterborne corticosterone over time and the impact of sampling procedures on animal corticosterone levels. We wanted to determine sampling parameters that yielded the least background corticosterone and had minimum negative impact on the animals. The second series of experiments (B) evaluated parameters linked to the biology of Xenopus, including the influence of circadian rhythm, sex and snout-vent length on waterborne corticosterone levels, since fundamental knowledge of a species' biology is essential for designing robust experiments and in the interpretation of the results.
RESULTS: We propose collecting corticosterone samples in deionised water in either plastic or glass containers. The filtering process does not impact the amount of corticosterone measured in the water sample. Levels of corticosterone collected in the water change over a 48-hr period so we advocate standardising time from hormone collection to storage at - 20 °C. Repeated transfer of frogs to sampling containers does not increase corticosterone, suggesting our methods are not cumulatively stressful. Corticosterone levels were not impacted by circadian phase, sex or snout-vent length.
CONCLUSION: We have developed and validated robust methods to quantify waterborne corticosterone. We hope they provide a template for researchers wishing to develop methods to measure waterborne corticosterone in aquatic amphibians.
Figure 1. . Background corticosterone levels in five water types.Scatterplot showing corticosterone (CORT) levels (pg/sample measured in 1 L water that was held in a tank for 1 hr) measured in five different types of water. The horizontal bar represents the median CORT pg per water type. Significant variations in log transformed CORT levels across the water types were determined using a between subject one factor ANOVA [F (4, 15) = 14.36, P < 0.0001]. Water types showing significant differences (P < 0.05) are denoted by *.
Figure 2. . Background corticosterone levels in water sampled in plastic and glass tanks.Scatterplot showing corticosterone (CORT) levels (pg/sample measured in 1 L water held in a plastic or glass vessels for 24 hrs). The horizontal bar represents the median CORT per vessel. An independent t test on log transformed CORT values revealed levels of background CORT did not vary across vessel type (t = 0.216, df 10, P = 0.800).
Figure 3. . Corticosterone levels in a sample held at ambient temperature change over 48 hours.Scatterplot showing corticosterone (CORT) levels (pg/per sample measured in sequential 500 ml water aliquots collected at six time points from a 5 L sample that had housed a frog for 48 hrs). The horizontal line represents the median CORT value. A mixed-effects ANOVA determined that log transformed waterborne CORT levels changed significantly across the 6 time periods [F (5, 35) = 3.872 P = 0.007]. Time periods showing significant differences (P < 0.05) are denoted by * and were determined using Tukey Honest Significant Difference test corrected for multiple comparisons.
Figure 4. . Corticosterone does not change with repeated transfer to a new tank over 4 hours.Scatterplot showing corticosterone (CORT) levels (pg/sample measured in 1 L water that had held a frog for 30 mins) from frogs experiencing eight repeated handling and transfer procedures to a new container containing 1 L of water. Frogs experienced one transfer and handling procedure every 30 minutes for a 4 hour period. The horizontal line represents the median CORT concentration at each 30 minute time period. A mixed effects ANOVA determined that concentrations of log transformed CORT did not vary significantly across the 8, transfer and handling procedures [F (7, 44) = 0.517, P = 0.816].
Figure 5. . Corticosterone does not change with repeated transfer to a new tank over 8 hours.Scatterplot showing corticosterone (CORT) levels (pg/sample measured in 1 L water that had held a frog for 1 hr) from frogs experiencing 8 repeated handling and transfer procedures to a new container containing 1 L water. Each frog experienced one handling and transfer procedure every 60 minutes for an 8 hour period and remained in each tank for a 1 hr period. The horizontal line represents the median CORT concentration at each 60 minute time period. A mixed effects ANOVA on log transformed CORT concentrations determined that CORT did not vary significantly across the 8 handling and transfer procedures [F (7, 48) = 1.875 P = 0.100].
Figure 6. . Corticosterone levels are comparable between males and females and during the day and night periods.Scatterplot showing individual corticosterone (CORT) (pg/sample measured from 1 L water taken from a 5 L volume which had held the frog for 12 hrs) levels for 10 females and 5 males measured over a 12 hour light and a 12 hour dark period. The horizontal line represents the median CORT value. A mixed 2-factor ANOVA determined there were no significant differences in cortisol levels between the sexes [F (1, 13) = 0.712, P = 0.414], or across the light and dark phases [F (1, 13) = 0.154, P = 0.701] or interaction between the two variables [F (1, 13) = 0.712, P = 0.414].
Figure 7. . No relationship between snout-vent length and corticosterone levels.Scatterplot showing individual snout-vent length plotted against the individual’s corresponding corticosterone (CORT) levels (pg/sample measured from 1 L water taken from a 5 L volume which had held the frog for 48 hrs). Data are shown for 10 females and 5 males. Spearman correlations determined there was no relationship between snout-vent length and CORT for either males (r = -0.154, P = 0.833) or females (r = -0.042, P = 0.900).
