Expression, Regulation, and Function of Paired-Box Gene 8 in the Human Placenta and Placental Cancer Cell Lines
http://www.100md.com
《内分泌学杂志》
Departments of Clinical Science (S.F., T.M., A.S., E.T., R.M.) and Experimental Medicine and Pathology (E.F., A.G.), University of Rome, La Sapienza, 00161 Rome, Italy
Neuromed Institute (A.G.), 86077 Pozzilli, Italy
Department of Clinical and Experimental Medicine (F.A., I.P., D.R.), G. Salvatore, and Department of Farmacobiological Sciences, University of Catanzaro, 88100 Catanzaro, Italy
Department of Science and Biomedical Tecnology (G.T., G.D.), University of Udine, 33100 Udine, Italy
Departments of Clinical Biology (L.L.), Institut Gustave-Roussy, 94805 Villejuif Cedex, France
Abstract
Pax proteins are transcriptional regulators that control a variety of developmental decisions in vertebrates. During development, the paired-box gene 8 (PAX8) is expressed in the thyroid, kidney, and several areas of the central nervous system. It is also expressed in the adult thyroid gland, in which it mediates TSH-induced modulation of the expression of important genes, such as those encoding thyroglobulin, thyroperoxidase, and the sodium/iodide symporter (NIS). Thus far, placental expression of PAX8 has been described only in mice. In the present study, we show that PAX8 is also expressed in the human placenta at term. In an in vitro model of placental cancer, the JAR choriocarcinoma cell line, human chorionic gonadotropin (hCG) increased levels of PAX8 mRNA and protein, and gel retardation assays indicated that the up-regulation of PAX8 protein expression is associated with an increase in its DNA-binding activity. The effects of hCG were mimicked by forskolin, indicating that they are cAMP dependent. Levels of mRNA for the Wilms’ tumor 1 (WT1) and NIS genes were increased in JAR cells by hCG treatment, whereas overexpression of PAX8 increased only levels of WT1 mRNA. In cells transfected with PAX8-specific small interfering RNA, the stimulatory effects of hCG on WT1 mRNA levels were abolished, but hormonal enhancement of NIS mRNA levels was unchanged. These findings indicate that, in JAR cells, hCG activates a cAMP-dependent pathway that can up-regulate WT1 expression through PAX8.
Introduction
THE PLACENTA DEVELOPS from two major cell lineages. The trophectoderm of the blastocyst is the precursor of the trophoblast cell lineage, which gives rise to the epithelial parts of the placenta, whereas stromal cells and those of the blood vessels develop from the extraembryonic mesoderm (1). The molecular mechanisms involved in the development and maintenance of the human placenta are still largely unknown. However, several of the transcription factors that regulate trophoblastic cell development have already been identified (1, 2, 3). mRNA for the paired-box transcription factor, PAX8, has recently been described in murine placental cells (4), but thus far there is no evidence of PAX8 expression in human placental cells.
PAX8 is involved in the control of development and maintenance of various human tissues (5, 6, 7). Its expression has been demonstrated in the developing thyroid gland (5, 8) and adult follicular thyroid cells, in which it determines the differentiated phenotype (9). In thyroid cells, PAX8 expression is regulated by TSH through cAMP-dependent mechanisms (10, 11, 12), and it, in turn, regulates together with thyroid transcription factors 1 and 2 (TITF1/NKx2.1 and FoxE1) the expression of several essential thyroid genes (10, 11, 13, 14), including the sodium iodide symporter (NIS). Recent reports suggest that PAX8 is also involved in modulating the expression of D2 deiodinase (15, 16), one of the enzymes responsible for the deiodination of iodothyronines. More recent evidence indicates that the NIS (17) and the iodothyronine deiodinases, D2 and D3, which were originally considered to be thyroid specific, are also expressed by placental trophoblasts (18, 19), in which their presence is essential for ensuring adequate supplies of iodide for the synthesis of fetal thyroid hormones. These discoveries suggested that placental expression levels of these genes might be modulated by PAX8, as they are in the thyroid.
The Wilms’ tumor 1 (WT1) gene, another well-known target of PAX8 regulation, is also known to be expressed by human trophoblasts (20). The transcription factor it encodes was originally identified as a tumor suppressor based on genetic analyses showing that loss of WT1 function can initiate (or promote) tumorigenesis and functional studies demonstrating that wild-type WT1 expression is usually associated with inhibition of cellular proliferation (21). It has now become clear, however, that WT1 also plays important roles in the prenatal development of a number of organs (i.e. kidney, gonads, spleen, adrenal glands, retina, and heart), as recently reviewed by Scholz and Kirschner (22). Its precise function in the human placenta is unknown, as are the mechanisms underlying the regulation of its expression by trophoblasts, but its up-regulation has been shown to involve cAMP-dependent mechanisms (20). The latter finding is of particular interest because, as noted above, cAMP dependency is also a feature of TSH-induced modulation of thyrocyte expression of PAX8 (10, 11, 12).
The aim of the present study was to analyze the expression of PAX8 and its regulation in the human placenta and investigate its effects on the transcription of two of its known gene targets, NIS and WT1.
Materials and Methods
Tissues samples, cell lines, and reagents
All tissue samples used in the study had been collected with full patient consent and institutional review board approval. They included specimens from 10 normal full-term human placentas and 10 specimens of normal thyroid tissue (used as controls). Immediately after removal, part of each tissue specimen was snap frozen in liquid nitrogen and stored at –80 C for RNA and protein isolation; the rest was formalin fixed for histological studies. All cell lines used in the study were from the American Type Culture Collection (Manassas, VA). Choriocarcinoma cells, JAR and JEG-3, were grown in 10% heat-inactivated fetal calf serum/RPMI 1640 medium. D283 medulloblastoma cells were cultured in 20% fetal calf serum/HEPES/glutamine MEM.
Cell treatments
Hormonal regulation of placental PAX8 expression was investigated in JAR cells before and after exposure to 5 UI/ml human chorionic gonadotropin (hCG) (obtained from Sigma Chemical Corp., St. Louis, MO). To determine whether the hCG-induced effects were cAMP mediated, cells were grown in medium containing 10 μM forskolin (Sigma).
In PAX8 overexpression experiments, JAR cells were transfected with full-length human PAX8 cDNA, kindly provided by R. Di Lauro (University of Naples, Naples, Italy), by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA), according to the instructions of the manufacturer.
For PAX8 gene-silencing studies, JAR cells were transfected (Oligofectamine agent, Invitrogen) with PAX8-specific small interfering mRNA (siRNA) (sequence: AAGTGCAGCAACCATTCAACC) synthesized with the QIAGEN system; control siRNA (GFP-22, QIAGEN S.p.A., Milan, Italy); or the Oligofectamine agent alone. Twelve hours after transfection, hCG (5 UI/ml) was added to culture medium, and the cells were incubated for up to 48 h.
Quantitative mRNA evaluation by real-time PCR
Total RNA was extracted from tissue or cells with the TRIzol reagent (Invitrogen) and reverse transcribed with oligo(dT) primers and 200 U Superscript II reverse transcriptase (Invitrogen), in accordance with the manufacturer’s instructions. Quantitative PCR (Q-PCR) analysis of PAX8, NIS, and WT1 mRNA expression was performed on each cDNA sample using the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA). Oligonucleotide primers and probes for the genes analyzed and the endogenous control were purchased from Applied Biosystems. A reaction mixture containing cDNA template, TaqMan Universal PCR master mix (Applied Biosystems), and primer probe mixture was amplified using standard Q-PCR thermal cycler parameters. Each amplification reaction was performed in triplicate, and the average of the three threshold cycles was used to calculate the amount of transcript in the sample (using SDS version 1.7a software; Applied Biosystems). mRNA quantification was expressed, in arbitrary units as the ratio of the sample quantity to the quantity of the calibrator. All values were normalized with two endogenous controls, glyceraldehyde-3-phosphate dehydrogenase and -actin, which yielded similar results.
Nuclear extract preparation, Western blot, and gel retardation assays
Nuclear extracts were prepared from frozen pelleted cells and tissues that had been homogenized in a buffer containing 10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl2, 0.1 mM EGTA, 0.1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 2 μg/ml leupeptin. After the nuclei had been released, the suspension was overlaid with the same buffer and centrifuged. The pelleted nuclei were rewashed with the same buffer and centrifuged. Nuclear proteins were extracted by a second buffer containing 10 mM HEPES (pH 7.9), 400 mM NaCl, 1.5 mM MgCl2, 0.1 mM EGTA, 5% glycerol, 0.1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 2 μg/ml leupeptin. The protein concentration of the supernatant was measured using the Bradford method (Bio-Rad Laboratories, Inc., Hercules, CA).
Immunoblot analysis of PAX8 expression was performed with polyclonal goat PAX8 antibody (catalog no. SC12679, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and monoclonal mouse WT1 antibody (DakoCytomation S.p.A, Milan, Italy). Proteins (25 μg/lane) were run on a 10% SDS-PAGE gel and transferred to nitrocellulose membranes (Amersham S.r.l., Milan, Italy). Membranes were blocked in 5% milk and incubated overnight with primary PAX8 antibody (1:100 dilution) and then for an additional 12 h with primary WT1 antibody (1:100 dilution). Polyclonal rabbit histone 2A antibody (Santa Cruz Biotechnology) was used as an internal control.
For gel retardation assays, 10 μg of each nuclear extract were incubated with DNA for 30 min at room temperature in a buffer containing 20 mM Tris-HCl (pH 7.6), 75 mM KCl, 10 mg/ml calf thymus DNA, and 10% glycerol. The oligonucleotides used were C and specificity protein 1 (SP1), which contain high-affinity binding sites for PAX8 and SP1, respectively (23). Oligonucleotides were labeled at the 5' end using polynucleotide kinase (Amersham) and [-32P]ATP and annealed with their respective complementary strands. At the end of the binding reaction, samples were loaded onto a 7.5% native polyacrylamide gel and run at 4 C in Tris-borate-EDTA buffer. After electrophoresis, gels were fixed in 10% acetic acid and subjected to autoradiography. In supershift and competition experiments, the labeled C oligonucleotide was added after the nuclear extracts had been preincubated with either antisera or unlabeled oligonucleotides for 30 min at room temperature.
Immunofluorescence
Sections of formalin-fixed paraffin-embedded normal placental samples were prepared on poly-L-lysine-coated slides and control stained with hematoxylin and eosin (Sigma). For immunohistochemistry (IHC), sections were deparaffinized, rehydrated, and microwave heated in 10 mM citrate buffer (pH 6) for antigen retrieval. In studies designed to detect coexpression of PAX8 with NIS and/or WT1, the sections were first immunostained overnight with goat polyclonal anti-PAX8 diluted 1:20 (SC12679) and, after three washes with PBS, they were incubated with fluorescein isothiocyanate-conjugated antigoat secondary antibody (Sigma-Aldrich) diluted 1:30. The sections were then immunostained with monoclonal anti-NIS or anti-WT1 antibodies [U.S. Biological (Swampscott, MA), and DakoCytomation, respectively] diluted 1:100 and incubated with a 1:40 dilution of Texas-RED-conjugated antimouse secondary antibody (Jackson Laboratory, Bar Harbor, ME). For negative controls, sections were incubated with PBS instead of primary antibodies. The samples were mounted and photographed under phase-contrast and florescence microscopy (BH2; Olympus, Tokyo, Japan).
Statistical analysis
Results are expressed as means ± SD. Differences between different points of stimulation were analyzed with the Mann-Whitney U test. P < 0.05 was considered statistically significant.
Results
Human placental expression of PAX8
PAX8 mRNA has been previously demonstrated in mouse placenta (4). Using RT-Q-PCR, we demonstrated that mRNA for PAX8 (along with that for its target genes, NIS and WT1) is also expressed in normal human placenta, as well as in the two widely used choriocarcinoma cell lines, JAR and JEG-3 (Fig. 1A). Western blot analysis was then used to evaluate PAX8 protein expression in nuclear extracts prepared from normal human placental tissue. As shown in Fig. 1B, placental expression of PAX8 was found to be comparable to that observed in normal human thyroid tissue, and immunohistochemical studies of the placental sections localized the PAX8 expression to the trophoblasts (Fig. 1, C and D). These same sections were subjected to immunofluorescence analysis after incubation with antibodies specific directed against the NIS and WT1 proteins. As shown in Fig. 1D, more than half (five of nine) of the cells displaying WT1 immunoreactivity were also positive for PAX8. In contrast, PAX8-positive staining was much less common (five of 25) in NIS-positive cells (Fig. 1C).
Regulation of PAX8 expression in JAR cells
The expression of a number of genes in JAR cells is known to be hormonally regulated (by hCG) as we previously reported for NIS (24). Thus, the possibility that human placental expression of PAX8 is hormonally regulated was investigated in JAR cells exposed to 5 IU/ml of hCG for up to 48 h. As shown in Fig. 2A, levels of PAX8 mRNA were significantly increased by the hormonal treatment. Peak effects (4- to 5-fold increases over basal expression) were observed after 6–8 h of hCG stimulation; expression decreased thereafter and returned to near basal levels after 24–48 h of stimulation. Less marked but detectable increases were also observed with lower hCG doses (3 IU/ml) (data not shown). Western blot experiments showed that PAX8 protein expression in JAR cells was also increased by hCG stimulation: maximal enhancement was observed after 8 h of hCG exposure, and expression was still significantly increased at 24 h (Fig. 2B).
Cell treatment with hCG (5 IU/ml) caused a similar increase in mRNA levels for WT1, but this effect was observed later, after 12, 24, and 48 h of hCG stimulation (Fig. 2A). Again, Western blot experiments confirmed that this increase was also reflected in levels of WT1 protein (Fig. 2C). The effects of hCG on expression of NIS were not tested because a previous study by our group (24) has already demonstrated that hCG (5 IU/ml) up-regulates JAR cell expression of NIS at both the mRNA and protein levels. Because the same study had also showed that the hormone’s effects on NIS expression involved a cAMP-dependent mechanism, we reevaluated WT1 mRNA levels in JAR cells treated with forskolin (10 μM), instead of hCG, and found that the cAMP analog also produced significant increases in WT1 transcript levels (data not shown).
We next attempted to determine whether the increased PAX8 protein expression corresponded to an increase in its DNA-binding activity. Nuclear extracts from JAR cells treated with hCG (5 UI/ml) were analyzed in a gel retardation assay performed with the PAX8-specific oligonucleotide C (23). As shown in Fig. 3A, PAX8 DNA-binding activity was indeed increased by the hormonal treatment, which produced maximum effects after 12 h stimulation. The hCG-induced increase was mimicked by cell treatment with forskolin (10 μM), indicating that the effect was cAMP dependent. The PAX8 specificity of the effects of hCG and forskolin was confirmed by the absence of changes in DNA binding to the control oligonucleotide, SP1 (Fig. 3B). The shift detected with the C oligonucleotide was due to PAX8 binding. In fact, when nuclear extracts from JAR cells treated for 12 h with forskolin were used, preincubation with PAX8 antiserum produced a supershift, which was not detected with preimmune serum (Fig. 3C). In addition, the band was competed by a 20-fold excess of unlabeled C oligonucleotide but not by 20-fold excess of the Cant2 oligonucleotide, which is a mutant of C no longer recognized by PAX8 (23). Similar results were obtained using nuclear extracts of JAR cells treated for 12 h with hCG (data not shown).
Transcriptional effects of PAX8 in JAR cells
The experiments described above clearly demonstrated that PAX8 is expressed in JAR cells and that this expression is regulated by hCG. To identify the effects of this expression, we overexpressed PAX8 in JAR cells by means of transfection and performed quantitative RT-PCR to measure the levels of mRNA for the two PAX8-target genes, WT1 (25, 26) and NIS (10), both of which are known to be expressed in human placenta (4, 17, 27). As shown in Fig. 4A, overexpression of PAX8 had no effect on NIS mRNA expression in JAR cells, but it markedly enhanced that of WT1 mRNA. Western blot experiments showing the PAX8 protein expression levels increased after transfection with PAX8 expression vector (0.1 and 1 μg) were used as experimental control (Fig. 4B). These results suggest a role for PAX8 in the control of WT1 expression in the human placenta.
The effects of hCG were then reassessed in JAR cells that had been subjected to PAX8 gene-silencing treatment (Fig. 5). Cells were transfected with PAX-8-specific or control siRNA, and 12 h later hCG (5 UI/ml) was added to the culture medium. As shown in Fig. 5A, in cells transfected with specific PAX8 siRNA, the hormone produced no increase in PAX8 mRNA levels. Indeed, these levels actually dropped, and maximal reduction was observed after 20 h of PAX8 siRNA. As expected, PAX8 gene silencing had no effect on hCG’s modulation of NIS mRNA levels (Fig. 5B), but it significantly diminished the hormones effect on WT1 mRNA levels (P < 0.05) (Fig. 5C). As control, siRNA-transfected cells were assayed for PAX8 protein levels by Western blot analysis 20 and 48 h after transfection (data not shown). These data strongly support the notion that PAX8 exerts direct control over JAR-cell expression of WT1 but not that of NIS.
Discussion
To date, evidence of placental expression of PAX gene transcription factors has been limited to a single report documenting PAX8 mRNA in the placental cells of mice (4). In the present study, we have demonstrated that: 1) PAX8 is expressed in the human placenta at term, mainly in the nuclei of the trophoblast cells; 2) placental PAX8 expression is controlled by hCG through a cAMP-dependent mechanism; and 3) in placental cells, PAX8 modulates the expression of the tumor-suppressor gene WT1 but has no effect on that of the NIS gene.
The transcription factors currently known to be expressed in the human placenta include basic helix-loop-helix proteins, such as HAND1 (heart and neural creast derivates expressed 1) and MASH2 (achaete-scute complex-like protein 2) [Ascl2 (achaete-scute complex homolog-like 2)], and the glial cells missing-1 transcription factor (2), whose down-regulation is thought to contribute to the development of preeclampsia (28). Several homeobox genes are also expressed in placental tissue, and it has been suggested that down-regulation of homeobox A11 may be necessary for the differentiation of cytotrophoblasts into syncytiotrophoblasts (3). In light of our findings, PAX8 must be added to the list of transcription factors involved in the regulation of gene expression in placental cells. Apart from PAX8, however, there is currently no evidence that any other PAX genes are expressed in the placenta.
PAX8 has been more thoroughly investigated in thyroid cells. Its expression in these cells is known to be hormonally regulated (by TSH) via cAMP-dependent mechanisms (10, 11, 12), and one of its main functions is the regulation of NIS gene expression (8). Our findings indicate that placental expression of PAX8 is also subject to cAMP-mediated hormonal regulation (in this case, by hCG), but in this tissue the transcription factor does not appear to exert any effect on the expression of the NIS. These data are based on experiments conducted in transformed cells, and whereas JAR cells are known to retain most of the features of normal placental cells (29), we cannot exclude the possibility that, in normal placental cells, PAX8 does indeed modulate NIS expression.
PAX8 regulation of NIS transcription requires the cooperative effects of other transcriptional regulators in thyrocytes (5, 8, 10, 30, 31). Thus, it was not fully surprising that overexpression of PAX8 in JAR placental cells did not lead to a modulation of NIS expression. Our findings could be a reflection of a complex molecular mechanism that controls NIS gene expression involving more than one factor, different between placental and thyroid cells. This hypothesis is consistent with previous reports showing that regulation of NIS promoter activity in the thyroid involves functional interaction between PAX8 and factors that bind a cAMP-responsive element-like sequence (10, 32). Thus, NIS is expressed by both thyroid and placental cells, but the molecular mechanisms that control its expression might be different in the two cell types. As various investigators have recently noted, tissue-specific gene regulation is not achieved with a few specific transcription factors: it is the result of complex combinatorial interactions between multiple general and tissue-specific proteins (33, 34).
In contrast, PAX8 regulation of the WT1 gene does not seem to be cell type dependent. Evidence of this type of regulation has been obtained in various cell types (25, 35), and some authors have recently suggested that PAX8 may be involved in the overexpression of WT1 that is often associated with acute myeloid leukemias in humans (36). Our findings are consistent with those of Feingold et al. (20), who showed that WT1 expression in human trophoblasts is regulated by means of cAMP-dependent mechanisms. The precise function(s) of WT1 in placental development is still unclear, but there is suggestive evidence of its possible role in the suppression of trophoblast growth and invasion (20).
The absence of PAX8 is known to have detrimental effects on the development of the human thyroid gland, but its impact on the development and function of the human placenta has never been investigated. Its placental effects cannot be investigated in Pax8-null mice because these animals are born without the follicular portion of the thyroid gland and die before reaching reproductive age (37). Nonetheless, our findings predict that inactivating mutations or functionally relevant polymorphisms of the PAX8 gene could influence placental function. Although PAX8 does not appear to be a direct regulator of NIS expression in human trophoblasts (as it is in thyrocytes), it does seem to be involved in the cAMP-dependent regulation of the Wilms’ tumor suppressor gene. Future studies on placental tissues will hopefully shed more light on the role of this transcription factor in the development and functions of placenta.
Acknowledgments
We thank D. De Sanctis for technical assistance and M. Kent for editorial assistance.
Footnotes
This work was supported by grants from the Italian Ministry of Health and the Italian Ministry of University and Research (to S. F.).
1 E.F. and F.A. contributed equally to this work.
Abbreviations: hCG, Human chorionic gonadotropin; IHC, immunohistochemistry; NIS, sodium iodide symporter; PAX8, paired-box gene 8; Q-PCR, quantitative PCR; siRNA, small interfering mRNA; SP1, specificity protein 1; WT1, Wilms’ tumor 1.
References
Cross JC, Baczyk D, Dobric N, Hemberger M, Hughes M, Simmons DG, Yamamoto H, Kingdom JC 2003 Genes, development and evolution of the placenta. Placenta 24:123–130
Hemberger M, Cross JC 2001 Genes governing placental development. Trends Endocrinol Metab 12:162–168
Zhang YM, Xu B, Rote N, Peterson L, Amesse LS 2002 Expression of homeobox gene transcripts in trophoblastic cells. Am J Obstet Gynecol 187:24–32
Kozmik Z, Kurzbauer R, Dorfler P, Busslinger M 1993 Alternative splicing of Pax-8 gene transcripts is developmentally regulated and generates isoforms with different transactivation properties. Mol Cell Biol 13:6024–6035
Damante G, Tell G, Di Lauro R 2001 A unique combination of transcription factors controls differentiation of thyroid cells. Prog Nucleic Acids Res Mol Biol 66:307–356
Poleev A, Fickenscher H, Mundlos S, Winterpacht A, Zabel B, Fidler A, Gruss P, Plachov D 1992 PAX8, a human paired box gene: isolation and expression in developing thyroid, kidney and Wilms’ tumors. Development 116:611–623
Plachov D, Chowdhury K, Walther C, Simon D, Guenet JL, Gruss P 1990 PAX8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110:643–651
Pasca di Magliano M, Di Lauro R, Zannini M 2000 PAX8 has a key role in thyroid cell differentiation. Proc Natl Acad Sci USA 97:13144–13149
Zannini M, Francis-Lang H, Plachov D, Di Lauro R 1992 Pax-8, a paired domain-containing protein, binds to a sequence overlapping the recognition site of a homeodomain and activates transcription from two thyroid-specific promoters. Mol Cell Biol 12:4230–4241
Ohno M, Zannini M, Levy O, Carrasco N, Di Lauro R 1999 The paired-domain transcription factor PAX8 binds to the upstream enhancer of the rat sodium/iodide symporter gene and participates in both thyroid-specific and cyclic-AMP-dependent transcription. Mol Cell Biol 19:2051–2060
Fabbro D, Pellizzari L, Mercuri F, Tell G, Damante G 1998 Pax-8 protein levels regulate thyroglobulin gene expression. J Mol Endocrinol 21:347–354
Van Renterghem P, Vassart G, Christophe D 1996 Pax 8 expression in primary cultured dog thyrocyte is increased by cyclic AMP. Biochim Biophys Acta 1307:97–103
Chun YS, Saji M, Zeiger MA 1998 Overexpression of TTF-1 and PAX-8 restores thyroglobulin gene promoter activity in ARO and WRO cell lines. Surgery 124:1100–1105
Pellizzari L, D’Elia A, Rustighi A, Manfioletti G, Tell G, Damante G 2000 Expression and function of the homeodomain-containing protein Hex in thyroid cells. Nucleic Acids Res 28:2503–2511
Friedrichsen S, Christ S, Heuer H, Schafer MK, Mansouri A, Bauer K, Visser TJ 2003 Regulation of iodothyronine deiodinases in the Pax8–/– mouse model of congenital hypothyroidism. Endocrinology 144:777–784
Ambroziak M, Pachucki J, Chojnowski K, Wiechno W, Nauman J, Nauman A 2003 Pax-8 expression correlates with type II 5' deiodinase expression in thyroids from patients with Graves’ disease. Thyroid 13:141–148
Bidart JM, Lacroix L, Evain-Brion D, Caillou B, Lazar V, Frydman R, Bellet D, Filetti S, Schlumberger M 2000 Expression of Na+/I– symporter and Pendred syndrome genes in trophoblast cells. J Clin Endocrinol Metab 85:4367–4372
Chan S, Kachilele S, Hobbs E, Bulmer JN, Boelaert K, McCabe CJ, Driver PM, Bradwell AR, Kester M, Visser TJ, Franklyn JA, Kilby MD 2003 Placental iodothyronine deiodinase expression in normal and growth-restricted human pregnancies. J Clin Endocrinol Metab 88:4488–4495
Huang SA, Dorfman DM, Genest DR, Salvatore D, Larsen PR 2003 Type 3 iodothyronine deiodinase is highly expressed in the human uteroplacental unit and in fetal epithelium. J Clin Endocrinol Metab 88:1384–1388
Feingold M, Zilberstein M, Srivastava RK, Seibel MM, Bar-Ami S, Hambartsoumian E 1998 Expression of Wilms’ tumor suppressor gene (WT1) in term human trophoblast: regulation by cyclic adenosine 3',5'-monophosphate. J Clin Endocrinol Metab 83:2503–2508
Madden SL, Cook DM, Morris JF, Gashler A, Sukhatme VP, Rauscher III FJ 1991 Transcriptional repression mediated by the WT1 Wilms’ tumor gene product. Science 253:1550–1553
Scholz H, Kirschner KM 2005 A role for the Wilms’ tumor protein WT1 in organ development. Physiology (Bethesda) 20:54–59
Pellizzari L, Fabbro D, Lonigro R, Di Lauro R, Damante G 1996 A network of specific minor-groove contacts is a common characteristic of paired-domain-DNA interactions. Biochem J 315:363–367
Arturi F, Lacroix L, Presta I, Scarpelli D, Caillou B, Schlumberger M, Russo D, Bidart JM, Filetti S 2002 Regulation by human chorionic gonadotropin of sodium/iodide symporter gene expression in the JAR human choriocarcinoma cell line. Endocrinology 143:2216–2220
Fraizer GC, Shimamura R, Zhang X, Saunders GF 1997 PAX8 regulates human WT1 transcription through a novel DNA binding site. J Biol Chem 272:30678–30687
Dehbi M, Pelletier J 1996 PAX8-mediated activation of the wt1 tumor suppressor gene. EMBO J 15:4297–4306
Mitchell AM, Manley SW, Morris JC, Powell KA, Bergert ER, Mortimer RH 2001 Sodium iodide symporter (NIS) gene expression in human placenta. Placenta 22:256–258
Chen CP, Chen CY, Yang YC, Su TH, Chen H 2004 Decreased placental GCM1 (glial cells missing) gene expression in pre-eclampsia. Placenta 25:413–421
Hochberg A, Rachmilewitz J, Eldar-Geva T, Salant T, Schneider T, de Groot N 1992 Differentiation of choriocarcinoma cell line (JAr). Cancer Res 52:3713–3717
Schmitt TL, Espinoza CR, Loos U 2001 Transcriptional regulation of the human sodium/iodide symporter gene by PAX8 and TTF-1. Exp Clin Endocrinol Diabetes 109:27–31
Costamagna E, Garcia B, Santisteban P 2004 The functional interaction between the paired domain transcription factor Pax8 and Smad3 is involved in transforming growth factor- repression of the sodium/iodide symporter gene. J Biol Chem 279:3439–3446
Taki K, Kogai T, Kanamoto Y, Hershman JM, Brent GA 2002 A thyroid-specific far-upstream enhancer in the human sodium/iodide symporter gene requires Pax-8 binding and cyclic adenosine 3',5'-monophosphate response element-like sequence binding proteins for full activity and is differentially regulated in normal and thyroid cancer cells. Mol Endocrinol 16:2266–2282
Kumar MS, Owens GK 2003 Combinatorial control of smooth muscle-specific gene expression. Arterioscler Thromb Vasc Biol 23:737–747
Chiu IM, Touhalisky K, Baran C 2001 Multiple controlling mechanisms of FGF1 gene expression through multiple tissue-specific promoters. Prog Nucleic Acids Res Mol Biol 70:155–174
Eccles MR, Yun K, Reeve AE, Fidler AE 1995 Comparative in situ hybridization analysis of PAX2, PAX8, and WT1 gene transcription in human fetal kidney and Wilms’ tumors. Am J Pathol 146:40–45
Siehl JM, Thiel E, Heufelder K, Snarski E, Schwartz S, Mailander V, Keilholz U 2003 Possible regulation of Wilms’ tumour gene 1 (WT1) expression by the paired box genes PAX2 and PAX8 and by the haematopoietic transcription factor GATA-1 in human acute myeloid leukaemias. Br J Haematol 123:235–242
Mansouri A Wendler F, Fickenscher H, Zannini MS, Yaginuma K, Abbott C, Plachov D 1995 Distinct functional properties of three human paired-box-protein, PAX8, isoforms generated by alternative splicing in thyroid, kidney and Wilms’ tumors. Eur J Biochem 228:899–911(Elisabetta Ferretti1, Fra)
Neuromed Institute (A.G.), 86077 Pozzilli, Italy
Department of Clinical and Experimental Medicine (F.A., I.P., D.R.), G. Salvatore, and Department of Farmacobiological Sciences, University of Catanzaro, 88100 Catanzaro, Italy
Department of Science and Biomedical Tecnology (G.T., G.D.), University of Udine, 33100 Udine, Italy
Departments of Clinical Biology (L.L.), Institut Gustave-Roussy, 94805 Villejuif Cedex, France
Abstract
Pax proteins are transcriptional regulators that control a variety of developmental decisions in vertebrates. During development, the paired-box gene 8 (PAX8) is expressed in the thyroid, kidney, and several areas of the central nervous system. It is also expressed in the adult thyroid gland, in which it mediates TSH-induced modulation of the expression of important genes, such as those encoding thyroglobulin, thyroperoxidase, and the sodium/iodide symporter (NIS). Thus far, placental expression of PAX8 has been described only in mice. In the present study, we show that PAX8 is also expressed in the human placenta at term. In an in vitro model of placental cancer, the JAR choriocarcinoma cell line, human chorionic gonadotropin (hCG) increased levels of PAX8 mRNA and protein, and gel retardation assays indicated that the up-regulation of PAX8 protein expression is associated with an increase in its DNA-binding activity. The effects of hCG were mimicked by forskolin, indicating that they are cAMP dependent. Levels of mRNA for the Wilms’ tumor 1 (WT1) and NIS genes were increased in JAR cells by hCG treatment, whereas overexpression of PAX8 increased only levels of WT1 mRNA. In cells transfected with PAX8-specific small interfering RNA, the stimulatory effects of hCG on WT1 mRNA levels were abolished, but hormonal enhancement of NIS mRNA levels was unchanged. These findings indicate that, in JAR cells, hCG activates a cAMP-dependent pathway that can up-regulate WT1 expression through PAX8.
Introduction
THE PLACENTA DEVELOPS from two major cell lineages. The trophectoderm of the blastocyst is the precursor of the trophoblast cell lineage, which gives rise to the epithelial parts of the placenta, whereas stromal cells and those of the blood vessels develop from the extraembryonic mesoderm (1). The molecular mechanisms involved in the development and maintenance of the human placenta are still largely unknown. However, several of the transcription factors that regulate trophoblastic cell development have already been identified (1, 2, 3). mRNA for the paired-box transcription factor, PAX8, has recently been described in murine placental cells (4), but thus far there is no evidence of PAX8 expression in human placental cells.
PAX8 is involved in the control of development and maintenance of various human tissues (5, 6, 7). Its expression has been demonstrated in the developing thyroid gland (5, 8) and adult follicular thyroid cells, in which it determines the differentiated phenotype (9). In thyroid cells, PAX8 expression is regulated by TSH through cAMP-dependent mechanisms (10, 11, 12), and it, in turn, regulates together with thyroid transcription factors 1 and 2 (TITF1/NKx2.1 and FoxE1) the expression of several essential thyroid genes (10, 11, 13, 14), including the sodium iodide symporter (NIS). Recent reports suggest that PAX8 is also involved in modulating the expression of D2 deiodinase (15, 16), one of the enzymes responsible for the deiodination of iodothyronines. More recent evidence indicates that the NIS (17) and the iodothyronine deiodinases, D2 and D3, which were originally considered to be thyroid specific, are also expressed by placental trophoblasts (18, 19), in which their presence is essential for ensuring adequate supplies of iodide for the synthesis of fetal thyroid hormones. These discoveries suggested that placental expression levels of these genes might be modulated by PAX8, as they are in the thyroid.
The Wilms’ tumor 1 (WT1) gene, another well-known target of PAX8 regulation, is also known to be expressed by human trophoblasts (20). The transcription factor it encodes was originally identified as a tumor suppressor based on genetic analyses showing that loss of WT1 function can initiate (or promote) tumorigenesis and functional studies demonstrating that wild-type WT1 expression is usually associated with inhibition of cellular proliferation (21). It has now become clear, however, that WT1 also plays important roles in the prenatal development of a number of organs (i.e. kidney, gonads, spleen, adrenal glands, retina, and heart), as recently reviewed by Scholz and Kirschner (22). Its precise function in the human placenta is unknown, as are the mechanisms underlying the regulation of its expression by trophoblasts, but its up-regulation has been shown to involve cAMP-dependent mechanisms (20). The latter finding is of particular interest because, as noted above, cAMP dependency is also a feature of TSH-induced modulation of thyrocyte expression of PAX8 (10, 11, 12).
The aim of the present study was to analyze the expression of PAX8 and its regulation in the human placenta and investigate its effects on the transcription of two of its known gene targets, NIS and WT1.
Materials and Methods
Tissues samples, cell lines, and reagents
All tissue samples used in the study had been collected with full patient consent and institutional review board approval. They included specimens from 10 normal full-term human placentas and 10 specimens of normal thyroid tissue (used as controls). Immediately after removal, part of each tissue specimen was snap frozen in liquid nitrogen and stored at –80 C for RNA and protein isolation; the rest was formalin fixed for histological studies. All cell lines used in the study were from the American Type Culture Collection (Manassas, VA). Choriocarcinoma cells, JAR and JEG-3, were grown in 10% heat-inactivated fetal calf serum/RPMI 1640 medium. D283 medulloblastoma cells were cultured in 20% fetal calf serum/HEPES/glutamine MEM.
Cell treatments
Hormonal regulation of placental PAX8 expression was investigated in JAR cells before and after exposure to 5 UI/ml human chorionic gonadotropin (hCG) (obtained from Sigma Chemical Corp., St. Louis, MO). To determine whether the hCG-induced effects were cAMP mediated, cells were grown in medium containing 10 μM forskolin (Sigma).
In PAX8 overexpression experiments, JAR cells were transfected with full-length human PAX8 cDNA, kindly provided by R. Di Lauro (University of Naples, Naples, Italy), by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA), according to the instructions of the manufacturer.
For PAX8 gene-silencing studies, JAR cells were transfected (Oligofectamine agent, Invitrogen) with PAX8-specific small interfering mRNA (siRNA) (sequence: AAGTGCAGCAACCATTCAACC) synthesized with the QIAGEN system; control siRNA (GFP-22, QIAGEN S.p.A., Milan, Italy); or the Oligofectamine agent alone. Twelve hours after transfection, hCG (5 UI/ml) was added to culture medium, and the cells were incubated for up to 48 h.
Quantitative mRNA evaluation by real-time PCR
Total RNA was extracted from tissue or cells with the TRIzol reagent (Invitrogen) and reverse transcribed with oligo(dT) primers and 200 U Superscript II reverse transcriptase (Invitrogen), in accordance with the manufacturer’s instructions. Quantitative PCR (Q-PCR) analysis of PAX8, NIS, and WT1 mRNA expression was performed on each cDNA sample using the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA). Oligonucleotide primers and probes for the genes analyzed and the endogenous control were purchased from Applied Biosystems. A reaction mixture containing cDNA template, TaqMan Universal PCR master mix (Applied Biosystems), and primer probe mixture was amplified using standard Q-PCR thermal cycler parameters. Each amplification reaction was performed in triplicate, and the average of the three threshold cycles was used to calculate the amount of transcript in the sample (using SDS version 1.7a software; Applied Biosystems). mRNA quantification was expressed, in arbitrary units as the ratio of the sample quantity to the quantity of the calibrator. All values were normalized with two endogenous controls, glyceraldehyde-3-phosphate dehydrogenase and -actin, which yielded similar results.
Nuclear extract preparation, Western blot, and gel retardation assays
Nuclear extracts were prepared from frozen pelleted cells and tissues that had been homogenized in a buffer containing 10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl2, 0.1 mM EGTA, 0.1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 2 μg/ml leupeptin. After the nuclei had been released, the suspension was overlaid with the same buffer and centrifuged. The pelleted nuclei were rewashed with the same buffer and centrifuged. Nuclear proteins were extracted by a second buffer containing 10 mM HEPES (pH 7.9), 400 mM NaCl, 1.5 mM MgCl2, 0.1 mM EGTA, 5% glycerol, 0.1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 2 μg/ml leupeptin. The protein concentration of the supernatant was measured using the Bradford method (Bio-Rad Laboratories, Inc., Hercules, CA).
Immunoblot analysis of PAX8 expression was performed with polyclonal goat PAX8 antibody (catalog no. SC12679, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and monoclonal mouse WT1 antibody (DakoCytomation S.p.A, Milan, Italy). Proteins (25 μg/lane) were run on a 10% SDS-PAGE gel and transferred to nitrocellulose membranes (Amersham S.r.l., Milan, Italy). Membranes were blocked in 5% milk and incubated overnight with primary PAX8 antibody (1:100 dilution) and then for an additional 12 h with primary WT1 antibody (1:100 dilution). Polyclonal rabbit histone 2A antibody (Santa Cruz Biotechnology) was used as an internal control.
For gel retardation assays, 10 μg of each nuclear extract were incubated with DNA for 30 min at room temperature in a buffer containing 20 mM Tris-HCl (pH 7.6), 75 mM KCl, 10 mg/ml calf thymus DNA, and 10% glycerol. The oligonucleotides used were C and specificity protein 1 (SP1), which contain high-affinity binding sites for PAX8 and SP1, respectively (23). Oligonucleotides were labeled at the 5' end using polynucleotide kinase (Amersham) and [-32P]ATP and annealed with their respective complementary strands. At the end of the binding reaction, samples were loaded onto a 7.5% native polyacrylamide gel and run at 4 C in Tris-borate-EDTA buffer. After electrophoresis, gels were fixed in 10% acetic acid and subjected to autoradiography. In supershift and competition experiments, the labeled C oligonucleotide was added after the nuclear extracts had been preincubated with either antisera or unlabeled oligonucleotides for 30 min at room temperature.
Immunofluorescence
Sections of formalin-fixed paraffin-embedded normal placental samples were prepared on poly-L-lysine-coated slides and control stained with hematoxylin and eosin (Sigma). For immunohistochemistry (IHC), sections were deparaffinized, rehydrated, and microwave heated in 10 mM citrate buffer (pH 6) for antigen retrieval. In studies designed to detect coexpression of PAX8 with NIS and/or WT1, the sections were first immunostained overnight with goat polyclonal anti-PAX8 diluted 1:20 (SC12679) and, after three washes with PBS, they were incubated with fluorescein isothiocyanate-conjugated antigoat secondary antibody (Sigma-Aldrich) diluted 1:30. The sections were then immunostained with monoclonal anti-NIS or anti-WT1 antibodies [U.S. Biological (Swampscott, MA), and DakoCytomation, respectively] diluted 1:100 and incubated with a 1:40 dilution of Texas-RED-conjugated antimouse secondary antibody (Jackson Laboratory, Bar Harbor, ME). For negative controls, sections were incubated with PBS instead of primary antibodies. The samples were mounted and photographed under phase-contrast and florescence microscopy (BH2; Olympus, Tokyo, Japan).
Statistical analysis
Results are expressed as means ± SD. Differences between different points of stimulation were analyzed with the Mann-Whitney U test. P < 0.05 was considered statistically significant.
Results
Human placental expression of PAX8
PAX8 mRNA has been previously demonstrated in mouse placenta (4). Using RT-Q-PCR, we demonstrated that mRNA for PAX8 (along with that for its target genes, NIS and WT1) is also expressed in normal human placenta, as well as in the two widely used choriocarcinoma cell lines, JAR and JEG-3 (Fig. 1A). Western blot analysis was then used to evaluate PAX8 protein expression in nuclear extracts prepared from normal human placental tissue. As shown in Fig. 1B, placental expression of PAX8 was found to be comparable to that observed in normal human thyroid tissue, and immunohistochemical studies of the placental sections localized the PAX8 expression to the trophoblasts (Fig. 1, C and D). These same sections were subjected to immunofluorescence analysis after incubation with antibodies specific directed against the NIS and WT1 proteins. As shown in Fig. 1D, more than half (five of nine) of the cells displaying WT1 immunoreactivity were also positive for PAX8. In contrast, PAX8-positive staining was much less common (five of 25) in NIS-positive cells (Fig. 1C).
Regulation of PAX8 expression in JAR cells
The expression of a number of genes in JAR cells is known to be hormonally regulated (by hCG) as we previously reported for NIS (24). Thus, the possibility that human placental expression of PAX8 is hormonally regulated was investigated in JAR cells exposed to 5 IU/ml of hCG for up to 48 h. As shown in Fig. 2A, levels of PAX8 mRNA were significantly increased by the hormonal treatment. Peak effects (4- to 5-fold increases over basal expression) were observed after 6–8 h of hCG stimulation; expression decreased thereafter and returned to near basal levels after 24–48 h of stimulation. Less marked but detectable increases were also observed with lower hCG doses (3 IU/ml) (data not shown). Western blot experiments showed that PAX8 protein expression in JAR cells was also increased by hCG stimulation: maximal enhancement was observed after 8 h of hCG exposure, and expression was still significantly increased at 24 h (Fig. 2B).
Cell treatment with hCG (5 IU/ml) caused a similar increase in mRNA levels for WT1, but this effect was observed later, after 12, 24, and 48 h of hCG stimulation (Fig. 2A). Again, Western blot experiments confirmed that this increase was also reflected in levels of WT1 protein (Fig. 2C). The effects of hCG on expression of NIS were not tested because a previous study by our group (24) has already demonstrated that hCG (5 IU/ml) up-regulates JAR cell expression of NIS at both the mRNA and protein levels. Because the same study had also showed that the hormone’s effects on NIS expression involved a cAMP-dependent mechanism, we reevaluated WT1 mRNA levels in JAR cells treated with forskolin (10 μM), instead of hCG, and found that the cAMP analog also produced significant increases in WT1 transcript levels (data not shown).
We next attempted to determine whether the increased PAX8 protein expression corresponded to an increase in its DNA-binding activity. Nuclear extracts from JAR cells treated with hCG (5 UI/ml) were analyzed in a gel retardation assay performed with the PAX8-specific oligonucleotide C (23). As shown in Fig. 3A, PAX8 DNA-binding activity was indeed increased by the hormonal treatment, which produced maximum effects after 12 h stimulation. The hCG-induced increase was mimicked by cell treatment with forskolin (10 μM), indicating that the effect was cAMP dependent. The PAX8 specificity of the effects of hCG and forskolin was confirmed by the absence of changes in DNA binding to the control oligonucleotide, SP1 (Fig. 3B). The shift detected with the C oligonucleotide was due to PAX8 binding. In fact, when nuclear extracts from JAR cells treated for 12 h with forskolin were used, preincubation with PAX8 antiserum produced a supershift, which was not detected with preimmune serum (Fig. 3C). In addition, the band was competed by a 20-fold excess of unlabeled C oligonucleotide but not by 20-fold excess of the Cant2 oligonucleotide, which is a mutant of C no longer recognized by PAX8 (23). Similar results were obtained using nuclear extracts of JAR cells treated for 12 h with hCG (data not shown).
Transcriptional effects of PAX8 in JAR cells
The experiments described above clearly demonstrated that PAX8 is expressed in JAR cells and that this expression is regulated by hCG. To identify the effects of this expression, we overexpressed PAX8 in JAR cells by means of transfection and performed quantitative RT-PCR to measure the levels of mRNA for the two PAX8-target genes, WT1 (25, 26) and NIS (10), both of which are known to be expressed in human placenta (4, 17, 27). As shown in Fig. 4A, overexpression of PAX8 had no effect on NIS mRNA expression in JAR cells, but it markedly enhanced that of WT1 mRNA. Western blot experiments showing the PAX8 protein expression levels increased after transfection with PAX8 expression vector (0.1 and 1 μg) were used as experimental control (Fig. 4B). These results suggest a role for PAX8 in the control of WT1 expression in the human placenta.
The effects of hCG were then reassessed in JAR cells that had been subjected to PAX8 gene-silencing treatment (Fig. 5). Cells were transfected with PAX-8-specific or control siRNA, and 12 h later hCG (5 UI/ml) was added to the culture medium. As shown in Fig. 5A, in cells transfected with specific PAX8 siRNA, the hormone produced no increase in PAX8 mRNA levels. Indeed, these levels actually dropped, and maximal reduction was observed after 20 h of PAX8 siRNA. As expected, PAX8 gene silencing had no effect on hCG’s modulation of NIS mRNA levels (Fig. 5B), but it significantly diminished the hormones effect on WT1 mRNA levels (P < 0.05) (Fig. 5C). As control, siRNA-transfected cells were assayed for PAX8 protein levels by Western blot analysis 20 and 48 h after transfection (data not shown). These data strongly support the notion that PAX8 exerts direct control over JAR-cell expression of WT1 but not that of NIS.
Discussion
To date, evidence of placental expression of PAX gene transcription factors has been limited to a single report documenting PAX8 mRNA in the placental cells of mice (4). In the present study, we have demonstrated that: 1) PAX8 is expressed in the human placenta at term, mainly in the nuclei of the trophoblast cells; 2) placental PAX8 expression is controlled by hCG through a cAMP-dependent mechanism; and 3) in placental cells, PAX8 modulates the expression of the tumor-suppressor gene WT1 but has no effect on that of the NIS gene.
The transcription factors currently known to be expressed in the human placenta include basic helix-loop-helix proteins, such as HAND1 (heart and neural creast derivates expressed 1) and MASH2 (achaete-scute complex-like protein 2) [Ascl2 (achaete-scute complex homolog-like 2)], and the glial cells missing-1 transcription factor (2), whose down-regulation is thought to contribute to the development of preeclampsia (28). Several homeobox genes are also expressed in placental tissue, and it has been suggested that down-regulation of homeobox A11 may be necessary for the differentiation of cytotrophoblasts into syncytiotrophoblasts (3). In light of our findings, PAX8 must be added to the list of transcription factors involved in the regulation of gene expression in placental cells. Apart from PAX8, however, there is currently no evidence that any other PAX genes are expressed in the placenta.
PAX8 has been more thoroughly investigated in thyroid cells. Its expression in these cells is known to be hormonally regulated (by TSH) via cAMP-dependent mechanisms (10, 11, 12), and one of its main functions is the regulation of NIS gene expression (8). Our findings indicate that placental expression of PAX8 is also subject to cAMP-mediated hormonal regulation (in this case, by hCG), but in this tissue the transcription factor does not appear to exert any effect on the expression of the NIS. These data are based on experiments conducted in transformed cells, and whereas JAR cells are known to retain most of the features of normal placental cells (29), we cannot exclude the possibility that, in normal placental cells, PAX8 does indeed modulate NIS expression.
PAX8 regulation of NIS transcription requires the cooperative effects of other transcriptional regulators in thyrocytes (5, 8, 10, 30, 31). Thus, it was not fully surprising that overexpression of PAX8 in JAR placental cells did not lead to a modulation of NIS expression. Our findings could be a reflection of a complex molecular mechanism that controls NIS gene expression involving more than one factor, different between placental and thyroid cells. This hypothesis is consistent with previous reports showing that regulation of NIS promoter activity in the thyroid involves functional interaction between PAX8 and factors that bind a cAMP-responsive element-like sequence (10, 32). Thus, NIS is expressed by both thyroid and placental cells, but the molecular mechanisms that control its expression might be different in the two cell types. As various investigators have recently noted, tissue-specific gene regulation is not achieved with a few specific transcription factors: it is the result of complex combinatorial interactions between multiple general and tissue-specific proteins (33, 34).
In contrast, PAX8 regulation of the WT1 gene does not seem to be cell type dependent. Evidence of this type of regulation has been obtained in various cell types (25, 35), and some authors have recently suggested that PAX8 may be involved in the overexpression of WT1 that is often associated with acute myeloid leukemias in humans (36). Our findings are consistent with those of Feingold et al. (20), who showed that WT1 expression in human trophoblasts is regulated by means of cAMP-dependent mechanisms. The precise function(s) of WT1 in placental development is still unclear, but there is suggestive evidence of its possible role in the suppression of trophoblast growth and invasion (20).
The absence of PAX8 is known to have detrimental effects on the development of the human thyroid gland, but its impact on the development and function of the human placenta has never been investigated. Its placental effects cannot be investigated in Pax8-null mice because these animals are born without the follicular portion of the thyroid gland and die before reaching reproductive age (37). Nonetheless, our findings predict that inactivating mutations or functionally relevant polymorphisms of the PAX8 gene could influence placental function. Although PAX8 does not appear to be a direct regulator of NIS expression in human trophoblasts (as it is in thyrocytes), it does seem to be involved in the cAMP-dependent regulation of the Wilms’ tumor suppressor gene. Future studies on placental tissues will hopefully shed more light on the role of this transcription factor in the development and functions of placenta.
Acknowledgments
We thank D. De Sanctis for technical assistance and M. Kent for editorial assistance.
Footnotes
This work was supported by grants from the Italian Ministry of Health and the Italian Ministry of University and Research (to S. F.).
1 E.F. and F.A. contributed equally to this work.
Abbreviations: hCG, Human chorionic gonadotropin; IHC, immunohistochemistry; NIS, sodium iodide symporter; PAX8, paired-box gene 8; Q-PCR, quantitative PCR; siRNA, small interfering mRNA; SP1, specificity protein 1; WT1, Wilms’ tumor 1.
References
Cross JC, Baczyk D, Dobric N, Hemberger M, Hughes M, Simmons DG, Yamamoto H, Kingdom JC 2003 Genes, development and evolution of the placenta. Placenta 24:123–130
Hemberger M, Cross JC 2001 Genes governing placental development. Trends Endocrinol Metab 12:162–168
Zhang YM, Xu B, Rote N, Peterson L, Amesse LS 2002 Expression of homeobox gene transcripts in trophoblastic cells. Am J Obstet Gynecol 187:24–32
Kozmik Z, Kurzbauer R, Dorfler P, Busslinger M 1993 Alternative splicing of Pax-8 gene transcripts is developmentally regulated and generates isoforms with different transactivation properties. Mol Cell Biol 13:6024–6035
Damante G, Tell G, Di Lauro R 2001 A unique combination of transcription factors controls differentiation of thyroid cells. Prog Nucleic Acids Res Mol Biol 66:307–356
Poleev A, Fickenscher H, Mundlos S, Winterpacht A, Zabel B, Fidler A, Gruss P, Plachov D 1992 PAX8, a human paired box gene: isolation and expression in developing thyroid, kidney and Wilms’ tumors. Development 116:611–623
Plachov D, Chowdhury K, Walther C, Simon D, Guenet JL, Gruss P 1990 PAX8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110:643–651
Pasca di Magliano M, Di Lauro R, Zannini M 2000 PAX8 has a key role in thyroid cell differentiation. Proc Natl Acad Sci USA 97:13144–13149
Zannini M, Francis-Lang H, Plachov D, Di Lauro R 1992 Pax-8, a paired domain-containing protein, binds to a sequence overlapping the recognition site of a homeodomain and activates transcription from two thyroid-specific promoters. Mol Cell Biol 12:4230–4241
Ohno M, Zannini M, Levy O, Carrasco N, Di Lauro R 1999 The paired-domain transcription factor PAX8 binds to the upstream enhancer of the rat sodium/iodide symporter gene and participates in both thyroid-specific and cyclic-AMP-dependent transcription. Mol Cell Biol 19:2051–2060
Fabbro D, Pellizzari L, Mercuri F, Tell G, Damante G 1998 Pax-8 protein levels regulate thyroglobulin gene expression. J Mol Endocrinol 21:347–354
Van Renterghem P, Vassart G, Christophe D 1996 Pax 8 expression in primary cultured dog thyrocyte is increased by cyclic AMP. Biochim Biophys Acta 1307:97–103
Chun YS, Saji M, Zeiger MA 1998 Overexpression of TTF-1 and PAX-8 restores thyroglobulin gene promoter activity in ARO and WRO cell lines. Surgery 124:1100–1105
Pellizzari L, D’Elia A, Rustighi A, Manfioletti G, Tell G, Damante G 2000 Expression and function of the homeodomain-containing protein Hex in thyroid cells. Nucleic Acids Res 28:2503–2511
Friedrichsen S, Christ S, Heuer H, Schafer MK, Mansouri A, Bauer K, Visser TJ 2003 Regulation of iodothyronine deiodinases in the Pax8–/– mouse model of congenital hypothyroidism. Endocrinology 144:777–784
Ambroziak M, Pachucki J, Chojnowski K, Wiechno W, Nauman J, Nauman A 2003 Pax-8 expression correlates with type II 5' deiodinase expression in thyroids from patients with Graves’ disease. Thyroid 13:141–148
Bidart JM, Lacroix L, Evain-Brion D, Caillou B, Lazar V, Frydman R, Bellet D, Filetti S, Schlumberger M 2000 Expression of Na+/I– symporter and Pendred syndrome genes in trophoblast cells. J Clin Endocrinol Metab 85:4367–4372
Chan S, Kachilele S, Hobbs E, Bulmer JN, Boelaert K, McCabe CJ, Driver PM, Bradwell AR, Kester M, Visser TJ, Franklyn JA, Kilby MD 2003 Placental iodothyronine deiodinase expression in normal and growth-restricted human pregnancies. J Clin Endocrinol Metab 88:4488–4495
Huang SA, Dorfman DM, Genest DR, Salvatore D, Larsen PR 2003 Type 3 iodothyronine deiodinase is highly expressed in the human uteroplacental unit and in fetal epithelium. J Clin Endocrinol Metab 88:1384–1388
Feingold M, Zilberstein M, Srivastava RK, Seibel MM, Bar-Ami S, Hambartsoumian E 1998 Expression of Wilms’ tumor suppressor gene (WT1) in term human trophoblast: regulation by cyclic adenosine 3',5'-monophosphate. J Clin Endocrinol Metab 83:2503–2508
Madden SL, Cook DM, Morris JF, Gashler A, Sukhatme VP, Rauscher III FJ 1991 Transcriptional repression mediated by the WT1 Wilms’ tumor gene product. Science 253:1550–1553
Scholz H, Kirschner KM 2005 A role for the Wilms’ tumor protein WT1 in organ development. Physiology (Bethesda) 20:54–59
Pellizzari L, Fabbro D, Lonigro R, Di Lauro R, Damante G 1996 A network of specific minor-groove contacts is a common characteristic of paired-domain-DNA interactions. Biochem J 315:363–367
Arturi F, Lacroix L, Presta I, Scarpelli D, Caillou B, Schlumberger M, Russo D, Bidart JM, Filetti S 2002 Regulation by human chorionic gonadotropin of sodium/iodide symporter gene expression in the JAR human choriocarcinoma cell line. Endocrinology 143:2216–2220
Fraizer GC, Shimamura R, Zhang X, Saunders GF 1997 PAX8 regulates human WT1 transcription through a novel DNA binding site. J Biol Chem 272:30678–30687
Dehbi M, Pelletier J 1996 PAX8-mediated activation of the wt1 tumor suppressor gene. EMBO J 15:4297–4306
Mitchell AM, Manley SW, Morris JC, Powell KA, Bergert ER, Mortimer RH 2001 Sodium iodide symporter (NIS) gene expression in human placenta. Placenta 22:256–258
Chen CP, Chen CY, Yang YC, Su TH, Chen H 2004 Decreased placental GCM1 (glial cells missing) gene expression in pre-eclampsia. Placenta 25:413–421
Hochberg A, Rachmilewitz J, Eldar-Geva T, Salant T, Schneider T, de Groot N 1992 Differentiation of choriocarcinoma cell line (JAr). Cancer Res 52:3713–3717
Schmitt TL, Espinoza CR, Loos U 2001 Transcriptional regulation of the human sodium/iodide symporter gene by PAX8 and TTF-1. Exp Clin Endocrinol Diabetes 109:27–31
Costamagna E, Garcia B, Santisteban P 2004 The functional interaction between the paired domain transcription factor Pax8 and Smad3 is involved in transforming growth factor- repression of the sodium/iodide symporter gene. J Biol Chem 279:3439–3446
Taki K, Kogai T, Kanamoto Y, Hershman JM, Brent GA 2002 A thyroid-specific far-upstream enhancer in the human sodium/iodide symporter gene requires Pax-8 binding and cyclic adenosine 3',5'-monophosphate response element-like sequence binding proteins for full activity and is differentially regulated in normal and thyroid cancer cells. Mol Endocrinol 16:2266–2282
Kumar MS, Owens GK 2003 Combinatorial control of smooth muscle-specific gene expression. Arterioscler Thromb Vasc Biol 23:737–747
Chiu IM, Touhalisky K, Baran C 2001 Multiple controlling mechanisms of FGF1 gene expression through multiple tissue-specific promoters. Prog Nucleic Acids Res Mol Biol 70:155–174
Eccles MR, Yun K, Reeve AE, Fidler AE 1995 Comparative in situ hybridization analysis of PAX2, PAX8, and WT1 gene transcription in human fetal kidney and Wilms’ tumors. Am J Pathol 146:40–45
Siehl JM, Thiel E, Heufelder K, Snarski E, Schwartz S, Mailander V, Keilholz U 2003 Possible regulation of Wilms’ tumour gene 1 (WT1) expression by the paired box genes PAX2 and PAX8 and by the haematopoietic transcription factor GATA-1 in human acute myeloid leukaemias. Br J Haematol 123:235–242
Mansouri A Wendler F, Fickenscher H, Zannini MS, Yaginuma K, Abbott C, Plachov D 1995 Distinct functional properties of three human paired-box-protein, PAX8, isoforms generated by alternative splicing in thyroid, kidney and Wilms’ tumors. Eur J Biochem 228:899–911(Elisabetta Ferretti1, Fra)