| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre d’Investigacions en Bioquimica i Biologia Molecular (CIBBIM) (N.T., M.S., N.R., A.M.), Institut de Recerca Vall d’Hebron, Pg. Vall d’Hebron 119-129, 08035 Barcelona, Spain; and Departament de Biologia Molecular i Cellular (J.B.), Institut de Biologia Molecular de Barcelona, Consejo Superior de Investigaciones Cientificas, Parc Científic de Barcelona, 08028 Barcelona, Spain
Address all correspondence and requests for reprints to: Anna Meseguer, Ph.D., Centre d’Investigacions en Bioquimica i Biologia Molecular (CIBBIM), Institut de Recerca Vall d’Hebron, Vall d’Hebron 119-129, 08035 Barcelona, Spain. E-mail: ameseguer{at}vhebron.net.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ß levels that, in turn, control KAP expression in PCT cells in a developmentally dependent manner. We demonstrated that these factors bind to sites in the proximal KAP promoter, thereby collaborating with androgens for full KAP expression. Finally, we propose that TH and GH, acting through CCAAT/enhancer binding protein, may constitute a general regulatory mechanism of androgen-dependent genes in mouse kidney. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Our laboratory has been interested in understanding the regulation of KAP mRNA that displays a tissue- and cell-specific expression restricted to epithelial cells of proximal convoluted tubules in mouse kidney (7). This gene is subjected to strict and exquisite regulation by thyroid and sexual steroid hormones in different segments of proximal tubules. In vivo studies have shown that S3 cells (pars recta), which correspond to the outer stripe of the outer medulla, express the gene in the presence of thyroid hormone (TH/T3) (8), and are also sensitive to androgen and estrogen action (9). Females and castrated males express the gene exclusively in S3 cells (7). Segments S2 and S1 (pars convoluta), which correspond to the cortical part of the kidney, express the gene only in the presence of androgens and a functional androgen receptor (AR); therefore, females, castrated males or AR-deficient mice of the Tfm/Y strain are negative for cortical expression of KAP mRNA (10). Accordingly, androgen replacement in castrated males or treatment of females with testosterone induces expression in the cortical segments of the tubule (7). We also observed that male mice genetically deficient for TH (hyt/hyt strain) exhibit lower cortical KAP mRNA expression than wild-type control males and that exogenous T3 injection prompts full expression recovery (8). Moreover, treatment of adult male control mice with potassium perchlorate (KClO4), which reversibly competes for entry of iodine into the thyroid gland, resulting in severe pharmacological hypothyroidism, produces the same effect (11). These results, and the fact that simultaneous treatment of KClO4 and TH restored KAP expression in the cortex, strongly suggested the involvement of TH in the androgenic control of KAP mRNA expression in cortical tubule segments. Presence of KClO4 did not modify plasma testosterone levels or expression of the androgen receptor itself (11). More remarkably, animals born to mothers exposed to KClO4 during pregnancy and after delivery were unable to express KAP mRNA in any compartment of the kidney, including the cortical convoluted segments, even at postnatal d 90 (11). Because T3 is unable to induce cortical expression in castrated males (8), we concluded that androgens are necessary but not sufficient to induce KAP expression in the cortex and that presence of TH is also required.
In the present work, we aimed to gain further insight into the molecular mechanisms that control TH involvement in the developmental sexual dimorphic pattern of KAP expression in mouse kidney by identifying the molecular elements that collaborate with androgens in the cortical expression of the KAP gene. To this end, in vivo and in vitro studies were conducted using the proximal tubule-derived cell line PKSV-PCT, originally developed in Dr. Vandewalle’s laboratory (12, 13) that, together with the PKSV-PR, are alone in supporting androgen-dependent expression of kidney-specific genes in an isolated cell system (14). For years, the lack of an appropriate cell line has hampered detailed characterization of the molecular elements mediating androgen-responsive gene expression in the kidney. Our studies proved the hormone-specific regulation of the KAP gene promoter in these cultured mouse renal proximal tubule-derived cells (15), by demonstrating that these cell lines constitute an excellent and valuable ex vivo model for analyzing the mechanisms and further characterizing the molecular elements controlling the intricate and complex regulation of KAP gene in mouse kidney.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Results obtained in groups 2 and 3 showed that only animals treated with 1% KClO4 during the gestational period (group 2) were able to express the gene like those in the control water-treated group.
|
, including table). For those under permanent treatment (group 4), weight and size were 2.5- and 1.5-fold lower, respectively, than in control animals (group 1), which suggests that an associated GH deficiency-related event to the strong hypothyroidism caused by the goitrogen treatment. As for KAP expression, effects on weight and size were clearer in postnatal treated mice (group 3) than in those exposed to treatment before birth (group 2). These results indicated that, at least in part, GH could be involved in KAP cortical expression. From previous work, we knew that when TH deficiency occurs in the absence of a dwarf-associated phenotype, i.e. in the hyt/hyt mouse model, cortical expression is diminished but not abolished (8); therefore, we hypothesized that, the combined absence of GH and TH would likely result in lack of KAP expression. To demonstrate this hypothesis, we used two genetically deficient mouse models, GH and prolactin (PRL)-deficient little mice and pituitary transcription factor-1 (Pit-1)-deficient C3H/HeJ dwarf mice from The Jackson Laboratory (Bar Harbor, ME). The specific pituitary transcription factor Pit-1 deficiency led to absence of GH, PRL, and TSH in the C3H/HeJ mice, resulting in a combined hypothyroid and dwarf phenotype. Results from this experiment showed that deficiency of GH and PRL does not compromise androgen-dependent cortical KAP expression (little) as does combined absence of GH, PRL, and TSH (Jackson dwarf) (Fig. 1B).
Goitrogen-treated male mice were injected with T3, GH, or both at the same time, from postnatal d 7–21, to further verify the involvement of these hormones in the pharmacological model and to demonstrate that the results obtained were not indirectly related to putative nephrotoxic effects of KClO4 in the kidney. In situ hybridization experiments showed that T3 administration restored KAP expression in both kidney compartments (outer medulla and cortex). This effect was enhanced by coinjection of GH, although GH administration alone produced no effect (Fig. 1C).
Punctual Presence of T3 Is Sufficient to Trigger KAP Expression
We next aimed to investigate at what time points in postnatal development TH was required to trigger KAP expression. To avoid stressing the animals with daily T3 injections, a different strategy was used based on replacing 1% KClO4 containing water by normal tap water during the time points studied. Before proceeding with this novel approach, we tested whether this treatment had equivalent effects to T3 injection in animals exposed to 1% KClO4 during the same period of time. Results on KAP expression when mice were euthanized (postnatal d 40) confirmed that both procedures had similar effects (data not shown); concomitantly, weights of animals under both treatments were practically identical: 13.31 g ± 0.1 for those exposed to T3 from postnatal d 7–21 and 13.98 g ± 0.1 for those exposed to tap water from postnatal d 7–21. Control males had a weight of 19.7 g ± 0.1 vs. 8 g ± 0.1 of those permanently exposed to 1% KClO4.
Once the procedure had been confirmed as effective, hypothyroid animals (potassium perchlorate-treated males) were exposed to regular tap water for different time periods (see Fig. 2A) and KAP mRNA expression was compared with that of age- and sex-matched animals exposed to tap water (nontreated control male) or potassium perchlorate (control) at 40 d postnatally. Results showed in Fig. 2A, indicated that the presence of T3 around postnatal d 12 and beyond facilitated KAP expression and that 48 h exposure to tap water seemed sufficient to produce this effect. Female mice treated with tap water from birth to d 14 were negative for KAP expression in cortex at d 40 (Fig. 2B), whereas males treated in the same manner exhibited KAP expression in this compartment. These results indicated that only in males was the presence of TH after postnatal d 12 required to trigger KAP expression in cortex and that once this had occurred, the presence of TH was no longer required.
|
and ß Genes under Different Hormonal Conditions and Correlation with KAP Expression in Mouse Kidney and ß gene expression (17) and also because of the significance of these transcription factors in development, we investigated their expression levels in mouse kidney, under different hormonal conditions and correlated them with those of KAP mRNA.
c/EBP and ß levels in kidneys of male mice exposed to 1% KClO4 or castration were assessed by Western blot assays. Results clearly showed c/EBP and ß expression to be residual in 1% KClO4-treated mice and levels of both factors to be reduced in castrated males (Fig. 3A). Semiquantitative RT-PCR in the same kidney samples revealed an almost undetectable expression of KAP mRNA in the presence of 1% KClO4 and an obvious decrease in castrated males (Fig. 3B), in agreement with previous in situ hybridization results. Changes in KAP mRNA expression correlated well with c/EBP and b protein levels, thereby indicating a possible functional relationship. To further test this hypothesis, c/EBP and ß protein levels were determined in kidneys of potassium perchlorate-treated males (controls), animals additionally given TH from postnatal d 7–21 (7–21 d T3) and controls (nontreated control males). Western blot assays indicated that c/EBP and ß were recovered in the TH-treated males (Fig. 3C) and that KAP mRNA was also expressed under these conditions (Fig. 3D). In summary, the clear correlation observed among hormonal status of the animals, c/EBP and ß levels and KAP mRNA expression strongly suggests involvement of c/EBPs in KAP gene expression.
|
and ß on KAP Promoter Transcription in PCT3 Cells and ß, transient transfection assays using the –638 KAP proximal promoter fragment fused to chloramphenicol acetyl transferase reporter gene (CAT) were performed (Fig. 4A). Experiments were carried out in PCT3 cells under different hormonal treatments and in the presence or not of cotransfected expression vectors for c/EBP and ß cDNAs.
|
or c/EBPß alone (lanes 7 and 8, respectively) exhibited the same transactivation capacity as DHT in the presence of a transfected functional androgen receptor (lane 3) on the –638 KAP promoter (
3-fold) (Fig. 4A). When androgens and androgen receptor were included together with c/EBP, ß or both, activity was around 5- to 6-fold higher (lanes 4–6, respectively). These results indicated a role for c/EBP and ß as well as an additive effect of androgens and these transcription factors on KAP gene transactivation.
Site-Directed Mutagenesis of Putative c/EBP Binding Sites in the KAP Promoter
Site-directed mutagenesis was performed on the 224K and 638K KAP constructs fused to the luciferase (LUC) reporter gene to further investigate whether the putative c/EBP binding sites identified in the KAP promoter were functionally relevant. The K224 construct contains two putative sites, from –66 to –56 and from –110 to –101 from the transcriptional initiation site, which were individually deleted and assayed by transient transfection in PCT3 cells. The same was done within the K638 construct where sites –66 to –56, –110 to –101, –429 to –420 and –457 to –448 were individually deleted and tested (henceforth referred to as: –66, –110, –429 and –457) (Fig. 5A). Although results from these experiments indicated no significance for any of the individually mutated sites (Fig. 5B), a second set of experiments revealed that the simultaneous mutation of the –457 and –429 sites in the K638 construct produced a strong decrease in their transcriptional activity, which dropped approximately 35- to 40-fold, compared with that of the K638 control construct (Fig. 5C). c/EBP or ß cotransfection did not affect the transcriptional capacity of the K638 double-mutated construct (Fig. 5C). Triple mutated constructs, including the other two sites at positions –110 or –66, did not produce additional effects to those obtained with the double mutated construct (Fig. 5C).
|
, upper left panel) showed that mutation of the ARE further decreased the relative activity of the –457 and –429 mut K638 construct that went from 40-fold to 100- to 150-fold. In the absence of steroid hormones (Fig. 5D, right side of the upper panel), the effects were comparable to mutation of the ARE; no significant differences were observed between the K638(
457–448, 429–420) or the K638(457–448, 429–420, 39–35) construct, indicating that androgens, acting through this androgen receptor binding site, together with factors binding to the –457 and –429 sites collaborated on the transcriptional activity of the KAP promoter.
The transcriptional activity of the pGL3 basic vector was influenced neither by the presence of transfected c/EBP expression vectors nor by the culture conditions used (Fig. 5D, lower panels), which proved that the effects observed were KAP promoter dependent.
c/EBP and ß Bind to the –457 Box
All evidence found in the above-described experiments, both in kidney and in PCT3 cells, pointed to c/EBP and ß as the putative transcription factors able to control KAP gene expression in the presence of androgens.
To reinforce this concept, gel retardation assays were performed with oligonucleotides corresponding to the 457 and 429 boxes, in the presence of nuclear extracts obtained from kidneys or PCT3 cells. Results (Fig. 6A) showed that there were nuclear factors in PCT3 cells able to bind to these two boxes and, also, to a consensus wild-type (WT) c/EBP box, but not to a mutant one (MUT). To determine the specificity of these retarded band, competition experiments including 30-fold excess of cold 457, 429, c/EBP WT, and c/EBP MUT oligonucleotides were performed. Results (Fig. 6B) showed that the retarded band on 457 box is competed out by cold WT c/EBP or by the same 457 sequence, but not by the c/EBP MUT. Because box 429 was only competed by its own sequence, our results suggested that the factor bound to the 457 sequence, but not that one bound to the 429 sequence, may belong to the c/EBP family. Next, we aimed to identify the nature of this factor by performing supershift assays with antibodies against c/EBP and ß. As expected, the retarded band for the 429 box was unaffected by the presence of the antibodies. Anti-c/EBPß, but not anti-c/EBP, produced a supershift of the 457 and c/EBP WT boxes (Fig. 6C). The lack of c/EBP supershift could not be attributed to absence of this isoform because its presence in the PCT3 nuclear extract was determined by Western blot analysis (not shown). Similar experiments performed with male mouse kidney nuclear extracts produced retarded bands for the 457 box that were competed by c/EBPWT but not by c/EBPMUT (Fig. 6D). When using these extracts, supershifts occurred with both anti-c/EBPß and anti-c/EBP antibodies (Fig. 6E) indicating that, contrary to what occurs in PCT3 cells, both c/EBP factors are able to bind to the 457 box in mouse kidney extracts. Note that kidney extracts contain c/EBPs from all their different cell types. Whether this difference in binding ability to the 457 box is due to a different cellular origin of the active c/EBP in whole kidney (i.e. not necessarily coming from cortex cells) or reflects an abnormal behavior of c/EBP of the PCT cells is not known. The nature of the factor that binds to the 429 box and the mechanisms that operate to make c/EBP incompetent to bind to the 457 box in PCT3 cells remain to be determined.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Although TH involvement was established, striking differences arose in KAP mRNA expression levels between congenital hypothyroid hyt/hyt adult males (8) and adult mice exposed to potassium perchlorate from d 10 of gestation until euthanized at postnatal d 90 (11). The hyt mice present primary hypothyroidism related to the hyporesponsiveness of the thyroid gland to TSH, which results in very low TH serum levels and abnormally high TSH levels (18). Perchlorate inhibits the transport of iodine into the thyroid in a competitive and reversible manner, which results in a decrease of TH production (T4 and T3) and an increase in TSH production (19). In the past, we had determined T4 and T3 concentrations in kidneys of hypothyroid mice (8, 11). Adult animals treated for 1 wk with KClO4 gave values that ranged from 0.258–0.524 ng of T3 and from 0.734–1.119 ng of T4/g tissue. Although very difficult because of their small size, congenital perchlorate-treated mice were also analyzed and values of T3 and T4 in kidney tissues ranged from 0.306–0.742 ng of T3 and from 0.643–2.260 ng of T4/g of tissue. Animals of the hyt/hyt strain had values ranging from 0.548–0.974 ng of T3 and from 0.857–2.85 ng of T4/g of tissue. We concluded that KClO4 treatment was effective to render the animals hypothyroid, with TH values being even lower than those in the hyt/hyt mice. Untreated control mice ranged from 1.3–2.4 or higher ng of T3/g of tissue.
Because we already knew that the KClO4 treatment was effective to render the animals hypothyroid, and animals in this study presented the same dwarf phenotype as before, hormone levels were not measured.
Although both models apparently showed the same defect, homozygous mice for the recessive hyt mutation exhibited lower cortical expression than controls, whereas perchlorate-treated animals showed complete absence of KAP mRNA from the early stages of development (11). In addition, in potassium perchlorate-treated adult males, cortical expression did not disappear but only decreased (11). In light of these data, our first hypothesis to explain the differences between models was that exposure to maternal TH, in obligated heterozygous hyt/+ mothers, was responsible for priming and enabling KAP cortical response in adult hyt/hyt males, whereas mice born of mothers exposed to goitrogen during gestation became unable to express KAP in adult life. Although the damage caused by lack of maternal and fetal TH in the developing brain had been reported (see Ref.20 for review), the results of this work do not support the concept that prenatal TH deprivation is responsible for KAP expression prevention, but rather demonstrate that its postnatal presence makes expression possible.
Another distinctive feature between both hypothyroidism models was the significant growth failure associated with perchlorate treatment, which has also been described in childhood hypothyroidism (21). Hyt mice, previously used in our laboratory (8), present slightly lower weight than normal littermates, approximately 20 g, whereas animals under postnatal perchlorate treatment are about half the weight, i.e. approximately 10 g. TH involvement in the regulation of the rat GH promoter (22), its interaction with retinoic acid receptors (23) and the participation of cell-specific transcription factors such as Pit-1/GH factor-1 (24) to fully accomplish GH expression have been reported.
This observation prompted us to hypothesize that, although absence of TH diminishes KAP cortical expression, combined TH and GH deficiency abolishes it. In situ hybridization analysis performed in the GH-deficient lit/lit mice (25) and Jackson dwarf mice, deficient in TSH, PRL, and GH due to an inactivating mutation in the anterior pituitary-specific transcription factor Pit-1 (26), revealed that GH and TH combined deficiency does compromise KAP gene expression in cortical segments in this genetically deficient mouse model. These results provided further evidence of the in vivo interactions between GH and T3 for the regulation of KAP gene and, also, that lack of KAP mRNA was not related to putative undesirable perchlorate side effects.
This concept was further challenged by administration of T3, GH or both hormones simultaneously to perchlorate-treated males on d 7–21. When mice were euthanized, on postnatal d 40, it became clear that T3 replacement overcomes the inhibitory effects of perchlorate and that this effect is even more prominent in the presence of GH. These results, together with our previous data (11), strongly support the idea that TH deprivation without consequences on severe growth retardation, i.e. in hyt mice and in adult perchlorate-treated mice, reduces but does not abolish KAP expression, whereas absence of TH associated with severe growth retardation, i.e. Jackson/dwarf mice and mice treated postnatally with perchlorate results in absolute absence of KAP mRNA. Transient transfection assays from our laboratory had demonstrated that the KAP promoter enhances its response to androgens by 2.5-fold, in the presence of IGF-I (15). These data, together with the in vivo results presented in this work, support the concept that GH can modulate androgen-dependent KAP cortical expression in mouse kidney.
Although poorly studied in mice, the participation of TH and GH in gonadal development and puberal maturation has been described in rats (27, 28, 29, 30, 31). According to those studies, the lack of expression of sexual dimorphic kidney markers such as KAP could, in part, be related to insufficient androgen levels; nevertheless, our data indicate that other factors must be involved because discrete presence of T3 for 48 h anytime after postnatal d 10, promotes cortical expression of the gene, when analyzed at postnatal d 40. This time, which is too short to reprogram gonadal development in males, might be sufficient to induce one or more T3/GH-dependent factors required to collaborate with androgens in KAP expression.
From the in vivo data presented here, we propose that postnatal expression of KAP mRNA in proximal convoluted tubules requires the presence of androgens and, also, the contribution of developmentally regulated factors triggered essentially by T3. Taking into account the punctual need of T3, we also hypothesize that, after TH priming and in the presence of some androgen levels, the one or more factors involved in KAP expression must be either autoregulated or under the control of hormones or stimuli other than T3.
The transcription factors CCAAT/enhancer-binding proteins (c/EBPs) are members of the bZIP (basic region leucine zipper) class of DNA-binding proteins. They play an important role in cellular differentiation and development (32, 33, 34, 35, 36) and fulfill many of the requirements of the hypothesized factor(s) involved in the regulation of KAP expression in kidney cortex. Congenital hypothyroidism causes a significant decrease in expression of both c/EBP and ß isoforms at early stages of postnatal liver development (17), and T3 positively regulates c/EBP gene expression through a functional TH response element found in c/EBP proximal promoter (37). c/EBPß has also been reported to contribute to regulating c-fos expression mediated by GH (38). It has recently been reported that GH promotes relocalization of c/EBPß to heterochromatin, in association with the activation of MAPK signaling, introducing a new level of transcriptional regulation mediated by this hormone. GH-mediated phosphorylation and nuclear redistribution of c/EBPß may be coordinated to achieve spatial-temporal control of gene expression (39).
Correlation of c/EBP and ß levels in nuclear and cytoplasmic kidney extracts of perchlorate-treated males with KAP mRNA expression, in contralateral kidneys of the same animals, showed that, in the absence of both forms of c/EBP, there is no expression of KAP. Moreover, T3 injection in perchlorate-treated males recovers levels of both c/EBPs and restores KAP expression. Direct evidence of c/EBPs involvement in KAP transactivation was obtained by performing transient transfection assays of the KAP proximal promoter (K638) fused to the CAT reporter gene in PCT3 cells. This promoter fragment (Fig. 4A) was selected for transfection because it contains four putative c/EBP binding sites at positions –66, –110, –429, and –457. Our results showed that transfection of c/EBP or ß (Fig. 4B, lanes 7 and 9) exerts the same effect on transcriptional activity as transfection of the AR in the presence of DHT (Fig. 4B, lane 3) and also that simultaneous cotransfection of and/or b c/EBPs and AR, in the presence of androgens, (Fig. 4B, lanes 4–6) produces an additive effect on K638 transcriptional activity. Altogether, these results support the concept that the effects of T3 on the androgenic response of KAP observed in kidney are, at least in part, mediated by c/EBP and ß.
Site-directed mutagenesis was performed at specific sites to elucidate which of the putative c/EBP binding sites on the promoter were functionally significant (Fig. 5A). Results (Fig. 4B) indicated that mutation of individual sites had no effect on promoter activity, whereas simultaneous mutation of boxes –429 and –457 decreased the transcriptional activity of the promoter approximately 40-fold (Fig. 4C), thereby indicating that these sites are functionally significant when they work in a cooperative fashion.
Deprivation of steroid hormones and/or mutation of the ARE at 39 bp from the transcriptional initiation site was performed to further investigate whether factors binding to these functional sites could cooperate with the androgenic response. The 100- to 150-fold decrease in promoter activity obtained in either situation clearly demonstrated a functional relationship between transcription factors bound to the –457 and –429 boxes and the AR.
Competitive shift and supershift assays performed to verify that the –457 and –429 sites were binding-specific nuclear factors and identify their nature demonstrated that c/EBP and ß bind to the –457 box but not to the –429. The nature of the factor bound to box 429 remains to be identified. Overall, the data presented here, together with previous contributions from our laboratory indicate that expression of the KAP gene in epithelial cells of proximal convoluted tubules depends not only on androgens and functional androgen receptors, but also on transcription factors that bind at boxes –457 and –429 in the proximal promoter. c/EBP and ß can bind to the –457 box, activate transcription and, together with a third party, bind to box –429 and trigger expression of the gene when androgen levels are sufficient in males. After birth, the presence of TH and GH is required for sexual maturity and expression of c/EBP and ß. Once these factors have been expressed and probably because they can be autoregulated (40, 41), TH becomes dispensable and cortical KAP expression occurs in the presence of androgens alone.
Study of the KAP gene has raised the theory that androgens are necessary, but not sufficient, to initiate an androgenic response in kidney and that other developmentally regulated factors, ultimately controlled by T3, are involved. Studies with the ornithine decarboxylase (ODC) gene demonstrate that the lack of renal androgenic response during the postnatal period cannot be related only to the low levels of plasma testosterone at that time, but rather to immaturity of renal mechanisms to respond to androgens, which follow a maturational process in which testosterone could be only one of the factors implicated (42). Induction of renal ODC by androgens in adult mice requires activation of the transduction cascade mediated by catecholamines binding to 1-adrenergic receptors (43). Reports describe the existence of a critical period for T3 to play a role in development of renal 1-adrenergic receptors (44); the influence of TH in the development of ß adrenergic control of ODC in rat kidney and heart (45) and evidence that c/EBP is required for transcription of ß-adrenergic receptors in adipose tissue (46) suggest that c/EBPs under T3 control might not only be involved in the androgenic control of KAP gene expression, but rather constitute a more general phenomenon for transcriptional control of androgen-regulated genes in kidney.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Animals
Jackson dwarf (strain C3H/HeJ-dwj/dwj), little mice (strain C57BL/6J-lit/lit) and their genetically matched normal controls (strain C3H/HeJ dwj/+ and C57BL/6J-lit/+) were obtained at 4–8 wk from The Jackson Laboratory. C57BL/6 control mice were obtained at 6 wk of age from Charles River España (Barcelona, Spain).
Mice were housed in the animal facilities in covered cages at 22 C with 12 h light and darkness cycles. Pelleted food and tap water were supplied ad libitum.
At the time of the experiment, animals were euthanized by cervical dislocation. Kidneys were removed and snap-frozen immediately in liquid nitrogen in the presence of 2-methylbutane. All animal experimental procedures were conducted in accordance with Institutional standards which fulfill the requirements established by the Spanish Government and the European Community (Real Decreto 223/1988: B.O.E. no. 67, 3/18/88) and B.O.E. no. 256, 10/25/90).
Hormone Treatment
T3 hormone was resuspended in 10 mM KOH and 1.2 µg/animal/d were administered sc human GH was reconstituted with Na2CO3 25 mM (pH 9.4) in 0.9% NaCl and 18 µg/animal/d were injected, sc. Pharmacologically induced congenital hypothyroidism was achieved through addition of 1% KClO4 to drinking water of pregnant dams at d 10 of gestation. From birth until being euthanized, animals were exposed to goitrogen, even when other treatments were simultaneously performed. The corresponding sham treatments were effected in all experimental situations. Hypothyroidism in newborn mice was induced using the same drugs at the same concentration until the animals were euthanized. When required, KClO4-treated water was replaced by normal tap water.
Probe Synthesis and in Situ Hybridization Histochemistry
35S-Labeled transcripts from pKAP3 or pKAP4 cDNA plasmids, which correspond to antisense and sense cRNA, respectively, were prepared as previously described (14). The preparation of renal sections, hybridization protocol and autoradiographic analysis were all performed as reported (8, 11). Hybridized sections were examined under a light microscopy. Positive silver grain staining (appearing in black) determined the site of KAP mRNA synthesis (using KAP3 probe) in proximal tubule cells. All sections hybridized with the mRNA KAP4-specific strand sense probe exhibited no detectable silver grain staining (data not shown).
The magnification chosen for the images shown in the figures (x40) permitted visualization of the three major compartments of the kidney (cortex, c; outer medulla, o; inner medulla, i) and determination of the spatial location of KAP mRNA. Kidney sections from same figure were each analyzed in a single experiment. Consequently, slides were exposed to the same conditions throughout the entire procedure, and KAP mRNA expression levels on different slides were comparable. Photomicrographs shown in the figures were selected from similar results obtained in different experiments using animals from different litters.
RNA Extraction and Semiquantitative RT-PCR
Total RNA was extracted from kidneys of animals using the guanidinium thiocyanate method (47). Different amounts of total RNA were used in each RT-PCR using the SuperScript One-Step RT-PCR System (Life Technologies, Inc.) under conditions previously reported. Primer sequences for KAP and CypA internal control genes have already been described (15).
Preparation of Protein Extracts and Western Blotting
Kidney tissues were homogenized in buffer A containing 20 mM Tris-HCl (pH 6.8), 50 mM NaCl, 3 mM MgCl2, 300 mM sucrose; 1% Triton X-100, 2 mM phenylmethanesulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM Na3OV4. Homogenates were centrifuged for 15 min at 13,000 rpm. Supernatant (cytosol and membrane fraction) was heated at 70 C for 5 min, centrifuged for 20 min at 14,000 rpm and supernatant (cytosol) frozen at –80 C. The pellet from the original kidney homogenates (including nuclear fraction) was resuspended in buffer B containing 25 mM Tris-HCl (pH 7.5), 1 mM EGTA, 2 mM EDTA, 1% SDS, and 2 mM phenylmethanesulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM Na3OV4. The suspension was sonicated until clarified, centrifuged for 15 min at 13,000 rpm and the supernatant (nuclear fraction) saved at –80 C. All steps were performed at 4 C. Protein concentration was determined by the micromethod of Bio-Rad (Hercules, CA) using BSA as a standard.
Protein extraction from cell cultures was performed using the same protocol except that, after washing of plates with PBS, cells were scraped with a rubber policeman and homogenized with buffer A on a rocking platform for 60 min at 4 C. The remaining procedures were the same as for kidney extracts.
Fifty micrograms of nuclear and cytosolic proteins from mouse kidney or cultured cells were size-fractionated by denaturing 12% SDS-PAGE. Proteins were electroblotted onto polyvinylidene difluoride membranes (Shleicher & Schuell, Keene, NH) in transfer buffer (50 mM Tris-HCl, 40 mM glycine, 20% methanol) and blots blocked overnight at 4 C in 4% BSA in TBST [Tris-HCl 20 mM (pH 7.6), NaCl 130 mM and Tween 20 at 0.1%]. Primary polyclonal rabbit antibodies anti-c/EBP (14AA) and anti-c/EBPß (C-19) from Santa Cruz Biotechnologies (Santa Cruz, CA) were diluted at 0.2 mg/ml in blocking buffer. Washes were performed following the membrane manufacturer’s instructions and secondary antibody (horseradish peroxidase-conjugated goat antirabbit; Dako A/S, Glostrup, Denmark) was diluted 1:5000 and incubated for 1 h at room temperature. After washing, bands were detected using the ECL+ chemiluminescence detection method (Amersham Pharmacia Biotech) and exposed to Hyperfilm. To assess specificity of the antibodies, several assays were performed using specific blocking peptides (5-fold excess) (sc-61) for c/EBP, (sc-150 P) for c/EBPß and nonspecific (sc-565P) for gp-130, from Santa Cruz Biotechnologies.
Plasmid Constructs
The strategy for obtaining the promoter fragments consisting of nucleotides (nt) –224 to +1 and –638 to +1 in the KAP gene (relative to the transcription start site) has already been described (15). Constructs were verified by extensive restriction mapping and partial DNA sequencing. All promoter lengths are available in both CAT and LUC promoterless vectors. The pCAT-Basic and pGL3-Basic vectors were obtained from Promega Corp. (Madison, WI).
The expression plasmid for ß-galactosidase pCH110 (Amersham Pharmacia. Biotech) and the Renilla LUC reporter plasmid pRL-TK (Promega Corp.) were used to normalize transfection efficiencies of CAT and LUC assays, respectively. Plasmid pSVAR0 containing a simian virus 40 promoter that directs transcription of the full-length human androgen receptor cDNA was kindly provided by Dr. C. López-Otín (Universidad de Oviedo, Spain) (48). TH receptor, consisting of cDNA-encoding rat T3R1 cloned in pCDM8, retinoid X receptor expression vector, human retinoid X receptor cDNA cloned in plasmid pSG-5 and the SaltK reporter construct, which corresponds to a TH-responsive element introduced into the reporter plasmid pBLCAT2 upstream of the heterologous Herpes simplex virus-thymidine kinase promoter, were kindly provided by Dr. A. Muñoz (Instituto de Investigaciones Biomédicas Alberto Sols, Centro Superior de Investigaciones Cientificas, Spain) (49). pMSV-c/EBP and PMSV-c/EBPß are expression vectors that contain the entire open reading frame of rat c/EBP and c/EBPß (p35 LAP), respectively, driven by the murine sarcoma virus promoter (50). They were kindly provided by Dr. M. Giralt (Universitat de Barcelona).
Transient Transfection Studies
The cloned PKSV-PCT, referred to as PCT3 cells, has previously been described (15). The culture conditions have also been reported (12, 13). Transient transfection experiments were performed as previously reported (15).
CAT and LUC Assays
These assays were performed as previously reported (15). Stimulation of CAT and LUC activities were expressed as fold increase over activities of noninduced transfected cells with the same reporter plasmids set to 1 and based on at least three independent experiments.
Site-Directed Mutagenesis of Putative c/EBP Binding Sites
K224 (110–101), K224 (66–56), K638 (457–448), K638 (429–420), K638 (110–101), and K638 (66–56) contain deleted putative c/EBP binding sites of the KAP promoter. K224 and K638 were used as the DNA mutagenesis template to anneal with the mutagenic primers. Additional ARE mutation at position –39 was performed as described (15).
Mutant strands were synthesized with Pfu Turbo DNA polymerase and used to transform Epicurian Coli XL1-Blue Supercompetent cells using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, Ca). Plasmid DNAs were isolated from the selection plates and mutants were identified by sequencing.
EMSA Experiments
Nuclear extracts from PCT3 cells were prepared as described by Dignam et al (51) without dialysis. Nuclear extracts from C57BL/6 mice kidney were prepared as described by Landschulz et al. (52) starting from 3.5 g of tissue and without dialysis. For EMSA experiments, the 32P-labeled DNA fragments (C/EBP WT: TGCAGATTGCGCAATCTGCAA, C/EBP MUT: TGCAGAGACTAGTCTCTGCAA, 429:CTTCACTGTTC-TCCAGCAATCTGCCAGGAT,457:GATAGTTCTGCTTTTGC-AATGAGCAGTTCT were incubated with 25 µg of kidney nuclear protein or 5 µg of cellular nuclear protein in a binding buffer [20 mM Tris-HCl (pH 7.4), 75 mM KCl, 0.5% Nonidet P-40, 10% glycerol, 1 mM DTT] in the presence of 100 ng of poly(deoxyinosine-deoxycytosine) and 50 ng of acetylated BSA for 30 min at 37 C in a final volume of 20 µl. Oligonucleotide competition experiments were performed under the same conditions by previous mixing of all the reagents and increasing amounts of corresponding cold oligonucleotides before nuclear protein addition. Supershift assays were performed by incubating the complexes for a further 10 min more at 37 C with anti-C/EBP and anti-C/EBPß before loading. The samples were electrophoresed on 5% acrylamide 0.5x TBE gels, dried, and developed with intensifier screen at –80 C.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
First Published Online September 8, 2005
1 N.T. and M.S. contributed equally to this work. 
Abbreviations: ARE, Androgen receptor response element; CAT, chloramphenicol acetyl transferase reporter gene; KAP, kidney androgen-regulated protein; LUC, luciferase; ODC, ornithine decarboxylase; PCT, pars convoluta; Pit-1, pituitary transcription factor-1; TH, thyroid hormone; PR, pars recta; PRL, prolactin.
Received for publication June 10, 2005. Accepted for publication September 1, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ß genes during liver development. Biochem Biophys Res Commun 234:605–610[CrossRef][Medline] gene and regulation by thyroid hormone in rat immortalized brown adipocytes. Endocrinology 141:4164–4170
contribute to growth hormone-regulated transcription of c-fos. J Biol Chem 274:31597–31604
