- Citado por SciELO
versão impressa ISSN 0100-8455versão On-line ISSN 1678-4502
Braz. J. Genet. v. 20 n. 4 Ribeirão Preto Dez. 1997
Prolactin inhibits auto- and cross-induction of thyroid hormone and estrogen receptor and vitellogenin genes in adult Xenopus (Amphibia) hepatocytes
Elida M.L. Rabelo1 and Jamshed R. Tata 2
1Departamento de Parasitologia, ICB, UFMG, Av. Antônio Carlos, 6627, 31270-901 Belo Horizonte, MG, Brasil. Send correspondence to E.M.L.R.
2 Laboratory of Developmental Biochemistry, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, UK.
It is well known that virtually every tissue of the amphibian larvae is highly sensitive to the mutually antagonistic actions of thyroid hormone (TH) and prolactin (PRL), but it is not known if adult amphibian tissues respond similarly to these two hormones. We have previously shown that very low doses of triiodothyronine (T3) rapidly and strongly potentiate the activation of silent vitellogenin (Vit) genes by estrogen (E2) and the autoinduction of estrogen receptor (ER) transcripts in primary cultures of adult Xenopus hepatocytes. This response to T3 is accompanied by the upregulation of thyroid hormone receptor b (TRb) mRNA. Using Northern blot and RNase protection assays, we now show that ovine PRL added for 12 h along with 2 x 10-9 M T3 will completely prevent potentiation of E2 induction of Vit mRNA in primary cultures of adult Xenopus hepatocytes. PRL also abolished the auto-upregulation of TRb mRNA and the cross-activation of autoinduction of ER mRNA. Thus, we show for the first time that the anti-TH action of PRL that is manifested in Xenopus tadpole tissues during metamorphosis is retained in adult liver, and suggest that the mutually antagonistic actions of the two hormones may be brought about by similar molecular mechanisms in larval and adult amphibian tissues.
Whereas virtually every tissue of amphibian larvae is a target for thyroid hormones (TH), there is no evidence that these hormones have any effect on adult amphibian tissues. However, we were able to demonstrate, in a previous work, that very low doses of triiodothyronine (T3) strongly potentiated vitellogenin (Vit) gene activation by estrogen (E2) in primary cultures of adult Xenopus hepatocytes (Rabelo and Tata, 1993). This unequivocal response of adult cells to T3 was later found to be reproducible in Xenopus tadpole liver (Rabelo et al., 1994).
One major characteristic of amphibian metamorphosis is inhibition or retardation of natural or TH-induced development by homologous or heterologous preparations of prolactin (PRL) (Beckingham Smith and Tata, 1976; Gilbert and Frieden, 1981; White and Nicoll, 1981; Kikuyama et al., 1993; Tata, 1993). This "juvenilizing" action of prolactin has proved to be useful in analyzing some of the mechanisms underlying both morphogenesis and extensive cell death that occur during metamorphosis. However, until now the ability of this hormone to block thyroid hormone action has only been established in larval amphibian tissues during postembryonic development, and it is not known if similar interaction between the two hormones occurs in adult tissues. In view of our previous observ tion of the rapid potentiation by T3 of E2-induced vitellogenin gene activation in adult Xenopus hepatocytes (Rabelo and Tata, 1993), we decided to explore the possibility of a PRL-T3 interaction in adult amphibian tissue.
The present study reports that the anti-TH action of PRL exhibited during late larval development is retained in adult Xenopus liver. By using primary cell cultures of adult male Xenopus liver, we confirmed that low doses of T3 (2 x 10-9 M) strongly and rapidly potentiated silent vitellogenin gene activation by estrogen while, at the same time, induced transcripts of its own receptor TRb and enhanced autoinduction of an estrogen receptor. When together with T3 and E2 in the culture medium, prolactin abolished T3-enhanced accumulation of all the above transcripts. Thus, we show, for the first time, that an adult amphibian tissue is responsive to both prolactin and thyroid hormone. It is most likely that adult tissues follow the same mechanism as larval tissues.
MATERIAL AND METHODS
3,3,5-Triiodo-L-thyronine Na salt was synthesized at the National Institute for Medical Research (London, UK), Estradiol-17b was purchased from Sigma Chemical Co. (Poole, UK), and ovine prolactin was a gift from the National Hormone and Pituitary Program, National Institutes of Health (Rockville, MD). For labelling complementary DNA (cDNA) and RNA (cRNA) probes, [a-32P]dCTP (3000 Ci/mmol) and [a-32P]UTP (450 Ci/mmol), respectively, were purchased from Amersham International (Amersham, UK). All other materials were of analytical grade and purchased from Sigma Chemical Co. (Poole, UK).
Primary cell cultures
Adult male Xenopus were purchased from Blade Biological (Cowden, Kent, UK) and maintained in our amphibian facility, as described earlier (Kawahara et al., 1991). Primary liver cultures were prepared following standard laboratory procedures (Perlman et al., 1984), where the cells were maintained in culture for 64 h before hormonal manipulation. Each culture dish had cells from the equivalent of ~800 mg of livers pooled from 8 male frogs. T3 and PRL were added to the cell cultures in different sequences 12 h before E2. After a further 12 h in culture, hepatocytes were washed, and RNA was extracted.
Total RNA was extracted from batches of 2 dishes of hepatocytes following the guanidinium isothiocyanate procedure (Chomczymski and Sacchi, 1987). After its purity was checked by gel electrophoresis and spectroscopy, RNA was analyzed either by Northern blot or RNase protection assay. Vitellogenin mRNA was measured by Northern blot with 32P- labelled Xenopus vitellogenin B1 cDNA (Baker and Tata, 1990). The blots were also probed with a 32P-labelled Xenopus actin cDNA as loading controls. TRb and ER mRNAs were detected and estimated by RNase protection assay with 32P-labelled riboprobes. For this purpose a 325-nt TRb fragment (Kawahara et al., 1991) and an ER cRNA derived from a 167-nt BglII-HindIII cDNA fragment (Weiler et al., 1987) were used as 32P- labelled cRNA probes. A 120-nt Xenopus 5S RNA cRNA was used as a loading control for RNase protection assay. Autoradiograms of Northern blot and protection assays were scanned and quantified in a Molecular Dynamics Imagequant.
The signal from each sample was normalized to actin mRNA or 5S RNA and expressed as arbitrary units.
RESULTS AND DISCUSSION
Silent vitellogenin gene activation in male Xenopus hepatocytes by exogenous estrogen added to primary cell cultures has allowed a more precise analysis of gene expression regulation by steroid hormones than is possible in intact animals (Wangh and Schneider, 1982; Perlman et al., 1984; Shapiro et al., 1989; Tata, 1991). Recently, we have shown that T3 added only a few hours before E2 strongly potentiates vitellogenin gene activation (Rabelo and Tata, 1993). Both these hormonal effects have been reproduced and the results are showed in Figure 1.
Exposure of male hepatocytes to 2 x 10-9 M T3 for 12 h prior to addition of 10-8 M E2 for 12 h resulted in strong Vit mRNA accumulation (lanes 2 and 3). Autoradiogram scanning showed that the increase was 4-fold. Pre-exposure of cell to PRL (0.5 IU/ml culture media) before T3 and E2 or together with T3 abolished or substantially diminished the T3 effect. (Figure 1, lanes 6 and 8). In data not shown, 0.2 IU PRL had the same inhibiting effect as 0.5 IU PRL. It acted much the same way as shown earlier for organ cultures of Xenopus tadpole tails (Baker and Tata, 1992). However, PRL did not inhibit the T3 effect in adult liver primary cell cultures, if it was not maintained throughout the duration of the experiment, i.e., included with E2 during the last 12 h of the culture period (Figure 1, lanes 5 and 7). This finding may either reflect PRL instability in our cultures, or an acceleration of Vit mRNA breakdown caused by PRL. The latter is improbable because PRL did not significantly diminish the modest amount of Vit mRNA in hepatocytes treated for 12 h with E2 without being pre-treated with T3 (Figure 1, lane 4). Thus, the most noticeable action of PRL is directed against the potentiation by T3 of E2 action and not against the induction of Vit genes by E2 (the genes remain silent when T3 is added without E2 (Rabelo and Tata, 1993).
Hitherto, PRL action has only been demonstrated in amphibian larval stages when undergoing TH-dependent metamorphosis (Gilbert and Frieden, 1981; Kikuyama et al., 1993; Tata, 1993). We have previously found in Xenopus tadpoles that in inhibiting T3-induced metamorphosis, PRL abolished the autoinduction of TR (Baker and Tata, 1992). In view of the above results concerning T3-antagonist effect of PRL in adult Xenopus hepatocytes, we wondered if PRL has the same effect on adult tissues regarding nuclear receptor mRNA induction as it does in tadpoles, i.e., if the PRL effect on Vit mRNA would also be accompanied by the inhibition of auto- and cross-induction of ER and TR by E2 and T3. The RNase protection assay (Figure 2, lanes 2 and 3) confirms the ability of E2 to upregulate its own receptor transcripts in primary cultures of male Xenopus hepatocytes. Pre-treatment of the cells for 12 h with 2 x 10-9 M or 10-7 M T3 potentiates ER mRNA autoinduction (compare lanes 4 and 6 with lane 3). Prolactin inhibited potentiation by 2 x 10-9 M T3 of estrogen action (lane 5). Quantitatively, the 4-fold potentiation by 2 x 10-9 M T3 of ER autoinduction was completely eliminated through continuous PRL presence (Table I). PRL, without T3 pre-treatment, did not inhibit the autoinduction of ER (Figure 2, lane 8). Interestingly, raising T3 concentration to 10-7 M overcame inhibition by PRL (lanes 6 and 7). Since the cellular response to PRL remains largely unknown (Kelly, 1990; Kelly et al., 1991), it is difficult to explain the dose-related competition between this hormone acting via membrane receptors and T3 which acts via nuclear receptors (Chin, 1991; Chatterjee and Tata, 1992). On the other hand, this phenomenon may explain why rising levels of thyroid hormones overcome the antimetamorphic action of prolactin during natural metamorphosis (Kikuyama et al., 1993). As for the accumulation of ER mRNA, PRL had to be added together with E2 in order to observe this effect. PRL inhibition of T3 cross-induction of ER is relatively significant in the context of de novo activation of Vit genes in male Xenopus liver by E2. Earlier studies of the close association between the upregulation of functional ER and transcription in vivo of Vit genes had suggested that the basal level of the receptor was not sufficient to activate Vit genes, but that additional ER was necessary for the induction of Vit mRNA (Perlman et al., 1984; Shapiro et al., 1989; Tata et al., 1993). Our present data are fully compatible with this suggestion.
Table I - Prolactin abolishes both TRb mRNA autoinduction and T3 enhancement of ER autoinduction in primary cultures of adult Xenopus hepatocytes.
T3 + PRL/E2 + PRL
Where indicated, the hepatocytes were exposed to 2 x 10-9 M T3 with or without 0.5 IU/ml PRL for 12 h prior to the addition of 10-8 M E2. After a further 12 h of incubation, total RNA was extracted and the relative concentration of TRb and ER mRNAs measured by scanning autoradiograms obtained from RNase protection assays. All other details found in Figures 2, 3 and in the text.
There is now increasing evidence from several laboratories that many receptors, and their transcripts, of the steroid/thyroid hormone/retinoic acid receptor family are autoinduced by their own ligands, particularly in a developmental context (Tata et al., 1993). Previous studies from our laboratory (Baker and Tata, 1992; Rabelo and Tata, 1993) have focused on the autoinduction of Xenopus TRa and b mRNA in larval tissues and in primary cultures of adult male Xenopus hepatocytes. Since the autoinduction of TRb mRNA is more pronounced than that of a isoform in the above tissues, we restricted the present study to measure only the steady-state levels of TRb mRNA as a function of the treatment of primary hepatocyte cultures with different combinations of the three hormones. As shown in Figure 3, T3 produced a strong upregulation of TRb mRNA (compare lanes 3 and 5 with 1 and 2), confirming an earlier observation (Rabelo and Tata, 1993). E2 did not alter the amount of TRb mRNA nor the extent of its autoinduction (compare lanes 1, 2, 3 and 5). however, if PRL was added in the same sequence as in the experiments shown in Figures 1 and 2, TRb mRNA was not upregulated by T3 (lane 4). The magnitude of autor cross-induction (or de-induction) of ER and TRb transcripts is depicted in Table I. This shows that the 4-fold induction of TRb mRNA by T3, irrespective of whether or not E2 was added, was almost completely eliminated by PRL. When the numbers in Table I are compared with the autoradiographic signals for Vit gene activation in Figure 1, it is obvious that the latter is closely associated with the auto- and cross-regulation of TR and ER gene regulation.
The present results are particularly significant when considering T3 and PRL action in larval and adult amphibia. Until very recently, both hormones have been recognized almost exclusively for the important role they play in amphibian metamorphosis, particularly in anurans (Beckingham Smith and Tata, 1976; Gilbert and Frieden, 1981; Kikuyama et al., 1993; Tata, 1993). However, there was no convincing evidence that low doses of TH had any effect on adult amphibian tissues nor that PRL would suppress its action. Studies on the effects of these hormones directly on primary cultures of adult hepatocytes allow one to show that the expression of three different genes, namely Vit, TRb and ER, responds the same way in adult and larval livers. Previously, Wangh found in longterm maintenance of primary cultures of adult Xenopus hepatocytes that 2-week exposure to T3 enhanced Vit induction by E2 (Wangh, 1982; Wangh and Schneider, 1982). What is so striking about our present results and previous findings (Rabelo and Tata, 1993) is the rapidity and extent of potentiation of E2 action by T3. It is difficult to say whether or not the acute and long-term responses to T3 are brought about by the same mechanisms, so it would be of considerable interest to determine what effect PRL would have on long-term cultures and in adult Xenopus in vivo.
Finally, our results highlight the general precept of endocrinology and cellular signalling that hormonal interplay is particularly important for the regulation of developmental processes and specific gene expression (Gorbman and Bern, 1962; Gilbert and Frieden, 1981; Tata, 1984; Baulieu and Kelly, 1990). More specifically, they reinforce the participation of thyroid hormone in the multihormonal networks that regulate diverse physiological processes (Pitt-Rivers and Tata, 1959; Gorbman and Bern, 1962; Oppenheimer and Samuels, 1983; Chin, 1991; Tata, 1993). Prolactin is also known to similarly participate in many developmental processes, particularly lactation and reproductive functions, via mechanisms that are not fully understood (White and Nicoll, 1981; Kelly, 1990). What our present data do establish is that now one has to seriously consider the possibility that both thyroid hormone and prolactin may play a role in adult amphibian life as, say, in reproductive functions. They also offer a novel approach to the analysis of how these important hormones, which have been put to diverse uses through evolution, exert their action as participants in hormonal networks that control morphogenesis and cell death.
We wish to thank Mrs. Betty Baker for help with many aspects of experimental work. E.M.L.R. was the recipient of a CAPES predoctoral fellowship from the Brazilian Government during the execution of this work.
É bem estabelecido que virtualmente todos os tecidos das larvas de anfíbios são altamente sensíveis à ação mutuamente antagonista dos hormônios tireoideanos (TH) e prolactina (PRL). Porém, não é sabido se os tecidos do anfíbio adulto respondem de uma forma semelhante a estes hormônios. Em um trabalho anterior, nós demonstramos que baixas doses de triiodotironina (T3) potenciam rapidamente e com alta intensidade a ativação pelo estrogênio (E2) de genes silenciosos da vitelogenina (Vit), bem como a autoindução de transcritos de receptores para estrogênio (ER) em culturas primárias de hepatócitos de Xenopus adultos. Esta resposta para o T3 é acompanhada pela ativação da transcrição do mRNA de receptores b para o hormônio tireoideano (TRb). Através das técnicas de Northern blotting e "RNase protection assay", nós agora estamos mostrando que prolactina ovina adicionada por 12 h, juntamente com 2 x 10-9 M de T3, em culturas primárias de hepatócitos adultos de Xenopus bloqueia completamente a ação de T3 na potenciação da indução de mRNA da Vit pelo hormônio E2. PRL também aboliu a autoregulação positiva do mRNA para TRb e a ativação cruzada da autoindução do mRNA do ER. Assim, nós estamos mostrando pela primeira vez que a ação anti-TH da PRL, manifestada em girinos de Xenopus durante a metamorfose, é mantida no fígado adulto, sugerindo que esta ação mutuamente antagonista dos dois hormônios pode ser desempenhada por mecanismos moleculares semelhantes nos tecidos de ambos os estágios do anfíbio.
Baker, B.S. and Tata, J.R. (1990). Accumulation of protooncogene c-erb-a related transcripts during Xenopus development: association with early acquisition of response to thyroid hormone and estrogen. EMBO J. 9: 879-885. [ Links ]
Baker, B.S. and Tata, J.R. (1992). Prolactin prevents the auto- induction of thyroid hormone receptor mRNAs during amphibian metamorphosis. Dev. Biol. 149: 463-467. [ Links ]
Baulieu, E.-E. and Kelly, P.A. (Eds.) (1990). Hormones: From Molecules to Disease. Hermann, Paris. [ Links ]
Beckingham Smith, K. and Tata, J.R. (1976). Amphibian metamorphosis. In: Developmental Biology of Plants and Animals (Graham, C.F. and Wareing, P.F., eds.). Blackwell, Oxford, pp. 232-245. [ Links ]
Chatterjee, V.K.K. and Tata, J.R. (1992). Thyroid hormone receptors and their role in development. Cancer Surv. 14: 147-167. [ Links ]
Chin, W.W. (1991). Nuclear thyreoid hormone receptors. In: Nuclear Hormone Receptors (Parker, M., ed.). Academic Press, London, pp. 79-102. [ Links ]
Chomczymski, P. and Sacchi, P. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanatephenol-chloroform extraction. Anal. Biochem. 162: 156- 159. [ Links ]
Gilbert, L.I. and Frieden, E. (Eds.) (1981). Metamorphosis: A Problem in Developmental Biology. Plenum Press, New York. [ Links ]
Gorbman, A. and Bern, A. (1962). A Textbook of Comparative Endocrinology. John Wiley, New York. [ Links ]
Kawahara, A., Baker, B.S. and Tata, J.R. (1991). Developmental and regional expression of thyreoid hormone receptor genes during Xenopus metamorphosis. Development 112: 933-943. [ Links ]
Kelly, P.A. (1990). Growth hormone and prolactin. In: Hormones: From Molecules to Disease (Baulieu, E.-E. and Kelly, P.A., eds.). Hermann, Paris, pp. 191-217. [ Links ]
Kelly, P.A., Djiane, J., Postel-Vinay, M.-C. and Edery, M. (1991). The prolactin/growth hormone family. Endocr. Rev. 12: 235-251. [ Links ]
Kikuyama, S., Kawamura, K., Tanaka, S. and Yamamoto, K. (1993). Aspects of amphibian metamorphosis: hormonal control. Int. Rev. Cytol. 145: 105-148.
Oppenheimer, J.H. and Samuels, H.H. (Eds.) (1983). Molecular Basis of Thyroid Hormone Action. Academic Press, New York. [ Links ]
Perlman, A.J., Wolffe, A.P., Champion, J. and Tata, J.R. (1984). Regulation by estrogen receptor of vitellogenin gene transcription in Xenopus hepatocyte cultures. Mol. Cell. Endocrinol. 38: 151-161. [ Links ]
Pitt-Rivers, R. and Tata, J.R. (1959). The Thyroid Hormones. Pergamon Press, London. [ Links ]
Rabelo, E.M.L. and Tata, J.R. (1993). Thyroid hormone potentiates estrogen activation of vitellogenin genes and autoinduction of estrogen receptor in adult Xenopus hepatocytes. Mol. Cell. Endocrinol. 96: 37-44. [ Links ]
Rabelo, E.M.L., Baker, B.S. and Tata, J.R. (1994). Interplay between thyroid hormone and estrogen in modulating expression of their receptor and vitellogenin genes during Xenopus metamorphosis. Mech. Dev. 45: 49-57. [ Links ]
Shapiro, D.J., Barton, M.C., McKearin, D.M., Chang, T.-C., Lew, D., Blume, J., Nielsen, D.A. and Gould, L. (1989). Oestrogen regulation of gene transcriptions and mRNA stability. Recent Prog. Horm. Res. 45: 29-58. [ Links ]
Tata, J.R. (1984). The action of growth and developmental hormones. In: Biological Regulation and Development (Goldberger, R.F. and Yamamoto, K.R., eds.). Plenum Publishing, New York, Vol. 3B, pp. 1-58. [ Links ]
Tata, J.R. (1991). Hormonal and developmental regulation of Xenopus estrogen receptor and egg protein gene expression. In: The New Biology of Steroid Hormones (Hochberg, R.B. and Naftolin, F., eds.). Serono Symposium Publications, Raven Press, New York, Vol. 74, pp. 213-225. [ Links ]
Tata, J.R. (1993). Gene expression during metamorphosis: an ideal model for post-embryonic development. BioEssays 15: 239-248. [ Links ]
Tata, J.R., Baker, B.S., Machuca, I., Rabelo, E.M.L. and Yamauchi, K. (1993). Autoinduction of nuclear receptor genes and its significance. J. Steroid Biochem. Mol. Biol. 46: 105-119. [ Links ]
Wangh, I.J. (1982). Glucocorticoids act together with estrogens and thyroid hormones in regulating the synthesis and secretion of Xenopus vitellogenin, serum albumin, and fibrogen. Dev. Biol. 89: 294-298. [ Links ]
Wangh, L.J. and Schneider, W. (1982). Thyroid hormones are corequisites for estradiol-17b in vitro induction of Xenopus vitellogenin synthesis and secretion. Dev. Biol. 89: 287-293. [ Links ]
Weiler, I.J., Lew, D. and Shapiro, D.J. (1987). The Xenopus laevis estrogen receptor: Sequence homology with human and avian receptors and identification of multiple estrogen receptor messenger ribonucleic acids. Mol. Endocrinol. 1: 355-362. [ Links ]
White, B.A. and Nicoll, C.S. (1981). Hormonal control of amphibian metamorphosis. In: Metamorphosis. A Problem in Developmental Biology (Gilbert, L.I. and Frieden, E., eds.). Plenum Press, New York, pp. 363-396. [ Links ]
(Received June 3, 1997)
Figure 1 - Prolactin inhibits the potentiation by T3 of E2 induced vitellogenin mRNA in primary cultures of adult Xenopus hepatocytes. Vitellogenin genes were activated by the addition of 10-8 M E2 to the cells for the last 12 h of incubation (lanes 2-8), following different pre-treatments for 12 h with 2 x 10-9 M T3 and 0.5 IU/ml PRL alone or together (lanes 3-8), as indicated below. At the end of the culture period, total RNA was extracted from the hepatocytes and the amount of Vit mRNA accumulated was determined with a Northern blot, using as probes a Xenopus vitellogenin B1 cDNA, and a Xenopus cytoplasmic actin cDNA as loading control. The autoradiogram of the filter blot is shown here. Lane 1: Control (no hormonal treatment); lane 2: E2 alone for the last 12 h. Pre-treatment before adding E2: lane 3: 2 x 10-9 M T3 for 12 h only; lane 4: PRL only for 12 h; lane 5: PRL for 12 h, then PRL + T3 for 12 h; lane 6: PRL for 12 h, then PRL + T3 for 12 h and PRL with E2 for the last 12 h. Lane 7: PRL + T3 together for 12 h; lane 8: PRL + T3 for 12 h, then PRL only with E2 for the last 12 h
Figure 2 - PRL inhibition of T3 potentiation of ER mRNA autoinduction in primary cultures of adult Xenopus hepatocytes. Liver cells were incubated for the last 12 h with 10-8 M E2 (lanes 3-8) following pre-treatment with 2 x 10-9 M or 10-7 M T3 and 0.5 IU/ml PRL in different combinations (lanes 4-8). Total RNA was extracted and the amount of ER mRNA determined by RNase protection assays with a 167-nt Xenopus ER cRNA probe and a 384-nt Xenopus 5S RNA cRNA as a loading control. Major protected bands corresponding to ER mRNA and 5S RNA are indicated by arrowheads. Lane 1: tRNA; lane 2: control (no hormone added); lane 3: E2 only for the last 12 h; lane 4: pre-treated with 2 x 10-9 M T3 for 12 h before E2 for another 12 h; lane 5: PRL + 2 x 10-9 M T3 for 12 h before PRL + E2 for the next 12 h; lane 6: 10-7 M T3 for 12 h followed by E2 for 12 h; lane 7: PRL + 10-7 M T3 for 12 h followed by PRL + E2 for 12 h; lane 8: PRL + E2 for 12 h.
Figure 3 - RNase protection assay showing that PRL prevents the autoinduction of TRb mRNA in primary cultures of adult Xenopus hepatocytes. The cells were incubated for the last 12 h with 10-8 M E2 (lanes 2-4) or 2 x 10-9 M T3 (lane 5) preceded or not by pre-treatment for 12 h with 2 x 10-9 M T3 and 0.5 IU/ml PRL. Total RNA was extracted and probed with Xenopus TRb cRNA, and a 5S cRNA as loading control. The cRNA probes and the protected bands of TRb mRNA and 5S RNA (arrowheads) are marked on the autoradiogram. Lane 1: Control (no hormone added); lane 2: E2 alone for the last 12 h; lane 3: pre-treated with T3 for 12 h before incubation with E2; lane 4: PRL + T3 for 12 h followed by PRL + E2 for 12 h; lane 5: T3 alone for the last 12 h; lane 6: tRNA control.