|
nif:isString
|
-
This experiment was done according to Swiss law, with ethical approval given by the relevant authority, i.e. the Swiss Animal Welfare Agency (permit numbers 30/2012 and 233/2014). No other permit was needed for handling the meerkats. This study did not involve any endangered or protected species. All details on sampling methods are given below.
We collected fecal samples from male and female adult meerkats (age range: 11–58 months) housed at the University of Zurich in an outdoor-indoor enclosure (250m2). During the study, the colony was exposed to natural fluctuations in photoperiod and natural outdoor temperatures, although heat lamps are available both indoors and outdoors for extra warmth. Water was provided ad libitum and individuals were fed three times daily, with a mix of fruit, vegetables, mealworms, crickets, hard-boiled eggs and dead chicks. The captive colony originally started with four adults (1 female and 3 males) in 2011 and increased in size to 17 individuals throughout the time of the study. Information regarding the sexes, dates of birth and manipulation participation of all study animals is presented in Table 1.
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0153161.t001 Individual ID, rank, sex and date of birth of all subjects used in the ACTH challenge tests I and II as treated individuals or as water-injected controls (water).F = female, M = male, ACTH I: April 2012, ACTH 2: May 2013. We conducted ACTH challenge tests in April 2012 (4 males, 1 female) and May 2013 (5 males, 5 females, Table 1). To establish basal FGCM levels we collected 1–3 fecal samples per day from each individual 5–9 days prior to the pharmacological challenge (pre-treatment levels). On the day of injection, we distracted the subjects by spreading mealworms on the ground and injected intramuscularly a single dose (5 IU per kg body weight) of synthetic ACTH (Synacthen Depot ampule 1mg/ml, Novartis Pharma Schweiz AG, Bern) on the thigh [52]. The focal individuals were well habituated to close human presence, which allowed ACTH administration without capture or restrain. On the day of injection and the three days thereafter, we attempted to collect three samples per day and individual. Subsequently, we collected two samples per day and study subject for up to two weeks after ACTH administration. Concurrently with the second ACTH challenge in 2013, we injected four males with distilled water (0.5ml/kg body weight) to assess the effect of the injection procedure on glucocorticoid production.
Natural stressor: subordinate male eviction: To assess FGCM alterations in response to biologically relevant stressors, we opportunistically collected samples from 10 adult individuals (six males and four females) around the time of intense attack on one group member, which culminated in its permanent eviction from the group and relocation to a local zoo. In total, we collected 12 samples (from five individuals) prior to the observed aggressive interactions and 12 and 32 samples (from 10 individuals) within 48 h and 5 weeks after the observed attack, respectively. In addition, we collected fecal samples from the attacked male before and during the time when he was attacked.
In total, we analyzed 406 samples, which were collected while the animals remained in their group. Before each meal we fed each individual a food item covered with a unique combination of different colored indigestible glitter or chrome dioxide [60] and food dye to enable fecal sample identification and reliable assignment of samples to individuals. We inspected the enclosure for new feces every couple of hours and, if we did not find any samples from a particular individual, before the next meal, we gave it another food item with its respective glitter/food dye combination to improve our chances of collecting the aimed number of samples per individual per day. Upon collection, samples were placed into labeled plastic bags and stored at -20°C within 15 min. The material remained at -20°C until shipment (on dry ice) to the Endocrinology Laboratory at the German Primate Center for FGCM analysis.
Hormone extraction and accounting for sample soil contamination: We processed and extracted all fecal samples following Heistermann et al. [61]. Briefly, samples were lyophilized for 72 h, pulverized, and 0.10–0.12 g of the fecal powder were extracted in 3 ml of 80% aqueous methanol by vortexing the suspension for 15 min. The applied protocol is strongly recommended by Palme and colleagues [53] who have shown that 80% aqueous methanol proved best suited for extraction of naturally occurring glucocorticoid (and other steroid hormone) metabolites from feces of virtually all mammalian species tested so far, yielding high recoveries (e.g. 80–90%) and precision [62–65]. Following extraction, we centrifuged the suspension, recovered the supernatant and stored it at -20°C until FGCM analysis [66]. Since feces were often deposited onto sandy ground, most samples were covered in a layer of soil. To minimize this “contamination” and thus aid in reducing non-biological variance in FGCM levels [67], we manually rubbed off the soil as best as possible before the samples were pulverized. We assessed the potential impact of the remaining sample “contamination” with soil on overall FGCM excretion patterns by determining in the samples from 2013 (n = 176 samples from 10 ACTH challenges, see “ACTH challenge” in Methods section) the amount of soil and subsequently comparing between fecal hormone concentrations (for both the CCST and 11β-hydroxyetiocholanolone EIA assays) uncorrected and corrected for soil content, i.e. expressed per fecal mass without soil (see [67]). Following centrifugation of the fecal suspensions for extraction (see above) and drying the fecal matter under a fume cabinet for >14 days, we separated the organic fecal material from the soil and determined a soil-free fecal dry weight for each sample. We then used this weight to calculate hormone concentrations per pure organic fecal mass.
Fecal extracts resulting from the ACTH challenge in 2012 were measured for immunoreactive FGCMs using three different enzyme immunoassays (EIAs) detecting corticosterone (CCST) [68] and cortisol metabolites with a 5β-3α-ol-11-one structure (11oxo-etiocholanolone, [63] and 5β-3α,11β-diol-structure (11β-hydroxyetiocholanolone, [69]. The CCST EIA used the same antibody that had been used in the CCST RIA system validated for stress hormone assessment in meerkats by Young et al. [52]. Detailed information on antibody characteristics, standards, and hormone labels as well as on other assay details, e.g., data on assay sensitivities, are given in Heistermann et al. [68]. Fecal extracts resulting from the ACTH challenge II and water injections from 2013 were analyzed with the CCST and 11β-hydroxyetiocholanolone EIA only. This decision was based on the results from 2012 (see Table 2), which showed that 1) the values generated by the 11β-hydroxyetiocholanolone and 11oxo-etiocholanolone assays correlated strongly with each other for each of the 5 tested individuals (all r = 0.71–0.98; p < 0.0001) thus providing very similar response patterns to adrenocortical stimulation, 2) time lags of responses were identical for the two assays, but 3) peak increase in levels in response to the challenge was, on average, twice as large (13.6 vs. 6.1 fold) for 11β-hydroxyetiocholanolone than for 11oxo-etiocholanolone, indicating a higher sensitivity of the former to stressful events. The remaining fecal extracts from the subordinate male eviction event (see “Natural stressor: subordinate male eviction” in Methods section) were assayed only in the 11β-hydroxyetiocholanolone EIA, because the data from the two ACTH challenge tests as well as the control water injection tests indicated that this assay is more sensitive than the CCST assay for monitoring FGCM patterns in meerkats (see “ACTH challenge” in Results section).
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0153161.t002 Fecal glucocorticoid concentrations in response to ACTH administration in individual meerkats measured using a 11β-hydroxyetiocholanolone, a corticosterone (CCST), and a 11β-oxo-etiocholanolone assay. aF = female, M = male,bpre-treatment levels (mean) in μg/g feces (see Methods),cpeak levels in response to ACTH administration in μg/g,dx-fold increase of peak levels above mean pre-treatment concentrations,elag-time in hours between administration of ACTH and peak CG response.
We performed all EIAs on microtiter plates according to the procedure described in detail by Heistermann et al. [68, 70]. Prior to steroid measurement, we diluted fecal extracts 1:10–1:1000 (depending on concentration and assay) in assay buffer. All samples were run in duplicate and samples with a coefficient of variation >7% between duplicates were re-measured. Serial dilutions of fecal extracts gave displacement curves parallel to those obtained with the respective standard in each assay. Intra- and inter-assay coefficients of variation of high- and low-value quality controls for each assay were <10% and <15%, respectively. All final hormone concentrations are given as μg/g fecal dry weight.
To characterize the immunoreactive metabolites measured by the CCST and 11β-hydroxy-etiocholanolone assays, we performed reverse-phase high-pressure liquid chromatography (RP-HPLC) using the procedure previously described in detail by Möhle et al. [71] and Heistermann et al. [68]. We performed HPLC on a male and a female sample, each representing peak FGCM response to the ACTH challenge, to evaluate possible sex differences in the characteristics of excreted cortisol metabolites (c.f. [39, 72]). HPLC also enabled us to assess whether certain fecal androgen metabolites, which could potentially be detected by antibodies raised against cortisol metabolites [31, 68, 69], were co-measured by the two EIAs. We measured each HPLC fraction in the two aforementioned assays to generate the profiles of immunoreactivity.
We generated composite (average) profiles of FGCM response to ACTH for the CCST and 11β-hydroxyetiocholanolone assay by calculating the percentage change in FGCM levels (relative to the mean pre-treatment baseline levels; set as 100%) across 8–12 h intervals following ACTH injection for each individual and averaging values across all individuals. We excluded M3, who clearly did not respond to the ACTH stimulation (see Table 2), from this and all other analyses. We determined the time course of FGCM excretion as the lag time from the administration of ACTH (or water) to peak FGCM response, separately for each individual and assay. To investigate for a possible sex effect in FGCM response to the ACTH challenge, we compared the magnitude of peak FGCM elevation between sexes using a Mann-Whitney U-test. To examine the effects of sex, age and time of day (morning (AM): 8-11am, midday (MD): 11:30-2pm, afternoon (PM): 3-6pm) on basal FGCM levels, we used all fecal samples collected during pre-ACTH challenge control periods (i.e. samples reflecting baseline FGCM levels), numbering 128 samples from 13 individuals. In this analysis, values were log-transformed to conform to assumptions of linearity, and linear mixed effects models (NLME package version 3.1–118 in R) were used to account for ACTH challenge (I and II) and individual identity as random factors, because two individuals were tested in both challenges (F1 and M4) and because we had multiple samples per individual which were not evenly distributed among the three day time periods. The effect of the subordinate eviction on FGCM levels was assessed in multiple steps to maximize the information from our available data, due to uneven sample sizes between time periods. We used a Friedman test to assess overall individual changes in FGCM levels. Pre and post-attack “control” levels, and during- and post-attack attack levels were compared using Wilcoxon signed-rank-tests. We determined the percentage of soil content of a sample by dividing the mass of soil per sample by the total mass of extracted fecal material. We assessed the correspondence between FGCM levels expressed per mass extracted feces and those expressed per “soil-free” fecal mass (see “Hormone extraction and accounting for sample soil contamination” in Methods section) by calculating Spearman rank correlations for each individual and separately for the CCST and 11β-hydroxyetiocholanolone assays. To examine whether the two assays differed in their degree of correspondence, we compared their correlation coefficients using the Wilcoxon signed-rank test. To further evaluate the influence of soil contamination on characteristics of FGCM pattern, we compared the magnitude of responses to ACTH (peak to baseline ratio) between uncorrected and “soil-corrected” FGCM profiles as well as the variability in pre-treatment FGCM baseline levels (calculated as coefficient of variation, CV) under the two conditions. Both comparisons were carried out using Wilcoxon signed-rank tests. All statistical tests, carried out in R Studio Version 0.98.1102 and SPSS (IBM SPSS Statistics for Macintosh, version 22.0), were two-tailed and the statistical significance level was set at 0.05.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}