0013-7227/02/$15.00/0 Printed in U.S.A.
Endocrinology 143(2):347–359 Copyright © 2002 by The Endocrine Society
Early Gene Expression Changes Preceding Thyroid Hormone-Induced Involution of a Thyrotrope Tumor WILLIAM M. WOOD, VIRGINIA D. SARAPURA, JANET M. DOWDING, WHITNEY W. WOODMANSEE, DANIELLE J. HAAKINSON, DAVID F. GORDON, AND E. CHESTER RIDGWAY Division of Endocrinology, Metabolism and Diabetes, Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado 80262 Treatment with thyroid hormone (TH) results in shrinkage of a thyrotropic tumor grown in a hypothyroid host. We used microarray and Northern analysis to assess the changes in gene expression that preceded tumor involution. Of the 1,176 genes on the microarray, 7 were up-regulated, whereas 40 were decreased by TH. Many of these were neuroendocrine in nature and related to growth or apoptosis. When we examined transcripts for cell cycle regulators only cyclin-dependent kinase 2, cyclin A and p57 were down-regulated, whereas p15 was induced by TH. Retinoblastoma protein, c-myc, and mdm2 were unchanged, but E2F1 was down-regulated. TH also decreased expression of brain-derived neurotrophic factor, its receptor trkB, and the receptor for TRH. These, in addition to
two other genes, neuronatin and PB cadherin, which were upand down-regulated, respectively, showed a more rapid response to TH than the cell cycle regulators and may represent direct targets of TH. Finally, p19ARF was dramatically induced by TH, and although this protein can stabilize p53 by sequestering mdm2, we found no increase in p53 protein up to 48 h of treatment. In summary, we have described early changes in the expression of genes that may play a role in TH-induced growth arrest of a thyrotropic tumor. These include repression of specific growth factor and receptors and cell cycle genes as well as induction of other factors associated with growth arrest and apoptosis. (Endocrinology 143: 347–359, 2002)
A
MONG THE MANY diverse actions of thyroid hormone (TH) on metabolism, development, and differentiation is its unique and vital role as a negative regulator of thyrotrope cell function. In addition to being responsible for inhibition of TSH secretion (1) and transcriptional repression of the ␣- and TSH-subunit genes (2), it also plays a regulatory role in the control of thyrotrope cell proliferation within the pituitary gland (3). Protracted hypothyroidism in humans as well as in mice causes high plasma levels of TSH and pituitary enlargement due to hyperplasia of thyrotrope cells as a result of lack of negative feedback regulation by TH (4). Treatment of hypothyroid patients with TH not only corrects the symptoms of hypothyroidism but also reverses the pituitary enlargement. The mechanism underlying this cell-specific antiproliferative effect of TH is currently unknown. The mouse TtT-97 thyrotropic tumor is a well differentiated thyrotrope hyperplasia that secretes TSH (5, 6). It is propagated by subscapular injection of dispersed tumor tissue into hypothyroid mice. When mice with established tumors are administered TH, there is a reduction of TSH secretion (1) and a gradual involution of the tumor (7). Both effects are reversed upon hormone withdrawal. Therefore, this tumor model can be used to study the mechanism underlying physiological regulation of the thyrotrope cell, specifically its growth inhibition by TH. We have previously shown, using this tumor as an in vivo thyrotrope model, that the inhibition of growth that results from prolonged treat-
ment with either pharmacological or physiological doses of TH, is associated with increased expression of the genes for somatostatin receptor (sst) subtypes 1 and 5 and somatostatin binding sites within the tumor (8, 9). Similar treatments of TtT-97 tumors with TH also resulted in opposing effects on the steady-state levels of mRNAs encoding the various TRs (7). TR1 levels were up-regulated with TH, whereas those for the pituitary-restricted TR2 isoform were downregulated. TR␣1 transcripts were unchanged or modestly decreased, whereas those encoding the nonhormone-binding ␣2 splice variant were down-regulated by TH. The existence of multiple TR isoforms and their differential autoregulation thus complicates an analysis of TH-regulated genes involved in growth. Growth stimulation by TH has long been recognized, as is manifested by the growth retardation and cretinism in humans deprived of TH during development. Patients with resistance to TH due to TR gene mutations and transgenic mice engineered to have similar dominant negative TR genotypes both exhibit growth defects (10, 11). These data underscore the critical role for TRs in TH regulation of growth and suggest that the inhibition of thyrotrope cell proliferation is also receptor mediated and therefore likely to be exerted at the level of gene expression. Cell growth is, for the most part, regulated by controlling entry into or exit from the cell cycle. An actively proliferating cell will progress through the various phases of the cycle including the DNA synthesis (S)-phase, mitosis (M)-phase, and eventual cell division. Whether a cell divides or not depends on its capacity to pass a certain irreversible checkpoint, after which it is committed to enter the S-phase. The critical determinant in the pathway that leads to S-phase commitment is the product of the gene encoding the retino-
Abbreviations: BDNF, Brain-derived neurotrophic factor; cdk, cyclindependent kinase; cdki, cdk inhibitor; M-phase, mitosis phase; oligo-dT, oligo-deoxythymidine; Rb, retinoblastoma protein; S-phase, synthesis phase; sst, somatostatin receptor; STAT, signal transduction activator of transcription factor; TH, thyroid hormone; TRHR, TRH receptor.
347
348
Endocrinology, February 2002, 143(2):347–359
blastoma protein (Rb). In its hypophosphorylated, or active, state, Rb binds to E2F transcription factor complexes and prevents them from activating several target genes required for initiation and progression through S-phase (12). Thus, phosphorylation of Rb is the key step in cell proliferation and its prevention results in growth arrest. Therefore, TH, acting via its receptor, most probably activates and/or represses the expression of one or more genes that subsequently leads to accumulation of hypophosphorylated Rb. Rb is phosphorylated by cyclin-dependent kinases that are composed of a catalytic subunit (a cdk) and a regulatory component (a cyclin). Different cyclins associate with different cdks to phosphorylate Rb at various stages of the cell cycle. Intracellular levels of cyclins D1, 2, and 3 are dependent on sustained growth factor stimulation and interact with cdk4 and cdk6 to initiate phosphorylation of Rb (13), whereas cyclin E and cyclin A combine with cdk2 to maintain Rb in a hyperphosphorylated state even after growth factor influences have been withdrawn (14). A decrease in expression in thyrotropes of one or more of these activating influences by TH would preclude S-phase entry and lead to growth arrest. Conversely, TH induction of one or more members of the Cip/Kip or Ink4 families of cdk inhibitors, which inhibit the activity of cyclin/cdk complexes (15), would also lead to repression of growth. In this report, we examined changes in gene expression as a consequence of TH treatment of TtT-97 tumors. Although a period of several weeks (our previously reported treatment time) was required to detect visible changes in mass between tumors treated or not with TH, we reasoned that effects of TH on gene expression would be manifested much earlier, and thus we chose to examine the effect on gene expression 24 h after TH administration. To maximize the number of genes studied, we initially used a high density microarray that contained more than 1,000 different mouse cDNAs comprising several functional categories including many of the cell cycle regulators. Although this approach uncovered several interesting genes, both up- and down-regulated by TH, the majority of the cDNAs on the microarray, including those within the category of cell cycle regulators, were not able to be quantified because of barely detectable hybridization signals. We then used conventional Northern blot analysis to evaluate changes in cell cycle regulators and other growth related molecules as a consequence of TH treatment. Together, the microarray and Northern analyses revealed that the expression of several factors associated with cell proliferation was altered by shorter treatment with TH. These included a limited cyclin/cdk/cdki (cdk inhibitor) array, several relevant growth factors, and their receptors, and other proteins associated with growth regulation as well as programmed cell death. Materials and Methods Propagation of thyrotropic tumors and treatment with TH TtT-97 thyrotropic tumor propagation and maintenance in male LAF1 hypothyroid mice have been described previously (16). To analyze the effect of TH on gene expression, mice were given a single injection of either TH (l-T3 at a dose of 10 g/100 g BW) (⫹T3) or vehicle (⫺T3). Twenty-four hours later, animals were anesthetized, weighed, and blood collected after killing by decapitation. Tumors were removed,
Wood et al. • TH-Induced Growth Arrest
weighed, and dissected pieces immediately processed for either RNA or protein extraction as described below. For time course studies, animals were killed also at 2 h, 6 h, 10 h, and 48 h following TH injection. In the case of the 48-h time point, a second TH injection of identical dose was administered at 24 h. Analysis of serum samples for TH levels showed that vehicle treated mice were hypothyroid (T3 levels ⬍28 ng/dl), whereas those given TH were hyperthyroid with T3 levels in excess of 170 ng/dl at all times measured.
RNA Northern blot analysis Total RNA was extracted from TtT-97 tumors excised from vehicleand T3-treated mice using the guanidinium isothiocyanate/CsCl gradient method as previously described (17). Purification of poly(A⫹)RNA by oligo-deoxythymidine (dT) cellulose chromatography was as described (18). Four paired sets of either 5- or 10-g aliquots of poly(A⫹)RNA from vehicle and T3-treated tumors were loaded in adjacent lanes of 0.8% denaturing agarose gels along with evenly positioned radiolabeled MW standards (a combination of HindIII digested phage DNA and HaeIII digested (⫻174 plasmid DNA) and subjected to electrophoresis and Northern blotting to Nytran nylon membranes as described previously (19). Filters were cut into strips each containing a standard and two lanes with RNA from vehicle- and T3-treated tumors, and each strip was hybridized overnight with a different fragment radiolabeled by nick translation using a commercial kit (Life Technologies, Inc.). The source and generation of the various fragments used as hybridization probes is described in the corresponding figure legend. After washing, which included a final stringent treatment at 65 C with 0.1⫻ SSC, the filters were then subjected to autoradiography for a period of time specified in the figure legends. When a desirable exposure was achieved the probe was removed by immersion in boiling 0.1⫻ SSC, 0.1% SDS three times for 15 min. After checking for lack of residual signal by a second exposure to photographic film for an appropriate length of time, the hybridization was repeated with a different probe. Finally, filters were probed with a radiolabeled mouse -actin fragment to check for loading uniformity.
Microarray analysis Poly(A⫹)RNA for use in the microarray analysis was prepared in a similar manner as for Northern blotting except that it was subjected to a DNase treatment before oligo(dT) cellulose chromatography. This step was included to ensure that contaminating genomic DNA would not act as a template for the primers used to make the radiolabeled probe for the subsequent microarray hybridization. Samples of total RNA (425 g) were resuspended in 850 l of 40 mm Tris-Cl (pH 7.5), 10 mm NaCl, and 6 mm MgCl2, and incubated with 42 U of DNase I (RQ1, Promega Corp., Madison, WI) for 30 min at 37 C to remove residual DNA contamination. Reactions were adjusted to 50 mm EDTA (pH 8) and 1 mg/ml mussel glycogen to terminate the reaction followed by sequential extraction with phenol:chloroform (1:1) and chloroform and ethanol precipitation. Poly(A⫹)RNA was then purified by oligo(dT) cellulose chromatography as described. Radiolabeled cDNA probes from vehicle- or T3-treated tumors were synthesized using Maloney murine leukemia virus reverse transcriptase in the presence of [␣32P]deoxy-ATP and a cDNA synthesis primer mix specific for the Atlas mouse 1.2 array (CLONTECH Laboratories, Inc., Palo Alto, CA) using the conditions specified by the supplier. Probes were purified from unincorporated nucleotides by NucleoSpin columns (CLONTECH Laboratories, Inc.) before use and were denatured by treatment at 68 C for 20 min with 100 mm NaOH, 1 mm EDTA followed by neutralization in 0.5 m sodium phosphate (pH 7.0). Duplicate Atlas 1.2 arrays consisting of 1,176 mouse cDNAs arranged on a nylon membrane were prehybridized in separate hybridization bottles, probes were added to 10 ml Express Hybridization mix (CLONTECH Laboratories, Inc.) supplemented with 1 mg of heat denatured and sheared salmon testes DNA and hybridization continued for 18 h at 68 C using a rotating Fisher hybridization oven. Filters were washed for 4 ⫻ 30 min in 2⫻ SSC, 1% SDS and 1 ⫻ 30 min in 0.1⫻ SSC, 0.5% SDS followed by autoradiography for 1 and then 8 d. An additional set of duplicate arrays were screened to allow reproducibility of the results from separate RNA samples.
Wood et al. • TH-Induced Growth Arrest
Isolation of whole cell lysates and immunoblot analysis TtT-97 tumors were excised from hypothyroid or T3-treated mice as above, dissected free of connective tissue, finely minced (2- to 3-mm pieces) on ice with a razor blade, and then force-filtered through a 1-mm coarse stainless steel wire mesh using a Cellector tissue sieve (E-C Apparatus Corp.) in 1–2 ml of ice-cold calcium and magnesium-free HBSS supplemented with 10 U penicillin and 10 g/ml streptomycin (HBSS). The filtrate was then passed sequentially through a series of increasingly finer mesh filters (250 m and 70 m) to generate a single cell suspension. Cells were washed 5 times with 50 ml of ice cold HBSS with centrifugation at 4 C (400 ⫻ g) between washes. Cell counts were determined with a hemocytometer and cells were resuspended in 1⫻ calcium and magnesium-free Dulbecco’s PBS (Life Technologies, Inc., Gaithersburg, MD) at a density of 5 ⫻ 106 cells/ml. Viable cells were assayed by trypan blue exclusion with viability approaching 90%. Cells were sedimented by centrifugation at 4 C for 10 min (400 ⫻ g) and resuspended on ice in TEA-SDS solubilization buffer (55 mm triethanolamine, pH 7.5), 111 mm NaCl, 2.2 mm EDTA, 0.44% SDS, 1 mm DTT, 10 g/ml leupeptin, 1 mm benzamidine, 1 mm sodium orthovanadate, 20 mm NaF, 1 mm PMSF, and a cocktail of protease inhibitors (Complete, Roche Molecular Biochemicals, Indianapolis, IN) at a density of 1.0 ⫻ 107 cells/ml. After a 10-min incubation on ice, the cells were disrupted by repeated aspirations (7⫻) through a 21-gauge needle and then with a 25-gauge needle. Cellular debris was removed by centrifugation at 400 ⫻ g at 4 C for 15 min. Aliquots were removed for protein determination using the DC Protein Assay (Bio-Rad Laboratories, Inc., Richmond, CA). Lysates were diluted an equal volume of 2⫻ treatment buffer [0.125 m Tris-Cl (pH 6.8), 4% SDS, 20% glycerol, and 10% 2-mercaptoethanol] and boiled for 3 min. Protein extracts were resolved by 10 or 12.5% SDS PAGE and electroblotted at 100 mA overnight at 4 C onto polyvinylidenedifluoride membranes. Membranes were blocked for 90 min with 5% nonfat milk in PBST [10 mm sodium phosphate (pH 7.4), 150 mm NaCl, and 0.1% Tween-20] and immobilized proteins incubated with a rabbit polyclonal p53 antibody (NCL-p53-CM5p, Novocastra Laboratories, Newcastle, UK), at a dilution of 1:5000, overnight at 4 C in blocking buffer, washed five times with PBST followed by incubation with goat antirabbit antibody conjugated to horseradish peroxidase (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at a dilution of 1:3000. For p19ARF detection, membranes were blocked in 10% milk in TBST [10 mm Tris-Cl (pH 8), 150 mm NaCl, 0.1% Tween-20] overnight at 4 C and then incubated with a rabbit polyclonal antibody (NB200-106, Novus Biologicals, Littleton, CO) at a dilution of 1:400 in TBST for 1 h at 25 C, followed by incubation with the antirabbit horseradish peroxidase secondary antibody at a dilution of 1:5000. Loss of the p19ARF signal was verified by coincubating the primary antibody with 50 g of a synthetic 19ARF blocking peptide (Novus Biologicals) in a parallel immunoblot. Specific proteins were detected by a chemiluminescent assay (Amersham Pharmacia Biotech, Piscataway, NJ).
Endocrinology, February 2002, 143(2):347–359 349
T3 for 24 h. The RNA from both tumors was reverse transcribed to generate radiolabeled cDNA that was subsequently used for probing the Atlas filters which were autoradiographed and developed after 1 and 8 d of exposure. The resultant hybridization signal was a measure of the expression level of the gene. Differential expression levels of individual genes were examined by comparing the relative hybridization signals from the control and T3-treated filters in two separate experiments. Forty-seven major targets of the 1,176 total genes represented on the array, the expression of which was either decreased (40) or increased (7) by TH are shown by numbered arrows in Fig. 1 (⫺T3, decreased) and B (⫹T3, increased), respectively. As a control for probe equivalence, the intensities of the signals for several abundant housekeeping genes including -actin, glyceraldehyde-3phosphate dehydrogenase and ubiquitin were shown to be
Animal treatment Studies on LAF1 mice bearing TtT-97 thyrotropic tumors were conducted with the highest standards of humane animal care in accordance with the NIH Guide for the Care and Use of Laboratory Animals. The protocols were approved by the Committee on Animal Care and Use of the University of Colorado Health Sciences Center (Denver, CO).
Results Changes in thyrotropic tumor gene expression assessed by microarray analysis after treatment for 24 h with TH
As an initial approach to analyze differences in gene expression levels in TtT-97 thyrotropic tumors treated with TH, we used a mouse Atlas high density cDNA expression array (CLONTECH Laboratories, Inc.) consisting of two identical filters containing 1,176 cDNAs encoding known genes of different functional classes, nine housekeeping control cDNAs as well as negative controls. We isolated poly(A⫹)RNA from TtT-97 tumors grown in hypothyroid mice treated for 24 h with ethanol vehicle and also from similar mice treated with
FIG. 1. Expression pattern of 1,176 genes in a microarray used to detect genes in thyrotrope cells regulated in response to TH. Separate microarray filters were hybridized to radiolabeled cDNA generated from poly(A⫹)RNA isolated from TtT-97 thyrotropic tumors treated with vehicle (⫺T3) or TH (⫹T3) for 24 h. After processing as outlined in Materials and Methods, the filters were subjected to autoradiography at ⫺80 C for 8 d. The genes represented on the filters are grouped by functional category into quadrants (A–F), as described in Results. Hybridizing signals representing genes down-regulated by TH are designated by numbered arrows in the top panel (⫺T3), whereas those up-regulated by TH are similarly labeled in the bottom panel. Arrows at the bottom of both panels identified with letters U, G, and A point to signals arising from the housekeeping genes ubiquitin, glyceraldehyde 3-phosphate dehydrogenase, and -actin, respectively.
350
Endocrinology, February 2002, 143(2):347–359
similar in the two tumor samples (A, G, and U, respectively, in Fig. 1). A listing of the identity of the TH-regulated genes with its GenBank accession number, its corresponding number designation from Fig. 1, and whether it is up or downregulated by TH is shown in Table 1. References describing the isolation and characterization of each TH-regulated gene can be found by accessing the appropriate GenBank number shown. As shown in Fig. 1 and described in Table 1, the 1,176 mouse cDNAs represented on the microarray are arranged in six quadrants (A–F in Fig. 1) and comprise products of genes involved in the control of cell surface antigenicity and gene transcription (quadrant A); cell cycle regulation and cell adhesion (quadrant B); extracellular transport and matrix formation, oncogenicity and tumor suppression, stress response, ion transport, trafficking, metabolic and translational processes (quadrant C); apoptosis and signal transduction (quadrant D and E); and finally protein turnover, cytoskeletal motility, and DNA synthesis, repair and recombination (quadrant F). Also in Table 1 is indicated the number of cDNAs represented in each category and the number that were above the limits of detection, even after an 8-d exposure, and thus subject to differential expression analysis. Unfortunately, most if not all of the genes within the cell cycle regulator category fell below the limits of detectability and thus were not analyzable by the microarray methodology. Those that showed the greatest up-regulation by TH (numbers 11, 20, 26, and 39 in Fig. 1) were the signal transduction activator of transcription factor family member (STAT1) (20), an Na-dependent serotonin transporter (21), a thymic maturation factor, thymosin -4 (22), and Crk-associated substrate, a protein involved in focal adhesion and apoptosis (23), respectively. In contrast, several genes showed striking suppression by TH (numbers 4, 5, 16, 22, 27, 28, 29, and 45 in Fig. 1). These included the neural E-box proteins nHLH1 and neurogenin 3 (24, 25); prohormone convertase and its substrate chromogranin B (26, 27), the ␣-subunit of the TNF family member, inhibin (28) and the neurotropin receptor trkB (29). Effect of TH on growth factors and their receptors
The finding that trkB expression was decreased prompted us to further explore the possibility that TH may regulate other growth factors within thyrotropes. Many growth factors have been described as being important in early pituitary development when specific cell types have still not been determined. These include members of the fibroblast growth factor and bone morphogenic protein families of growth factors (30, 31) as well as epidermal growth factor (32, 33). Because these growth factors are generally involved in the early maturation of all pituitary cell types, we chose initially to focus our studies on those potential growth-regulating factors and their receptors that might be more relevant in thyrotrope cells. We previously reported that transcript levels of TRs and both sst 1 and 5 were altered by TH treatment for 21–28 d (8). We measured mRNA levels, by Northern analysis, for the TR isoforms and both ssts after a shorter 24-h treatment (Fig. 2, A and B) and showed that 6.4-kb TR1 and 2.6-kb sst5 transcripts were up-regulated as before but messages encoding sst1 (3.8 kb), which was previously in-
Wood et al. • TH-Induced Growth Arrest
duced, and TR2 (6.4 kb), previously reduced, were unchanged after only 24 h of treatment. This suggests that while the sst5 and TR1 genes could be direct targets of TH, sst1 and TR2 changes are probably secondary. mRNA encoding preprosomatostatin (⬃1-kb message), the precursor of the sst ligand, was present in the tumor (Fig. 2C), suggesting an autocrine/paracrine loop. However, its level was not changed by TH treatment for 24 h (Fig. 2C). Another potential candidate involved specifically in thyrotrope cell growth is TRH, which has been shown to be expressed in (34) and secreted by (35) the pituitary gland. Although well characterized as a secretagogue and regulator of gene expression (36), its role as a growth factor is still controversial, although earlier data suggested that it increases the proliferation of anterior pituitary cells (37) and specifically of thyrotropes (38). To address the role of TRH in the thyrotrope model, we obtained cDNA probes for preproTRH and both TRH receptors; TRHR1 (39), and the recently characterized TRHR2 (40) and examined their expression and regulation by TH in TtT-97 tumors by Northern blots. Despite prolonged autoradiographic exposure, we were unable to detect appropriately sized transcripts of 1.6 kb encoding mouse preproTRH (41) in RNA samples from either hypothyroid or TH-treated TtT-97 tumors (not shown). However, two transcripts encoding TRHR1 of 4.3 and 1.8 kb were present and decreased by TH in TtT-97 tumors (Fig. 3A), whereas a 2.5-kb mRNA for the TRHR2 isoform was unaffected (Fig. 3B). Differently sized transcripts encoding mouse TRHR1 have not been reported, although there is evidence in the murine species for splicing within the 3⬘ untranslated region (42). Another growth factor potentially involved in the growth of thyrotropes is the neurotrophin brain-derived neurotrophic factor (BDNF), which specifically acts through the trkB receptor (43) whose transcripts were decreased in the earlier microarray analysis (Table 1). Interestingly, BDNF expression within the pituitary gland appears to be restricted to TSH staining cells (44). When we used Northern blotting to examine the transcript levels of both BDNF and its receptor trkB in TtT-97 tumors following TH treatment for 24 h, we found that levels of both the 3.9-kb and 2-kb messages due to different polyadenylation sites (45) corresponding to BDNF were reduced in abundance by TH (Fig. 4A), whereas only one transcript (⬃2.5 kb) of four detectable trkB mRNAs was significantly decreased (Fig. 4B). Figure 5 shows that levels of mRNAs encoding TRHR1 and BDNF are already reduced by 6 h, and possibly also by as early as 2 h, of TH treatment, suggesting that they could be primary targets of TH. TrkB transcripts, specifically the 2.5-kb species, are also modestly decreased by TH at these early time points. Therefore, again we see that TH decreases the expression of genes encoding a mitogen receptor and its ligand in our thyrotrope model suggesting that BDNF synthesized within the pituitary gland, in addition to TRH from the hypothalamus, could be contributing to thyrotrope cell expansion in the hypothyroid state. Effect of TH on the cell cycle regulators
During the G1 phase of the cell cycle, cells respond to extracellular cues such as the previously described growth factors that determine whether they will make the decision
Wood et al. • TH-Induced Growth Arrest
Endocrinology, February 2002, 143(2):347–359 351
TABLE 1. TH-responsive genes identified using differential hybridization of Atlas mouse cDNA expression arrays #
GenBank Accession No.
Gene Name
Regulation
Cell surface antigens (25 represented; 13 detectable) 1 X17320 Brain-specific polypeptide PEP-19 (Brain specific antigen PCP-4) 2 M18934 CD2 antigen
Up Down
Transcription factors and DNA-binding proteins (290 represented; 130 detectable) 3 AF039567 Msx-interacting zinc finger protein 1 (MIZ1) 4 M97506 Helix-loop-helix protein 1 (HEN1) (NSCL); NHLH1; HEN1 5 U76208 Neurogenin 3 (NGN3); MATH4B (mammalian atonal homolog 4B) 6 U36340 CACCC box-binding protein BKLF 7 M74517 GA-binding protein  2 subunit (GABBP2; GABPB) 8 U09419 RXR interacting protein (RIP15) 9 U41671 Zinc finger transcription factor RU49 10 X72310 Transcription factor E2F dimerization partner 1 (TFDP1); DRTF 1 11 U06924 Signal transducer and activator of transcription 1 (STAT1)
Down Down Down Down Down Down Up Down Up
Cell cycle regulators (45 represented; 0 detectable) Cell adhesion receptors and proteins (56 represented; 30 detectable) 12 D14888 Cadherin 4 (CDH4); retinal cadherin (R-cadherin; R-CAD) 13 U90331 Neural plakophilin-related arm-repeat protein (NPRAP) 14 AF036585 Semaphorin N 15 U43512 Dystroglycan 1 16 X97818 Semaphorin G precursor (Samaphorin G)
Up Down Down Down Down
Extracellular transporters (4 represented; 2 detectable) 17 D89076 Transthyretin precursor (Prealbumin) (TBPA)(TTR)(ATTR)
Down
Oncogenes and tumor suppressors (71 represented; 20 detectable) 18 X83974 Transcription termination factor 1 (TTF1)
Down
Stress response proteins (20 represented; 9 detectable) 19 M36829 84-kDa heat shock protein (HSP84); HSP 90-; 84-kDa antigen; HSPCB
Down
Ion channels and transport proteins (36 represented; 2 detectable) 20 AF013604 Sodium-dependent serotonin transporter; 5HT transporter (5HTT) 21 U34920 ATP-binding cassette 8; ABC8; homolog of Drosophila white
Up Down
Extracellular matrix proteins (2 represented; 0 detectable) Trafficking and targeting proteins (1 represented; 0 detectable) Metabolic pathways (12 represented; 2 detectable) 22 M58589 Neuroendocrine convertase 1 (NEC 1); prohormone convertase 1 (PC1)
Down
Posttranslational modification and folding (2 represented; 2 detectable) Translation (1 represented; 0 detectable) Apoptosis-associated proteins (60 represented; 14 detectable) 23 U83628 Defender against cell death 1 (DAD1)
Down
Receptors (by ligands) (95 represented; 36 detectable) 24 M98547 Tyrosine-protein kinase ryk precursor; kinase vik; nyk-R
Down
Extracellular cell signaling and communication (116 represented; 24 detectable) 25 X68837 Secretogranin II precursor (SGII); chromogranin C 26 X16053 Thymosin -4 27 X69618 Inhibin ␣-subunit precursor (INHA) 28 X53028 Chromogranin B
Down Up Down Down
Modulators, effectors and intracellular transducers (160 represented; 30 detectable) 29 M33385 TRK-B; bdnf/nt-3 growth factors tyrosine kinase receptor 30 U43205 Frizzled-3; Drosophila frizzled homolog 3; dishevelled receptor 31 U43319 Frizzled homolog 6 (FZD6) 32 U10871 Mitogen-activated protein kinase p38 (MAP kinase 38) CRK1; CSBP1; CSBP2 33 U51866 Casein kinase II ␣-related sequence 4 (CSNK2A1-RS4) 34 D28117 Protein phosphatase 2C ␣ isoform (PP2C-␣) 35 M96653 Adenylate cyclase 6 36 X95403 Rab2 ras-related protein 37 M16465 Calpactin I light chain; p10 protein; cellular ligand of annexin II 38 X61432 Calmodulin 39 U48853 Cas; Crk-associated substrate; focal adhesion kinase substrate 40 U28423 58-kDa inhibitor of RNA-activated protein kinase
Down Down Down Down Down Down Down Down Down Down Up Down
Protein turnover (39 represented; 12 detectable) 41 AF042822 Epithin 42 X70296 Protease nexin 1
Down Down
Receptors (by activities) (2 represented; 2 detectable) Cytoskeleton and mobility proteins (57 represented, 9 detectable) 43 X51438 Vimentin
Down
DNA synthesis, repair and recombination proteins (48 represented, 10 detectable) 44 X96618 Recombination activating protein 1 (RAG1); RGA; novel stromal cell protein 45 D49429 HR21spA; protein involved in DNA double-strand break repair; PW29 46 S50213 Structure-specific recognition protein 1 (SSRP1); T160
Down Down Down
Other (25 represented; 5 detectable) 47 U92437
Down
Putative protein-tyrosine phosphatase PTEN; MMAC1
The differential expression analysis generated in this table was from comparative microarrays using RNA generated from tumors excised from one hypothyroid mouse and one T3-treated mouse as shown in Fig. 1.
352
Endocrinology, February 2002, 143(2):347–359
Wood et al. • TH-Induced Growth Arrest
FIG. 3. Effect of TH on transcripts encoding TRH receptor isoforms TRHR1 (A) and TRHR2 (B). Ten micrograms of poly(A⫹)RNA from vehicle (⫺T3) and TH-treated tumors (⫹T3) were analyzed by Northern blotting as described in Fig. 2. Probes for both mouse TRHR 1 and 2 isoforms were excised from plasmids provided by Dr. M. Gershengorn. Autoradiography was for 19 h.
FIG. 2. Effect of TH on transcripts encoding TR isoforms (A), somatostatin receptors 1 and 5 (B), and preprosomatostatin (C). Five micrograms of poly(A⫹)RNA isolated from TtT-97 tumors excised from mice treated with vehicle (⫺T3) or TH (⫹T3) for 24 h were size fractionated on denaturing agarose gels, transferred to nylon membranes, and incubated overnight with the following 32P-labeled DNA fragment hybridization probes. A, Probes for mouse TR1 and TR2 were fragments G and H, respectively, described in Wood et al. (17). B, Preparation of probe comprising full length coding region cDNAs for rat sst5 and mouse sst1 was as previously described (8). C, A fragment corresponding to a 361-bp sequence of the mouse preprosomatostatin coding region (98) was amplified from TtT-97 poly(A⫹)RNA using RT-PCR and the following sense and antisense primers; 5⬘ GGGAGATGCTGTCCTGCCGTCTCCAGTG 3⬘ and 5⬘ TAAAGCTAACAGGATGTGAATGTCTTCCA 3⬘. Following insertion into pCR2.1 (Invitrogen, Carlsbad, CA) it was reexcised with EcoRI and used as a hybridization probe. Autoradiographic exposure at ⫺80 C for both TR1 and TR2 was for 6 d, sst5 for 5 d, sst1 for 4 wk and preprosomatostatin for 2 d. Signals were stripped and reprobed with a mouse -actin fragment that was subjected to autoradiography as before for 4 –24 h. Radioactive bands corresponding to specific mRNAs are shown labeled with arrows denoting their approximate sizes in kb, derived using molecular weight standards loaded in a parallel lane.
to enter the S-phase, replicate their DNA, and ultimately divide. Alternatively, if these cues are withdrawn or inhibited, the cells will exit the cell cycle and growth will be arrested. We wanted to see whether TH, in addition to increasing TR1 and sst5 and reducing the effect of extracellular mitogens such as BDNF and TRH, also affected the expression of the genes encoding cell cycle regulators and thus lead to more sustained growth arrest in TtT-97 thyrotropes. Because the microarray analysis did not prove useful for the cell cycle molecules because of detection limitations, we obtained mouse cDNA coding for several molecules involved in cell cycle control and performed Northern blot analyses to determine whether the level of their transcripts
FIG. 4. Effect of TH on transcripts encoding BDNF (A) and its receptor trkB (B). Five micrograms of poly(A⫹)RNA were isolated from vehicle and TH-treated tumors and analyzed as described in Fig. 2. Full-length coding region probes for mouse BDNF and trkB were excised from plasmids supplied by Drs. K. Jones and L. Reichardt, respectively. Filters were exposed to photographic film for 4 wk (BDNF) and 18 h (trkB).
changed with TH treatment. We examined the mRNA transcript levels of 4 cdks (2, 4, 5, 6), 6 cyclins (D1–3, E, A1 and A2), and 7 cdki (p15, p16, p18, p19, p21, p27, and p57) 24 h after TH treatment. Of the cdks, the expression of 1.9-kb cdk4, 1.6-kb cdk5, and 2.6-kb cdk6 mRNAs was unaffected (Fig. 6). However, levels of a 2.5-kb transcript encoding cdk2 were dramatically down-regulated by TH (Fig. 6). When we analyzed cyclin transcripts (Fig. 7) those encoding cyclin D3 (2.3 kb) and cyclin E (2.1 kb) were not changed. Although we saw modest changes in a 4.2-kb cyclin D1 mRNA, which was slightly reduced, and a 6.8-kb cyclin D2 transcript, which was moderately increased, the greatest effect was on both cyclins A1 and A2 where a 1.8-kb (A1) and 3- and 1.9-kb (A2) mRNAs were markedly reduced by TH (Fig. 7). Therefore, it
Wood et al. • TH-Induced Growth Arrest
FIG. 5. Effect of time of TH treatment on transcripts encoding TRHR1, BDNF, and trkB. Five micrograms of poly(A⫹)RNA were isolated from tumors from animals killed after treatment with vehicle (Hypo) or 2 or 6 h after a single injection of TH and analyzed by Northern blotting as in Fig. 2. Radiolabeled probes were as for Figs. 3 and 4 and exposures were for 11 d (TRHR1), 3 d (BDNF and trkB).
appears that thyrotrope growth is predominantly driven by cyclin A activation of cdk2 and that down-regulation of this cell cycle pathway is a major target of TH regulation. When cdkis were examined only a single 2.2-kb transcript encoding p57 of the Cip/Kip family, which specifically inhibits cdk2 (46), was down-regulated by TH (Fig. 8A); the others, a 2.3 kb p21 mRNA and two p27 messages of 4.5 and 2.4 kb, were not changed at the transcript level. Of the INK4 proteins that preferentially inhibit cdk4/6 (47), a 1-kb p16 mRNA, two p18 transcripts of 2.3- and 1.1-kb p19 mRNA were unaffected, whereas a 1.9-kb transcript encoding p15 was significantly stimulated by TH (Fig. 8B). We carried out a time course to see when TH-induced changes in cell cycle regulator expression were first detected. We previously demonstrated that the steady-state levels of mRNAs for the TSH and ␣-subunit, which are known to be regulated at the level of transcription by TH in TtT-97 tumors (2), were significantly decreased within 4 h of TH treatment (48). Figure 9 shows that of those cell cycle regulators that were affected by 24 h (cdk2, cyclin A1 and A2, and cdk inhibitors p15 and p57), a change in transcript levels was not
Endocrinology, February 2002, 143(2):347–359 353
FIG. 6. Effect of TH on transcripts for various cyclin-dependent kinases. Poly(A⫹)RNA was isolated from tumors treated for 24 h with vehicle (⫺T3) or TH (⫹T3) and 5 g for cdk 4, cdk5, and cdk6, or 10 g for cdk2 were analyzed by Northern blotting as in Fig. 2. Probes for hybridization were derived from plasmids encoding mouse cDNAs for cdk2 (Dr. S-Y. Cheng) cdk4 (Dr. C. Sherr), cdk5 (Dr. D. Wolgemuth), and cdk6 (Dr. P. Zhang). Films were exposed for 5 d (cdk2), 18 h (cdk4 and cdk5), and 11 d (cdk6).
seen at 2 h or 6 h, suggesting that, unlike the situation described earlier for TRHR1, BDNF and trkB (Fig. 5), the genes coding for these more ubiquitous cell cycle regulators are not likely to be direct targets of TH. Interestingly, the p16 gene alternate transcript, p19ARF, was strikingly induced by TH treatment. This is shown as a 1.1-kb minor band above the p16 band in Fig. 8B. To verify the identity of this unexpected TH-responsive gene, we prepared a p19ARF-specific probe by PCR, amplifying the unique 5⬘ exon (49), and repeated the analysis. A single band of 1.1 kb is now shown dramatically appearing within 24 h of TH treatment in Fig. 10A. When we looked to see when the signal was first detected, it was not visible at 2 h or 6 h, but a signal was detectable at 10 h and was further increased at 24 h (Fig. 10B), again suggesting it was not a direct target of TH. Because the classical role for p19ARF is to sequester mdm2 to the nucleolus (50) and thus stabilize p53, enabling it to activate many genes involved in growth arrest and apoptosis (51), we examined the levels of p53 protein after TH treatment. Figure 11A shows that a 53-kDa protein present in TtT-97 cell extracts and recognized by Western analysis using a p53 antibody is not changed in amount at 24 or even 48 h after TH treatment, suggesting that p19ARF may be contributing to TH mediated inhibition of thyrotrope growth in a p53-independent manner. Protein levels of p19ARF, also measured by Western blotting using specific
354
Endocrinology, February 2002, 143(2):347–359
FIG. 7. Effect of TH on transcripts for various cyclins. Northern blotting was performed on poly(A⫹)RNA samples from tumors as for Fig. 2. Ten micrograms were used for cyclin A1 and A2 and 5 g was used for the others. Specific mouse hybridization probes were derived from plasmids for cyclin A1 and A2 (Dr. Wolgemuth), cyclin E (Dr. Cheng), and cyclins D1, D2, and D3 (Dr. Sherr). Exposure was for 18 h (cyclin A2), 2 wk (cyclin A1), 2 d (cyclin E), 18 h (cyclin D1), 4 d (cyclin D2), and 3 d (cyclin D3).
FIG. 8. Effect of TH on transcripts encoding cdk inhibitors of the Cip/Kip (A) and INK4 (B) families. Northern analysis was carried out as described in Fig. 2 using 10 g of poly(A⫹)RNA for p18 and 5 g for the remaining Northerns. Mouse cdk inhibitor probes used for hybridization were excised from vectors obtained from the following investigators; Dr. Sherr (p15, p16, and p19), Dr. Zhang (p21, p27, and p57), and Dr. J. DeGregori (p18). Exposure times were 2 d (p18, p19 and p57), 3 d (p16 and p27), and 4 d (p15 and p21).
antibodies, emerged after 10 h of TH treatment and increased even further at 24 h (Fig. 11B), in parallel with the message levels (Fig. 10B). It was reported recently that p19ARF protein levels were increased in mouse embryo fibroblasts by inducing expression from a conditionally active B-Raf allele (52). However, steady-state levels of B-Raf transcripts, which were detectable on the microarray, were not affected by TH
Wood et al. • TH-Induced Growth Arrest
FIG. 9. Effect of time of TH treatment on transcripts encoding THresponsive cell cycle regulators. Five micrograms of poly(A⫹)RNA were isolated from tumors from animals killed after treatment with vehicle (Hypo) or 2, 6, or 24 h after a single injection of TH and analyzed by Northern blotting as in Fig. 5. Hybridization was carried out using the corresponding probes described in Figs. 6 – 8. Exposures were 6 d (cdk2), 3 wk (cyclin A1), 3 d (cyclin A2 and p57), and 18 h (p15).
FIG. 10. Effect on p19ARF transcripts after treatment with TH for 24 h (A) and at earlier time points (B). Five micrograms of poly(A⫹)RNA were isolated from tumors from animals killed after treatment with (A) vehicle (⫺T3) or TH (⫹T3) for 24 h or (B) vehicle (Hypo) or 2, 6, 10, or 24 h after a single injection of TH and analyzed by Northern blotting as in Fig. 5. A region specific for the unique N-terminal exon of mouse p19ARF (49) was amplified from purified TtT-97 mouse genomic DNA by PCR using the following sense and antisense primers; 5⬘ ATGGGTCGCAGGTTCTTGG 3⬘ and 5⬘ CTGGTCCAGGATTCCGGTG 3⬘. Following cloning into PCR2.1 (Invitrogen), it was excised with EcoRI and used as a p19ARF-specific hybridization probe on both filters. Autoradiographic exposure was for 3 d for both A and B.
treatment. We also looked at TH’s effect on the transcript levels of other cell cycle proteins. These included 4- and 2.3-kb Rb transcripts, a single 1.3-kb c-myc message, and a 3-kb mRNA for mdm2, all of which were not affected by TH (Fig. 12) and E2F1, an E2F family member, where 2.5- and
Wood et al. • TH-Induced Growth Arrest
FIG. 11. Effect of time of TH treatment on protein levels of p53 (A) and p19ARF (B). Whole cell extracts (100 g protein), prepared from TtT-97 tumors injected with vehicle (Hypo) or 2 h, 10 h, 24 h after a single injection of TH, or 48 h after two injections of TH on successive days, were resolved on 10% (A) or 12.5% (B) polyacrylamide-SDS gels and transferred to a polyvinylidenedifluoride membrane. The blots were probed with a 1:5000 dilution of a rabbit polyclonal antibody raised against p53 (A) or a 1:400 dilution of a rabbit polyclonal antibody for p19ARF (B) as described in Materials and Methods. Shown labeled on the left with their size in kilodaltons are the bands corresponding to p53 and p19ARF. The identity of the p19ARF band was verified by competition with a blocking peptide.
2.2-kb transcripts were noticeably down-regulated by TH (Fig. 12). Effect of TH on other genes expressed in thyrotropes
Other genes whose expression and response to TH in thyrotropes we have studied are neuronatin, which was first described as being highly expressed in developing brain but declined in abundance following birth (53) and PB-cadherin, a novel member of the cell adhesion family of cadherins expressed only in the brain and pituitary gland (54). Both were shown to be present in a TSH-expressing cell line (T␣T-1) but not in a precursor cell line (␣T-1), which expressed only the ␣-subunit but had not yet committed to either the thyrotrope or gonadotrope phenotype (55). As shown in Fig. 13A, the levels of a 1.4-kb neuronatin mRNA were dramatically increased in the thyrotrope tumor with TH administration and a shorter time course showed that this increase could be first detected as early as 2 h and was significantly increased by 6 h (Fig. 13B). When we measured levels of PB cadherin mRNA, a 4.4-kb transcript was decreased in amount by 24 h after TH treatment (Fig. 14A), and this was also shown to occur rapidly as its level was already lowered by 2 h (Fig. 14B). Therefore, neuronatin and PB cadherin are excellent candidates for thyrotrope-restricted primary targets for TH. However, their role in growth arrest, if any, remains undefined. Discussion
The molecular mechanism underlying thyroid hormone control of thyrotrope cell proliferation is poorly understood.
Endocrinology, February 2002, 143(2):347–359 355
As an initial approach to shed light on this important function of TH, we used the TtT-97 tumor model of thyrotrope hyperplasia and have identified, using a microarray approach, together with conventional Northern blotting, multiple genes whose expression is regulated by TH. Our analyses looked at times of TH treatment of 24 h or less, as we reasoned that although several weeks of treatment was necessary to see changes in tumor mass, TH action via its receptors, in keeping with their role as ligand-dependent transcription factors, was probably being exerted as early as a few hours following administration. Although our original intent was to examine TH’s effects on growth, we decided, in order not to overlook potentially important genes, to cast a wide net and to adopt a high density microarray approach with more than 1,000 genes. However, this strategy proved disappointing for analyzing important proteins relevant to cell proliferation including growth factors and cell cycle regulators that were either not represented on the filters or below the levels of detectability. Potential autocrine growth stimulatory pathways shown to be present in pituitary thyrotrope-derived cells are those for TRH and BDNF (34, 35, 44), and indeed we show abundant transcripts encoding BDNF and its receptor, trkB, as well as the major TRH receptor (TRHR1) in the thyrotrope tumor model and these are all down-regulated by TH (Figs. 3 and 4). Although not shown specifically in thyrotropes previously, expression of both BDNF and trkB in brain (56, 57), and TRHR1 in the pituitary (58), has been demonstrated to be regulated by thyroid hormone status and, in fact, in the case of trkB this has been shown to be at the promoter level in transient transfection studies in neuroblastoma cells (57). Therefore, withdrawal of positive growth stimuli by decreasing the expression of critical components of stimulatory pathways most likely is playing a role in the inhibition of thyrotrope growth by TH. Inhibition of cell growth mediated through sst5 has been shown to result in a reduction of intracellular cAMP levels (59, 60), as well as induction of hypophosphorylated Rb, leading to inactivation of E2F transcription factors and cell cycle arrest (61). Although inhibition of thyrotrope cell proliferation by dephosphorylation of Rb could be potentially being mediated through increased sst signaling by THinduced expression of sst5 (8) (also shown in this report), the phosphorylation state of Rb, the major determinant of cell cycle progression, is mainly controlled by the prevailing expression levels of cell cycle regulators, the cyclin-dependent kinases (cdks), and their regulatory activators (cyclins) and inhibitors (cdkis). When we looked at the effect of TH on the expression of these growth controllers in the thyrotrope model, we were intrigued to find that only a limited subset of each of these regulators was affected by TH. Of the cdks examined, only cdk2 expression was down-regulated by TH (Fig. 5), as were both cyclin A isoforms (Fig. 6). In addition, only one member of each family of inhibitors, p15 of the Ink4 family was up-regulated, whereas p57 of the Cip/Kip family was down-regulated by TH (Fig. 7). TH decreased Cdk2 activity by activating the expression of p27 and c-myc leading to growth arrest of a neuroblastoma cell line (62). Increases in both p27 and p21 expression were also reported to occur in chondrocytes induced to terminally differentiate with TH
356
Endocrinology, February 2002, 143(2):347–359
Wood et al. • TH-Induced Growth Arrest
FIG. 12. Effect of TH on transcripts for other cell cycle proteins. Poly(A⫹)RNA was isolated from tumors treated for 24 h with vehicle (⫺T3) or TH (⫹T3) and 5 g for Rb, c-myc and mdm2, or 10 g for E2F1 were analyzed by Northern blotting as in Fig. 2. Probes for hybridization were derived from plasmids encoding mouse cDNAs for Rb and E2F1 (Dr. DeGregori) and a human cDNA for c-myc (Dr. A. Gutierrez-Hartmann). A plasmid containing an EST, from which a fragment encoding the mouse mdm2 sequence was excised, was obtained from the IMAGE Consortium (dbEST no. 3588758). Autoradiographic exposure was for 3 d (Rb), 4 d (c-myc and E2F1), and 18 h (mdm2).
FIG. 13. Effect on transcripts encoding neuronatin after treatment with TH for 24 h (A) and at earlier time points (B). Five micrograms of poly(A⫹)RNA were isolated from tumors from animals killed after treatment with (A) vehicle (⫺T3) or TH (⫹T3) for 24 h or (B) vehicle (Hypo) or 2 and 6 h after a single injection of TH and analyzed by Northern blotting as in Fig. 5. A probe for mouse neuronatin was excised from a plasmid containing the complete coding region obtained from Dr. P. Gruss. Exposures were for 3 h (A) and 45 min (B).
FIG. 14. Effect on transcripts encoding PB cadherin after treatment with TH for 24 h (A) and at earlier time points (B). Five micrograms of poly(A⫹)RNA were isolated from tumors from animals killed after treatment with (A) vehicle (⫺T3) or TH (⫹T3) for 24 h or (B) vehicle (Hypo) or 2 and 6 h after a single injection of TH and analyzed by Northern blotting as in Fig. 5. A probe for PB cadherin was excised from a plasmid containing the mouse cDNA provided by Dr. T. Nakamura. Exposures were for 6 d (A) and 11 d (B).
(63). In this report p27, p21 and c-myc expression were unaffected by TH (Figs. 7 and 11, respectively) suggesting an alternate mechanism for TH in thyrotrope cell growth inhibition. Interestingly, in two models of TH stimulation of growth, liver repopulation, and GC pituitary cells, cdk2 expression was shown to be increased (64, 65). TH-stimulated GC cell growth was also associated with changes in cyclin E
and D1 expression (64), which was not the case for thyrotrope cells in this study. Although there are no published reports of effects of TH on cyclin A or the two cdki regulated in this report (p15 and p57), it has been shown that TH increases the promoter activity of mdm2 (66), which directly activates cyclin A expression (67). Again in the TtT-97 thyrotrope model, we saw no change in mdm2 transcripts with TH treatment (Fig. 11). However, we did observe a dramatic induction of transcripts and protein corresponding to p19ARF (Figs. 9 and 10), which can interact with and sequester mdm2 to the nucleolus (50), thus nullifying its activation of the cyclin A promoter. Surprisingly, this putative sequestration of mdm2 by p19ARF did not result in increased p53 stability (68), as we did not see an increase in p53 protein levels even after 48 h of TH treatment (Fig. 10). Thus, p19ARF may be acting through a p53- and perhaps even mdm2independent pathways, to arrest the growth of thyrotrope cells as is suggested by studies showing that reintroduction of exogenous p19ARF into fibroblasts from p53/mdm2/ p19ARF null mice results in cell cycle arrest (69). Thus, a model for TH inhibition of thyrotrope growth can be envisaged whereby cyclin A/cdk2 complexes, which are required for maintenance of S-phase progression, are greatly reduced. Furthermore, cyclin E, although not affected by TH (Fig. 7), is rendered ineffective as the kinase it activates, cdk2, is deficient in its ability to form a functional complex. The cyclinD/cdk4/6 complex components, which again are not affected at the expression level by TH, can be inactivated by potential withdrawal of TRH and BDNF growth stimuli as a result of decreases in expression of BDNF and the receptors for both BDNF (trkB) and TRH (Figs. 3 and 4). Further destabilization of cyclinD/cdk4 complexes can also be brought about by reduction in Cip/Kip family members such as p57, which have recently been shown to be required for these complexes to assemble (70). In addition, remaining cdk4/6 activity can be directly inhibited by the increased p15 levels as members of the Ink4 family specifically affect these kinases (15). Finally, we showed that TH treatment also reduces transcripts for E2F1 that, along with its dimerization partner DRTF1 earlier shown by microarray to be reduced, would result in a decrease in transcription of genes required for DNA replication (71). A suppression of E2F1 promoter activity by TH in transfected embryonal carcinoma cells has recently been reported (72). A model consistent with growth
Wood et al. • TH-Induced Growth Arrest
arrest as a result of the above gene expression changes induced by TH is shown in Fig. 15. Although inadequate for the assessment of classical cell cycle genes, the microarray strategy did uncover several interesting genes previously not reported to be expressed in the pituitary gland or to be targets of TH regulation and whose altered expression might be playing a role in mediating growth arrest in the thyrotrope. Perhaps not surprisingly, several of the genes identified as TH responsive are neural or neuroendocrine in origin and some have been reported to be associated with cell adhesion and calciummediated growth regulation, particularly via Wnt signaling, which regulates cell polarity and specification of cell fate (73). Among those down-regulated by TH were cadherin 4, NPRAP (␦-catenin), the protease nexin I, semaphorins, and calpactin light chain, all of which are expressed during neurite outgrowth (74 –78). In addition, two Wnt receptor homologs (frizzled 3 and 6) were also down-regulated, which along with catenins and cadherins are intermediates in Wnt signaling pathways (73). Regulation of Wnt pathway components was recently reported in GC cells after TH stimulation of growth (79). Calmodulin, a calcium-dependent kinase, which stimulates cdk2 and Rb phosphorylation as well as cyclin A expression (80), was decreased by TH. The expression of other growth-associated proteins including the ␣-subunit of inhibin, a member of the transforming growth factor  superfamily (81) and defender against cell death protein DAD1, an antiapoptotic factor (82), was also repressed by TH. Unexpectedly, two genes involved in protection against apoptosis in neurons, PEP19 (83) and Crkassociated substrate, which is also a substrate for focal adhesion kinase (84), were up-regulated by TH. Thus, apoptosis may not be playing a major role in tumor regression at least at early times following TH treatment. Future studies will address the role of apoptosis pathways at later times
FIG. 15. Schematic illustrating the effects of TH on the cell cycle in thyrotropes. Shown is a pictorial representation of the cell cycle (comprising G1-, S-, G2-, and M-phases) and the various cell cycle regulators that impact it in the absence (Growth Progression) and the presence of TH (Growth Arrest). The size of the type reflects the relative abundance of the various molecules in the two states based on the gene expression levels determined in the current report. The various cyclins are abbreviated to a single letter (A, D, and E) symbolizing all isoforms. Arrows denote positive effects, whereas blocked lines represent inhibitory influences and the size of each represents the magnitude of the influence on the molecules that they impinge upon. The X on the cycle on the right denotes that progression through S-phase is blocked.
Endocrinology, February 2002, 143(2):347–359 357
when tumor involution is more noticeable. Other signal transduction molecules affected by TH were ryk, a nonclassical tyrosine kinase (85) involved in stimulating the STAT1 pathway (86), which was down-regulated and STAT1 itself, an apoptosis initiator (87), which was increased by TH. The expression of several brain-specific transcription factors was considerably decreased by TH including neurogenin 3, which colocalizes with neurons that stain positive for the proliferation marker Ki67 (88) and nHLH1 whose misexpression is associated with neuronal cell death (89), again a puzzling example of TH regulation in an unexpected direction. Finally, a neuroendocrine protease (PC1) responsible for prohormone cleavage, which was previously reported to be TH regulated (90), and two of its substrates, chromogranin B and C (also known as secretogranin II) were all decreased with TH. Chromogranins have been described in thyrotropes and TSH adenomas (91, 92), and they have been reported to increase their expression in response to epidermal growth factor and basic fibroblast growth factor (93) and through a cAMP-stimulated pathway (94). A TH-induced reduction in cAMP levels as discussed above could also be a direct consequence of decreased expression of adenyl cyclase type 6 seen on the microarray (Table 1). Except for TRHR1 and BDNF and its receptor trkB, which are likely regulated by TH at the level of transcription even in thyrotrope cells as shown by their more rapid response to TH (Figs. 3 and 4), the other regulated genes, specifically the cell cycle genes, which are expressed in many cell types whose growth is not regulated by TH, are probably not primary targets of TH ,which is also suggested by their later response (⬎6 h, see Fig. 8). A possible candidate is sst5, although it is expressed in other cell types whose growth is not inhibited by TH (60). It is therefore likely that the key primary targets of TH are genes that display an expression pattern mainly in thyrotropes and that respond rapidly to TH indicative of a transcriptional effect. Two genes that meet these criteria are those encoding neuronatin and PB cadherin, which we showed were changed in expression, albeit in opposing directions, within 2 h of TH administration (Figs. 12 and 13). Both are restricted to only the brain and pituitary gland (54, 95). Neuronatin expression declines in the brain after birth (96) leaving the pituitary as the only adult tissue where it is expressed, and PB cadherin showed evidence of pituitary-specific transcripts (54). Furthermore, both were shown, by subtractive hybridization analyses, to be present in a TSH-expressing cell line but not in an ␣-subunit expressing precursor cell line (55), possibly implying mature thyrotrope specificity. A unique cadherin could play a role in modulating a specific Wnt pathway unique to thyrotropes, whereas neuronatin, a small proteolipid related to phospholamban, a subunit of a Ca2⫹-ATPase (97), may be modulating a thyrotrope-specific calcium-dependent signal transduction pathway. Obviously, further studies will be necessary to elucidate the function of these poorly characterized, but intriguing, TH-regulated genes. Acknowledgments We would like to thank P. Zhang, G. Johnson, A. Arnold, and A. Gutierrez-Hartmann for helpful discussions. We are grateful to the following people for generous provision of plasmids: C. Sherr (Mem-
358 Endocrinology, February 2002, 143(2):347–359
phis, TN), S.-Y. Cheng (Bethesda, MD), D. Wolgemuth (New York, NY), P. Zhang (Denver, CO), K. Jones (Boulder, CO), L. Reichardt (San Francisco, CA), A. Gutierrez-Hartmann (Denver, CO), M. Gershengorn (New York, NY), J. DeGregori (Denver, CO), P. Gruss (Gottingen, Germany), S. Lee, and T. Nakamura (Osaka, Japan). We acknowledge the technical help of A. Wood and S. Nagy.
Wood et al. • TH-Induced Growth Arrest
22. 23. 24.
Received August 16, 2001. Accepted October 17, 2001. Address all correspondence and requests for reprints to: William M. Wood, Ph.D., Department of Medicine University of Colorado Health Sciences Center (B-151), 4200 East Ninth Avenue, Box B151, Denver, Colorado 80262. E-mail:
[email protected]. This research was supported by NIH Grants CA-47411 and DK-36843 (to E.C.R.).
25.
26.
References 1. Ross DS, Downing MF, Chin WW, Kieffer JD, Ridgway EC 1983 Divergent changes in murine pituitary concentration of free ␣ and thyrotropin- subunits in hypothyroidism and after thyroxine administration. Endocrinology 112: 187–193 2. Shupnik MA, Chin WW, Habener JF, Ridgway EC 1985 Transcriptional regulation of the thyrotropin subunit genes by thyroid hormone. J Biol Chem 260:2900 –2903 3. Brown NS, Smart A, Sharma V, Brinkmeier ML, Greenlee L, Camper SA, Jensen DR, Eckel RH, Krezel W, Chambon P, Haugen BR 2000 Thyroid hormone resistance and increased metabolic rate in the RXR-␥-deficient mouse. J Clin Invest 106:73–79 4. Marshall Jr MC, Williams D, Weintraub BD 1981 Regulation of de novo biosynthesis of thyrotropin in normal, hyperplastic, and neoplastic thyrotrophs. Endocrinology 108:908 –915 5. Furth J 1955 Experimental pituitary tumors. Recent Prog Horm Res 11:221–257 6. Furth J, Moy P, Hershman JM, Ueda G 1973 Thyrotropic tumor syndrome. A multiglandular disease induced by sustained deficiency of thyroid hormones. Arch Pathol 96:217–226 7. Sarapura VD, Wood WM, Gordon DF, Ridgway EC 1993 Effect of thyroid hormone on T3-receptor mRNA levels and growth of thyrotropic tumors. Mol Cell Endocrinol 91:75– 81 8. James RA, Sarapura VD, Bruns C, Raulf F, Dowding JM, Gordon DF, Wood WM, Ridgway EC 1997 Thyroid hormone-induced expression of specific somatostatin receptor subtypes correlates with involution of the TtT-97 murine thyrotrope tumor. Endocrinology 138:719 –724 9. Woodmansee WW, Gordon DF, Dowding JM, Stolz B, Lloyd RV, James RA, Wood WM, Ridgway EC 2000 The effect of thyroid hormone and a long-acting somatostatin analogue on TtT-97 murine thyrotropic tumors. Thyroid 10: 533–541 10. Refetoff S, Weiss RE, Usala SJ 1993 The syndromes of resistance to thyroid hormone. Endocr Rev 14:348 –399 11. Wong R, Vasilyev VV, Ting YT, Kutler DI, Willingham MC, Weintraub BD, Cheng S 1997 Transgenic mice bearing a human mutant thyroid hormone  1 receptor manifest thyroid function anomalies, weight reduction, and hyperactivity. Mol Med 3:303–314 12. Sherr CJ 2000 The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 60:3689 –3695 13. Sherr CJ 1994 Growth factor-regulated G1 cyclins. Stem Cells 12(Suppl 1): 47–55 14. Lundberg AS, Weinberg RA 1998 Functional inactivation of the retinoblastoma protein requires sequential modification by at least two distinct cyclincdk complexes. Mol Cell Biol 18:753–761 15. Sherr CJ, Roberts JM 1995 Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 9:1149 –1163 16. Wood WM, Kao MY, Gordon DF, Ridgway EC 1989 Thyroid hormone regulates the mouse thyrotropin  subunit gene promoter in transfected primary thyrotropes. J Biol Chem 264:14840 –14847 17. Wood WM, Ocran KO, Gordon DF, Ridgway EC 1991 Isolation and characterization of mouse complementary DNAs encoding ␣ and  thyroid hormone receptors from thyrotrope cells: the mouse pituitary specific -2 isoform differs at the amino terminus from the corresponding species from rat pituitary tumor cells. Mol Endocrinol 5:1049 –1061 18. Wei LL, Krett NL, Francis MD, Gordon DF, Wood WM, O’Malley BW, Horwitz KB 1988 Multiple human progesterone receptor messenger RNAs and their autoregulation by progestin agonists and antagonists in breast cancer cells. Mol Endocrinol 2:62–72 19. Gordon DF, Lewis SR, Haugen BR, James A, McDermott MT, Wood WM, Ridgway EC 1997 Pit-1 and GATA-2 interact and functionally cooperate to activate the thyrotropin  subunit promoter. J Biol Chem 272:24339 –24347 20. Zhong Z, Wen Z, Darnell Jr JE 1994 Stat3 and Stat4: members of the family of signal transducers and activators of transcription. Proc Natl Acad Sci USA 91:4806 – 4810 21. Chang AS, Chang SM, Starnes DM, Schroeter S, Bauman AL, Blakely RD
27. 28. 29. 30.
31.
32. 33.
34.
35.
36. 37. 38.
39. 40.
41.
42. 43.
44. 45.
1996 Cloning and expression of the mouse serotonin transporter. Brain Res Mol Brain Res 43:185–192 Rudin CM, Engler P, Storb U 1990 Differential splicing of thymosin 4 mRNA. J Immunol 144:4857– 4862 Polte TR, Hanks SK 1995 Interaction between focal adhesion kinase and Crk-associated tyrosine kinase substrate p130Cas. Proc Natl Acad Sci USA 92:10678 –10682 Brown L, Espinosa R, III, Le Beau MM, Siciliano MJ, Baer R 1992 HEN1 and HEN2: a subgroup of basic helix-loop-helix genes that are coexpressed in a human neuroblastoma. Proc Natl Acad Sci USA 89:8492– 8496 Sommer L, Ma Q, Anderson DJ 1996 neurogenins, a novel family of atonalrelated bHLH transcription factors, are putative mammalian neuronal determination genes that reveal progenitor cell heterogeneity in the developing CNS and PNS. Mol Cell Neurosci 8:221–241 Seidah NG, Marcinkiewicz M, Benjannet S, Gaspar L, Beaubien G, Mattei MG, Lazure C, Mbikay M, Chretien M 1991 Cloning and primary sequence of a mouse candidate prohormone convertase PC1 homologous to PC2, Furin, and Kex2: distinct chromosomal localization and messenger RNA distribution in brain and pituitary compared to PC2. Mol Endocrinol 5:111–122 Pohl TM, Phillips E, Song KY, Gerdes HH, Huttner WB, Ruther U 1990 The organisation of the mouse chromogranin B (secretogranin I) gene. FEBS Lett 262:219 –224 Albano RM, Groome N, Smith JC 1993 Activins are expressed in preimplantation mouse embryos and in ES and EC cells and are regulated on their differentiation. Development 117:711–723 Klein R, Conway D, Parada LF, Barbacid M 1990 The trkB tyrosine protein kinase gene codes for a second neurogenic receptor that lacks the catalytic kinase domain. Cell 61:647– 656 Dasen JS, O’Connell SM, Flynn SE, Treier M, Gleiberman AS, Szeto DP, Hooshmand F, Aggarwal AK, Rosenfeld MG 1999 Reciprocal interactions of Pit1 and GATA2 mediate signaling gradient-induced determination of pituitary cell types. Cell 97:587–598 Ericson J, Norlin S, Jessell TM, Edlund T 1998 Integrated FGF and BMP signaling controls the progression of progenitor cell differentiation and the emergence of pattern in the embryonic anterior pituitary. Development 125: 1005–1015 LeRiche VK, Asa SL, Ezzat S 1996 Epidermal growth factor and its receptor (EGF-R) in human pituitary adenomas: EGF-R correlates with tumor aggressiveness. J Clin Endocrinol Metab 81:656 – 662 Chaidarun SS, Eggo MC, Sheppard MC, Stewart PM 1994 Expression of epidermal growth factor (EGF), its receptor, and related oncoprotein (erbB-2) in human pituitary tumors and response to EGF in vitro. Endocrinology 135: 2012–2021 Pagesy P, Croissandeau G, Le Dafniet M, Peillon F, Li JY 1992 Detection of thyrotropin-releasing hormone (TRH) mRNA by the reverse transcriptionpolymerase chain reaction in the human normal and tumoral anterior pituitary. Biochem Biophys Res Commun 182:182–187 Le Dafniet M, Lefebvre P, Barret A, Mechain C, Feinstein MC, Brandi AM, Peillon F 1990 Normal and adenomatous human pituitaries secrete thyrotropin-releasing hormone in vitro: modulation by dopamine, haloperidol, and somatostatin. J Clin Endocrinol Metab 71:480 – 486 Gershengorn MC 1989 Mechanism of signal transduction by TRH. Ann NY Acad Sci 553:191–196 Pawlikowski M, Slowinska-Klencka D 1994 Effects of TRH and TRH-like peptides on anterior pituitary cell proliferation in rats. Cytobios 79:117–122 Simard M, Mirell CJ, Pekary AE, Drexler J, Kovacs K, Hershman JM 1988 Hormonal control of thyrotropin and growth hormone secretion in a human thyrotrope pituitary adenoma studied in vitro. Acta Endocrinol (Copenh) 119:283–290 Straub RE, Frech GC, Joho RH, Gershengorn MC 1990 Expression cloning of a cDNA encoding the mouse pituitary thyrotropin- releasing hormone receptor. Proc Natl Acad Sci USA 87:9514 –9518 Cao J, O’Donnell D, Vu H, Payza K, Pou C, Godbout C, Jakob A, Pelletier M, Lembo P, Ahmad S, Walker P 1998 Cloning and characterization of a cDNA encoding a novel subtype of rat thyrotropin-releasing hormone receptor. J Biol Chem 273:32281–32287 Yamada M, Saga Y, Shibusawa N, Hirato J, Murakami M, Iwasaki T, Hashimoto K, Satoh T, Wakabayashi K, Taketo MM, Mori M 1997 Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin-releasing hormone gene. Proc Natl Acad Sci USA 94:10862–10867 Jones KE, Brubaker JH, Chin WW 1996 An alternative splice variant of the mouse TRH receptor mRNA is the major form expressed in the mouse pituitary gland. J Mol Endocrinol 16:197–204 Klein R, Nanduri V, Jing SA, Lamballe F, Tapley P, Bryant S, Cordon-Cardo C, Jones KR, Reichardt LF, Barbacid M 1991 The trkB tyrosine protein kinase is a receptor for brain-derived neurotrophic factor and neurotrophin-3. Cell 66:395– 403 Conner JM, Yan Q, Varon S 1996 Distribution of brain-derived neurotrophic factor in the rat pituitary gland. Neuroreport 7:1937–1940 Ohara O, Gahara Y, Teraoka H, Kitamura T 1992 A rat brain-derived neurotrophic factor-encoding gene generates multiple transcripts through alternative use of 5⬘ exons and polyadenylation sites. Gene 121:383–386
Wood et al. • TH-Induced Growth Arrest
46. Hashimoto Y, Kohri K, Kaneko Y, Morisaki H, Kato T, Ikeda K, Nakanishi M 1998 Critical role for the 310 helix region of p57(Kip2) in cyclin-dependent kinase 2 inhibition and growth suppression. J Biol Chem 273:16544 –16550 47. Hirai H, Roussel MF, Kato JY, Ashmun RA, Sherr CJ 1995 Novel INK4 proteins, p19 and p18, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6. Mol Cell Biol 15:2672–2681 48. Chin WW, Shupnik MA, Ross DS, Habener JF, Ridgway EC 1985 Regulation of the ␣ and thyrotropin -subunit messenger ribonucleic acids by thyroid hormones. Endocrinology 116:873– 878 49. Quelle DE, Zindy F, Ashmun RA, Sherr CJ 1995 Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 83:993–1000 50. Weber JD, Taylor LJ, Roussel MF, Sherr CJ, Bar-Sagi D 1999 Nucleolar Arf sequesters Mdm2 and activates p53. Nat Cell Biol 1:20 –26 51. Levine AJ 1997 p53, the cellular gatekeeper for growth and division. Cell 88:323–331 52. Ries S, Biederer C, Woods D, Shifman O, Shirasawa S, Sasazuki T, McMahon M, Oren M, McCormick F 2000 Opposing effects of Ras on p53: transcriptional activation of mdm2 and induction of p19ARF. Cell 103:321–330 53. Joseph R, Dou D, Tsang W 1995 Neuronatin mRNA: alternatively spliced forms of a novel brain-specific mammalian developmental gene. Brain Res 690:92–98 54. Sugimoto K, Honda S, Yamamoto T, Ueki T, Monden M, Kaji A, Matsumoto K, Nakamura T 1996 Molecular cloning and characterization of a newly identified member of the cadherin family, PB-cadherin. J Biol Chem 271:11548 – 11556 55. Pernasetti FM, Guirand DM, Holley SJ, Mellon PL, Characterization of factors involved in thyrotrope cell lineage commitment. Program of the 80th Annual Meeting of The Endocrine Society, New Orleans, LA, 1998 (Abstract P3-98) 56. Koibuchi N, Fukuda H, Chin WW 1999 Promoter-specific regulation of the brain-derived neurotropic factor gene by thyroid hormone in the developing rat cerebellum. Endocrinology 140:3955–3961 57. Pombo PM, Barettino D, Espliguero G, Metsis M, Iglesias T, Rodriguez-Pena A 2000 Transcriptional repression of neurotrophin receptor trkB by thyroid hormone in the developing rat brain. J Biol Chem 275:37510 –37517 58. Schomburg L, Bauer K 1995 Thyroid hormones rapidly and stringently regulate the messenger RNA levels of the thyrotropin-releasing hormone (TRH) receptor and the TRH-degrading ectoenzyme. Endocrinology 136:3480 –3485 59. Chisholm C, Greenberg GR 2000 Somatostatin receptor subtype-5 mediates inhibition of peptide YY secretion from rat intestinal cultures. Am J Physiol Gastrointest Liver Physiol 279:G983–G989 60. Patel YC 1999 Somatostatin and its receptor family. Front Neuroendocrinol 20:157–198 61. Sharma K, Patel YC, Srikant CB 1999 C-terminal region of human somatostatin receptor 5 is required for induction of Rb and G1 cell cycle arrest. Mol Endocrinol 13:82–90 62. Perez-Juste G, Aranda A 1999 The cyclin-dependent kinase inhibitor p27(Kip1) is involved in thyroid hormone-mediated neuronal differentiation. J Biol Chem 274:5026 –5031 63. Ballock RT, Zhou X, Mink LM, Chen DH, Mita BC, Stewart MC 2000 Expression of cyclin-dependent kinase inhibitors in epiphyseal chondrocytes induced to terminally differentiate with thyroid hormone. Endocrinology 141: 4552– 4557 64. Barrera-Hernandez G, Park KS, Dace A, Zhan Q, Cheng SY 1999 Thyroid hormone-induced cell proliferation in GC cells is mediated by changes in G1 cyclin/cyclin-dependent kinase levels and activity. Endocrinology 140:5267– 5274 65. Oren R, Dabeva MD, Karnezis AN, Petkov PM, Rosencrantz R, Sandhu JP, Moss SF, Wang S, Hurston E, Laconi E, Holt PR, Thung SN, Zhu L, Shafritz DA 1999 Role of thyroid hormone in stimulating liver repopulation in the rat by transplanted hepatocytes. Hepatology 30:903–913 66. Qi JS, Yuan Y, Desai-Yajnik V, Samuels HH 1999 Regulation of the mdm2 oncogene by thyroid hormone receptor. Mol Cell Biol 19:864 – 872 67. Leveillard T, Wasylyk B 1997 The MDM2 C-terminal region binds to TAFII250 and is required for MDM2 regulation of the cyclin A promoter. J Biol Chem 272:30651–30661 68. Kamijo T, Weber JD, Zambetti G, Zindy F, Roussel MF, Sherr CJ 1998 Functional and physical interactions of the ARF tumor suppressor with p53 and Mdm2. Proc Natl Acad Sci USA 95:8292– 8297 69. Weber JD, Jeffers JR, Rehg JE, Randle DH, Lozano G, Roussel MF, Sherr CJ, Zambetti GP 2000 p53-independent functions of the p19(ARF) tumor suppressor. Genes Dev 14:2358 –2365 70. Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM, Sherr CJ 1999 The p21(Cip1) and p27(Kip1) CDK ’inhibitors’ are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 18:1571–1583 71. Lavia P, Jansen-Durr P 1999 E2F target genes and cell-cycle checkpoint control. Bioessays 21:221–230 72. Bondesson Bolin M, Nygard M, Wahlstrom G 2000 Thyroid hormoneinduced cell cycle arrest through transcriptional repression of E2F-1. Endocr J 47:240
Endocrinology, February 2002, 143(2):347–359 359
73. Cadigan KM, Nusse R 1997 Wnt signaling: a common theme in animal development. Genes Dev 11:3286 –3305 74. Liu Q, Sanborn KL, Cobb N, Raymond PA, Marrs JA 1999 R-cadherin expression in the developing and adult zebrafish visual system. J Comp Neurol 410:303–319 75. Ho C, Zhou J, Medina M, Goto T, Jacobson M, Bhide PG, Kosik KS 2000 ␦-Catenin is a nervous system-specific adherens junction protein which undergoes dynamic relocalization during development. J Comp Neurol 420: 261–276 76. Houenou LJ, D’Costa AP, Li L, Turgeon VL, Enyadike C, Alberdi E, Becerra SP 1999 Pigment epithelium-derived factor promotes the survival and differentiation of developing spinal motor neurons. J Comp Neurol 412:506 –514 77. Sakai T, Furuyama T, Ohoka Y, Miyazaki N, Fujioka S, Sugimoto H, Amasaki M, Hattori S, Matsuya T, Inagaki S 1999 Mouse semaphorin H induces PC12 cell neurite outgrowth activating Ras-mitogen-activated protein kinase signaling pathway via Ca2⫹ influx. J Biol Chem 274:29666 –29671 78. Fox MT, Prentice DA, Hughes JP 1991 Increases in p11 and annexin II proteins correlate with differentiation in the PC12 pheochromocytoma. Biochem Biophys Res Commun 177:1188 –1193 79. Miller LD, Park KS, Guo QM, Alkharouf NW, Malek RL, Lee NH, Liu ET, Cheng SY 2001 Silencing of Wnt signaling and activation of multiple metabolic pathways in response to thyroid hormone-stimulated cell proliferation. Mol Cell Biol 21:6626 – 6639 80. Taules M, Rius E, Talaya D, Lopez-Girona A, Bachs O, Agell N 1998 Calmodulin is essential for cyclin-dependent kinase 4 (Cdk4) activity and nuclear accumulation of cyclin D1-Cdk4 during G1. J Biol Chem 273:33279 –33286 81. Lopez P, Vidal F, Rassoulzadegan M, Cuzin F 1999 A role of inhibin as a tumor suppressor in Sertoli cells: down-regulation upon aging and repression by a viral oncogene. Oncogene 18:7303–7309 82. Hong NA, Flannery M, Hsieh SN, Cado D, Pedersen R, Winoto A 2000 Mice lacking Dad1, the defender against apoptotic death-1, express abnormal Nlinked glycoproteins and undergo increased embryonic apoptosis. Dev Biol 220:76 – 84 83. Erhardt JA, Legos JJ, Johanson RA, Slemmon JR, Wang X 2000 Expression of PEP-19 inhibits apoptosis in PC12 cells. Neuroreport 11:3719 –3723 84. Kook S, Shim SR, Choi SJ, Ahnn J, Kim JI, Eom SH, Jung YK, Paik SG, Song WK 2000 Caspase-mediated cleavage of p130cas in etoposide-induced apoptotic Rat- 1 cells. Mol Biol Cell 11:929 –939 85. Hovens CM, Stacker SA, Andres AC, Harpur AG, Ziemiecki A, Wilks AF 1992 RYK, a receptor tyrosine kinase-related molecule with unusual kinase domain motifs. Proc Natl Acad Sci USA 89:11818 –11822 86. Zong C, Yan R, August A, Darnell Jr JE, Hanafusa H 1996 Unique signal transduction of Eyk: constitutive stimulation of the JAK- STAT pathway by an oncogenic receptor-type tyrosine kinase. EMBO J 15:4515– 4525 87. Chin YE, Kitagawa M, Kuida K, Flavell RA, Fu XY 1997 Activation of the STAT signaling pathway can cause expression of caspase 1 and apoptosis. Mol Cell Biol 17:5328 –5337 88. Jensen J, Heller RS, Funder-Nielsen T, Pedersen EE, Lindsell C, Weinmaster G, Madsen OD, Serup P 2000 Independent development of pancreatic ␣- and -cells from neurogenin3-expressing precursors: a role for the notch pathway in repression of premature differentiation. Diabetes 49:163–176 89. Li CM, Yan RT, Wang SZ 1999 Misexpression of cNSCL1 disrupts retinal development. Mol Cell Neurosci 14:17–27 90. Li QL, Jansen E, Brent GA, Friedman TC 2001 Regulation of prohormone convertase 1 (PC1) by thyroid hormone. Am J Physiol Endocrinol Metab 280:E160 –E170 91. Vallet VS, Li JY, Duval J 1997 Secretogranin II (SgII) distribution and processing studies in human normal and adenomatous anterior pituitaries using new polyclonal antibodies. Regul Pept 68:155–163 92. Riva C, Leutner M, Capella C, Usellini L, la Rosa S, Brianza MR, Buffa R 1993 Different expression of chromogranin A and chromogranin B in various types of pituitary adenomas. Zentralbl Pathol 139:165–170 93. Weiss C, Winkler H, Laslop A 2001 Regulation of chromogranin biosynthesis by neurotrophic growth factors in neuroblastoma cells. Neurochem Int 38: 43–52 94. Mahapatra NR, Mahata M, Datta AK, Gerdes HH, Huttner WB, O’Connor DT, Mahata SK 2000 Neuroendocrine cell type-specific and inducible expression of the chromogranin B gene: crucial role of the proximal promoter. Endocrinology 141:3668 –3678 95. Usui H, Morii K, Tanaka R, Tamura T, Washiyama K, Ichikawa T, Kumanishi T 1997 cDNA cloning and mRNA expression analysis of the human neuronatin. High level expression in human pituitary gland and pituitary adenomas. J Mol Neurosci 9:55– 60 96. Joseph R, Dou D, Tsang W 1994 Molecular cloning of a novel mRNA (neuronatin) that is highly expressed in neonatal mammalian brain. Biochem Biophys Res Commun 201:1227–1234 97. Tada M 1992 Molecular structure and function of phospholamban in regulating the calcium pump from sarcoplasmic reticulum. Ann NY Acad Sci 671:92–102 98. Fuhrmann G, Heilig R, Kempf J, Ebel A 1990 Nucleotide sequence of the mouse preprosomatostatin gene. Nucleic Acids Res 18:1287
