- Email: [email protected]
Influence of Recombinant Chicken Prolactin on Thyroid Hormone Metabolism in the Chick Embryo
General and Comparative Endocrinology 103, 349–358 (1996) Article No. 0130
Influence of Recombinant Chicken Prolactin on Thyroid Hormone Metabolism in the Chick Embryo E. R. Ku¨hn,* K. Shimada,† T. Ohkubo,‡ L. M. Vleurick,* L. R. Berghman,§ and V. M. Darras* *Laboratory of Comparative Endocrinology, and §Laboratory of Neuroendocrinology and Immunological Biotechnology Zoological Institute, Catholic University of Leuven, B-3000 Leuven, Belgium; †Laboratory of Animal Physiology, School of Agricultural Sciences, Nagoya University, Chikusa, Nagoya 464-01, Japan; and ‡Center for Molecular Biology and Genetics, Mie University, Tsu, Mie 514, Japan Accepted May 13, 1996
The influence of a recombinant chicken prolactin (rcPRL) preparation on thyroid function was studied in 18- and 19-day-old chicken embryos. Displacement studies on hepatic microsomes indicate that this preparation does not compete with radiolabeled chicken growth hormone (cGH) for hepatic GH-receptor binding. In a first series of experiments rcPRL or immunoaffinity-purified cGH was injected intravenously in 19-day-old chicken embryos. After 2 hr, cGH increased plasma T3 in a dosedependent way by inhibiting hepatic inner ring type III deiodination (IRD-III) and consequently T3 degradation. Outer ring deiodination (ORD-I) was not influenced confirming previous results. The rcPRL preparation (2 and 10 mg) did not influence plasma T3, but depressed T4 and raised hepatic IRD-III activity simultaneously, whereas no influence on hepatic ORD-I activity could be found. In a second experiment on 18-day-old embryos, it could be demonstrated that the effect of 2.5 mg cGH on plasma T3 and liver IRD-III lasted up to 6 hr after injection, whereas 2.5 mg cPRL affected plasma T4 and liver IRD-III up to 2 hr. Both rcPRL and cGH depressed rT3 up to 6 hr, whereas an injection of rcPRL, but not of cGH, elevated plasma concentrations of corticosterone. These results indicate that prolactin may have a role, together with GH, in controlling peripheral thyroid hormone metabolism. r 1996 Academic Press, Inc.
0016-6480/96 $18.00 Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
In recent years it has become obvious that peripheral thyroid hormone metabolism in the chick embryo may be controlled by the somatotrophic and corticotrophic axis. Growth hormone (GH) increases plasma T3 and decreases plasma T4 in 18-day-old chicken embryos and some evidence has been presented that GH is the only hypophyseal factor responsible for this effect (Berghman et al., 1989; Darras et al., 1990). However, administration of the adrenocorticotrophic hormone (ACTH) or glucocorticoids also was able to increase the circulating concentrations of T3 and the T3 to T4 ratio, without consistent changes in plasma T4 or GH concentrations (Decuypere et al., 1983). Specific type I and type III deiodinase tests indicated that GH has no effect at all on the in vitro hepatic outer ring deiodinase (ORD-I) activity, but acutely decreases the in vitro inner ring deiodinase (IRD-III) activity, suggesting that the GH-induced increase in plasma T3 is not due to an increased T3 production, but to a decreased T3 breakdown (Darras et al., 1992a). Recent studies on 18-day-old embryos confirmed that ACTH and glucocorticoids increased plasma concentrations of T3, but decreased T4. Hepatic ORD-I activity was increased, but only after 24 and 48 hr, while renal ORD-I was not significantly influenced. IRD-III activity, however, was strongly depressed after 4 and 24 hr in the liver and after 4 hr in the kidney (Darras et al., 1995).
349
350
Ku¨hn et al.
At the end of incubation, prolactin (PRL) is also increased in plasma and pituitary, next to GH, ACTH, and cortisol (Harvey et al., 1979). This increase is associated with a marked rise in the amount of pituitary PRL mRNA (Ishida et al., 1991). However, no information is available concerning the physiological role of these high prolactin levels at the end of
FIG. 1.
incubation. In the present study the influence of a recombinant chicken PRL (rcPRL) preparation (Ohkubo et al., 1993; Shimada et al., 1993) on thyroid hormone metabolism was therefore studied in 18-dayold chicken embryos. Since the adrenal axis also seems to be involved in the control of peripheral thyroid function (Geris et al., 1995; Darras et al., 1995), the
Displacement of cGH in liver microsomes of 18-day-old chicken embryos (A) and of adult hens (B).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
351
Influence of rcPRL on Thyroid Hormone Metabolism
influence of rcPRL on plasma concentrations of corticosterone was studied as well.
TABLE 1 Effective Dose (ng) for 50% Displacement of 125I-Labeled cGH by Several Hormone Preparations in the Chicken Embryo and Adult Chicken
Embryo (18 days) Adult hens
rcGH
oGH
cGH
oPRL
rcPRL
0.33 0.69
0.61 1.53
2.06 3.67
6.93 9.08
— —
MATERIALS AND METHODS Eggs of the Hisex White Strain were purchased from a commercial breeder (Euribrid, Aarschot, Belgium) and kept in a forced-draft laboratory incubator under standard conditions.
FIG. 2. Influence of saline and different injection doses of rcPRL (2 and 10 µg) and cGH (0.4, 2, and 10 µg) on plasma concentrations of T3 (A) and T4 (B) (pmol/ml) after 2 hr in 19-day-old chicken embryos. Sal 1 and sal 2, control injections at start and end of experiment. Values are means 6 SEM for 12 individual samples. Groups with different superscripts are significantly different (P , 0.05, DUNCAN).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
352
Ku¨hn et al.
FIG. 3. Influence of saline and different injection doses of rcPRL (2 and 10 µg) and cGH (0.4, 2, and 10 µg) on liver ORD-I (A) and IRD-III (B) activity in 18-day-old chicken embryos. Sal 1 and sal 2, injections at start and end of experiment. Values are means 6 SEM for 12 individual samples. Groups with different superscripts are significantly different (P , 0.05, DUNCAN).
Injections using 19- (first experiment) and 18- (second experiment) day-old embryos were made into one of the peripheral chorioallantoic blood vessels lying close to a window cut in the shell at the lateral side. Blood samples (of 0.5 to 1 ml) were taken by cardiac puncture
and collected in heparinized tubes after 2 hr in a first experiment and after 2, 6, and 24 hr in a second one. In a first series of experiments (n 5 12 for each group) 2 or 10 µg rcPRL or 0.4, 2, or 10 µg of hypophyseal chicken GH (cGH) were injected and all injections were made over
FIG. 4. Plasma concentrations of T3 (A), T4 (B), and rT3 (C) (pmol/ml) of 18-day-old chicken embryos 2, 6, and 24 hr following injection of saline, 2.5 µg of rcPRL and 2.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
354
a period of 1.5 hr starting with saline (sal 1), 0.4 µg GH, 2 µg GH, 2 µg rcPRL, 10 µg GH, 10 µg rcPRL, and again saline (sal 2). Since obviously injection time after the time interval between removal from the incubators and injection may have some influence on results, this experiment was repeated using 2.5 µg of rcPRL and 2.5 µg of cGH but injections were randomized in groups of six. In addition, blood and tissue samples were taken after 2, 6, and 24 hr and the effects on plasma rT3 and corticosterone were also included. Following centrifugation the plasma was kept at 220° until assay of hormones. Livers and kidneys were also excised for deiodinase assays and stored frozen at 280° until homogenization. In vitro hepatic ORD-I and IRD-III activities were measured in microsomal (Mx) preparations as described previously (Darras et al., 1992b). Briefly, hepatic ORD-I activity was measured using 200 µg Mx protein/ml with 1 µM 3,38,58-triiodothyronine (rT3) as substrate and 5 mM dithiothreitol (DTT) as cofactor. Hepatic IRD-III activity was measured using 50 µg Mx protein/ml with 10 nM T3 as substrate in the presence of 1 µM rT3, 50 mM DTT as cofactor, and 1 mM 6-propyl-2-thiouracil (PTU) to block type I interference. For displacement studies several hormone preparations were used to compete with 125I-labeled cGH for binding to liver microsomes of 18-day-old embryos (Hisex White) or to adult hens (Warren). The cGH used in this study and for injections was purified following the method described by Berghman et al. (1988). The recombinant cGH (rcGH) was obtained through the courtesy of Dr. T. Hall (Ciba Geigy, Basel), whereas the rcPRL preparation was made by Ohkubo et al. (1993). Ovine GH (oGH) was obtained from NIH (NIADDK-NIH (oGH-15)) and ovine PRL (oPRL) from Sigma. GH receptors were determined as described by Vanderpooten et al. (1991). Slopes and intercepts of the linearized curves (logit (B/Bo) 5 intercept 1 (slope 3 logdose)) have been calculated by linear regression. These parameters have been used to calculate the effective dose for 50% displacement. Plasma T3, T4, and rT3 were measured by radioimmunoassay (RIA) using antisera for T3 and T4 from Byk-Belga (Belgium) and the antiserum described by Visser et al. (1977) for rT3. [125I]T3, [125I]T4 and, [125I]rT3 were from Amersham (Belgium) and the standards consisted of T3, T4, and rT3 diluted in hormone-free human serum (Byk-Belga).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
Ku¨hn et al.
Corticosterone was measured using a commercially available RIA kit (ICN, Belgium) after validation for use in chicken plasma. For all RIAs chicken plasma dilution tests and loading tests showed good parallelism with the standard curve. Intraassay variabilities were 4.5% (T3), 5.4% (T4), and 6.9% (corticosterone), whereas interassay variabilities were 15, 17, and 7.3%, respectively. The assay of rT3 was based on the instruction as described by Visser et al. (1977). Statistical analysis of the results was made using the GLM procedure (SAS Institute, Inc., 1985) followed by the Duncan multiple range test.
RESULTS GH Displacement Curves in Liver Microsomes (Fig. 1) Displacement studies on hepatic microsomes of 18-dayold chicken embryos and adult hens indicated that parallel inhibition curves could be obtained with cGH, rcGH, and oGH for displacement of 125I-labeled cGH. Ovine PRL could also compete for binding 125I-labeled cGH, but no parallel curve was obtained. The effective doses (ng) for 50% displacement are summarized in Table 1. The recombinant cPRL preparation, however, was totally ineffective in this regard in embryos as well as in adult hens.
Injection Studies In a first series of experiments no difference in both saline injections was noted for plasma T3, plasma T4, and hepatic IRD-III, but liver ORD-I activity was significantly elevated in saline 2. The rcPRL injections decreased plasma concentrations of T4 but had no influence on T3, whereas cGH increased plasma T3 at every injection dose used and decreased T4 at the 2and 10-µg doses (Fig. 2). Liver ORD-I of the sal 1 group was lower than for all other groups including sal 2. These groups however did not differ among each other. Liver IRD-III, however, was increased following injection of both PRL concentrations compared to both saline injections and decreased following injection of all GH doses when compared to sal 1 but only at the 0.4- and 2-µg dose when compared to sal 2 (Fig. 3).
Influence of rcPRL on Thyroid Hormone Metabolism
Results of the second series of experiments indicated that control plasma concentrations of T4 were increased after 6 and 24 hr compared to that at 2 hr, whereas no significant changes were present for rT3 and T3. After 2 hr, 2.5 µg of rcPRL decreased T4 and had no influence on T3. On the contrary, an injection of 2.5 µg of cGH did not affect T4 concentrations, but
355
raised T3 after 2 and 6 hr. This was accompanied by decreases in plasma rT3, a decrease which was also present in the PRL-injected group but here only after 6 hr (Fig. 4). Hepatic ORD-I activities were increased after 24 hr and this increase was more pronounced in the cGH group compared to other groups. Hepatic IRD-III
FIG. 5. Liver ORD-I (A) and IRD-III (B) activity of 18-day-old chicken embryos 2, 6, and 24 hr following injection of saline, 0.5 µg of rcPRL, and 0.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
356
activity increased firmly following injections of 2.5 µg rcPRL after 2 hr, whereas a profound decrease is present following injection of 2.5 µg cGH after 2 and 6 hr (Fig. 5). Plasma corticosterone levels were elevated after 2 hr compared to controls following injection of 2.5 µg of rcPRL. An injection of 2.5 µg cGH however did not influence corticosterone plasma concentrations (Fig. 6).
DISCUSSION The present results indicate that in the chicken embryo rcPRL and cGH have different hepatic receptors and opposite effects on hepatic IRD-III activity. They also differ as to their influence on plasma concentrations of T3. However, both hormones are capable of depressing plasma concentrations of T4 and rT3, and data indicate that these effects are not mediated by any changes in hepatic ORD-I levels. Earlier studies using oPRL as a chicken prolactin substitute have shown that oPRL binds to hepatic GH receptors as well (Leung et al., 1984). This has been
Ku¨hn et al.
confirmed in the present study, which explains why oPRL (Ku¨hn et al., 1983) and cGH (Berghman et al., 1989) have similar effects on plasma concentrations of thyroid hormones. No binding to chicken hepatic GH receptors occurs however with the present rcPRL preparation, confirming results of Burnside and Cogburn (1993) and results on thyroid hormone metabolism differ profoundly between both hormones. The observed effects following injection of cGH in the chicken embryo confirm previous observations (Darras et al., 1992a; Berghman et al., 1989). Briefly, in control animals hepatic IRD-III activity decreases more than 10-fold between Day 17 and hatching, whereas the hepatic ORD-I activity is increasing slowly up to the period of pipping and hatching (Darras et al., 1992b). This is accompanied by a sudden increase in plasma concentrations of T3 at the end of incubation. cGH, when injected during this period, provokes a profound decrease in hepatic IRD-III, without changing in vitro ORD-I activity. The concurrent marked increase in plasma concentrations of T3 and decrease of rT3 is likely to be the result of the above enzymatic activities (Darras et al., 1990, 1992a).
FIG. 6. Plasma corticosterone concentrations (ng B/ml) in 18-day-old chicken embryos 2, 6, and 24 hr following injections of saline, 0.5 µg of rcPRL, and 0.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
357
Influence of rcPRL on Thyroid Hormone Metabolism
It has been shown that endogenous GH is an important factor in the control of plasma T3 levels in growing chickens due to its influence on the activity of the T3-degrading type III deiodinase (Darras et al., 1993). On the other hand, growth rate is closely related to plasma concentrations of T3 and growth retardation as a result of a lower incubation temperature is paralleled by ‘‘retarded’’ T3 levels (Ku¨hn et al., 1982). In the present study PRL is capable of stimulating the hepatic T3-degrading activity (though plasma levels of T3 were not affected) and it is remarkable that in the earlier experiments on chicken embryos incubated at a lower temperature higher levels of prolactin (and lower levels of T3) were found immediately after hatching together with the observed growth inhibition (Ku¨hn et al., 1985). However, together with an increase in the hepatic IRD-III activity following injections of rcPRL, no effect on plasma T3 was observed and lower T4 and rT3 levels were found. These results are difficult to explain since one would expect decreased plasma T3 and possibly T4 concentrations together with an increase in rT3. It is possible that the effect of PRL on thyroid metabolism is complicated by interference with the adrenal axis. In mammals the adrenal cortical tissue is known to have high concentrations of PRL receptors (Horrobin, 1977) and in the chicken an increase in mitogenic activity of the adrenals was observed following PRL injections (Maiti and Chakraborti, 1980). It is also known that oPRL counteracts the corticosteroneinduced suppression of net ACTH-stimulated corticosterone production of intact domestic fowl adrenocortical cells. However, oPRL does not stimulate basal or ACTH-induced corticosterone synthesis (Carsia et al., 1987). The involvement of the adrenal axis in thyroid metabolism is well-known (Wada et al., 1975; Ku¨hn et al., 1983, 1985; Decuypere et al., 1983) and recent studies have extended the earlier observations on the histology of the thyroid gland and circulating thyroid hormones with new observations on hepatic and renal deiodinating enzymes (Geris et al., 1995; Darras et al., 1995). These results indicate that corticosterone has no short-term influence on ORD-I but decreases IRD-III and plasma concentrations of rT3 and T4 and increases plasma T3. The present observation that PRL is capable of increasing corticosterone concentrations in the plasma of 18-day-old embryos seems to support the
hypothesis that part of the present results may be explained by interference of prolactin with adrenal secretion. This way the suppressive effect of rcPRL on plasma concentration of T4 and rT3 may be mediated by the release of corticosterone.
ACKNOWLEDGMENTS We thank Mrs. F. Voets and L. Noterdaeme and Mr. W. Van Ham for their valuable technical assistance. We thank Dr. T. Hall (CibaGeigy, Basel) for the gift of recombinant cGH and Dr. T. Visser (Erasmus University, Rotterdam, The Netherlands) for the gift of rT3 antiserum. V. Darras and L. Berghman were supported by the National Fund for Scientific Research. This work was also supported by grants from the National Fund for Scientific Research (2.0114.90 and 2.0114.94).
REFERENCES Berghman, L. R., Darras, V. M., Huybrechts, L. M., Decuypere, E., Vandesande, F., and Ku¨hn, E. R. (1989). Evidence for chicken GH as the only hypophyseal factor responsible for the stimulation of the hepatic 58-monodeiodination activity in the chick embryo. Reprod. Nutr. Dev. 29, 197–202. Berghman, L. R., Van Beeumen, J., Decuypere, E., Ku¨hn, E. R., and Vandesande, F. (1988). One-step purification of chicken growth hormone from a crude pituitary extract by use of a monoclonal immunoadsorbent. J. Endocrinol. 118, 381–387. Burnside, J., and Cogburn, L. A. (1993). Molecular biology of the chicken growth hormone receptor. In ‘‘Avian Endocrinology’’ (P. J. Sharp, Ed.), pp. 161–176. Journal of Endocrinology Ltd., Bristol, UK. Carsia, R. V., Scanes, C. G., and Malamed, S. (1987). Polyhormonal regulation of avian and mammalian corticosteroidogenesis in vitro. Comp. Biochem. Physiol. A 88, 131–140. Darras, V. M., Berghman, L. R., Vanderpooten, A., and Ku¨hn, E. R. (1992a). Growth hormone acutely decreases type III iodothyronine deiodinase in chicken liver. FEBS Lett. 300, 5–8. Darras, V. M., Huybrechts, L. M., Berghman, L. R., Ku¨hn, E. R., and Decuypere, E. (1990). Ontogeny of the effect of purified chicken growth hormone on the liver 58-monodeiodination activity in the chicken: Reversal of the activity after hatching. Gen. Comp. Endocrinol. 77, 212–220. Darras, V. M., Kotanen, S., Geris, K. L., and Ku¨hn, E. R. (1995). Differential effects of hormones of the adrenal axis on thyroid function in embryonic chicken. Abstract 497, 11th International Thyroid Congress, September 10–15, Toronto, Canada. Darras, V. M., Rudas, P., Visser, T. J., Hall, T. R., Huybrechts, L. M., Vanderpooten, A., Berghman, L. R., Decuypere, E., and Ku¨hn, E. R. (1993). Endogenous growth hormone controls high plasma levels
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
358 of 3,38,58-triiodothyronine (T3) in growing chickens by decreasing the T3-degrading type III deiodinase activity. Dom. Anim. Endocrinol. 10, 55–65. Darras, V. M., Visser, T. J., Berghman, L. R., and Ku¨hn, E. R. (1992b). Ontogeny of type I and type III deiodinase activity in embryonic and posthatch chicks: Relationship with changes in plasma triiodothyronine and growth hormone levels. Comp. Biochem. Physiol. A 103, 131–136. Decuypere, E., Scanes, C. G., and Ku¨hn, E. R. (1983). Effects of glucocorticoids on circulating concentrations of thyroxine (T4) and triiodothyronine (T3) and on peripheral monodeiodination in preand post-hatching chickens. Horm. Metab. Res. 15, 233–236. Geris, K. L., Darras, V. M., Berghman, L. R., and Ku¨hn, E. R. (1995). Influence of corticotropin-releasing factor on the in vitro thyroxine and thyrotropin secretion in newly hatched fowl. Belg. J. Zool. 125, 143–156. Harvey, S., Davison, T., and Chadwick, A. (1979). Ontogeny of growth hormone and prolactin secretion in the domestic fowl (Gallus domesticus). Gen. Comp. Endocrinol. 39, 270–273. Horrobin, D. F. (1977). In ‘‘Prolactin,’’ (D. F. Horrobin, Ed.), Vol. 5, pp. 105–108. Annual Research Reviews. Livingstone, Edinburgh. Ishida, H., Shimada, K., Sato, K., Seo, H., Matsui, N., and Zadworny, D. (1991). Developmental expression of the prolactin gene in the chicken. Gen. Comp. Endocrinol. 83, 463–467. Ku¨hn, E. R., Decuypere, E., Colen, L. M., Chadwick, A., Heyns, W., Michels, H., and Berghman, L. R. (1985). Circadian rhythms of corticosterone, prolactin and iodohormone secretion in the posthatch chick, and their influence on growth. In ‘‘Current Trends in Comparative Endocrinology’’ (B. Lofts and W. N. Holmes, Eds.), pp. 671–675. Hong Kong Univ. Press, Hong Kong. Ku¨hn, E. R., Decuypere, E., Colen, L. M., and Michels, H. (1982). Posthatch growth and development of a circadian rhythm for
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
Ku¨hn et al.
thyroid hormones in chicks incubated at different temperatures. Poultry Sci. 61, 540–549. Ku¨hn, E. R., Decuypere, E., Hemschoote, K., Berghman, L., and Paulussen, J. (1983). Antagonism of serum tri-iodothyronine changes after injection of prolactin in the domestic fowl before and after hatching. J. Endocrinol. 99, 401–407. Leung, F. C., Taylor, J. E., Steelman, S. L., Bennett, C. D., Rodkey, J. A., Long, R. A., Serio, R., Weppelman, R. N., and Olson, G. (1984). Purification and properties of chicken growth hormone and the development of a homologous radioimmunoassay. Gen. Comp. Endocrinol. 56, 389–400. Maiti, B. R., and Chakraborty, S. (1980). The mitogenic action of prolactin on some endocrine organs of chicks. Arch. Histol. Jap. 43, 341–346. Ohkubo, T., Tanaka, M., Nakashima, K., Shimada, K., Saito, N., and Sato, K. (1993). High level expression of biological active chicken prolactin in E. coli. Comp. Biochem. Physiol. A 105, 123–128. SAS Institute, Inc. (1985). ‘‘SAS User’s Guide: Statistics Version 5 Edition.’’ SAS Institute Inc. Cary, NC. 956 pp. Shimada, K., Ohkubo, T., Saito, N., Talbot, R. T., and Sharp, P. J. (1993). The molecular biology of prolactin. In ‘‘Avian Endocrinology’’ (P. J. Sharp, Ed.), pp. 135–147. Journal of Endocrinology Ltd., Bristol, UK. Vanderpooten, A., Darras, V. M., Huybrechts, L. M., Rudas, P., Decuypere, E., and Ku¨hn, E. R. (1991). Effect of hypophysectomy and acute administration of growth hormone (GH) on GHreceptor binding in chick liver membranes. J. Endocrinol. 129, 275–281. Visser, T. J., Docter, R., and Hennemann, G. (1977). Radioimmunoassay of reverse triiodothyronine. J. Endocrinol. 73, 395–396. Wada, M., Arimatsu, Y., and Kobayashi, H. (1975). Effect of prolactin on thyroid function in Japanese quail (Coturnix coturnis japonica). Gen. Comp. Endocrinol. 27, 28–33.
Influence of Recombinant Chicken Prolactin on Thyroid Hormone Metabolism in the Chick Embryo E. R. Ku¨hn,* K. Shimada,† T. Ohkubo,‡ L. M. Vleurick,* L. R. Berghman,§ and V. M. Darras* *Laboratory of Comparative Endocrinology, and §Laboratory of Neuroendocrinology and Immunological Biotechnology Zoological Institute, Catholic University of Leuven, B-3000 Leuven, Belgium; †Laboratory of Animal Physiology, School of Agricultural Sciences, Nagoya University, Chikusa, Nagoya 464-01, Japan; and ‡Center for Molecular Biology and Genetics, Mie University, Tsu, Mie 514, Japan Accepted May 13, 1996
The influence of a recombinant chicken prolactin (rcPRL) preparation on thyroid function was studied in 18- and 19-day-old chicken embryos. Displacement studies on hepatic microsomes indicate that this preparation does not compete with radiolabeled chicken growth hormone (cGH) for hepatic GH-receptor binding. In a first series of experiments rcPRL or immunoaffinity-purified cGH was injected intravenously in 19-day-old chicken embryos. After 2 hr, cGH increased plasma T3 in a dosedependent way by inhibiting hepatic inner ring type III deiodination (IRD-III) and consequently T3 degradation. Outer ring deiodination (ORD-I) was not influenced confirming previous results. The rcPRL preparation (2 and 10 mg) did not influence plasma T3, but depressed T4 and raised hepatic IRD-III activity simultaneously, whereas no influence on hepatic ORD-I activity could be found. In a second experiment on 18-day-old embryos, it could be demonstrated that the effect of 2.5 mg cGH on plasma T3 and liver IRD-III lasted up to 6 hr after injection, whereas 2.5 mg cPRL affected plasma T4 and liver IRD-III up to 2 hr. Both rcPRL and cGH depressed rT3 up to 6 hr, whereas an injection of rcPRL, but not of cGH, elevated plasma concentrations of corticosterone. These results indicate that prolactin may have a role, together with GH, in controlling peripheral thyroid hormone metabolism. r 1996 Academic Press, Inc.
0016-6480/96 $18.00 Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
In recent years it has become obvious that peripheral thyroid hormone metabolism in the chick embryo may be controlled by the somatotrophic and corticotrophic axis. Growth hormone (GH) increases plasma T3 and decreases plasma T4 in 18-day-old chicken embryos and some evidence has been presented that GH is the only hypophyseal factor responsible for this effect (Berghman et al., 1989; Darras et al., 1990). However, administration of the adrenocorticotrophic hormone (ACTH) or glucocorticoids also was able to increase the circulating concentrations of T3 and the T3 to T4 ratio, without consistent changes in plasma T4 or GH concentrations (Decuypere et al., 1983). Specific type I and type III deiodinase tests indicated that GH has no effect at all on the in vitro hepatic outer ring deiodinase (ORD-I) activity, but acutely decreases the in vitro inner ring deiodinase (IRD-III) activity, suggesting that the GH-induced increase in plasma T3 is not due to an increased T3 production, but to a decreased T3 breakdown (Darras et al., 1992a). Recent studies on 18-day-old embryos confirmed that ACTH and glucocorticoids increased plasma concentrations of T3, but decreased T4. Hepatic ORD-I activity was increased, but only after 24 and 48 hr, while renal ORD-I was not significantly influenced. IRD-III activity, however, was strongly depressed after 4 and 24 hr in the liver and after 4 hr in the kidney (Darras et al., 1995).
349
350
Ku¨hn et al.
At the end of incubation, prolactin (PRL) is also increased in plasma and pituitary, next to GH, ACTH, and cortisol (Harvey et al., 1979). This increase is associated with a marked rise in the amount of pituitary PRL mRNA (Ishida et al., 1991). However, no information is available concerning the physiological role of these high prolactin levels at the end of
FIG. 1.
incubation. In the present study the influence of a recombinant chicken PRL (rcPRL) preparation (Ohkubo et al., 1993; Shimada et al., 1993) on thyroid hormone metabolism was therefore studied in 18-dayold chicken embryos. Since the adrenal axis also seems to be involved in the control of peripheral thyroid function (Geris et al., 1995; Darras et al., 1995), the
Displacement of cGH in liver microsomes of 18-day-old chicken embryos (A) and of adult hens (B).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
351
Influence of rcPRL on Thyroid Hormone Metabolism
influence of rcPRL on plasma concentrations of corticosterone was studied as well.
TABLE 1 Effective Dose (ng) for 50% Displacement of 125I-Labeled cGH by Several Hormone Preparations in the Chicken Embryo and Adult Chicken
Embryo (18 days) Adult hens
rcGH
oGH
cGH
oPRL
rcPRL
0.33 0.69
0.61 1.53
2.06 3.67
6.93 9.08
— —
MATERIALS AND METHODS Eggs of the Hisex White Strain were purchased from a commercial breeder (Euribrid, Aarschot, Belgium) and kept in a forced-draft laboratory incubator under standard conditions.
FIG. 2. Influence of saline and different injection doses of rcPRL (2 and 10 µg) and cGH (0.4, 2, and 10 µg) on plasma concentrations of T3 (A) and T4 (B) (pmol/ml) after 2 hr in 19-day-old chicken embryos. Sal 1 and sal 2, control injections at start and end of experiment. Values are means 6 SEM for 12 individual samples. Groups with different superscripts are significantly different (P , 0.05, DUNCAN).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
352
Ku¨hn et al.
FIG. 3. Influence of saline and different injection doses of rcPRL (2 and 10 µg) and cGH (0.4, 2, and 10 µg) on liver ORD-I (A) and IRD-III (B) activity in 18-day-old chicken embryos. Sal 1 and sal 2, injections at start and end of experiment. Values are means 6 SEM for 12 individual samples. Groups with different superscripts are significantly different (P , 0.05, DUNCAN).
Injections using 19- (first experiment) and 18- (second experiment) day-old embryos were made into one of the peripheral chorioallantoic blood vessels lying close to a window cut in the shell at the lateral side. Blood samples (of 0.5 to 1 ml) were taken by cardiac puncture
and collected in heparinized tubes after 2 hr in a first experiment and after 2, 6, and 24 hr in a second one. In a first series of experiments (n 5 12 for each group) 2 or 10 µg rcPRL or 0.4, 2, or 10 µg of hypophyseal chicken GH (cGH) were injected and all injections were made over
FIG. 4. Plasma concentrations of T3 (A), T4 (B), and rT3 (C) (pmol/ml) of 18-day-old chicken embryos 2, 6, and 24 hr following injection of saline, 2.5 µg of rcPRL and 2.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
354
a period of 1.5 hr starting with saline (sal 1), 0.4 µg GH, 2 µg GH, 2 µg rcPRL, 10 µg GH, 10 µg rcPRL, and again saline (sal 2). Since obviously injection time after the time interval between removal from the incubators and injection may have some influence on results, this experiment was repeated using 2.5 µg of rcPRL and 2.5 µg of cGH but injections were randomized in groups of six. In addition, blood and tissue samples were taken after 2, 6, and 24 hr and the effects on plasma rT3 and corticosterone were also included. Following centrifugation the plasma was kept at 220° until assay of hormones. Livers and kidneys were also excised for deiodinase assays and stored frozen at 280° until homogenization. In vitro hepatic ORD-I and IRD-III activities were measured in microsomal (Mx) preparations as described previously (Darras et al., 1992b). Briefly, hepatic ORD-I activity was measured using 200 µg Mx protein/ml with 1 µM 3,38,58-triiodothyronine (rT3) as substrate and 5 mM dithiothreitol (DTT) as cofactor. Hepatic IRD-III activity was measured using 50 µg Mx protein/ml with 10 nM T3 as substrate in the presence of 1 µM rT3, 50 mM DTT as cofactor, and 1 mM 6-propyl-2-thiouracil (PTU) to block type I interference. For displacement studies several hormone preparations were used to compete with 125I-labeled cGH for binding to liver microsomes of 18-day-old embryos (Hisex White) or to adult hens (Warren). The cGH used in this study and for injections was purified following the method described by Berghman et al. (1988). The recombinant cGH (rcGH) was obtained through the courtesy of Dr. T. Hall (Ciba Geigy, Basel), whereas the rcPRL preparation was made by Ohkubo et al. (1993). Ovine GH (oGH) was obtained from NIH (NIADDK-NIH (oGH-15)) and ovine PRL (oPRL) from Sigma. GH receptors were determined as described by Vanderpooten et al. (1991). Slopes and intercepts of the linearized curves (logit (B/Bo) 5 intercept 1 (slope 3 logdose)) have been calculated by linear regression. These parameters have been used to calculate the effective dose for 50% displacement. Plasma T3, T4, and rT3 were measured by radioimmunoassay (RIA) using antisera for T3 and T4 from Byk-Belga (Belgium) and the antiserum described by Visser et al. (1977) for rT3. [125I]T3, [125I]T4 and, [125I]rT3 were from Amersham (Belgium) and the standards consisted of T3, T4, and rT3 diluted in hormone-free human serum (Byk-Belga).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
Ku¨hn et al.
Corticosterone was measured using a commercially available RIA kit (ICN, Belgium) after validation for use in chicken plasma. For all RIAs chicken plasma dilution tests and loading tests showed good parallelism with the standard curve. Intraassay variabilities were 4.5% (T3), 5.4% (T4), and 6.9% (corticosterone), whereas interassay variabilities were 15, 17, and 7.3%, respectively. The assay of rT3 was based on the instruction as described by Visser et al. (1977). Statistical analysis of the results was made using the GLM procedure (SAS Institute, Inc., 1985) followed by the Duncan multiple range test.
RESULTS GH Displacement Curves in Liver Microsomes (Fig. 1) Displacement studies on hepatic microsomes of 18-dayold chicken embryos and adult hens indicated that parallel inhibition curves could be obtained with cGH, rcGH, and oGH for displacement of 125I-labeled cGH. Ovine PRL could also compete for binding 125I-labeled cGH, but no parallel curve was obtained. The effective doses (ng) for 50% displacement are summarized in Table 1. The recombinant cPRL preparation, however, was totally ineffective in this regard in embryos as well as in adult hens.
Injection Studies In a first series of experiments no difference in both saline injections was noted for plasma T3, plasma T4, and hepatic IRD-III, but liver ORD-I activity was significantly elevated in saline 2. The rcPRL injections decreased plasma concentrations of T4 but had no influence on T3, whereas cGH increased plasma T3 at every injection dose used and decreased T4 at the 2and 10-µg doses (Fig. 2). Liver ORD-I of the sal 1 group was lower than for all other groups including sal 2. These groups however did not differ among each other. Liver IRD-III, however, was increased following injection of both PRL concentrations compared to both saline injections and decreased following injection of all GH doses when compared to sal 1 but only at the 0.4- and 2-µg dose when compared to sal 2 (Fig. 3).
Influence of rcPRL on Thyroid Hormone Metabolism
Results of the second series of experiments indicated that control plasma concentrations of T4 were increased after 6 and 24 hr compared to that at 2 hr, whereas no significant changes were present for rT3 and T3. After 2 hr, 2.5 µg of rcPRL decreased T4 and had no influence on T3. On the contrary, an injection of 2.5 µg of cGH did not affect T4 concentrations, but
355
raised T3 after 2 and 6 hr. This was accompanied by decreases in plasma rT3, a decrease which was also present in the PRL-injected group but here only after 6 hr (Fig. 4). Hepatic ORD-I activities were increased after 24 hr and this increase was more pronounced in the cGH group compared to other groups. Hepatic IRD-III
FIG. 5. Liver ORD-I (A) and IRD-III (B) activity of 18-day-old chicken embryos 2, 6, and 24 hr following injection of saline, 0.5 µg of rcPRL, and 0.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
356
activity increased firmly following injections of 2.5 µg rcPRL after 2 hr, whereas a profound decrease is present following injection of 2.5 µg cGH after 2 and 6 hr (Fig. 5). Plasma corticosterone levels were elevated after 2 hr compared to controls following injection of 2.5 µg of rcPRL. An injection of 2.5 µg cGH however did not influence corticosterone plasma concentrations (Fig. 6).
DISCUSSION The present results indicate that in the chicken embryo rcPRL and cGH have different hepatic receptors and opposite effects on hepatic IRD-III activity. They also differ as to their influence on plasma concentrations of T3. However, both hormones are capable of depressing plasma concentrations of T4 and rT3, and data indicate that these effects are not mediated by any changes in hepatic ORD-I levels. Earlier studies using oPRL as a chicken prolactin substitute have shown that oPRL binds to hepatic GH receptors as well (Leung et al., 1984). This has been
Ku¨hn et al.
confirmed in the present study, which explains why oPRL (Ku¨hn et al., 1983) and cGH (Berghman et al., 1989) have similar effects on plasma concentrations of thyroid hormones. No binding to chicken hepatic GH receptors occurs however with the present rcPRL preparation, confirming results of Burnside and Cogburn (1993) and results on thyroid hormone metabolism differ profoundly between both hormones. The observed effects following injection of cGH in the chicken embryo confirm previous observations (Darras et al., 1992a; Berghman et al., 1989). Briefly, in control animals hepatic IRD-III activity decreases more than 10-fold between Day 17 and hatching, whereas the hepatic ORD-I activity is increasing slowly up to the period of pipping and hatching (Darras et al., 1992b). This is accompanied by a sudden increase in plasma concentrations of T3 at the end of incubation. cGH, when injected during this period, provokes a profound decrease in hepatic IRD-III, without changing in vitro ORD-I activity. The concurrent marked increase in plasma concentrations of T3 and decrease of rT3 is likely to be the result of the above enzymatic activities (Darras et al., 1990, 1992a).
FIG. 6. Plasma corticosterone concentrations (ng B/ml) in 18-day-old chicken embryos 2, 6, and 24 hr following injections of saline, 0.5 µg of rcPRL, and 0.5 µg of cGH. Values represent means 6 SEM for 12 individuals. Means of control groups with the same letter are not significantly different (P , 0.05, DUNCAN within each time group).
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
357
Influence of rcPRL on Thyroid Hormone Metabolism
It has been shown that endogenous GH is an important factor in the control of plasma T3 levels in growing chickens due to its influence on the activity of the T3-degrading type III deiodinase (Darras et al., 1993). On the other hand, growth rate is closely related to plasma concentrations of T3 and growth retardation as a result of a lower incubation temperature is paralleled by ‘‘retarded’’ T3 levels (Ku¨hn et al., 1982). In the present study PRL is capable of stimulating the hepatic T3-degrading activity (though plasma levels of T3 were not affected) and it is remarkable that in the earlier experiments on chicken embryos incubated at a lower temperature higher levels of prolactin (and lower levels of T3) were found immediately after hatching together with the observed growth inhibition (Ku¨hn et al., 1985). However, together with an increase in the hepatic IRD-III activity following injections of rcPRL, no effect on plasma T3 was observed and lower T4 and rT3 levels were found. These results are difficult to explain since one would expect decreased plasma T3 and possibly T4 concentrations together with an increase in rT3. It is possible that the effect of PRL on thyroid metabolism is complicated by interference with the adrenal axis. In mammals the adrenal cortical tissue is known to have high concentrations of PRL receptors (Horrobin, 1977) and in the chicken an increase in mitogenic activity of the adrenals was observed following PRL injections (Maiti and Chakraborti, 1980). It is also known that oPRL counteracts the corticosteroneinduced suppression of net ACTH-stimulated corticosterone production of intact domestic fowl adrenocortical cells. However, oPRL does not stimulate basal or ACTH-induced corticosterone synthesis (Carsia et al., 1987). The involvement of the adrenal axis in thyroid metabolism is well-known (Wada et al., 1975; Ku¨hn et al., 1983, 1985; Decuypere et al., 1983) and recent studies have extended the earlier observations on the histology of the thyroid gland and circulating thyroid hormones with new observations on hepatic and renal deiodinating enzymes (Geris et al., 1995; Darras et al., 1995). These results indicate that corticosterone has no short-term influence on ORD-I but decreases IRD-III and plasma concentrations of rT3 and T4 and increases plasma T3. The present observation that PRL is capable of increasing corticosterone concentrations in the plasma of 18-day-old embryos seems to support the
hypothesis that part of the present results may be explained by interference of prolactin with adrenal secretion. This way the suppressive effect of rcPRL on plasma concentration of T4 and rT3 may be mediated by the release of corticosterone.
ACKNOWLEDGMENTS We thank Mrs. F. Voets and L. Noterdaeme and Mr. W. Van Ham for their valuable technical assistance. We thank Dr. T. Hall (CibaGeigy, Basel) for the gift of recombinant cGH and Dr. T. Visser (Erasmus University, Rotterdam, The Netherlands) for the gift of rT3 antiserum. V. Darras and L. Berghman were supported by the National Fund for Scientific Research. This work was also supported by grants from the National Fund for Scientific Research (2.0114.90 and 2.0114.94).
REFERENCES Berghman, L. R., Darras, V. M., Huybrechts, L. M., Decuypere, E., Vandesande, F., and Ku¨hn, E. R. (1989). Evidence for chicken GH as the only hypophyseal factor responsible for the stimulation of the hepatic 58-monodeiodination activity in the chick embryo. Reprod. Nutr. Dev. 29, 197–202. Berghman, L. R., Van Beeumen, J., Decuypere, E., Ku¨hn, E. R., and Vandesande, F. (1988). One-step purification of chicken growth hormone from a crude pituitary extract by use of a monoclonal immunoadsorbent. J. Endocrinol. 118, 381–387. Burnside, J., and Cogburn, L. A. (1993). Molecular biology of the chicken growth hormone receptor. In ‘‘Avian Endocrinology’’ (P. J. Sharp, Ed.), pp. 161–176. Journal of Endocrinology Ltd., Bristol, UK. Carsia, R. V., Scanes, C. G., and Malamed, S. (1987). Polyhormonal regulation of avian and mammalian corticosteroidogenesis in vitro. Comp. Biochem. Physiol. A 88, 131–140. Darras, V. M., Berghman, L. R., Vanderpooten, A., and Ku¨hn, E. R. (1992a). Growth hormone acutely decreases type III iodothyronine deiodinase in chicken liver. FEBS Lett. 300, 5–8. Darras, V. M., Huybrechts, L. M., Berghman, L. R., Ku¨hn, E. R., and Decuypere, E. (1990). Ontogeny of the effect of purified chicken growth hormone on the liver 58-monodeiodination activity in the chicken: Reversal of the activity after hatching. Gen. Comp. Endocrinol. 77, 212–220. Darras, V. M., Kotanen, S., Geris, K. L., and Ku¨hn, E. R. (1995). Differential effects of hormones of the adrenal axis on thyroid function in embryonic chicken. Abstract 497, 11th International Thyroid Congress, September 10–15, Toronto, Canada. Darras, V. M., Rudas, P., Visser, T. J., Hall, T. R., Huybrechts, L. M., Vanderpooten, A., Berghman, L. R., Decuypere, E., and Ku¨hn, E. R. (1993). Endogenous growth hormone controls high plasma levels
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
358 of 3,38,58-triiodothyronine (T3) in growing chickens by decreasing the T3-degrading type III deiodinase activity. Dom. Anim. Endocrinol. 10, 55–65. Darras, V. M., Visser, T. J., Berghman, L. R., and Ku¨hn, E. R. (1992b). Ontogeny of type I and type III deiodinase activity in embryonic and posthatch chicks: Relationship with changes in plasma triiodothyronine and growth hormone levels. Comp. Biochem. Physiol. A 103, 131–136. Decuypere, E., Scanes, C. G., and Ku¨hn, E. R. (1983). Effects of glucocorticoids on circulating concentrations of thyroxine (T4) and triiodothyronine (T3) and on peripheral monodeiodination in preand post-hatching chickens. Horm. Metab. Res. 15, 233–236. Geris, K. L., Darras, V. M., Berghman, L. R., and Ku¨hn, E. R. (1995). Influence of corticotropin-releasing factor on the in vitro thyroxine and thyrotropin secretion in newly hatched fowl. Belg. J. Zool. 125, 143–156. Harvey, S., Davison, T., and Chadwick, A. (1979). Ontogeny of growth hormone and prolactin secretion in the domestic fowl (Gallus domesticus). Gen. Comp. Endocrinol. 39, 270–273. Horrobin, D. F. (1977). In ‘‘Prolactin,’’ (D. F. Horrobin, Ed.), Vol. 5, pp. 105–108. Annual Research Reviews. Livingstone, Edinburgh. Ishida, H., Shimada, K., Sato, K., Seo, H., Matsui, N., and Zadworny, D. (1991). Developmental expression of the prolactin gene in the chicken. Gen. Comp. Endocrinol. 83, 463–467. Ku¨hn, E. R., Decuypere, E., Colen, L. M., Chadwick, A., Heyns, W., Michels, H., and Berghman, L. R. (1985). Circadian rhythms of corticosterone, prolactin and iodohormone secretion in the posthatch chick, and their influence on growth. In ‘‘Current Trends in Comparative Endocrinology’’ (B. Lofts and W. N. Holmes, Eds.), pp. 671–675. Hong Kong Univ. Press, Hong Kong. Ku¨hn, E. R., Decuypere, E., Colen, L. M., and Michels, H. (1982). Posthatch growth and development of a circadian rhythm for
Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
Ku¨hn et al.
thyroid hormones in chicks incubated at different temperatures. Poultry Sci. 61, 540–549. Ku¨hn, E. R., Decuypere, E., Hemschoote, K., Berghman, L., and Paulussen, J. (1983). Antagonism of serum tri-iodothyronine changes after injection of prolactin in the domestic fowl before and after hatching. J. Endocrinol. 99, 401–407. Leung, F. C., Taylor, J. E., Steelman, S. L., Bennett, C. D., Rodkey, J. A., Long, R. A., Serio, R., Weppelman, R. N., and Olson, G. (1984). Purification and properties of chicken growth hormone and the development of a homologous radioimmunoassay. Gen. Comp. Endocrinol. 56, 389–400. Maiti, B. R., and Chakraborty, S. (1980). The mitogenic action of prolactin on some endocrine organs of chicks. Arch. Histol. Jap. 43, 341–346. Ohkubo, T., Tanaka, M., Nakashima, K., Shimada, K., Saito, N., and Sato, K. (1993). High level expression of biological active chicken prolactin in E. coli. Comp. Biochem. Physiol. A 105, 123–128. SAS Institute, Inc. (1985). ‘‘SAS User’s Guide: Statistics Version 5 Edition.’’ SAS Institute Inc. Cary, NC. 956 pp. Shimada, K., Ohkubo, T., Saito, N., Talbot, R. T., and Sharp, P. J. (1993). The molecular biology of prolactin. In ‘‘Avian Endocrinology’’ (P. J. Sharp, Ed.), pp. 135–147. Journal of Endocrinology Ltd., Bristol, UK. Vanderpooten, A., Darras, V. M., Huybrechts, L. M., Rudas, P., Decuypere, E., and Ku¨hn, E. R. (1991). Effect of hypophysectomy and acute administration of growth hormone (GH) on GHreceptor binding in chick liver membranes. J. Endocrinol. 129, 275–281. Visser, T. J., Docter, R., and Hennemann, G. (1977). Radioimmunoassay of reverse triiodothyronine. J. Endocrinol. 73, 395–396. Wada, M., Arimatsu, Y., and Kobayashi, H. (1975). Effect of prolactin on thyroid function in Japanese quail (Coturnix coturnis japonica). Gen. Comp. Endocrinol. 27, 28–33.