Research communication — Navorsingsberig
The measurement of glucocorticoid concentrations in the serum and faeces of captive African elephants (Loxodonta africana) after ACTH stimulation
S K Steada, D G A Meltzera and R Palmeb
INTRODUCTION In 1998, 30 juvenile elephants were
captured in Botswana and transported to training facilities at African Game Services (AGS) ( 25°47’ S, 27°46’E) in South Africa. The move ignited a heated debate between animal rights and conservation organisations concerning the welfare of the so-called ‘Tuli Elephants’. On a broader scale, the psychological well- being of elephants maintained in zoos and circuses was highlighted, and the need to optimise husbandry conditions for behaviour, health and well-being was reiterated. As Garaï7 noted, it is reason- able to assume that juvenile elephants separated from their families, captured and translocated have experienced a
certain degree of stress. Stress is a subjec- tive experience, and thus the extent to which individuals are ‘stressed’ is difficult to quantify5.
Although there is no universal scientific agreement on the definition of stress, stress responses cause an increase in glucocorticoids, primarily cortisol and corticosterone, in the blood9. Conven- tionally, the assessment of adrenal re- sponses to stress relies on collection of blood samples and measurement of corticosteroids10. However, the process of blood collection is impossible without the use of capture drugs when studying free-ranging wild animals and will, in itself, elicit elevated cortisol levels16.
Metabolism of glucocorticoids occurs primarily in the liver3. There are large inter-species differences with respect to the metabolites formed and their route of excretion1,11–13,18, therefore the efficacy of measuring faecal cortisol metabolites should be evaluated for each species.
Palme and Möstl13 have established an 11-oxoaetiocholanolone enzyme im- munoassay (EIA) that measures 11,17-dioxoandrostanes (11,17-DOA), a group of faecal cortisol metabolites. The biological relevance of this method has been proven in ruminants following ACTH stimulation of cortisol release by the adrenal cortex15 and used to monitor transport stress in cattle14. This non- invasive technique has been applied to a number of domestic, zoo and wildlife species1,11,18,19.
In elephants, cortisol has been mea- sured in the saliva from 2 Asian elephants4
and in urine from 1 African and 1 Asian elephant2 as a means of assessing adreno- cortical activity in a non-invasive manner. However, the collection of faecal samples is more practical, especially when dealing with free-ranging animals, and provides measurements that are independent of short-term fluctuations2,15. The aim of our study was to validate a method for measuring glucocorticoid metabolites in elephant faeces, and to conduct a prelimi- nary investigation into the method’s biological relevance.
MATERIALS AND METHODS
Animals Twenty elephants at AGS made up a
large proportion of the animals studied. Of these, 14 animals, referred to as Group 1, were being trained by Indonesian mahouts and were kept in a relatively large enclosure of approximately 2500 m2. The remaining 6 animals, Group 2, were kept some distance away and out of sight and sound from the main hub of activity centred around the training process. Two of these animals were in an enclosure of 150 m2 and 4 in an enclosure 350 m2 in extent. Both groups had been in their respective enclosures for 12–14 months when the study was undertaken. The elephants were of a similar age, estimated between 5 and 7 years.
Three further elephants, Group 3, which were kept on the farm of the Glen Afric Lodge, Broederstroom (25°49’S, 27°51E), approximately 10 km distant from the AGS premises, were included in
192 0038-2809 Tydskr.S.Afr.vet.Ver. (2000) 71(3): 192–196
aWildlife Unit, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa.
bInstitute of Biochemistry and Ludwig Boltzmann Institute of Veterinary Endocrinology, University of Veterinary Medicine, Veterinärplatz 1, A-1210 Vienna, Austria.
Received: February 2000. Accepted: August 2000.
ABSTRACT Conventionally, the assessment of adrenal responses to stress relies on blood sample collec- tion. However, blood collection from animals is impossible without restraint or immobilisa- tion that influences results. This study was undertaken to validate recently established enzyme immunoassays that measure faecal glucocorticoid metabolites in elephants, and to perform a preliminary investigation into the biological relevance of this non-invasive method for use in assessing the degree of stress in this species. Four juvenile African elephants were injected i.m. with 2.15 mg synthetic adrenocorticotrophic hormone (Synacthén, Novartis, Switzerland). Blood and faecal samples were collected over 4 h and 7 d respectively. Concentrations of serum cortisol and faecal cortisol metabolites were determined using immunoassay. Variability of basal and peak values in blood and faeces was observed among the elephants. After ACTH injection, serum cortisol concentrations increased by 400–700 %. An 11-oxoaetiocholanolone enzyme immunoassay (EIA) proved best suited to measure cortisol metabolites (11,17-dioxoandrostanes) when compared to a cortisol and corticosterone EIA in faecal samples. Concentrations of faecal 11,17-dioxoandrostanes increased by 570–1070 %, reaching peak levels after 20.0–25.5 h. Greater levels of glucocorticoid metabolites were measured in faecal samples from elephants kept in small enclosures compared to levels in the faeces of animals ranging over a larger area. The results of this preliminary study suggest that non-invasive faecal monitor- ing of glucocorticoid metabolites is useful in investigating adrenal activity in African elephants.
Key words: ACTH, animal welfare, cortisol, EIA, elephant, faeces, glucocorticoids, non-invasive.
Stead S K, Meltzer D G A, Palme R The measurement of glucocorticoid concentrations in the serum and faeces of captive African elephants (Loxodonta africana) after ACTH stimu- lation. Journal of the South African Veterinary Association (2000) 71(3): 192–196 (En.). Wildlife Unit, Faculty of Veterinary Science, University of Pretoria, Private Bag X04, Onderstepoort, 0110 South Africa.
the study. These animals had been reared from an early age on the farm, were habit- uated to humans and roamed about in a 750 ha enclosure during the day accom- panied by a game guard. One animal, a male, was approximately 10 years old and the 2 females were 4 and 6 years of age.
None of the elephants studied were pregnant.
Management of the elephants Husbandry of all the AGS elephants,
Groups 1 and 2, was the responsibility of AGS under the supervision of the National Council of the Society for the Prevention of Cruelty to Animals. The Group 1 ele- phants were tethered by a front and a back leg in a barn overnight (between 17:00 and 10:00) and provided with fresh fruit, vegetables, Eragrostis curvula, tef or oat hay, lucerne and bedding. During the day they were released into their enclosure and had free access to a similar variety of feed.
The elephants in Group 2 remained free in their respective enclosures. They had little contact with humans, who only attended to them when providing feed and when the enclosures were cleaned.
Group 3 elephants spent the day wan- dering about in the 750 ha enclosure feed- ing as they pleased and were tethered by 1 leg in a barn overnight between 16:00 and 07:00. They were given 2–3 kg horse cubes, lucerne and bedding during the evening.
ACTH administration and sample collection
Only 4 elephants were available for the purposes of this experiment. As a result, an experimental design in which control animals would have been given an injec- tion of saline solution instead of ACTH could not be used.
Each elephant was injected intramuscu- larly with 300 mg azaperone. After 15 min, an 18-gauge catheter was inserted into an ear vein and a blood sample col- lected using a 10 m syringe. Two further blood samples were collected at 15 min intervals before the intramuscular admin- istration of 2.15 mg ACTH (Synacthén, Novartis, Switzerland). Thereafter, a venous blood sample was collected every 30 min for 4 h.
Faecal samples from almost all defaeca- tions were collected for 3 days before and 4 days after the ACTH injection.
Blood samples Five-millilitre samples of whole blood
were placed in plain vacutainer tubes (Becton Dickenson, USA). Blood was allowed to clot for 1 h and centrifuged at 1700 × g. Serum was placed in cryotubes
(Amersham, Johannesburg) and stored at –20 °C until analysis.
Measurement of serum cortisol Serum concentrations of cortisol were
determined using a Clinical Assays™ GammaCoat™ Cortisol 1 2 5 I Radio- immunoassay kit (DiaSorin; SA Scientific, Johannesburg).
Faecal samples Samples were collected within 30 min of
defaecation. A single faecal bolus was mixed by hand and then a handful of faeces was placed into a plastic freezer bag and stored at –20 °C until the prepara- tion for extraction and EIA analysis.
Faecal samples were collected from each group of elephants.
Analysis of faecal cortisol Frozen faecal samples were oven-dried
at 100 °C. Each sample was powdered and mixed thoroughly. A 0.5 g subsample was mixed with 10 m 80 % ethanol, shaken for 30 min and centrifuged at 1700 × g for 15 min. One millilitre of the supernatant was drawn off and stored at –20 °C until EIA analysis. Aliquots of the extract were analysed with 3 EIA systems (cortisol, corticosterone and 11-oxoaetiocholano- lone) as described by Palme and Möstl13.
High-performance liquid chromatography (HPLC)
HPLC of the faecal metabolites was per- formed at the Institute of Biochemistry, Vienna, as described by Teskey-Gerstl
et al.18. Faecal extracts containing peak 11,17-DOA concentrations were sub- jected to a clean-up procedure (Sep-Pak C18). Separation was performed on a reverse-phase Nova-Pak C18 column (3.9 × 150 mm, Millipore Corporation, Milford, Massachusetts, USA) using a linear gradient starting at 50 % methanol. Three fractions per minute were collected, dried under a stream of nitrogen, and reconstituted in assay buffer. Immuno- reactive glucocorticoid metabolites were quantified with the cortisol, corticos- terone and 11-oxoaetiocholanolone EIAs as previously described13.
RESULTS
HPLC analysis HPLC separations revealed a number of
immunoreactive substances present in elephant faeces. They showed a chro- matographic mobility between cortisol and 17,20-dihydroxyprogesterone (Fig. 1). The main metabolite determined with the 11-oxoaetiocholanolone-EIA eluted around cortisol. Lower amounts of immunoreactive substances were detected by testing the HPLC fractions with the corticosterone-EIA and negligi- ble amounts with the cortisol-EIA (detec- tion limit = 2 nmol/kg faeces).
ACTH challenge Injection of ACTH resulted in an in-
crease of serum cortisol (Fig. 2) and faecal cortisol metabolite concentrations (Fig. 3). Serum cortisol levels began to rise after
0038-2809 Jl S.Afr.vet.Ass. (2000) 71(3): 192–196 193
Fig 1: HPLC separation of immunoreactive glucocorticoid metabolites in 1 faecal sample from an elephant as tested in a cortisol-, corticosterone- and 11-oxoaetiocholanolone-EIA. Fractions marked with represent the approximate elution time of respective standards (17a,20aP = 17a,20a-dihydroxy-progesterone).
injection with azaperone and insertion of catheters. Following ACTH administration serum cortisol increased between 4- and 7-fold, reaching highest recorded values (526–652 nmol/ ) after 2 h. No distinct peaks were observed.
Individual differences in basal and peak values of faecal cortisol metabolites were observed. Basal values of faecal 11,17- DOA and corticosterone equivalents ranged from 21 to 168 nmol/kg (median: 48 nmol/kg) and 33 to 133 nmol/kg
(median: 50 nmol/kg) respectively. ACTH-induced peaks were between 572–1104 % (11,17-DOA) and 160–353 % (corticosterone) higher than basal values. These peak concentrations occurred 20–25.5 h after the injection. Additional
194 0038-2809 Tydskr.S.Afr.vet.Ver. (2000) 71(3): 192–196
Fig. 2: Time course of serum cortisol concentrations (nmol/ ) in 4 elephants before and after intramuscular injection of 2.15 mg ACTH at time zero. Individual elephants were identified as SW, BB, TZ and SH.
Fig. 3: Time course of concentrations of faecal 11,17-dioxoandrostanes and corticosterone equivalents (nmol/kg) in 4 elephants before and after intramuscular injection of 2.15 mg ACTH at time zero. Individual elephants were identified as SW, BB, TZ and SH.
peaks of varied height were observed for both groups of metabolites before and after the ACTH-induced peaks.
Faecal glucocorticoids in elephant groups
The results of faecal glucocorticoid analy- ses are summarised in Table 1.
DISCUSSION There have been few studies to investi-
gate the possibility of using non-invasive methods to assess adrenocortical activity in elephants2,4. Methods to identify and measure faecal glucocorticoid metabo- lites have been successfully used in a number of domestic livestock11,13,15 and some wild species1,7,18,19. The aim of this study was to assess whether the measure- ment of faecal cortisol metabolites is more suitable than using blood cortisol values to monitor adrenocortical activity in elephants.
Serum cortisol levels began to rise be- fore ACTH was injected, suggesting that handling or the injection of azaperone affected levels of serum cortisol within 30 min. These findings are in accordance with Sire et al.17 and Fulkerson and Jamison,6 who reported that physical restraint and blood-sampling can pro- duce an increase of blood cortisol levels within 15 min of handling. The lowest values recorded before ACTH injection (73–131 nmol/ ) were similar to those found by Hattingh et al.8 in plasma from 5 undisturbed adult female elephants that had been shot (mean = 111 nmol/ ). How- ever, it must be noted that their results are associated with a high standard deviation (24.8). The results are not directly compa- rable to those in this study owing to the elephants’ age, unknown reproductive status and undefined environmental stressors at the time of sample collection. Morton et al.10 collected blood samples from 27 elephants immobilised with etor- phine/xylazine. A mean plasma cortisol concentration of 347± 0.95 nmol/ was es- timated. It is likely that these values were a result of capture procedures and are not representative of baseline cortisol levels.
After ACTH injection, serum cortisol levels continued to rise, and in 3 of 4 cases reached a plateau. No further blood sam- ples were taken, as it was regarded as un-
ethical to collect blood for longer than 4 hours. The tranquillising effect of azaperone began to wear off and it became increasingly difficult to prevent the elephants from pulling the indwelling catheters out of the ear veins. Sampling from other sites, such as the tail, had been attempted previously and proved unsuc- cessful.
The highest recorded cortisol concen- trations were within the range of those found in 5 elephants that had been herded for 6–20 minutes and darted with succinyldicholine before sample collec- tion (mean: 688 nmol/ , SD: 269.5)8. This gives an indication of the type of stressor that may produce such elevated levels.
A number of glucocorticoid metabolites were detected in elephant faeces using HPLC analysis. Although the exact iden- tity of the metabolites was not deter- mined due to the cross-reactions of the EIA, a group of them may be collectively described as 11,17-dioxoandrostanes (11,17-DOA)13,15. As reported in domestic livestock11,13, negligible amounts of cortisol and low amounts of cortico- sterone were found in elephant faeces. These findings support the suggestion that the recently-developed 11-oxo- aetiocholanolone EIA, which measures 11,17-DOA, is the most suitable EIA to use for the non-invasive monitoring of adrenocortical activity in elephants.
As observed in other species, the time course of faecal cortisol metabolite con- centrations reflected the ACTH-induced stimulation of glucocorticoid produc- tion11,15. Concentrations peaked 20–25.5 h after the ACTH injection. Similar times were found by Möstl et al.11 in ponies. It has been suggested that the delay in faecal glucocorticoid excretion is correlated with the transit time of digesta from the duodenum to the rectum12. Our findings fit well with the total passage time from mouth to rectum of 33 h reported for Indian elephants20. Differences in diet and individual adaptations in hepatic or gastrointestinal function may explain differences in excretion rates21.
The large increase of 11,17 DOA (572–1104 % above basal levels) after ACTH injection was higher than that observed in ponies (200–660 %) by Möstl et al.11, but within the range of reported
increases in cattle during transport (400–1100 %)14. Lower percentage in- creases in corticosterone were measured, supporting the earlier suggestion that 11,17-DOA is a more suitable group of me- tabolites to measure.
Additional peaks after the ACTH- induced peak could be due to an entero- hepatic circulation of the metabolites12. Alternatively, additional peaks may have been caused by stressful events approxi- mately 24 h before they were recorded. More prolonged periods of behavioural observations conducted before, during and after the trial may have made it possi- ble to identify events that had caused these responses.
As reported in other species11,13,18, indi- vidual variation in basal and peak values was observed. This may be due to differ- ences in previous experiences, body mass, metabolism, age, diet or sex. Further investigations with greater numbers of animals are necessary to identify the influence of these confounding factors. Preliminary investigations into the application of the technique to assess welfare showed good correlat ion between behavioural observations, envi- ronmental stressors and faecal gluco- corticoid metabolite concentrations. Concentrations of 11,17-DOA from elephants allowed to range in 750 ha fell within the lower end of the range of basal values measured in Group 1. These Glen Afric elephants were exposed to fewer stressors than those housed at AGS and had more opportunity to perform species-specific behavioural activities such as foraging in a large area. The elephants kept in the small enclosures at AGS, Group 2 animals, had 11,17-DOA concentrations 450 % higher than the elephants at Glen Afric.
The primary advantages of faecal sample collection are that the collector does not require special skills and there is no need to handle the animals. Secondly, concen- trations of faecal glucocorticoid metabo- lites probably more closely reflect the amounts of cortisol produced and ex- creted than cortisol measured in blood, which only reflects a point in time during a dynamic process of absorption, metabo- lism and excretion15. We conclude that measuring faecal 11,17-DOA is a valuable
0038-2809 Jl S.Afr.vet.Ass. (2000) 71(3): 192–196 195
Table 1: The effect of enclosure size on faecal 11,17-dioxoandrostane concentration.
Group Number of elephants Enclosure size Range of 11,17-DOA Median Number of samples (m2) (nmol/kg) analysed
1 14 2 500 21–168 48 42
2 6 350 62–1000 176 6
3 3 7 500 000 15–47 39 6
196 0038-2809 Tydskr.S.Afr.vet.Ver. (2000) 71(3): 192–196
tool for non-invasive monitoring of adrenocortical activity in African ele- phants. This could help to optimise the capture, transport and husbandry of Afrian elephants and be useful in investi- gating stress in free-ranging situations.
ACKNOWLEDGEMENTS We thank Mrs A Kuchar-Schulz, Mrs M
Stark, Mrs A Human and Mrs M Mulders for excellent assistance in the laboratory and Dr C Speedy for her assistance in the field. The financial support of World Wildlife Fund and the Taeuber Manage- ment Trust Pty (Ltd) is gratefully ac- knowledged. The study was conducted with the permission of Mr R Giazza (Afri- can Game Services), Mr J Brooker (Glen Afric Country Lodge) and the National Council of the Society for the Prevention of Cruelty to Animals.
REFERENCES 1. Bahr N I, Palme R, Möhle U, Hodges J K,
Heistermann M 2000 Comparative aspects of the metabolism and excretion of cortisol in three individual nonhuman primates. General and Comparative Endocrinology 117: 427–438
2. Brown R L, Wemmer C M, Lehnhardt J 1995 Urinary cortisol analysis for monitoring adrenal activity in elephants. Zoo Biology 14: 533–542
3. Brownie A C 1992 The metabolism of adre- nal cortical steroids. In James V H (ed.) The adrenal gland. Raven Press, New York: 209–224
4. Dathe H H, Kuckelkorn B, Minnemann D 1992 Salivary cortisol assessment for stress detection in the Asian elephant. Zoo Biology 11: 285–289
5. Dawkins M 1980 Animal suffering. The science of animal welfare. Chapman & Hall, London
6. Fulkerson W J, Jamison P A 1982 Pattern of cortisol release in sheep following adminis- tration of synthetic ACTH or imposition of stressor agents. Australian Journal of Biologi- cal Sciences 35: 215–222
7. Goymann W, Möstl E, Van’t Hof T, East M, Hofer H 1999 Noninvasive fecal monitoring of glucocorticoids in spotted hyenas (Crocuta crocuta). General and Comparative Endocrinology 114: 340–348
8. Hattingh J, Wright P G, De Vos V, McNairn I S, Canhao M F, Silove M, Wolverson G, Cornelius S T 1984 Blood composition in culled elephants and buffaloes. Journal of the South African Veterinary Association 55: 157–180
9. Hofer H, East M L 1998 Biological conserva- tion and stress. In Moeller A P, Millinsky M, Slater P L (eds) Stress and behaviour: advances in the study of animal behaviour. Academic Press, San Diego: 405–426
10. Morton D J, Anderson E, Foggin C M, Kock M D, Tiran E P 1995 Plasma cortisol as an in- dicator of stress due to capture and trans- location in wildlife species. The Veterinary Record 136: 60–63
11. Möstl E, Meßmann S, Bagu E, Robia C, Palme R 1999 Measurement of gluco- corticoid metabolite concentrations in faeces of domestic livestock. Journal of Veter- inary Medicine 46: 621–632
12. Palme R, Fischer P, Schildorfer H, Ismail M N 1996 Excretion of infused 14C-steroid hormones via faeces and urine in domestic livestock. Animal Reproduction Science 43: 43–63
13. Palme R, Möstl E 1997 Measurement of cortisol metabolites in faeces of sheep as a parameter of cortisol concentration in blood. International Journal of Mammalian Biology 62 (Suppl. II): 192–19
14. Palme R, Robia C, Baumgartner W, Möstl E 2000 Transport stress in cattle as reflected by an increase in faecal cortisol metabolites. The Veterinary Record 146: 108–109
15. Palme R, Robia C, Meßmann S, Hofer H, Möstl E 1999 Measurement of faecal cortisol metabolites in ruminants: a non-invasive parameter of adrenocortical function. Wiener Tierärztliche Monatsschrift 86: 237–241
16. Reinhardt V, Cowley D, Eisele S G, Scheffler J 1991 Avoiding undue cortisol responses to venipuncture in adult rhesus macaques. Animal Technology 42: 83–86
17. Sire J E, Blom A, Sjaatad O V, Ropstad E, Bergdahl Nilsen T A 1995 The effect of blood sampling on plasma cortisol in female reindeer (Rangifer tarandus tarandus). Acta Veterinaria Scandinavia 36: 583–587
18. Teskey-Gerstl A, Bamberg E, Steineck T, Palme R 2000 Excretion of corticosteroids in urine and faeces of hares. Journal of Compar- ative Physiology B 170: 163–168
19. Wallner B, Möstl E, Dittami J, Prossinger H 1999 Fecal glucocorticoids document stress in female Barbary macaques. General and Comparative Endocrinology 113: 80–86
20. Warner A C 1981 Rate of passage of digesta through the gut of mammals and birds. Commonwealth Bureau of Nutrition Abstracts and Reviews, Series B 51: 809–818
21. Wasser S, Thomas R, Nair P, Guidry C, Southers J, Lucas J W D E, Monfort S L 1993 Effects of dietary fibre on faecal steroid measurements in baboons (Pa