Glucocorticoids, thyroid hormones and iodothyronine. deiodinases in embryonic saltwater crocodiles

Transcription

1 AJP-Regu Articles in PresS. Published on June 20, 2002 as DOI /ajpregu Glucocorticoids, thyroid hormones and iodothyronine deiodinases in embryonic saltwater crocodiles Caroline A. Shepherdley 1, Christopher B. Daniels 2, Sandra Orgeig 2, Samantha J. Richardson 3, Barbara K. Evans 1 and Veerle M. Darras 4 1 Department of Zoology, University of Melbourne, 3010, Victoria, Australia 2 Department of Environmental Biology, Adelaide University, 5005, South Australia, Australia 3 Department of Biochemistry and Molecular Biology, University of Melbourne, 3010, Victoria, Australia 4 Laboratory of Comparative Endocrinology, Zoological Institute, K.U. Leuven, B-3000 Leuven, Belgium Running head: Embryonic crocodile adrenal and thyroid interactions Address for correspondence and reprints: Caroline A Shepherdley Department of Biochemistry and Molecular Biology University of Melbourne, 3010 Victoria Australia Tel: , Fax: sjrich@unimelb.edu.au Copyright 2002 by the American Physiological Society.

2 1 Abstract We investigated the relationship between glucocorticoids, thyroid hormones and outer ring and inner ring deiodinases (ORD and IRD) during embryonic development in the saltwater crocodile (Crocodylus porosus). We treated the embryos with the synthetic glucocorticoid dexamethasone (DEX), triiodothyronine (T 3 ) and a combination of these two hormones (DEX + T 3 ). The effects of these treatments were specific in different tissues and at different stages of development, and also brought about changes in plasma concentrations of free thyroid hormones and corticosterone. Administration of DEX to crocodile eggs resulted in a decrease in T 4 ORD activities in liver and kidney microsomes, and a decrease in the high-km rt 3 ORD activity in kidney microsomes, on day 60 of incubation. DEX treatment increased the T 4 ORD activity in liver microsomes, but not kidney microsomes, on day 75 of incubation. DEX administration decreased T 3 IRD activity in liver microsomes; however, this decrease did not change plasma free T 3 concentrations, which suggests that free thyroid hormone levels are likely to be tightly regulated during development. Keywords: reptile, development, outer ring deiodination, inner ring deiodination, liver, kidney

3 2 Introduction Thyroid hormones and glucocorticoids have important roles during development. Both thyroid hormone and glucocorticoid concentrations in plasma increase around the time of birth in mammals (16, 48), hatching in birds, (18, 41), metamorphosis in anurans (23, 40), (46), early in fish development (6, 7, 17, 25) and during smoulting in salmonid fishes (39, 38). During early development, thyroid hormones are of maternal origin. In eutherian mammals, they are provided by placental transfer (45), and in the case of egg laying vertebrates, are deposited in the egg yolk (20). Later in development, embryos synthesise and release thyroid hormones from their own developed thyroid gland. Glucocorticoids of maternal origin are also received across the placenta by eutherian embryos (33). Glucocorticoids are synthesised and released from the developed adrenal gland late in development. Another important event in development is the onset of the regulation of these hormones through the hypothalamopituitary-adrenal (HPA) axis or the hypothalamo-pituitary-thyroid (HPT) axis. The peak in plasma thyroid hormones during the final stages of development is characterized by an increase in 3, 3, 5, 5 tetraiodothyronine (T 4 ), mediated by the HPT axis, and a concurrent increase in 3, 3, 5-triiodothyronine (T 3 ), which is largely the result of an increase in deiodination of T 4 to T 3 and a decrease in T 3 degradation. Deiodination is the removal of iodine from iodothyronine molecules. Outer ring deiodination converts T 4 into T 3, and reverse T 3 (3, 3, 5 -triiodothyronine or rt 3 ) to 3, 3 -diiodothyronine (T 2 ). Inner ring deiodination converts T 4 to rt 3 and T 3 to T 2. T 3 is responsible for most of the biological effects of thyroid hormones and is the ultimate ligand for the thyroid hormone receptor (31). Therefore, the conversion of T 4 to T 3 is an important step in the regulation of thyroid hormone bioactivity. Outer ring deiodinase (ORD) activities are detectable already during embryonic stages and have a general increase in activity towards the end of development (mammals: 51, 19, 30, 13, 28, 1; birds: 2, 5; amphibians: 12, 10). In contrast to ORD activities, inner ring

4 3 deiodinase (IRD) activity is highest during early developmental stages but generally declines towards the end of embryonic development (mammals: 29, 28; birds: 2, 5, 43; amphibians: 12). During development, glucocorticoids affect deiodination. In the last days prior to spontaneous labour in the sheep, there is a coincident rise in plasma cortisol and T 3 in the fetal sheep. Cortisol administration increases T 4 to T 3 converting activity in the liver of the fetal lamb (51) and therefore may contribute to the increase in type I ORD and increase in T 3 plasma levels at the end of gestation. In cultured fetal mouse liver cells, the addition of hydrocortisone induced type I ORD activity (32). In the embryonic chicken, endogenous corticosterone levels have been observed to rise late in development (34), around which time there is also an increase in the conversion of T 4 to T 3 (8). Administration of glucocorticoids (dexamethasone (DEX) or corticosterone) or adrenocorticotropic hormone (ACTH) to embryonic chickens increased plasma T 3, decreased rt 3 concentrations, and decreased, or did not alter, plasma T 4 concentrations (9, 4). These treatments also increased conversion of T 4 to T 3 in the liver (9, 15, 4) and decreased hepatic type III deiodinase (4). Interaction between corticosterone and thyroid hormones has also been demonstrated during amphibian metamorphosis. Corticosteroids increased the net production of T 3 by stimulating ORD activity (T 3 production) and inhibiting IRD activity (T 3 degradation) in pre-metamorphic tadpoles. This increased the whole body concentration of T 3 in the tadpoles and lead to premature metamorphosis (11). It was therefore postulated that glucocorticoids have a role in increasing the activity of ORD and decreasing the activity of IRD enzymes during development of the saltwater crocodile. This study had three aims: (i) to measure corticosterone and free thyroid hormones in the blood plasma of the saltwater crocodile embryo, (ii) to investigate the control of thyroid hormones and glucocorticoids by the HPT and HPA axis, respectively, and (iii) to

5 4 investigate the relationship between glucocorticoids, thyroid hormones and deiodinases in the embryonic crocodile. These aims were tested by treating the embryos with the synthetic glucocorticoid, DEX, T 3, and a combination of these two hormones (DEX + T 3 ). We hypothesized that DEX treatment would decrease endogenous corticosterone concentrations through interaction with the HPA axis, and that T 3 treatment would decrease endogenous free T 4 concentrations through interaction with the HPT axis. We also hypothesized that DEX treatment would increase T 4 ORD and rt 3 ORD activities in liver and kidney microsomes, and decrease T 3 IRD activity in liver microsomes. As deiodinase activities in the liver and kidney are responsible for contributing to circulating concentrations of thyroid hormones, we hypothesized that changes in deiodinase activities in these tissues in response to glucocorticoid treatment would decrease plasma free T 4 concentrations, and increase plasma free T 3 concentrations. Materials and methods Incubation of eggs. Crocodile eggs were purchased from a commercial crocodile farm, Crocodylus Park, Darwin, Northern Territory, Australia. Eggs were from two clutches and were incubated at Crocodylus Park for 40 days, then transferred to Adelaide University, South Australia. Upon arrival, eggs were weighed and placed in plastic storage boxes filled with moistened vermiculite substrate. Eggs from both clutches were placed in each box, half buried in substrate, 1 to 2 cm apart. Box lids were left slightly ajar to allow air to circulate. The storage boxes were placed in an incubator set at 32 C. Eggs were sprayed every 2 to 3 days with MilliQ water to maintain moisture levels. At day 40 of incubation the mean weight of eggs in clutch 1 was ± 0.62 g and the mean weight of eggs in clutch 2 was ± 0.47 g.

6 5 Experimental design. The relationship between glucocorticoids, thyroid hormones and deiodinases in embryonic saltwater crocodiles was investigated by injection into the eggs of solutions containing T 3 (Sigma Chemicals, Sydney, Australia), DEX (Sigma Chemicals, Sydney, Australia), DEX + T 3, and control injections consisting mostly of isotonic saline. The doses of these treatments given to saltwater crocodile embryos were based on DEX and T 3 treatment of snapping turtle eggs (Chelydra serpentina) (26) and chicken eggs (9, 50, 4) and scaled up according to differences in body masses of embryos. Therefore, in crocodile embryos, the treatments consisted of: two 50 µl injections of 5 µg T 3 (dissolved in 0.1 M NaOH and diluted 1:50 with 0.15 M NaCl, ph 7.5); two 50 µl injections of 50 µg DEX (dissolved in 0.15 M NaCl, ph 7.5), and two 50 µl injections of 50 µg DEX + 5 µg T 3 combined (each prepared as above). Control injections were two 50 µl injections of 0.1 M NaOH diluted 1:50 with 0.15 M NaCl, ph 7.5. The four treatment groups consisted of six eggs per group. Eggs were randomly assigned to treatment groups, the only stipulation being that each group contained 4 eggs from clutch 2 and two eggs from clutch 1. This was to account for possible clutch variation in embryo size and response to treatments. Injection of treatments took place 48 and 24 hours before embryos were euthenased for organ collection. The effects of the treatments were studied at three time points: animals were euthenased on day 60 (75% incubation), day 68 (85% incubation) and day 75 (94% incubation). One group of six eggs was allowed to develop through to one day after hatching and did not receive any hormone treatments. Injection protocol. Eggs were removed from the incubator and the top surface was swabbed with ethanol. A small hole was drilled in the shell above the embryo using a dental drill and 50 µl of treatment solution was injected into the egg through the hole. The hole was sealed by pipetting a drop of candle wax over the hole. Eggs were returned to the incubator. Twenty four hours later, the wax was removed, another 50 µl of hormone solution was injected. The

7 6 hole was resealed and the eggs were returned to the incubator. At day 58, it was necessary to remove approximately 100 µl of egg fluid before the injection of 50 µl of treatment. This was rarely necessary at subsequent injection times, as the volume of egg albumen decreased. On the day following the second injection, the eggs were opened with scissors and the embryos were removed from the shell. Embryos were injected intraperitoneally with 0.2 ml Lethabarb anaesthetic (Arnolds of Reading, Sydney, Australia) and blood was immediately sampled from the post occipital venous sinus using a heparinized syringe. Liver and kidney tissues were collected, weighed, frozen in liquid nitrogen and stored at -80 C until used for the preparation of microsomes. Blood was stored at 4 C until centrifugation at 850 xg for 5 min. Plasma was removed and stored at -20 C. Corticosterone radioimmunoassay. The free fraction of corticosterone is probably more physiologically important than the total concentration, as defined by the free hormone hypothesis (22). However, an assay for the measurement of free corticosterone was unavailable. Therefore, total corticosterone concentrations were measured in this study. Corticosterone was extracted from blood plasma (100 µl) with 1 ml AR grade ethanol by mixing for 1 min on a vortex mixer. Tubes were centrifuged for 5 min at 1400 xg to separate the precipitate from the supernatant. 200 µl of these sample extracts and 50 µl of corticosterone standards in the range of 0.5 to 16 ng/ml, were transferred into duplicate assay tubes. Samples and standards were evaporated under a stream of air. Radioactive tracer (100 µl) consisting of approximately 6000 cpm [1,2,6,7-3 H] corticosterone, (Amersham, Australia) and corticosterone antibody (100 µl) (Endocrine Sciences, USA) were added. Tubes were briefly vortexed and allowed to equilibrate at 4 C overnight. Unbound steroid was removed by addition of 500 µl of cold (4 C) charcoal (Norit A), 5 mg/ml, and dextran (T-70), 0.5 mg/ml. After briefly vortexing, the mixture was kept on ice for 15 min followed by centrifugation at 4 C, 15 min at 1400 xg. For each of the tubes, 150 µl of the supernatant was

8 7 transferred to scintillation tubes and 2.5 ml Aqueous Counting Scintillant (Amersham, Australia) was added. To measure the total amount of radioactivity, 100 µl of tracer was added to 2.5 ml scintillant. The radioactivity in each tube was counted using a scintillation counter (Packard, TriCarb 2100TR) for 5 min. Non-specific binding of the tracer was on average 0.63% and maximum binding of the antibody was 30%. The sensitivity of the assay was 0.1 ng/ml and recovery of tritiated corticosterone after extraction was 92%. Free T 4 and free T 3 radioimmunoassays. There were insufficient volumes of plasma samples to measure both total and free thyroid hormones. Measurement of free T 3 and free T 4 during development should represent the physiologically relevant fraction of thyroid hormones, as defined by the free hormone hypothesis (22). Free thyroid hormone radioimmunoassays were carried out using Coat-a-Count kits (Diagnostic Products Corporation, USA). It is a solid phase radioimmumoassay for the measurement of non-protein bound T 4 and T 3 levels in human plasma. These assay kits were validated for use with crocodile blood plasma. Charcoal (Sigma, Australia) was used to strip the plasma of endogenous thyroid hormones. Charcoal-stripped crocodile plasma was mixed with T 3 or T 4 standards and assayed to determine the accuracy of the assays by comparing linear relationships between expected and observed free T 3 and free T 4 concentrations. The recovery of standards from stripped plasma was 93% (free T 3 ) and 103% (free T 4 ). These results were compared to the standard curve, showing good parallelism. In both assays the sample volumes suggested by the manufacturer (100 µl for free T 3, and 50 µl for free T 4 ) were not large enough to easily measure free T 3 and free T 4 in crocodile plasma. Presumably, this was because the levels of free thyroid hormones in crocodile plasma are much lower than in human plasma. Therefore, the volume of saltwater crocodile plasma sample used in the assays was increased. This was also done in parallel with human serum samples as a control. Increasing the volume of human serum and crocodile plasma to 200 µl in both assays still gave accurate measurements of free T 3 and free

9 8 T 4, showing good parallelism to the standard curve. Results obtained with crocodile plasma were corrected for the use of a larger sample volume. Except for this modification of the crocodile plasma sample volume the assays were conducted as per the manufacturer s instructions. Standards in the range of 1 to 100 pg/ml for the free T 4 assay (50 µl) and in the range of 0.5 to 42 pg/ml for the free T 3 assay (100 µl), and crocodile plasma samples, were added to duplicate tubes. Occasionally only single sample determinations were made due to insufficient sample volume. One milliliter of tracer ([ 125 I T 4 ] corresponding to approximately cpm or [ 125 I T 3 ] corresponding to approximately cpm) was added, the tubes vortexed briefly and incubated at 37 o C for 1 hour for the free T 4 assay, and 3 hours for the free T 3 assay. The tubes were decanted thoroughly and the radioactivity remaining was counted in a gamma counter (Wallac, 1270 Rackgamma II) for ten minutes. For both assays, maximum binding of the antibody was 50%. For the free T 4 assay, the sensitivity of the assay was 0.1 pg/ml and the non-specific binding of the tracer was 1.6%. For the free T 3 assay, the sensitivity of the assay was 0.2 pg/ml and the non-specific binding of the tracer was 1.3%. Deiodinase assays. The characteristics of deiodinases in saltwater crocodile tissues have been recently investigated (36). Due to the small amounts of embryonic liver and kidney able to be collected, the deiodinase assay conditions in this study were based on the optimal deiodinase assay conditions found for juvenile crocodile tissue microsomal preparations. Microsome preparation and protein determination. Microsomal fractions were prepared as previously described (36). Total protein concentrations of the microsomal fractions were determined by a Bio-Rad protein assay as previously described (36). Radioiodinated iodothyronines. Radioactive iodothyronines were prepared in the laboratory as previously described (36). Measurement of high-km ORD activity using rt 3 as substrate and low-km ORD activity using T 4 as substrate. Assays for the measurement of high-km rt 3 ORD and low-km T 4 ORD were

10 9 carried out as previously described (36). Each sample was tested in duplicate and all assays included blanks for measurement of non-enzymatic degradation of tracer. The incubation temperature was chosen based on the recent study of the characterisation of deiodinases in tissues of the juvenile saltwater crocodile (36). The specific reaction conditions for the rt 3 ORD activity assays were: the substrate concentration was nm [3,5-125 I] rt 3 (corresponding to cpm) + 5 µm rt 3 with 10 mm DTT. Incubation was for 30 min, at 32 C. The final protein concentrations in the assays were 0.1 mg protein/ml for liver microsomes and 0.4 mg protein/ml for kidney microsomes. The specific reaction conditions for the T 4 ORD activity assays were: the substrate concentration was nm [3,5-125 I] T 4 (corresponding to cpm) + 1 nm T 4. Liver microsomes were incubated with 10 nm T 3 (to reduce interference of T 3 IRD activity in liver microsomes) and 15 mm DTT for 120 min at 32 C. Kidney microsomes were incubated with 20 mm DTT for 120 min at 25 C. All microsomal samples were diluted to a final protein concentration of 1 mg/ml. A second substrate mixture was prepared containing a tracer ( nm [3,5-125 I] T 4 ) and a high substrate concentration of 100 nm T 4 and incubated with liver and kidney microsomal samples. This was to test for the possibility of both low- Km and high-km enzymes being present in each tissue. Measurement of IRD activity using T 3 as substrate. Assays for the measurement of T 3 IRD were carried out as previously described (36). The specific reaction conditions for the T 3 IRD activity assays were: microsome samples from livers of saltwater crocodiles from days 60 and 68 were diluted to a final protein concentration of 0.1 mg/ml. The substrate concentration was 0.19 nm [3-125 I] T 3 (corresponding to cpm) + 10 nm T 3 and 50 mm DTT. Incubation was for 60 min at 35 C. Liver microsomal samples from day 75 of incubation and from hatchlings were diluted to a final protein concentration of 1 mg/ml. The substrate

11 10 concentration was 0.19 nm [3-125 I] T 3 (corresponding to cpm) + 1 nm T 3 and 50 mm DTT. Incubation was for 60 min at 35 C. Statistical analysis. Plasma concentrations of corticosterone and free thyroid hormones during development were analysed using a one-way ANOVA with a Tukey-Kramer post test. Comparisons between plasma hormone concentrations and deiodinase activities in liver and kidney microsomes in the control groups, and those of treatment groups, were analysed using a one-way ANOVA with a Dunnett s post test. Significance was assumed at p<0.05. Results Effects of hormone treatments on plasma corticosterone concentrations. The average concentrations of plasma corticosterone on days 60, 68 and 75 of incubation in animals receiving control injections, and at one day post hatching, are presented in figure 1A. Plasma corticosterone increased significantly between days 68 and 75 (p<0.01) and decreased significantly between day 75 and hatching (p<0.05). Plasma corticosterone concentrations in embryos receiving hormone treatments are also presented in figure 1A. T 3 increased corticosterone levels compared to the control treatment on day 60 (p<0.05) and on day 68 (p<0.01), but not on day 75. Treatment of embryos with DEX suppressed plasma corticosterone on day 75 (p<0.05) as did treatment with DEX + T 3 (p<0.05). Effects of hormone treatments on plasma free T 3 concentrations. Free T 3 levels in plasma on day 60 of incubation were not measured due to lack of sample. Moreover, free T 3 was not detectable by our assay before day 67 in embryonic crocodiles (Shepherdley et al., unpublished observations). The concentrations of free T 3 in plasma on days 68, 75 in embryos receiving the control injections, and at one day after hatching, are presented in figure 1B. Plasma free T 3 levels after hatching were significantly higher than on day 75 (p<0.001). Plasma free T 3 concentrations in embryos receiving hormone treatments are also presented in figure 1B. Treatment with T 3 significantly increased free T 3 levels in the plasma on days 68

12 11 and 75 of incubation (p<0.05). DEX treatment did not affect the concentration of free T 3 in plasma on days 68 or 75. The concentration of plasma free T 3 was significantly increased by treatment with DEX + T 3 only on day 75 (p<0.05). Effects of hormone treatments on plasma free T 4 concentrations. The average free T 4 concentrations on days 60, 68, 75 in animals receiving the control injections, and at one day post hatching, are presented in figure 1C. Plasma free T 4 levels after hatching were significantly higher than on day 75 (p<0.001). Plasma free T 4 concentrations in embryos receiving hormone treatments are also presented in figure 1C. Treatment with T 3 significantly reduced free T 4 concentrations in plasma on day 68 (p<0.05). Treatment with DEX significantly reduced free T 4 concentrations in plasma on days 68 (p<0.01) and 75 (p<0.05). Treatment with DEX + T 3 also significantly reduced free T 4 concentrations in plasma on days 68 (p<0.01) and 75 (p<0.05). Effects of hormone treatments on rt 3 ORD activity in liver and kidney microsomes. On days 60 and 68 of development, treatment with T 3 did not affect rt 3 ORD activity in liver; however, on day 75, treatment with T 3 increased rt 3 ORD activity (p<0.05) (figure 2A). Treatment of embryos with DEX or DEX + T 3 did not have any effect on rt 3 ORD activity in liver microsomes on days 60, 68 or 75 of development (figure 2A). Treatment of embryos with T 3 did not have any effect on rt 3 ORD activity in kidney microsomes (figure 2B). On day 60, kidney microsomal rt 3 ORD activity was reduced by treatment with DEX (p<0.05) (figure 2B). Treatment with DEX + T 3 also reduced kidney microsomal rt 3 ORD activity on day 60 (p< 0.01) (figure 2B). Effects of hormone treatments on T 4 ORD activity in liver and kidney microsomes. Treatment of embryos with T 3 did not have any significant effects on T 4 ORD activity in liver microsomes (figure 3A). Treatment with DEX reduced liver microsomal T 4 ORD activity on day 60 of incubation (p<0.01) (figure 3A). On day 68 of incubation, DEX treatment had not

13 12 changed T 4 ORD activity from that of control levels. On day 75 of incubation treatment with DEX significantly increased T 4 ORD (p<0.01). Treatment with DEX + T 3 reduced T 4 ORD activity on day 60 of incubation (p<0.01) but not on days 68 and 75 (figure 3A). Treatment with T 3 significantly increased kidney microsomal T 4 ORD activity on days 60 (p<0.05), and 68 (p<0.05), but had no effect on day 75 (figure 3B). Treatment with DEX decreased T 4 ORD activity in kidney microsomes significantly only on day 60 (p<0.05) (figure 3B). Treatment of embryos with DEX + T 3 had no effect on kidney microsomal T 4 ORD on days 60, 68 and 75 of development (figure 3B). Liver microsomes incubated with the tracer and the high T 4 substrate concentration of 100 nm had approximately 50% of the activity measured compared with assays using a low T 4 substrate concentration of 1 nm T 4 on days 60 and 68 (data not shown). The high substrate tracer demonstrated 40% of the activity on day 75 and at hatching, hence both low and high- Km enzymes were present in this tissue (data not shown). Incubation of the tracer together with a high T 4 substrate concentration of 100 nm inhibited all deiodination activity in kidney microsomes, indicating the presence of only a low-km enzyme (data not shown). Effects of hormone treatments on T 3 IRD activity in liver microsomes. Treatment of embryos with T 3 had no effect on T 3 IRD activity in liver microsomes on days 60, 68 or 75 of development (figure 4). Treatment with DEX suppressed T 3 IRD activity on days 60 (p<0.01) and 68 (p<0.05) (figure 4). Treatment with DEX + T 3 suppressed T 3 IRD activity on day 60 of incubation (p<0.01) but had no effect on days 68 and 75 (figure 4). Discussion Plasma corticosterone and free thyroid hormones in embryonic and hatchling saltwater crocodiles. Like the situation in other vertebrates, corticosteroids in the plasma of embryonic saltwater crocodiles increased during development. However, plasma corticosterone declined

14 13 at hatching. When compared to the developmental profile of plasma corticosterone in the American alligator (Alligator mississippiensis), another Crocodilian species, the decrease in plasma corticosterone late in development is unique to saltwater crocodiles. In alligator embryos, plasma corticosterone levels of males and females increased from the last third of incubation, continued to rise until hatching (62 to 70 days of incubation), and did not decline until three weeks after hatching (21). The decrease in plasma corticosterone prior to hatching in the present study could be due to rapid metabolism of glucocorticoids, a decrease in the rate of synthesis or a decrease in secretion, or a decrease in binding proteins in blood. Plasma corticosteroids in bird embryos show a decrease during (47), or preceding hatching (18). The plasma binding capacity for corticosterone and corticosteroid binding globulin (CBG) concentrations have been shown to decrease after days 15 or 16 of incubation (hatching at day 21) in the embryonic chicken, resulting in increased free corticosterone in the blood (37, 14). This result suggested that higher levels of free glucocorticoids in blood were required and were being rapidly utilized by tissues in the last stages of in ovo development. It still remains to be determined whether there is a CBG equivalent in reptiles, and what percentage of plasma corticosterone is bound to this and other plasma binding proteins, such as albumin. Successful development also depends on precisely controlled changes in plasma thyroid hormone levels. We hypothesized that, as in other vertebrates, free thyroid hormones would increase in plasma during development. This was observed during the last 20 days of development, the most significant increase occurring between day 75 of incubation and hatching. The concentrations of plasma free T 3 and free T 4 one day post hatching was 22 times greater and 10 times greater, respectively, than free T 3 and free T 4 levels in plasma of juvenile (6 months old) saltwater crocodiles (Shepherdley et al., unpublished observations). The magnitude of the increases in free thyroid hormones late in development was clearly an

15 14 indication of the many developmental processes relying on thyroid hormones to regulate gene transcription. Effects of hormone treatment on rt 3 ORD, T 4 ORD and T 3 IRD activities in liver and kidney microsomes. In the embryonic chicken, glucocorticoid administration has been related to changes in plasma thyroid hormone concentrations and deiodinase activities, particularly in the liver. Twenty-four hours following a single DEX treatment in the chicken embryo, hepatic high-km rt 3 ORD activity (type I) was increased (4). We hypothesized that two DEX injections given to saltwater crocodile embryos over a 48 hour period would increase rt 3 ORD activity in liver microsomes. This effect on rt 3 ORD activity in liver microsomes was not demonstrated with DEX or DEX + T 3 treatment. Treatment with DEX and DEX + T 3 suppressed rt 3 ORD in kidney at day 60, but no other effects were observed on days 68 or 75. We hypothesized here that low-km T 4 ORD would increase in liver and kidney microsomes following glucocorticoid treatment, as has been shown recently for type II deiodinase activity in embryonic chicken brain (44). The effects of glucocorticoids on a low- Km, T 4 to T 3 converting enzyme in peripheral tissues have not been well studied, as a distinction has not always been made between high-km and low-km ORD activities in previous studies. In the present study we attempted to measure specifically the low-km T 4 ORD activity in liver microsomes; however, this was very difficult. Results on T 4 ORD activity in liver microsomes also included some contribution of a high-km ORD enzyme because it was not possible to completely inhibit high-km ORD activity in these assays. Additionally, the presence of very high levels of a low-km IRD activity in liver microsomes (particularly on days 60 and 68 of incubation) may have also interfered with the assay, despite the addition of 10 nm T 3 in an attempt to reduce low-km IRD interference. Therefore, the results of T 4 ORD assays in liver microsomes are difficult to interpret. DEX and DEX + T 3 treatments decreased T 4 ORD activity in liver microsomes at day 60. It is possible that

16 15 glucocorticoids could have a role in decreasing T 4 ORD activity in liver microsomes by speeding up maturation of the deiodinases, as T 4 ORD activity in liver microsomes in untreated crocodile embryos decreased during the last 20 days of development (35). However, DEX and DEX + T 3 treatments stimulated T 4 ORD activity in liver microsomes at day 75. Therefore, DEX treatment earlier in development (i.e. around days 60 and 68) could have been inhibiting the low-km ORD present, and treatment later in development (i.e. day 75) could have been increasing the activity of the high-km enzyme. However, an increase in high-km ORD due to DEX treatment was not demonstrated in the rt 3 ORD assays for measuring high-km ORD activity. In contrast to T 4 ORD activity in liver microsomes, the results from assays using kidney microsomes appeared to show the presence of only low-km T 4 ORD activity. DEX treatment decreased T 4 ORD activity in kidney microsomes at day 60, and T 4 ORD activity had a tendency to remain low in kidney microsomes later in development. T 4 ORD activity measured in kidney microsomes in the present experiments was due to exclusively low-km ORD activity. As for liver, glucocorticoids may play a role in decreasing T 4 ORD activity in kidney microsomes by speeding up maturation of the deiodinases, as T 4 ORD in kidney microsomes in untreated crocodile embryos was also observed to decrease during the last 20 days of development (35). Glucocorticoid treatment in embryonic chickens resulted in a decrease in hepatic T 3 IRD activity (9, 4). This inhibition of T 3 IRD activity has recently been shown to parallel a decrease in hepatic type III deiodinase mrna levels (42), indicating that the decrease in T 3 IRD activity in response to DEX treatment is regulated predominantly at the transcriptional level (44). In the present study, a decrease in T 3 IRD activity in liver microsomes was demonstrated on days 60 and 68, after DEX treatment. The combined DEX + T 3 treatment had a similar result on day 60.

17 16 Effects of hormone treatment on corticosterone, free T 4 and free T 3 concentrations in plasma. Treatment with T 3 increased plasma corticosterone concentrations on days 60 and 68. In the post natal rat, where significant development occurs during the first few weeks post partum, hyperthyroidism induced by T 4 administration resulted in a significant rise in both total corticosterone concentration and the capacity of CBG to bind corticosterone (3). Thyroid hormone administration also induced an increase in corticotropin releasing hormone (CRH) mrna synthesis (3). This could explain the increase in plasma corticosterone in response to T 3 treatment in saltwater crocodile embryos. In the present study treatment with DEX had a powerful suppressive effect on free T 4 concentrations in the plasma of saltwater crocodile embryos. The lack of change in rt 3 ORD activity in liver and kidney microsomes in response to DEX treatment suggested that the decrease in plasma free T 4 concentrations on days 68 and 75 following DEX and DEX + T 3 treatment was not due to increased conversion of T 4 to T 3 by a high-km ORD. Interestingly, Darras et al., (4) recorded a decrease in plasma total T 4 four hours after glucocorticoid administration in the chicken embryo, which did not correspond to the increase in type I ORD in liver, which occurred later. These authors suggested that the decrease in plasma T 4 in chicken embryos following glucocorticoid treatment was due to lower thyroidal T 4 secretion due to decreased thyrotropin (TSH) secretion, as indicated by decreased plasma α subunit levels and a decrease in the calculated TSH index. Although these parameters were not measured in the saltwater crocodile embryos, it is possible that glucocorticoid treatment could be having a similar effect. Glucocorticoids are also observed to have a suppressive effect on TSH in post-natal mammals (49, 24, 27). Again, the decrease in plasma free T 4 levels does not seem to be related to changes in deiodinase activities. In embryonic chickens receiving glucocorticoid treatment, plasma total T 3 concentrations increase, corresponding to the decrease in T 3 IRD activity (9, 4). In the

18 17 embryonic crocodiles, there was no increase in plasma free T 3 concentrations following DEX or DEX + T 3 treatments that corresponded with decreases in T 3 IRD activity. The increase in plasma free T 3 concentrations between day 75 and hatching occurred when T 3 IRD activity had almost completely diminished. It is possible therefore that plasma free T 3 levels are tightly controlled during development, however, this remains a topic for further investigation. These experiments also demonstrated that both the HPA axis and HPT axis were functioning. Treatment with DEX and DEX + T 3 significantly decreased endogenous corticosterone concentrations on day 75. Hence, DEX appeared to be mimicking the actions of endogenous glucocorticoids in the embryonic saltwater crocodiles by affecting the HPA axis. Treatment with T 3 increased circulating free T 3 concentrations on days 68 and 75, and decreased free T 4 concentrations significantly on day 68, but not on day 75. Changes in circulating plasma thyroid hormone concentrations probably resulted from negative feedback regulation by the HPT axis. In summary, the effects of hormone treatment on free thyroid hormones and corticosterone in plasma and deiodinase activities in saltwater crocodile embryos varied depending on the stage of development and the tissue type. DEX treatment had inhibiting effects on T 4 ORD activity in liver and kidney microsomes and on rt 3 ORD activity in kidney microsomes around day 60 of incubation. The stimulating effect of DEX treatment on T 4 ORD activity in liver microsomes occurred during the later stages of development in this study, on day 75, suggesting specific timing of the responses of the deiodinases to glucocorticoids. A decrease in free T 4 was not related to changes in rt 3 ORD or T 4 ORD activities, and was possibly produced by changes in TSH secretion. Glucocorticoid treatment had a suppressive action on T 3 IRD activity in liver microsomes, even though there was not a corresponding increase in free T 3 concentrations in plasma, suggesting that free T 3 levels are tightly regulated during development.

19 18 Acknowledgements The authors would like to thank L. Sullivan, Dr. S. Munns, Dr. P. Wood and J. Griffiths from Adelaide University and W. Van Ham, F. Voets, L. Noterdaeme from the K.U. Leuven for technical assistance. Crocodile eggs were maintained at Adelaide University under permit from the National Parks and Wildlife (South Australia) (#Q20168). Animal surgery was performed under permit from the Adelaide University Animal Ethics Committee (S/45/00). This research was funded by an Australian Research Council (ARC) grants to C.B. Daniels and S.J. Richardson.

20 19 References 1. Bates JM, St Germain D and Galton VA. Expression profiles of the three iodothyronine deiodinases, D1, D2, and D3, in the developing rat. Endocrinol. 140: , Borges M, LaBourene J and Ingbar SH. Changes in hepatic iodothyronine metabolism during ontogeny of the chick embryo. Endocrinol. 107: , Dakine N, Oliver C and Grino M. Thyroxine modulates corticotropin-releasing factor but not arginine vasopressin gene expression in the hypothalamic paraventricular nucleus of the developing rat. J. Neuroendocrinol. 12: , Darras VM, Kotanen SP, Geris KL, Berghman LR and Kühn ER. Plasma thyroid hormone levels and iodothyronine deiodinase activity following an acute glucocorticoid challenge in embryonic compared with post hatch chickens. Gen. Comp. Endo. 104: , Darras VM, Visser TJ, Berghman LR and Kühn ER. Ontogeny of type 1 and type 3 deiodinase activities in embryonic and post hatch chicks: relationship with changes in plasma triiodothyronine and growth hormone levels. Comp. Biochem. Physiol. 103A: , de Jesus EG, Hirano T and Inui Y. Changes in cortisol and thyroid hormone concentrations during early development and metamorphosis in the Japanese flounder, Paralichthys olivaceus. Gen. Comp. Endo. 82: , de Jesus EGT and Hirano T. Changes in whole body concentrations of cortisol, thyroid hormones, and sex steroids during early development of the Chum salmon, Oncorhynchus keta. Gen. Comp. Endo. 85: 55-61, Decuypere E, Kühn ER, Clijmans B, Nouwen EJ and Michels H. Prenatal peripheral monodeiodination in the chick embryo. Gen. Comp. Endo. 47: 15-17, 1982.

21 20 9. Decuypere E, Scanes CG and Kühn ER. Effects of glucocorticoids on circulating concentrations of thyroxine (T4) and triiodothyronine (T3) and on peripheral monodeiodination in pre and post hatching chickens. Horm. Metab. Res. 15: , Galton VA. Iodothyronine 5`-deiodinase activity in the amphibian Rana catesbeiana at different stages of the life cycle. Endocrinol. 122: , Galton VA. Mechanisms underlying the acceleration of thyroid hormone induced tadpole metamorphosis by corticosterone. Endocrinol. 127: , Galton VA and Hiebert A. The ontogeny of iodothyronine 5'-monodeiodinase activity in Rana catesbeiana tadpoles. Endocrinol. 122: , Galton VA, McCarthy PT and St Germain DL. The ontogeny of iodothyronine deiodinase systems in liver and intestine of the rat. Endocrinol. 128: , Gasc J-M and Martin B. Plasma corticosterone binding capacity in the partially decapitated chick embryo. Gen. Comp. Endo. 35: , Hughes TE and McNabb FMA. Avian hepatic T3 production by two pathways of 5'- monodeiodination: effects of fasting and patterns during development. J. Exp. Zool. 238: , Hulbert AJ. Thyroid hormones and their effects: a new perspective. Biol. Rev. 75: , Hwang P-P, Wu S-M, Lin J-H and Wu L-S. Cortisol content of eggs and larvae of teleosts. Gen. Comp. Endo. 86: , Kalliecharan R and Hall BK. A developmental study of the levels of progesterone, corticosterone, cortisol and cortisone circulating in the plasma of chick embryos. Gen. Comp. Endo. 24: , 1974.

22 Kaplan MM and Yaskoski KA. Maturational patterns of iodothyronine phenolic and tyrosyl ring deiodinase activities in rat cerebrum, cerebellum, and hypothalamus. J. Clin. Invest. 67: , McNabb FMA and King DB. Thyroid hormone effects on growth, development, and metabolism. In: The endocrinology of growth, development and metabolism in vertebrates, edited by Schreibman MP, Scanes CG and Pang PKT. USA: Academic Press, 1993, p Medler KF and Lance VA. Sex differences in plasma corticosterone levels in alligator (Alligator mississippiensis) embryos. J. Exp. Zool. 280: , Mendel CM. The free hormone hypothesis: a physiologically based mathamatical model. Endocr. Rev. 10: , Mondou PM and Kaltenbach JC. Thyroxine concentrations in the blood serum and pericardial fluid of metamorphosing tadpoles and of adult frogs. Gen. Comp. Endo. 39: , Nicoloff JT, Fisher DA and Appleman Jnr MD. The role of glucocorticoids in the regulation of thyroid function in man. J. Clin. Invest. 49: , Okimoto DK, Weber GM and Grau EG. The effects of thyroxine and propylthiouracil treatment on changes in body form associated with a possible developmental thyroxine surge during post-hatching development of the tilapia, Oreochromis mossanbicus. Zool. Sci. 10: , O'Steen S and Janzen FJ. Embryonic temperature affects metabolic compensation and thyroid hormones in hatchling snapping turtles. Physiol. Biochem. Zool. 75: , Re RN, Kourides IA, Ridgway EC, Weintraub BD and Maloof F. The effect of glucocorticoid administration on human pituitary secretion of thyrotropin and prolactin. J. Clin. Endocrinol. Metab. 43: , 1976.

23 Richard K, Hume R, Kaptein E, Sanders JP, Van Toor H, De Herder WW, Den Hollander JC, Krenning EC and Visser TJ. Ontogeny of iodothyronine deiodinases in human liver. J. Clin. Endocrinol. Metab. 83: , Roti E, Braverman E, Fang S-L, Alex S and Emerson CH. Ontogenesis of placental inner ring thyroxine deiodinase and amniotic fluid 3,3',5'-triiodothyronine concentration in the rat. Endocrinol. 111: , Ruiz de Ona C, Obregon MJ, Escobar del Ray F and Morreale de Escobar G. Developmental changes in rat brain 5'-deiodinase and thyroid hormones during the fetal period: the effects of fetal hypothyroidism and maternal thyroid hormones. Pediatr. Res. 24: , Samuels HH, Stanley F and Casanova J. Relationship of receptor affinity to the modulation of thyroid hormone nuclear receptor levels and growth hormone synthesis by L- triiodothyronine and iodothyronine analogues in cultured GH cells. J. Clin. Invest. 63: , Sato K, Mimura H, Han DC, Tsushima T and Shizume K. Ontogenesis of iodothyronine-5'-deiodinase: induction of 5'-deiodinating activity by insulin, glucocorticoid, and thyroxine in cultured fetal mouse liver. J. Clin. Invest. 74: , Schreck CB. Glucocorticoids: metabolism, growth, and development. In: The endocrinology of growth, development, and metabolism in vertebrates, edited by Schreibman MP, Scanes CG and Pang PKT. USA: Academic Press, 1993, p Scott TR, Johnson WA, Satterlee DG and Gildersleeve RP. Circulating levels of corticosterone in the serum of developing chick embryos and newly hatched chicks. Poultry Sci. 60: , 1981.

24 Shepherdley CA, Richardson SJ, Evans BK, Kühn ER and Darras VM. Thyroid hormone deiodinases during embryonic development of the saltwater crocodile (Crocodylus porosus). Gen. Comp. Endo. in press, Shepherdley CA, Richardson SJ, Evans BK, Kühn ER and Darras VM. Characterisation of outer ring iodothyronine deiodinases in tissues of the saltwater crocodile (Crocodylus porosus). Gen. Comp. Endo. 125: , Siegel HS and Gould NR. Chick embryonic plasma proteins and binding capacity for corticosterone. Dev. Biol. 50: , Specker JL, DiStephano JJ, Grau EG, Nishioka RS and Bern HA. Developmentassociated changes in thyroxine kinetics in juvenile salmon. Endocrinol. 115: , Specker JL and Schreck CB. Changes in plasma corticosteroids during smoultification in Coho salmon, Oncorhynchus kisutch. Gen. Comp. Endo. 46: 53-58, Suzuki S and Suzuki M. Changes in thyroidal and plasma iodine compounds during and after metamorphosis of the bullfrog, Rana catesbeiana. Gen. Comp. Endo. 45: 74-81, Thommes RC and Hylka VW. Plasma iodothyronines in the embryonic and post hatch chick. Gen. Comp. Endo. 32: , Van der Geyten S, Buys NB, Sanders JP, Decuypere E, Visser TJ, Kühn ER and Darras VM. Acute pretranslational regulation of type III iodothyronine deiodinase by growth hormone and dexamethasone in chicken embryos. Mol. Cell. Endocrinol. 147: 49-56, Van der Geyten S, Sanders JP, Kaptein E, Darras VM, Kühn ER, Leonard JL and Visser TJ. Expression of chicken hepatic type I and type III iodothyronine deiodinases during development. Endocrinol. 138: , Van der Geyten S, Seges I, Gereben B, Bartha T, Rudas P, Larsen PR, Kühn ER and Darras VM. Transcriptional regulation of iodothyronine deiodinases during embryonic development. Mol. Cell. Endocrinol. 183: 1-9, 2001.

25 Vulsma T, Gons MH and de Vijlder JJM. Maternal-fetal transfer of thyroxine in congenital hypothyroidism due to a total organification defect or thyroid agenesis. New Engl. J. Med. 321: 13-16, Weber GM, Farrar ES, Tom CKF and Grau EG. Changes in whole-body thyroxine and triiodothyronine concentrations and total content during early development and metamorphosis of the toad Bufo marinus. Gen. Comp. Endo. 94: 62-71, Wentworth BC and Hussein MO. Serum corticosterone levels in embryos, newly hatched and young turkey poults. Poultry Sci. 64: , Whittle WL, Patel FA, Alfaidy N, Holloway AC, Fraser M, Gyomorey S, Lye SJ, Gibb W and Challis JRG. Glucocorticoid regulation of human and ovine parturition: the relationship between fetal hypothalamic-pituitary-adrenal axis activation and intrauterine prostaglandin production. Biol. Reprod. 64: , Wilber JF and Utiger RD. The effect of glucocorticoids on thyrotropin secretion. J. Clin. Invest. 48: , Wittmann J, Steib A, Liebich HG and Schmidt P. Acceleration and retarding effects of dexamethasone on the development of the avian lung. Dev. Pharm. Therapy 12: , Wu S-Y, Klein AH, Chopra IJ and Fisher DA. Alterations in tissue thyroxine-5'- monodeiodinating activity in perinatal period. Endocrinol. 103: , 1978.

26 25 Figure legends Figure 1: Plasma concentrations of (A) corticosterone (ng/ml ± SEM), (B) free T 3 (pg/ml ± SEM) and (C) free T 4 (pg/ml ± SEM) on days 60, 68 and 75 of incubation in embryonic saltwater crocodiles receiving the control injections, and at one day post hatching (black bars). ( p<0.05; p<0.01; p<0.001 versus the previous time point) Plasma concentrations of (A) corticosterone, (B) free T 3 and (C) free T 4 on days 60, 68 and 75 of incubation following two treatments of 5 µg T 3 (striped bars), 50 µg DEX (spotted bars), and DEX (50 µg) + T 3 (5 µg) (white bars). (* p<0.05; ** p<0.01 versus controls) Figure 2: rt 3 ORD activity in (A) liver microsomes and (B) kidney microsomes (pmol rt 3 deiodinated/mg protein/min ± SEM) on days 60, 68 and 75 of incubation, in animals receiving the control injections (black bars) and following two treatments of 5 µg T 3 (striped bars), 50 µg DEX (spotted bars), and DEX (50 µg) + T 3 (5 µg) (white bars). (* p<0.05; ** p<0.01 versus controls) Figure 3: T 4 ORD activity in (A) liver microsomes and (B) kidney microsomes (fmol T 4 deiodinated/mg protein/min ± SEM) on days 60, 68 and 75 of incubation, in animals receiving the control injections (black bars) and following two treatments of 5 µg T 3 (striped bars), 50 µg DEX (spotted bars), and DEX (50 µg) + T 3 (5 µg) (white bars). (** p<0.01 versus controls) Figure 4: T 3 IRD activity in liver microsomes (fmol T 3 deiodinated/mg protein/min ± SEM) on days 60, 68 and 75 of incubation, in animals receiving the control injections (black bars) and following two treatments of 5 µg T 3 (striped bars), 50 µg DEX (spotted bars), and DEX (50 µg) + T 3 (5 µg) (white bars). (* p<0.05 versus controls)

27 26 Figure 1A corticosterone (ng/ml) * ** 2 0 day 60 day 68 day 75 hatch day of incubation * * Figure 1B * free T3 (pg/ml) * * day 68 day 75 hatch day of incubation Figure 1C free T4 (pg/ml) * ** ** * * day 60 day 68 day 75 hatch day of incubation

28 27 Figure 2A pmol rt3 deiodinated/mg protein/min day 60 day 68 day 75 day of incubation * Figure 2B pmol rt3 deiodinated/mg protein/min * ** day 60 day 68 day 75 day of incubation

29 28 Figure 3A fmol T4 deiodinated/mg protein/min ** ** ** day 60 day 68 day 75 day of incubation Figure 3B fmol T4 deiodinated/mg protein/min * * * day 60 day 68 day 75 day of incubation

30 29 Figure 4 fmol T3 deiodinated/mg protein/min ** ** * day 60 day 68 day 75 day of incubation