General and Comparative Endocrinology 177 (2012) 93–97
Contents lists available at SciVerse ScienceDirect
General and Comparative Endocrinology journal homepage: www.elsevier.com/locate/ygcen
Fecal glucocorticoid response to environmental stressors in green iguanas (Iguana iguana) Otto Kalliokoski b,⇑, Jeanette A. Timm a,c, Ida B. Ibsen a,d, Jann Hau b, Anne-Marie B. Frederiksen a,c, Mads F. Bertelsen a a
Center for Zoo and Wild Animal Health, Copenhagen Zoo, Copenhagen, Denmark Department of Experimental Medicine, Faculty of Health Sciences, University and University Hospital of Copenhagen, Copenhagen, Denmark Department of Small Animal Clinical Science, Faculty of Life Sciences, Copenhagen University, Copenhagen, Denmark d Department of Ecology and Evolution, Faculty of Biology, Copenhagen University, Copenhagen, Denmark b c
a r t i c l e
i n f o
Article history: Received 3 February 2012 Revised 23 February 2012 Accepted 25 February 2012 Available online 6 March 2012 Keywords: Corticosterone Stress Reptile Squamate
a b s t r a c t Quantification of glucocorticoid metabolites in feces has been shown to be a powerful tool in evaluating well-being in vertebrates. Little is known however about the hypothalamic–pituitary–adrenal axis response to stressors, and consequent glucocorticoid excretion, in reptiles. In a longitudinal study, fecal corticosterone metabolite (FCM) levels in green iguanas (Iguana iguana) were quantified during periods of rest and exposure to hypothesized stressors. FCM quantification was combined with behavioral analysis to further contextualize the measured increases. It was shown that both daily 5-minute handling/ restraint, as well as housing devoid of climbing opportunity, resulted in increased FCM excretion. Behavioral analysis suggested that the iguanas were chronically stressed by the lack of climbing opportunity, whereas handling may have induced only a transient stress response. The experimental design, using repeated periods of stressor-exposure, also revealed a facilitating effect, where the two stressors potentiated one another. Furthermore, the order of the two stressors was found to be important. The study provides insight into the functioning of the hormonal stress response in green iguanas, and to the refining of their housing and handling. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction In captivity, reptiles are typically housed in one of two main conditions; thematic semi-natural enclosures mimicking conditions in the wild for display purposes, or barren enclosures with minimal furniture for breeding, holding, or research. While it may be speculated that animals may be stressed under the latter conditions, the effects of housing on corticosterone levels are largely unknown. More importantly, the tools to evaluate the impact of housing or other potential chronic stressors are lacking. A number of studies have investigated the serum corticosterone response in reptiles, when faced with an acute stressor (e.g. [12,14,25]). Obtaining unbiased measures in this fashion is however problematic, as blood sampling in itself may constitute a stressful procedure; restraint and handling are known to generate a serum glucocorticoid response in reptilian species [2,16]. Blood samples can therefore not be used to effectively assess the impact of environment, where multiple samples over time would be needed. In other animal species,
⇑ Corresponding author. Address: Department of Experimental Medicine, 19 Blegdamsvej 3B, 2000 Copenhagen, Denmark. Fax: +45 35 32 73 99. E-mail address:
[email protected] (O. Kalliokoski). 0016-6480/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ygcen.2012.02.017
non-invasive options, such as measuring urinary or fecal output of corticosterone and corticosterone metabolites, have shown promise. After the first successful attempts of characterizing the hypothalamic–pituitary-adrenal (HPA) axis activity in wild and captive species non-invasively [15,24], the study of immunoreactive glucocorticoid metabolites in animal feces has been applied successfully in a number of species [20,21]. At present however information on the applicability in reptiles is lacking. To our knowledge, only two studies in turtles [1,17] – one of which presented inconclusive results – have been published. Specifically, little is known about the physiological stress response of squamates such as the green iguana, Iguana iguana. However, as in most vertebrates, a strong connection between circulating glucocorticoids and overall health has been established in iguanids [19]. The purpose of this study was to evaluate fecal corticosterone metabolite (FCM) levels in green iguanas during periods of rest and stress. Our hypothesis was that the iguanas would have significantly higher levels of FCM during periods of stress, than during periods of rest. As glucocorticoid excretion is not synonymous with a stress response [11], an ethogram assessing behavioral consequences of the manipulations was applied to the animals. Combining behavioral methods with FCM analysis allows for contextualizing the HPA axis activity.
94
O. Kalliokoski et al. / General and Comparative Endocrinology 177 (2012) 93–97
2. Material and methods
and acclimatization periods (n = 1440), periods of handling (n = 364), and periods of deprived climbing opportunity (n = 416).
2.1. Animals and husbandry 2.4. Obtaining and analyzing fecal samples Seven male and three female immature green iguanas were individually housed in steel wire cages (180 96 91 cm; H W L) with solid bottoms but no bedding material. Each cage contained one tree branch, with a water dish and a food tray, placed in the upper third of the cage. The temperature under a 400 W UV-lamp was maintained at 30 °C at the level of the tree branch and 24.5 °C at the level of the floor. Full spectrum lights as well as UV light were provided on an 11 h on, 13 h off cycle. Water was provided ad libitum, and the iguanas were fed 6 days a week with various fresh vegetables and fruit, occasionally mixed with mealworms or wax worms. The iguanas were naïve to handling prior to the study. 2.2. Experimental design Following an initial acclimatization period of 8 days, the animals were subjected to two experimental periods of 8 days, each followed by an equal-length resting period. The experimental periods consisted of a handling period and a period of denied climbing opportunity. During the handling period, the animals were manually restrained for 5 minutes each day for 8 days (one hand around the thorax and front legs, one hand around the hind legs and abdomen/tail base). During the 8 day period of deprived climbing opportunity, the lower parts of the cage walls were lined with smooth boards (to a height of 70 cm above ground level) preventing access to the elevated tree branch. The UV-lamps were lowered, now providing 30 °C at the center of the floor, and foodand water dishes were placed on the floor. The order of the two experimental periods was randomized for each animal. Due to limited cage space, the experiment was conducted in two consecutive 7-week periods where five iguanas were tested at a time. From a pilot study, and due to the long intestinal transit time and irregular defecation patterns of iguanas, the FCM response of a stressor was expected to extend into the subsequent resting period. Therefore the FCM excretion in response to a stressor was judged from feces obtained both during the stressor period and in the following resting period. Unlike with the behavior observations the resting periods could not be used for reference values; the stressor responses could only be compared to the acclimatization period. All procedures were performed in compliance with Danish legislation for experimental animals, and the protocol was reviewed and approved by the Copenhagen Zoo Research Committee. 2.3. Behavioral analysis To monitor the behavioral consequences of the stressors applied, the animals were observed for 1 h on alternate days. Observations were made after the cages had been cleaned and the iguanas had been provided fresh food and water. Cages were subdivided into five areas, and the position and activity of the iguanas were noted (see Table 1). The observations, made with 5 min intervals, were instant registrations and interval registrations carried out by a single observer. In total, each 8 day experimental period provided 52 instant registrations and 48 interval registrations for each subject. The first set of iguanas were monitored during the last 2 weeks of the experiment (one experimental period, and one resting period), whereas the second set were monitored for the duration of the experiment. As the 2 month observation period did not encompass the entire experiment, and due to each experimental period being followed by an equal length resting period the number of observations was inequally distributed between the rest
Fecal samples were collected daily from cages in the morning and kept in small plastic bags at 20 °C until analysis. A total of 258 fecal samples were obtained during the course of the experiment. Extraction and analysis of FCM was performed as previously described [22]. Briefly, fecal samples from each iguana were pooled, when possible, into 4-day periods, weighed with 0.1 mg accuracy, and mixed with 96% ethanol (5 ml/g feces, with a maximum of 10 ml to an amount of pooled feces). Due to the iguanas’ irregular defecation pattern, longer periods sometimes had to be resorted to (the longest time between two defecations was recorded at 16 days). Wherever overlap occurred, the excreted FCM was attributed the preceding period. The mixed feces and ethanol were then incubated at room-temperature for 24 h on a mini-gyro rocker and then centrifuged (20 min at 2000 g). The supernatant was collected and centrifuged again to remove particulates (15 min at 10,000 g). FCM was measured using a Corticosterone ELISA Kit (DRG Instruments, Marburg, Germany) according to the manufacturer’s instructions. The included standards were substituted for custom standards of corticosterone (Sigma–Aldrich, St-Louis, USA) prepared in 96% ethanol (1–50 ng/ml). Results are presented as nanograms of corticosterone equivalents, i.e. the amount of corticosterone needed to produce identical results. To rule out the possibility of matrix effects interfering with assay accuracy, parallelism was tested for using serial dilutions of a random set of samples. 2.5. Statistical testing Significant differences in behavioral observation data were tested for using the Kruskal–Wallis one-way analysis of variance. Data are presented as a chi-square with subscripted degrees of freedom and a Student’s p. Where significant differences were found, pair-wise testing using the Mann–Whitney U test was utilized for determining, post hoc, where the significant difference lied. Data are presented as a Studentized asymptotic p. Log-transformed data were tested for significant differences using analysis of variance with Tukey’s post hoc test. Treatment type and internal testing order were used as inter-individual factors and cases were weighted using the number of days encompassed by the individual measurement. The reasoning behind weighting the cases is that a measurement spanning 12 days is roughly twelve times as reflective of the overall HPA activity of a certain experimental period as a measurement spanning a single day. Several samples were taken from each individual violating the underlying assumption of independence. To resolve this issue and test whether data were biased as a result of this, individual identities were offered as explanatory categorical covariates (implemented using dummy variables) when testing the effect of type of treatment and order of treatments. Data are presented as a Fisher’s F with between and within groups degrees of freedom subscripted and a Student’s p. In addition, the effect of sex on FCM excretion was tested using a heteroscedastic t-test. Throughout, differences were considered significant for Student’s p-values less than 0.05. Statistical testing was carried out in PASW v. 18 (SPSS Inc., Chicago, USA). 3. Results 3.1. Behavioral analysis When comparing the periods of daily handling with the periods of rest, the iguanas did not seem to make use of their cage space
95
O. Kalliokoski et al. / General and Comparative Endocrinology 177 (2012) 93–97
Table 1 Ethogram of iguana behavior. Positions, postures and activities were recorded as instant registrations (n = 13 for every 1-hour session), whereas events were recorded as interval registrations (n = 12 for every 1-hour session). Category
Recorded
Definition
Position
On tree branch In water dish In feed tray On the cage walls On the floor Under the light
Posture
Standing Lying flat Movement (partial) Movement (full body) Inactive Displaying dewlap Eye(s) shut Eating
The iguana The iguana The iguana The iguana The iguana The iguana branch) The iguana The iguana The iguana The iguana
Activity
Events
The The The The
was was was was was was
observed observed observed observed observed observed
having having having having having having
its front legs on the tree branch its front legs in the water dish its front legs in the feed tray its front legs on the cage walls at least three legs on the floor more than 50% of its body under the UV light (this was possible only on the floor or on the tree
was was was was
observed observed observed observed
standing on its front legs, the abdomen clear of the surface lying flat, not using its legs to support the upper body moving parts of its body moving all of its body
iguana did not move in the observed period dewlap is displayed to where at least 50% is visible eyes – or eye closest to the observer, if both were not visible – were closed for a minimum of 30 s in the observed period iguana was observed with its head in the feed tray, chewing food
differently or display altered behavior (no significant differences were found; Fig. 1). The iguanas spent on average 90% of their time on the elevated tree branch, whereof 60% was spent basking under the UV light. When climbing was made impossible there was an expected shift as the favored position in the cage was now off limits. In the experimental periods, subtle changes in behavior were recorded. A shift towards increased vigilance could be seen when climbing was withheld (Fig. 2), as the iguanas, in the time span when observed, were more likely to be found standing than lying down (Standing posture – Kruskal–Wallis: v22 = 6.8, p < 0.05; Mann–Whitney: p < 0.05 over reference period). In addition, the iguanas were never found with their eyes closed when observed in these periods (Kruskal–Wallis: v22 = 9.1, p < 0.05; Mann– Whitney: p < 0.01 compared to reference period).
F1,317 = 8.9, p < 0.005). Essentially, an additive effect of treatments was found – i.e. iguanas would have a higher level of HPA axis activity in the second experimental period than in the first. Furthermore iguanas challenged first with handling, and then denial of high ground, would be more stressed in the end than iguanas challenged in the opposite order (Fig. 4). Since the reference period was always the first period, these measures had to be excluded for the internal order testing. The significant effects of both the type of treatment and the time dependency persisted when checking for biased data by allowing for an inter-individual effect using individual covariates (Type – F2,405 = 130, p < 0.001; Order – F1,309 = 29, p < 0.001). In addition, a significant effect (t415 = 2.7, p < 0.01) of sex was found, with the males, on average, excreting more FCM (95% CI: 2.6–340 ng/day) than the females (95% CI: 3.6–130 ng/ day).
3.2. FCM analysis 4. Discussion FCM analyses confirmed the initial hypothesis; significantly higher levels of corticosterone metabolites were found in droppings collected during the experimental periods than during the baseline establishing period (Fig. 3, F2,414 = 99, p < 0.001). FCM were found to conform well to, and all of the statistical testing was conducted on log-transformed data. Parallelism of a random selection of serially diluted samples to the standard curve of the assay was confirmed (data not shown). Post hoc testing revealed no significant differences in excreted FCM between the two types of experimental treatment. In addition to a significant effect of treatments, a time dependent effect and an interaction term was identified (Order – F1,317 = 21, p < 0.001; Order Type –
In f eed tray; 4%
On tree branc h; 31%
A.
In water dish; 3%
On t he f loo r; 2%
In f eed tray; 8%
In water dish; 2%
Much like previous studies in other vertebrate species, this study demonstrates that it is possible to accurately quantify corticosterone metabolites in green iguana feces, opening new avenues for non-invasive stress research in this species and probably other squamates. During the handling period, all the iguanas had significantly higher levels of FCM, compared to the period of acclimatization. FCM were found to conform well to the log-normal distribution, which has previously also been found to be the case in many mammalian and avian species (e.g. [9,10]). All The elevated FCM levels are in accordance with previous studies of plasma corticosterone
In f eed tray; 1%
On the f loor; 4%
On tree branch, under light; 39% On tree branch, under light; 59%
On the f loor, under light; 46%
In water dish; 3%
On cage walls;
