Thyroid Hormone-Dependent Gene Expression in Differentiated Embryonic Stem Cells and Embryonal Carcinoma Cells: Identification of Novel Thyr
http://www.100md.com
内分泌学杂志 2005年第2期
Molecular Endocrinology Laboratory, Departments of Medicine and Physiology, David Geffen School of Medicine at University of California, Los Angeles, and Veterans Affairs Greater Los Angeles Healthcare System, Los Angeles, California 90073
Address all correspondence and requests for reprints to: Gregory A. Brent, Molecular Endocrinology Laboratory, VA Greater Los Angeles Healthcare System, Building 114, Room 230, 11301 Wilshire Boulevard, Los Angeles, California 90073. E-mail: gbrent@ucla.edu.
Abstract
T3 is required for normal early development, but relatively few T3-responsive target genes have been identified. In general, in vitro stem cell differentiation techniques stimulate a wide range of developmental programs, including thyroid hormone receptor (TR) pathways. We developed several in vitro stem cell models to more specifically identify TR-mediated gene expression in early development. We found that embryonic carcinoma (EC) cells have reduced T3 nuclear binding capacity and only modestly express the known T3 target genes, neurogranin (RC3) and Ca2+/calmodulin-dependent protein kinase IV (CaMKIV), in response to T3. Full T3 induction in transient transfection of EC cells was restored with cotransfection of a TR expression vector. We, therefore, performed gene expression profiles in wild-type embryonic stem (ES) cells compared with expression in cells with deficient (EC) or mutant TR (TR P398H mutant ES cells), to identify T3 target genes. T3 stimulation of wild-type ES cells altered mRNA expression of 610 known genes (26% of those studied), although only approximately 60 genes (1%) met criteria for direct T3 stimulation based on the magnitude of induction and requirement for the presence of TR. We selected five candidate T3 target genes, neurexophilin 2, spermatid perinuclear RNA-binding protein (SPNR), kallikrein-binding protein (KBP), prostate-specific membrane antigen (PSMA), and synaptotagmin II, for more detailed study. T3 responsiveness of these genes was evaluated in both in vitro endogenous gene expression and in vivo mouse model systems. These genes identified in a novel stem cell system, including those induced and repressed in response to T3, may mediate thyroid hormone actions in early development.
Introduction
TRIIODOTHYRONINE (T3) AND the thyroid hormone receptor (TR) isoforms, TR and TR?, are required for normal embryonic development, especially of the central nervous system (1, 2). Because many of the key genes are transiently regulated by T3, however, it is difficult to study T3 developmental pathways from adult tissue. The primary models for studying the mechanism of thyroid hormone action in early development have been Xenopus (3, 4), embryonic stem (ES), and embryonic carcinoma (EC) cell systems (5, 6, 7). These models have been studied with selective TR isoform agonists and antagonists, TR gene deletions, and TR gene mutations to identify selective T3-mediated developmental pathways. Despite the findings that thyroid hormone is essential for early development, however, only a modest number of T3 target genes have been identified.
TR isoform expression is a major mechanism for developmental regulation. TR is expressed first in all models of development and TR? follows, generally detected with the onset of endogenous thyroid hormone production (8, 9, 10, 11). The differential influence of TR isoforms in development is shown in the Xenopus system. A TR antagonist impairs normal thyroid hormone-dependent development in Xenopus (3). Selective stimulation of only the TR? isoform results in normal tail resorption but defective limb formation (4). TR? knockout mice have profound hearing loss because of abnormal cochlear development (12, 13) but otherwise have normal brain development. TR knockout models have no obvious signs of impaired central nervous system development (14, 15, 16, 17), but TR mutant mouse models show impaired sympathetic nervous system function (18). TR PV mutant mice have reduced brain mass and glucose use (19) as well as dwarfism and impaired fertility (20). TR R384C mutant mice have significant reduction in levels of neurogranin/RC3 mRNA, T3-responsive genes expressed in the brain, in addition to postnatal growth retardation and reduced cardiac function (21).
ES cells are totipotent and can undergo differentiation into all three embryonic layers, forming embryoid bodies in vitro or whole genetically modified mice when injected into blastocysts. In vitro, the commitment of ES cells into specific lineage (hepatocytes, neural cells, astroglial cells, epithelial cells, and cardiac muscle cells) is achieved by inducing agents, including hormones (22, 23, 24, 25, 26, 27, 28). Although ES cells have been widely used for the in vitro study of neural differentiation, the most popular model has been EC cells, such as F9, pheochromocytoma (PC) 12, and P19. Retinoic acid (RA) stimulation of F9 cells results in differentiation into three types of extra-embryonic endoderm (primitive, parietal, and visceral), depending on the RA concentration and the culture conditions used (29, 30). Rat PC12 cells require T3 and TR1 for nerve growth factor-dependent neural differentiation and expression of neuronal genes (31, 32, 33). Human P19 cells respond to RA and differentiate into neuronal, glial, and fibroblast phenotypes (34). EC (F9) cells are rapidly differentiated into neural-like or astroglial cells by RA. Unlike ES cells, formation of embryoid bodies is not required for RA-stimulated EC cell neural differentiation.
Previously, we established a neural differentiation system that used T3-induced neural differentiation in mouse ES cells (7). T3-induced neural differentiation was less extensive compared with that achieved with RA (7). A P398H point mutation of the TR gene impaired T3-induced gene expression but did not influence neural differentiation or morphology.
To understand the mechanism of thyroid hormone action in early development and identify T3 target genes, we performed DNA microarray analysis of differentiated stem cells comparing cells with wild-type TR with those with TR deficiency or mutation. DNA microarray analysis has been used to identify TR isoform-specific gene targets in adult liver (35, 36) using TR?- and TR-deficient mouse models with and without T3 treatment. We demonstrated that murine EC cells (F9) express less TR, resulting in significant reduction in T3 responsiveness. To analyze T3-induced neural differentiation, we performed DNA microarray analysis and real-time PCR determination of mRNA expression in cell lines of ES, EC, and ES cells with an introduced P398H TR mutation. We identified five T3 target genes: neurexophilin 2, spermatid perinuclear RNA-binding protein (SPNR), kallikrein-binding protein (KBP), prostate-specific membrane antigen (PSMA), and synaptotagmin II. We demonstrated that endogenous expression of these genes was T3 responsive in both in vitro systems and, for all but synaptotagmin II, hypothyroid and TR P398H mutant mice (18). Stem cell models with selective TR deficiency or mutation can be used to identify novel T3 target genes and are potential tools for more detailed mechanistic studies.
Materials and Methods
Cell culture and differentiation
J1 ES cells were maintained and differentiated as described previously (7). Murine-derived F9 EC cells were routinely cultured in DMEM with 10% fetal bovine serum (FBS). For differentiation, EC cells were plated onto a 10-cm dish at a density of 1 x 105 cells/ml and grown in DMEM/F-12 medium with hormone supplement (1 nM T3). Cells were treated with T3 for 3–5 d with media changed every other day. Control cells (ES and EC) were undifferentiated and not treated with T3. After differentiation, cells were harvested and total RNA was isolated. HepG2 cells were grown in MEM with 10% FBS and TM4 cells in DMEM/F-12 with 5% horse serum and 2.5% FBS. Before T3 treatment, HepG2 cells and TM4 cells were grown in serum-free medium for 24 h. T3 was added to the medium to a final concentration of 3 nM. HepG2 cells were harvested 24 h after T3 treatment and RNA was isolated. TM4 cells were treated with T3 for 4 h, and then RNA was harvested.
Animal preparation
All animal experiments were approved by the institutional committee for animal protection. Mice were kept under a normal light/dark cycle and feeding conditions. Wild-type and TR mutant (P398H) mice (n = 6), previously described (18), were euthanized at age 3 months. In a separate group of wild-type mice, hypothyroidism was induced by feeding an iodine-deficient diet supplemented with 0.15% propylthiouracil (PTU) for a period of 8.5 wk starting at age 3 wk. Hypothyroid mice (n = 3) were euthanized at age 11.5 wk. Total RNA was isolated from liver, testis, and brain and quantified by RT.
Quantitative real-time PCR analysis of gene expression
Total RNA (5 μg) was reverse transcribed using Superscript II (Invitrogen Inc., Carlsbad, CA). Quantitative real-time PCR was performed as previously described (18). In brief, cDNA was diluted at a ratio of 1:5, 1:25, and 1:125 to generate a standard curve for each pair of primers. For ?-actin mRNA, cDNA was diluted at a ratio of 1:10, 1:100, and 1:1000 and used for the standard curve. The data were normalized to the ?-actin mRNA level and expressed as arbitrary units. The quantitative real-time PCR was performed on Opticon DNA Engine (M.J. Research, Incline Village, NV). The primers used in real-time PCR are listed in Table 1.
TABLE 1. Primers used in real-time quantitative PCR
T3-binding assay
The binding assay sample preparation was described previously (7). In brief, the nuclear extract (15 μg) was incubated in buffer B (without glycerol) with increasing concentrations of [125I]T3 (0.01–0.08 pmol) either in the presence or absence of a 500-fold molar excess of cold T3. The incubation was carried out at 4 C for 4 h. The bound and free [125I]T3 was separated by filtration through a nitrocellulose membrane. The membrane was washed three times before being counted in a -counter.
RNA isolation and microarray analysis
Total RNA was isolated from T3-treated and control cells using Trizol (Invitrogen), digested with DNase I (Ambion Inc., Austin, TX) and cleaned using RNeasy (QIAGEN Biotech, Valencia, CA). The quality of RNA was analyzed by an RNA analyzer before DNA microarray analysis. Two individual P398H ES clones were used for DNA microarray analysis. DNA microarray analysis was performed using GeneChip containing 12,489 mouse genes (MU74Av2 gene chip, Affymetrix Inc., Santa Clara, CA). Data analysis was performed using Microarray Suite 5.0 (Affymetrix). RNA isolation from animal tissues (1–5 mg) was as described previously (18).
Transient transfection
Cells were plated in 24-well dishes at a density of 5 x 104 cells per well and grown 24 h in serum-free medium. The rat Ca2+/calmodulin-dependent protein kinase IV (CaMKIV) promoter-CAT-reporter construct was a gift from Dr. Anthony R. Means (Duke University). Cells were either transfected with rat CaMKIV promoter-CAT-reporter constructs (0.2 μg) alone or cotransfected with constructs expressing mouse TR1 (0.2 μg). The empty vector pCDNA 3.0 was added to the transfection to maintain constant DNA concentration in all transfections. After transfection, cells were grown in serum-free medium supplemented with 1 nM T3. CAT activity was analyzed 48 h after transfection.
Data analysis
All data are expressed as mean ± SE. Statistical analysis used ANOVA for multigroup comparison and Student’s t test for pairs with significance at P < 0.05.
Results
Differential response to T3 in EC and ES cells
We previously demonstrated that T3 induces neural differentiation in wild-type ES and TR P398H mutant J1 ES cells (P398H), as identified by morphology, neural marker gene expression, and immunostaining of neural marker (7). The related EC stem cells are induced toward neural differentiation in response to RA, but T3-induced differentiation has not been tested. We first characterized the capacity of murine-derived F9 EC cells to respond to T3 treatment. Both ES and EC cells were grown to the stage of embryoid bodies before hormone-induced differentiation. Embryoid bodies formed from ES cells were uniform with a round shape and tightly packed, whereas those from EC cells were loosely packed, irregular in shape, and reduced in number. Both ES and EC cells were plated in a 10-cm plate and treated with T3 for 3 d to induce neural differentiation. The morphology of both ES and EC cells reflected neural differentiation, with neuron progenitors predominating in the ES cells and astroglial cells in the EC cells (Fig. 1).
FIG. 1. Morphology of T3-induced neural differentiation of ES and EC cells. Differentiation was initiated by plating embryoid bodies to 10-cm dish without gelatin coating in conditioned DMEM/F-12 media supplemented with 1 nM T3. Wild-type ES and EC cells were differentiated for 3 d. Control cells did not receive hormone treatment. The differentiated cells were photographed with a light contrast microscope. Magnification, x250.
The differentiation markers for neuron progenitor (nestin) and glial cells (glial fibrillary acidic protein, GFAP) were analyzed. Nestin mRNA was induced 10-fold in ES cells but only 2.5-fold in EC cells (Fig. 2A), consistent with the morphology shown (Fig. 1). GFAP mRNA was not expressed in T3-treated ES cells but induced 5-fold in T3-treated EC cells (Fig. 2B). This pattern of gene expression is consistent with the morphological observations and demonstrates T3-induced astroglial differentiation in EC cells and neuron progenitors in ES cells.
FIG. 2. Nestin and GFAP mRNA detection in differentiated ES and EC cells. The differentiation conditions were the same as those described in Fig. 1. Total RNA was isolated from differentiated ES and EC cells and reverse transcribed. The mRNA levels of nestin (A) and GFAP (B) were PCR amplified with 33P-labled primers, resolved in a 5% polyacrylamide gel, and quantified by phosphor-imager. The data are shown as mean value ± SE of three separate experiments. *, P < 0.001 compared with control.
We analyzed the endogenous expression of the previously described T3 target genes, RC3 and CamKIV in ES and EC cells. In T3-treated ES cells, RC3 was induced 21-fold and CamKIV 4.7-fold (Fig. 3). In contrast, T3 treatment of EC cells did not result in significant induction of RC3 or CamKIV expression compared with control cells.
FIG. 3. Differential mRNA expression of neurogranin (RC3) and CaMKIV in ES and EC cells. ES and EC cells were differentiated for 3 d with T3 treatment. Total RNA was isolated and used for RT-PCR to determine the mRNA levels of RC3 (A) and CaMKIV mRNA (B). PCR amplification was done in 25 cycles with 2 μl cDNA. The expression level of ?-actin mRNA was quantified using diluted cDNA (1:5). Quantification was performed by phosphor-imager and controlled for the level of ?-actin mRNA. The data shown are from three separate experiments. **, P < 0.001 compared with control; *, P < 0.05 compared with control.
EC cells are derived from a carcinoma, which may have reduced TR1 expression, as has been reported in PC12 cells (33). To test this hypothesis, TR1 mRNA content was determined by RT-PCR and was reduced 5.7-fold in EC cells compared with ES cells (Fig. 4). We directly analyzed functional TR protein by T3-binding activity in nuclear extracts from EC and ES cells. The maximum binding capacity of EC cells was reduced 35% compared with ES cells. The affinity for T3 was 1.3 nM in nuclear extracts from EC cells and 1.8 nM in nuclear extracts from ES cells (Fig. 5).
FIG. 4. TR1 mRNA expression in ES and EC cells. Cell differentiation and PCR analysis were the same as those described in Fig. 3.
FIG. 5. [125I]T3 binding to nuclear extracts of ES and EC cells and Scatchard analysis. Nuclear extracts of ES cells (15 μg) (A) and EC cells (15 μg) (B) were incubated with various concentrations of [125I]T3 either in the presence or absence of a 500-fold molar excess of cold T3 for 4 h at 4 C. The bound and free T3 was separated and radioactivity counted (as described in Materials and Methods). The assays were done in triplicates at each concentration, and the mean of three determinations with SE is shown.
A number of factors, including overexpression of COUPTF1, have been shown to modulate T3 responsiveness in developing neurons (37). To determine whether the reduced TR content in EC cells was limiting for the T3 response, we performed a transient transfection assay. A CamKIV promoter reporter construct (37, 38) was transfected into ES and EC cells. In ES cells, 5-fold T3 induction was observed; however, in EC cells, no T3 induction was detected. Cotransfection of a TR1 expression vector in EC cells restored 4.3-fold T3 induction (Fig. 6). These data indicate that the reduced TR content in EC cells is likely to be responsible for reduced T3-induced gene expression. EC cells also provide a neural differentiation model that is not T3 responsive to compare with wild-type ES cells and identify T3 target genes.
FIG. 6. T3 induction of rat CaMKIV-1060 reporter construct in ES and EC cells. The CaMKIV-1060 CAT reporter construct (0.2 μg) was transfected into ES cells and EC cells with T3 treatment and with and without cotransfection of a TR expression vector (0.2 μg). CAT activity was determined 48 h after transfection and normalized for protein concentration used in the assay. The assays were performed in duplicate. *, P < 0.05.
Effects of wild-type TR and T3 treatment on gene expression
Despite the profound effects of T3 on development and the clear requirement for expression of TR, relatively few T3 gene targets have been identified. We previously developed an ES cell line by introducing a mutation (P398H) into the TR gene. The morphology of neural differentiation of TR P398H ES cells was indistinguishable from neural differentiated wild-type ES cells. The T3 target genes (RC3 and CaMKIV), however, were significantly down-regulated in TR P398H ES cells (7).
We used differentiated TR P398H ES cells to compare with differentiated wild-type ES cells, profile gene expression, and identify T3 target genes. These cell lines differ only by whether functional TR was present. EC cells, although different in origin from ES cells, are functionally TR deficient and provide another comparison with wild-type ES cells. We treated cells with T3 for 3 d under differentiation conditions. The gene expression profiles of the three cell models (ES, P398H ES, and EC) and undifferentiated control cells were analyzed using DNA microarray (MU74Av2 gene chip, Affymitrx). The gene chip (MU74Av2) contains 12,489 mouse genes. The comparison analysis used Microarray Suite, and wild-type ES cells (undifferentiated) were used as a baseline to filter out the genes that were not present and unchanged when compared with T3-treated wild-type ES cells. Similarly, T3-treated wild-type ES cells were used as the baseline when compared with P398H ES cells. Data analysis showed that approximately 45% of genes in the gene chip were not present. In the present genes, in wild-type ES cells, T3-induced neural differentiation resulted in altered mRNA expression of 610 known genes (26.2% of those tested). Using a threshold of greater than 5-fold stimulation or repression, 88 genes were identified with 5- to 26-fold up-regulation or 5- to 18-fold down-regulation. The gene with the greatest induction was myosin light chain (MLC), a known T3 target gene, with 26-fold induction.
Because differentiation alone alters gene expression, many of the genes identified are likely to be only indirectly regulated by T3. We, therefore, compared gene expression in wild-type ES cells with expression in TR P398H ES cells, a cell line with a dominant-negative TR P398H mutation known to impair T3-dependent gene expression (7, 18), and EC cells with low nuclear TR content. The results from DNA microarray analysis of two mutant ES cell lines derived from two individual TR P398H ES clones showed that 75–79% of known genes had no detectable change compared with wild-type ES cells; 20–24% of known genes had modest changes and 0.7–1% of known genes (45–60 genes) were significantly up- or down-regulated (>5-fold). The expression of most unknown genes, 85–90%, was unchanged, and 0.2–0.3% (12–18 genes) had equal or greater than 5-fold change in mRNA level. These results indicate that T3 treatment influenced a wide range of genes in a stem cell differentiation model (>20% of known genes) either directly or indirectly. However, the number of genes directly targeted by T3 was small. These findings are similar to studies of T3-dependent gene expression in other T3 target tissues, such as liver (35, 36). The T3 treatment with 3-d duration was sufficient for detecting the changes in early differentiation events.
Identification of T3 target genes
Based on the results of gene profiles in wild-type ES, compared with the TR P398H ES and EC cells, we focused on 40 putative T3-responsive genes to characterize endogenous expression. We characterized mRNA expression by quantitative real-time PCR in wild-type ES cells, TR P398H ES cells, and EC cells. Consistent endogenous T3 regulation was confirmed in five genes, with response in wild-type ES, but not TR P398H ES cells, and EC cells. The regulated genes included neurexophilin 2, SPNR, synatotagmin II, KBP, and PSMA. PSMA mRNA was negatively regulated by T3, and the other genes were induced. In wild-type ES cells, T3 stimulated neurexophilin 2 mRNA 22-fold, SPNR 20-fold, synaptotagmin II 13-fold, and KBP 5.3-fold compared with baseline (untreated ES cells). In TR P398H ES cells, T3 also stimulated mRNA expression of these genes, but the extent was significantly lower compared with wild-type ES cells (Table 2). In P398H ES cells, neurexophilin 2 mRNA was stimulated 8.3-fold, SPNR 6.7-fold, synaptotagmin II 7.2-fold, and KBP 3-fold. PSMA expression in response to T3 was reduced 90% in wild-type ES cells and 41% in P398H ES cells, compared with baseline. The target genes were not detected, or they had less than 2-fold induction, in T3-treated EC cells compared with control EC cells (Table 2).
TABLE 2. mRNA expression of potential T3-responsive genes (arbitrary units)
Endogenous gene expression in both in vivo and in vitro models
The T3-responsive genes identified in stem cells are endogenously expressed in a range of tissues, including brain, liver, and testes. T3 can regulate genes at the transcriptional and posttranscriptional levels. T1-tubulin, for example, is T3 inducible but does not have a thyroid hormone response element in the gene. It is thought to be regulated by the influence of T3 on mRNA stability (39). We used cell lines from the tissues relevant to the identified genes and tissue samples from hypothyroid mice and mice with a TR P398H mutation. Neurexophilin 2, a gene dominantly expressed in mouse liver (40, 41), was examined in liver-derived HepG2 cells. The cells were grown in serum-free medium and treated with 3 nM T3 for 24 h. The mRNA level of neurexophilin 2 was stimulated 3.6-fold (P < 0.05) (Fig. 7A). Both SPNR and KBP are expressed in testis and were examined in a testis cell line, TM4. After 4 h of T3 treatment, the mRNA levels of SPNR and KBP were stimulated 2.3-fold and 1.2-fold compared with the levels without T3 (Fig. 7B).
FIG. 7. T3 responsiveness of neurexophilin 2, SPNR, and KBP in cell culture. HepG2 and TM4 cells were gown in serum-free media for 24 h before T3 treatment. T3 was added to culture at final concentration 3 nM. The treatment for HepG2 cells was 24 h and for TM4 cells was 4 h. mRNA levels were determined by quantitative real-time PCR and normalized for ?-actin mRNA levels. Nerexophilin 2 (A) was determined using HepG2 cells and SPNR (B) and KBP (C) using TM4 cells. The assays were performed in triplicate. *, P < 0.05.
We further examined mRNA expression in euthyroid, hypothyroid, and TR P398H mutant mice. Mice were made hypothyroid by feeding a PTU-containing iodine-deficient diet for 8.5 wk. We confirmed hypothyroidism by analyzing mRNA levels of well-known T3-targeted genes, MLC, and CamKIV. The mRNA level of MLC in heart was reduced 11-fold and CaMKIV expression in testes reduced 3.1-fold (Fig. 8). The mRNA expression of the T3 target genes was determined using tissues isolated from three different animal models: wild-type, hypothyroid, and TR P398H mutant mice. In liver, neurexophilin 2 was reduced 3.7-fold in P398H mice and 2.7-fold in hypothyroid mice compared with euthyroid mice (Fig. 9A). In testis, SPNR was reduced 5.4-fold in the P398H mice and 2.2-fold in hypothyroid mice (Fig. 9B); KBP was reduced 2.3-fold in testes of the P398H mice and 3.3-fold in hypothyroid mice compared with wild-type mice (Fig. 9C). Levels of PSMA mRNA were increased 2-fold in P398H mice and 3-fold in hypothyroid mice compared with wild-type mice (Fig. 9D). Synaptotagmin II mRNA had only modest reduction in the brain of hypothyroid or mutant TR mice compared with wild-type mice (data not shown).
FIG. 8. mRNA levels of known T3-responsive genes in hypothyroid mice. Hypothyroidism in mice was induced by feeding iodine-deficient diet supplement with 0.15% PTU for 8.5 wk. Total RNA was isolated from heart and testis of wild-type (n = 4), P398H (n = 4), and hypothyroid male mice (n = 3). RNA was treated with DNase I before RT. The level of mRNA expression was determined using quantitative real-time PCR and normalized for ?-actin mRNA levels of MLC (A) in heart and CaMKIV (B) in testis. The data shown are the mean values ± SE. Statistical comparison is between the wild-type and P398H mice or wild-type and hypothyroid mice: *, P < 0.05; **, P < 0.01.
FIG. 9. In vivo T3 response of identified genes using animal models (described in Fig. 8). Total RNA was isolated from liver and testis. The conditions for quantitative real-time PCR were as described (Fig. 8). Neurexophilin 2 (A) was analyzed using RNA from liver; SPNR (B), KBP (C), and PSMA (D) were from testis. The assays were done in triplicate. *, P < 0.05; **, P < 0.01.
We tested the response of these genes to T3 treatment in EC cells. SPNR, PSMA, and KBP were not detected by quantitative real-time PCR at baseline or with T3 treatment. Neurexophilin 2 and synaptotagmin II mRNA were detected at baseline but did not significantly change with T3 treatment, consistent with reduced TR content (data not shown).
These data confirm that the five genes, including four stimulated by T3 and one repressed, are regulated by T3 and require wild-type TR for the response. T3 regulation was confirmed for all genes in ES or other in vitro cell models and for four of the five genes in in vivo mouse models. In the case of synaptotagmin II, it is likely that the T3 response is present developmentally but not in the adult brain. This will require further investigation.
Discussion
We have demonstrated that ES and EC cells have a differential response to T3-induced differentiation with respect to cell morphology, endogenous gene expression, and T3-dependent gene expression. The restoration of the T3 response to EC cells with the addition of TR indicates that reduced nuclear TR content contributes to the observed reduced T3-induced gene expression. In addition to the reduction in nuclear TR, however, altered expression of TR coactivators and corepressors may also play a role in reduced T3-dependent gene expression in EC cells. Reduced TR expression and expression of mutant TRs that do not bind T3 have been reported in a variety of tumors and tumor cell lines (33, 42, 43). In PC12 neural stem cells, TR1 and T3 are critical for maintaining nerve growth factor-dependent neural differentiation, expression of neuronal genes, and normal neural morphology (39). This supports the role of TR in maintaining differentiated function in stem cells. The morphological phenotype of EC cell differentiation and relative sensitivity to T3 and RA is likely because of a number of factors in addition to TR content, related to the neoplastic tissue origin of these cells.
The observation that expression of 26% of the mRNAs from the genes studied are altered as a consequence of T3-induced differentiation of ES cells indicates a large fraction of genes required for this developmental program. The value of comparison with the expression profile in the TR mutant ES cells and EC cells was to determine those genes most likely to be directly mediated by TR. These comparative models, and a relatively high threshold for significant induction, narrowed the number of genes studied.
We identified five genes (neurexophilin 2, SPNR, KBP, PSMA, and synaptotagmin II) that meet the criteria for a T3/TR-mediated response that have not previously been shown to be regulated by T3. We demonstrated that the expression of neurexophilin 2, SPNR, KBP, and PSMA was reduced in TR P398H mice (14) and hypothyroid mice but induced in euthyroid mice. The synaptotagmin II gene was T3 responsive in ES cells but not in brain from animal models. It is possible that synaptotagmin II transiently responds to T3 in an early developmental stage but not in adult tissue.
The physiological function of these T3-responsive genes is not fully understood. SPNR is a microtube-associated RNA-binding protein, and its mRNA is expressed at high levels in the testis, exclusively in postmeiotic germ cells (44, 45). Low-level expression of SPNR was also seen in ovarian and brain tissue. Mice deficient in SPNR have a high rate of mortality and abnormalities in neurological development, spermatogenesis, and sperm morphology (45). Indeed, SPNR was reduced 5.4-fold in testis of the P398H mice, and the male mice had a significant reduction in fertility (18). Other TR mutant mice (PV and R384C) have been reported to have an increased mortality rate and reduced fertility (20, 21). KBP, a serine proteinase inhibitor, binds to and inhibits tissue kallikrein activity and is found in endothelial and smooth muscle cells of blood vessels (46, 47). Neurexophilin 2 is a member of the neurexophilin family, the ligand for neurexins, and is exclusively expressed in mouse liver (41).
PSMA gene expression has been linked to recurrence of prostate cancer (48). Down-regulation of PSMA gene expression by thyroid hormone may be associated with an effect of differentiating prostate cancer. A number of studies have looked at the influence of thyroid hormone on prostate cancer growth and differentiation. Growth and differentiation of the LNCaP prostate cancer cell line has been associated with T3 treatment (49). Thyroid hormone directly stimulates the expression of the prostate-specific antigen, and a response element has been identified (50).
Thyroid hormone regulation of development requires stimulation of a number of genes in a time- and location-specific manner. The majority of T3-regulated genes that have been identified and studied are in adult tissues, which are much more accessible for study. Models of thyroid hormone action in mammalian development have been difficult to develop, because thyroid hormone has such a broad range of actions. The models we have used permit the identification of those genes that require T3 and TR mediation for their effects. The genes identified in this way were all regulated by T3 in vitro and, in all but one gene, in the in vivo system. This approach holds promise for further understanding of T3 regulation of development.
References
Oppenheimer JH, Schwartz HL 1997 Molecular basis of thyroid hormone-dependent brain development. Endocr Rev 18:462–475
Lazarus JH 1999 Thyroid hormones and neurodevelopment. Clin Endocrinol (Oxf) 50:147–148
Lim W, Nguyen NH, Yang HY, Scanlan TS, Furlow JD 2002 A thyroid hormone antagonist that inhibits thyroid hormone action in vivo. J Biol Chem 277:35664–35670
Furlow JD, Yang HY, Hsu M, Lim W, Ermio DJ, Chiellini G, Scanlan TS 2004 Induction of larval tissue resorption in Xenopus laevis tadpoles by the thyroid hormone receptor agonist GC-1. J Biol Chem 279:26555–26562
Lee LR, Mortensen RM, Larson CA, Brent GA 1994 Thyroid hormone receptor- inhibits retinoic acid-responsive gene expression and modulates retinoic acid-stimulated neural differentiation in mouse embryonic stem cells. Mol Endocrinol 8:746–756
Münoz A, Wrighton C, Seliger B, Bernal J, Beug H 1993 Thyroid hormone receptor/c-erbA: control of commitment and differentiation in the neuronal/chromaffin progenitor line PC12. J Cell Biol 121:23–38
Liu YY, Tachiki KH, Brent GA 2002 Targeted thyroid hormone receptor gene dominant negative mutation (P398H) selectively impairs gene expression in differentiated embryonic stem cells. Endocrinology 143:2664–2672
Forrest D, Hallbook F, Persson H, Vennstrom B 1991 Distinct functions for thyroid hormone receptors and ? in brain development indicated by differential expression of receptor genes. EMBO J 10:269–275
Porterfield SP, Hendrich CE 1993 The role of thyroid hormones in prenatal and neonatal neurological development: current perspectives. Endocr Rev 14:94–106
Dowling AL, Zoeller RT 2000 Thyroid hormone of maternal origin regulates the expression of RC3/neurogranin mRNA in the fetal rat brain. Brain Res Mol Brain Res 82:126–132
Bernal J, Guadano-Ferraz A, Morte B 2003 Perspectives in the study of thyroid hormone action on brain development and function. Thyroid 13:1005–1012
Forres D, Erway LC, Ng L, Altschuler R, Curran T 1996 Thyroid hormone receptor ? is essential for development of auditory function. Nat Genet 13:354–357
Forrest D, Hanebuth E, Smeyne RJ, Everds N, Stewart CL, Wehner JM, Curran T 1996 Recessive resistance to thyroid hormone in mice lacking thyroid hormone receptor ?: evidence for tissue-specific modulation of receptor function. EMBO J 15:3006–3015
Brent GA 2000 Tissue-specific actions of thyroid hormone: insights from animal models. Rev Endocr Metab Disord 1:27–33
Fraichard A, Chassande O, Plateroti M, Roux JP, Trouillas J, Dehay C, Legrand C, Gauthier K, Kedinger M, Malaval L, Rousset B, Samarut J 1997 The T3R gene encoding a thyroid hormone receptor is essential for post-natal development and thyroid hormone production. EMBO J 16:4412–4420
Gaüthier K, Plateroti M, Harvey CB, Williams GR, Weiss RE, Refetoff S, Willott JF, Sundin V, Roux JP, Malaval L, Hara M, Samarut J, Chassande O 2001 Genetic analysis reveals different functions for the products of the thyroid hormone receptor locus. Mol Cell Biol 21:4748–4760
Wikstrom L, Johansson C, Salto C, Barlow C, Barros AC, Baas F, Forrest D, Thoren P, Vennstrom B 1998 Abnormal heart rate and body temperature in mice lacking thyroid hormone receptor . EMBO J 17:455–461
Liu Y-Y, Schultz JJ, Brent GA 2003 A thyroid hormone receptor gene mutation (P398H) is associated with visceral adiposity and impaired catecholamine-stimulated lipolysis in mice. J Biol Chem 278:38913–38920
Itoh Y, Esaki T, Kaneshige M, Suzuki H, Cook M, Sokoloff L, Cheng SY, Nunez J 2001 Brain glucose utilization in mice with a targeted mutation in the thyroid hormone or ? receptor gene. Proc Natl Acad Sci USA 98:9913–9918
Kaneshige M, Suzuki H, Kaneshige K, Cheng J, Wimbrow H, Barlow C, Willingham MC, Cheng S 2001 A targeted dominant negative mutation of the thyroid hormone 1 receptor causes increased mortality, infertility, and dwarfism in mice. Proc Natl Acad Sci USA 98:15095–15100
Tinnikov A, Nordstrom K, Thoren P, Kindblom JM, Malin S, Rozell B, Adams M, Rajanayagam O, Pettersson S, Ohlsson C, Chatterjee K, Vennstrom B 2002 Retardation of post-natal development caused by a negatively acting thyroid hormone receptor 1. EMBO J 21:5079–5087
Keller GM 1995 In vitro differentiation of embryonic stem cells. Curr Opin Cell Boil 7:862–869
Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M, Gepstein A, Livne E, Binah O, Itsk-vitz-Eldor J, Gesptein L 2001 Human embryonic stem cells can differentiate into myocytes with structural and functional properties of cardiomyocytes. J Clin Invest 108:407–414
Balasubramaniyan V, de Haas AH, Bakels R, Koper A, Boddeke HW, Copray JC 2004 Functionally deficient neuronal differentiation of mouse embryonic neural stem cells in vitro. Neurosci Res 49:261–265
Pera MF, Filipczyk AA, Hawes SM, Laslett AL 2003 Isolation, characterization, and differentiation of human embryonic stem cells. Methods Enzymol 365:429–446
Heins N, Englund MC, Sjoblom C, Dahl U, Tonning A, Bergh C, Lindahl A, Hanson C, Semb H 2004 Derivation, characterization, and differentiation of human embryonic stem cells. Stem Cells 22:367–376
Kania G, Blyszczuk P, Czyz J, Navarrete-Santos A, Wobus AM 2003 Differentiation of mouse embryonic stem cells into pancreatic and hepatic cells. Methods Enzymol 365:287–303
Roberston EJ 1997 Derivation and maintenance of embryonic stem cell cultures. Methods Mol Biol 75:173–184
Strickland S, Smith KK, Marotti KR 1980 Hormonal induction of differentiation in teratocarcinoma stem cells: generation of parietal endoderm by retinoic acid and dibutyryl cAMP. Cell 21:347–355
Clifford J, Chiba H, Sobieszczuk D, Metzger D, Chambon P 1996 RXR-null F9 embryonal carcinoma cells are resistant to the differentiation, anti-proliferative and apoptotic effects of retinoids. EMBO J 15:4142–4155
Greene LA, Tischler AS 1976 Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc Natl Acad Sci USA 73:424–2428
Weber E, Jilling T, Kirk KL 1996 Distinct functional properties of Rab3A and Rab3B in PC12 neuroendocrine cells. J Biol Chem 271:6963–6971
Munoz A, Wrighton C, Seliger B, Bernal J, Beug H 1993 Thyroid hormone receptor/c-erbA: control of commitment and differentiation in the neuronal/chromaffin progenitor line PC12. J Cell Biol 121:423–438
Bain G, Ray WJ, Yao M, Gottlieb DI 1994 From embryonal carcinoma cells to neurons: the P19 pathway. Bioessays 6:343–348
Flores-Morales A, Gullberg H, Fernandez L, Stahlberg N, Lee NH, Vennstrom B, Norstedt G 2002 Patterns of liver gene expression governed by TR?. Mol Endocrinol 16:1257–1268
Yen PM, Feng X, Flamant F, Chen Y, Walker RL, Weiss RE, Chassande O, Samarut J, Refetoff S, Meltzer PS 2003 Effects of ligand and thyroid hormone receptor isoforms on hepatic gene expression profiles of thyroid hormone receptor knockout mice. EMBO Rep 4:581–587
Liu YY, Brent GA 2002 A complex deoxyribonucleic acid response element in the rat Ca2+/calmodulin-dependent protein kinase IV gene 5'-flanking region mediates thyroid hormone induction and chicken ovalbumin upstream promoter transcription factor 1 repression. Mol Endocrinol 16:2439–2451
Means AR, Ribar TJ, Kane CD, Hook SS, Anderson KA 1997 Regulation and properties of the rat Ca2+/calmodulin-dependent protein kinase IV gene and its protein products. Recent Prog Horm Res 52:389–407
Lorenao PI, Menard C, Miller FD, Bernal J 2002 Thyroid hormone-dependent regulation of T1 -tubulin during brain development. Mol Cell Neurosci 19:333–343
Missler M, Sudhof TC 1998 Neurexophilins form a conserved family of neuropeptide-like glycoproteins. J Neurosci 18:3630–3638
Petrenko AG, Ullrich B, Missler M, Krasnoperov V, Rosahl TW, Sudhof TC 1996 Structure and evolution of neurexophilin. J Neurosci 16:4360–4369
Zhang XY, Kaneshige M, Kamiya Y, Kaneshige K, McPhie P, Cheng SY 2002 Differential expression of thyroid hormone receptor isoforms dictates the dominant negative activity of mutant ? receptor. Mol Endocrinol 16:2077–2092
Gittoes NJ, McCabe CJ, Verhaeg J, Sheppard MC, Franklyn JA 1998 An abnormality of thyroid hormone receptor expression may explain abnormal thyrotropin production in thyrotropin-secreting pituitary tumors. Thyroid 8:9–14
Schumacher JM, Lee K, Edelhoff S, Braun RE 1995 Spnr, a murine RNA-binding protein that is localized to cytoplasmic microtubules. J Cell Biol 129:1023–1032
Pires-daSilva A, Nayernia K, Engel W, Torres M, Stoykova A, Chowdhury K, Gruss P 2001 Mice deficient for spermatid perinuclear RNA-binding protein show neurologic, spermatogenic, and sperm morphological abnormalities. Dev Biol 233:319–328
Chao J, Miao RQ, Chen V, Chen LM, Chao L 2001 Novel roles of kallistatin, a specific tissue kallikrein inhibitor, in vascular remodeling. Biol Chem 382:15–21
Evans BA, Drinkwater CC, Richards RI 1987 Mouse glandular kallikrein genes. Structure and partial sequence analysis of the kallikrein gene locus. J Biol Chem 262:8027–8034
Ross JS, Sheehan CE, Fisher HA, Kaufman Jr RP, Kaur P, Gray K, Webb I, Gray GS, Mosher R, Kallakury BV 2003 Correlation of primary tumor prostate-specific membrane antigen expression with disease recurrence in prostate cancer. Clin Cancer Res 9:6357–6362
Esquenet M, Swinnen JV, Heyns W, Verhoeven G 1995 Triiodothyronine modulates growth, secretory function and androgen receptor concentration in the prostatic carcinoma cell line LNCaP. Mol Cell Endocrinol 109:105–111
Zhu W, Young CY 2001 Androgen-dependent transcriptional regulation of the prostate-specific antigen gene by thyroid hormone 3,5,3'-L-triiodothyronine. J Androl 22:136–141(Yan-Yun Liu and Gregory A)
Address all correspondence and requests for reprints to: Gregory A. Brent, Molecular Endocrinology Laboratory, VA Greater Los Angeles Healthcare System, Building 114, Room 230, 11301 Wilshire Boulevard, Los Angeles, California 90073. E-mail: gbrent@ucla.edu.
Abstract
T3 is required for normal early development, but relatively few T3-responsive target genes have been identified. In general, in vitro stem cell differentiation techniques stimulate a wide range of developmental programs, including thyroid hormone receptor (TR) pathways. We developed several in vitro stem cell models to more specifically identify TR-mediated gene expression in early development. We found that embryonic carcinoma (EC) cells have reduced T3 nuclear binding capacity and only modestly express the known T3 target genes, neurogranin (RC3) and Ca2+/calmodulin-dependent protein kinase IV (CaMKIV), in response to T3. Full T3 induction in transient transfection of EC cells was restored with cotransfection of a TR expression vector. We, therefore, performed gene expression profiles in wild-type embryonic stem (ES) cells compared with expression in cells with deficient (EC) or mutant TR (TR P398H mutant ES cells), to identify T3 target genes. T3 stimulation of wild-type ES cells altered mRNA expression of 610 known genes (26% of those studied), although only approximately 60 genes (1%) met criteria for direct T3 stimulation based on the magnitude of induction and requirement for the presence of TR. We selected five candidate T3 target genes, neurexophilin 2, spermatid perinuclear RNA-binding protein (SPNR), kallikrein-binding protein (KBP), prostate-specific membrane antigen (PSMA), and synaptotagmin II, for more detailed study. T3 responsiveness of these genes was evaluated in both in vitro endogenous gene expression and in vivo mouse model systems. These genes identified in a novel stem cell system, including those induced and repressed in response to T3, may mediate thyroid hormone actions in early development.
Introduction
TRIIODOTHYRONINE (T3) AND the thyroid hormone receptor (TR) isoforms, TR and TR?, are required for normal embryonic development, especially of the central nervous system (1, 2). Because many of the key genes are transiently regulated by T3, however, it is difficult to study T3 developmental pathways from adult tissue. The primary models for studying the mechanism of thyroid hormone action in early development have been Xenopus (3, 4), embryonic stem (ES), and embryonic carcinoma (EC) cell systems (5, 6, 7). These models have been studied with selective TR isoform agonists and antagonists, TR gene deletions, and TR gene mutations to identify selective T3-mediated developmental pathways. Despite the findings that thyroid hormone is essential for early development, however, only a modest number of T3 target genes have been identified.
TR isoform expression is a major mechanism for developmental regulation. TR is expressed first in all models of development and TR? follows, generally detected with the onset of endogenous thyroid hormone production (8, 9, 10, 11). The differential influence of TR isoforms in development is shown in the Xenopus system. A TR antagonist impairs normal thyroid hormone-dependent development in Xenopus (3). Selective stimulation of only the TR? isoform results in normal tail resorption but defective limb formation (4). TR? knockout mice have profound hearing loss because of abnormal cochlear development (12, 13) but otherwise have normal brain development. TR knockout models have no obvious signs of impaired central nervous system development (14, 15, 16, 17), but TR mutant mouse models show impaired sympathetic nervous system function (18). TR PV mutant mice have reduced brain mass and glucose use (19) as well as dwarfism and impaired fertility (20). TR R384C mutant mice have significant reduction in levels of neurogranin/RC3 mRNA, T3-responsive genes expressed in the brain, in addition to postnatal growth retardation and reduced cardiac function (21).
ES cells are totipotent and can undergo differentiation into all three embryonic layers, forming embryoid bodies in vitro or whole genetically modified mice when injected into blastocysts. In vitro, the commitment of ES cells into specific lineage (hepatocytes, neural cells, astroglial cells, epithelial cells, and cardiac muscle cells) is achieved by inducing agents, including hormones (22, 23, 24, 25, 26, 27, 28). Although ES cells have been widely used for the in vitro study of neural differentiation, the most popular model has been EC cells, such as F9, pheochromocytoma (PC) 12, and P19. Retinoic acid (RA) stimulation of F9 cells results in differentiation into three types of extra-embryonic endoderm (primitive, parietal, and visceral), depending on the RA concentration and the culture conditions used (29, 30). Rat PC12 cells require T3 and TR1 for nerve growth factor-dependent neural differentiation and expression of neuronal genes (31, 32, 33). Human P19 cells respond to RA and differentiate into neuronal, glial, and fibroblast phenotypes (34). EC (F9) cells are rapidly differentiated into neural-like or astroglial cells by RA. Unlike ES cells, formation of embryoid bodies is not required for RA-stimulated EC cell neural differentiation.
Previously, we established a neural differentiation system that used T3-induced neural differentiation in mouse ES cells (7). T3-induced neural differentiation was less extensive compared with that achieved with RA (7). A P398H point mutation of the TR gene impaired T3-induced gene expression but did not influence neural differentiation or morphology.
To understand the mechanism of thyroid hormone action in early development and identify T3 target genes, we performed DNA microarray analysis of differentiated stem cells comparing cells with wild-type TR with those with TR deficiency or mutation. DNA microarray analysis has been used to identify TR isoform-specific gene targets in adult liver (35, 36) using TR?- and TR-deficient mouse models with and without T3 treatment. We demonstrated that murine EC cells (F9) express less TR, resulting in significant reduction in T3 responsiveness. To analyze T3-induced neural differentiation, we performed DNA microarray analysis and real-time PCR determination of mRNA expression in cell lines of ES, EC, and ES cells with an introduced P398H TR mutation. We identified five T3 target genes: neurexophilin 2, spermatid perinuclear RNA-binding protein (SPNR), kallikrein-binding protein (KBP), prostate-specific membrane antigen (PSMA), and synaptotagmin II. We demonstrated that endogenous expression of these genes was T3 responsive in both in vitro systems and, for all but synaptotagmin II, hypothyroid and TR P398H mutant mice (18). Stem cell models with selective TR deficiency or mutation can be used to identify novel T3 target genes and are potential tools for more detailed mechanistic studies.
Materials and Methods
Cell culture and differentiation
J1 ES cells were maintained and differentiated as described previously (7). Murine-derived F9 EC cells were routinely cultured in DMEM with 10% fetal bovine serum (FBS). For differentiation, EC cells were plated onto a 10-cm dish at a density of 1 x 105 cells/ml and grown in DMEM/F-12 medium with hormone supplement (1 nM T3). Cells were treated with T3 for 3–5 d with media changed every other day. Control cells (ES and EC) were undifferentiated and not treated with T3. After differentiation, cells were harvested and total RNA was isolated. HepG2 cells were grown in MEM with 10% FBS and TM4 cells in DMEM/F-12 with 5% horse serum and 2.5% FBS. Before T3 treatment, HepG2 cells and TM4 cells were grown in serum-free medium for 24 h. T3 was added to the medium to a final concentration of 3 nM. HepG2 cells were harvested 24 h after T3 treatment and RNA was isolated. TM4 cells were treated with T3 for 4 h, and then RNA was harvested.
Animal preparation
All animal experiments were approved by the institutional committee for animal protection. Mice were kept under a normal light/dark cycle and feeding conditions. Wild-type and TR mutant (P398H) mice (n = 6), previously described (18), were euthanized at age 3 months. In a separate group of wild-type mice, hypothyroidism was induced by feeding an iodine-deficient diet supplemented with 0.15% propylthiouracil (PTU) for a period of 8.5 wk starting at age 3 wk. Hypothyroid mice (n = 3) were euthanized at age 11.5 wk. Total RNA was isolated from liver, testis, and brain and quantified by RT.
Quantitative real-time PCR analysis of gene expression
Total RNA (5 μg) was reverse transcribed using Superscript II (Invitrogen Inc., Carlsbad, CA). Quantitative real-time PCR was performed as previously described (18). In brief, cDNA was diluted at a ratio of 1:5, 1:25, and 1:125 to generate a standard curve for each pair of primers. For ?-actin mRNA, cDNA was diluted at a ratio of 1:10, 1:100, and 1:1000 and used for the standard curve. The data were normalized to the ?-actin mRNA level and expressed as arbitrary units. The quantitative real-time PCR was performed on Opticon DNA Engine (M.J. Research, Incline Village, NV). The primers used in real-time PCR are listed in Table 1.
TABLE 1. Primers used in real-time quantitative PCR
T3-binding assay
The binding assay sample preparation was described previously (7). In brief, the nuclear extract (15 μg) was incubated in buffer B (without glycerol) with increasing concentrations of [125I]T3 (0.01–0.08 pmol) either in the presence or absence of a 500-fold molar excess of cold T3. The incubation was carried out at 4 C for 4 h. The bound and free [125I]T3 was separated by filtration through a nitrocellulose membrane. The membrane was washed three times before being counted in a -counter.
RNA isolation and microarray analysis
Total RNA was isolated from T3-treated and control cells using Trizol (Invitrogen), digested with DNase I (Ambion Inc., Austin, TX) and cleaned using RNeasy (QIAGEN Biotech, Valencia, CA). The quality of RNA was analyzed by an RNA analyzer before DNA microarray analysis. Two individual P398H ES clones were used for DNA microarray analysis. DNA microarray analysis was performed using GeneChip containing 12,489 mouse genes (MU74Av2 gene chip, Affymetrix Inc., Santa Clara, CA). Data analysis was performed using Microarray Suite 5.0 (Affymetrix). RNA isolation from animal tissues (1–5 mg) was as described previously (18).
Transient transfection
Cells were plated in 24-well dishes at a density of 5 x 104 cells per well and grown 24 h in serum-free medium. The rat Ca2+/calmodulin-dependent protein kinase IV (CaMKIV) promoter-CAT-reporter construct was a gift from Dr. Anthony R. Means (Duke University). Cells were either transfected with rat CaMKIV promoter-CAT-reporter constructs (0.2 μg) alone or cotransfected with constructs expressing mouse TR1 (0.2 μg). The empty vector pCDNA 3.0 was added to the transfection to maintain constant DNA concentration in all transfections. After transfection, cells were grown in serum-free medium supplemented with 1 nM T3. CAT activity was analyzed 48 h after transfection.
Data analysis
All data are expressed as mean ± SE. Statistical analysis used ANOVA for multigroup comparison and Student’s t test for pairs with significance at P < 0.05.
Results
Differential response to T3 in EC and ES cells
We previously demonstrated that T3 induces neural differentiation in wild-type ES and TR P398H mutant J1 ES cells (P398H), as identified by morphology, neural marker gene expression, and immunostaining of neural marker (7). The related EC stem cells are induced toward neural differentiation in response to RA, but T3-induced differentiation has not been tested. We first characterized the capacity of murine-derived F9 EC cells to respond to T3 treatment. Both ES and EC cells were grown to the stage of embryoid bodies before hormone-induced differentiation. Embryoid bodies formed from ES cells were uniform with a round shape and tightly packed, whereas those from EC cells were loosely packed, irregular in shape, and reduced in number. Both ES and EC cells were plated in a 10-cm plate and treated with T3 for 3 d to induce neural differentiation. The morphology of both ES and EC cells reflected neural differentiation, with neuron progenitors predominating in the ES cells and astroglial cells in the EC cells (Fig. 1).
FIG. 1. Morphology of T3-induced neural differentiation of ES and EC cells. Differentiation was initiated by plating embryoid bodies to 10-cm dish without gelatin coating in conditioned DMEM/F-12 media supplemented with 1 nM T3. Wild-type ES and EC cells were differentiated for 3 d. Control cells did not receive hormone treatment. The differentiated cells were photographed with a light contrast microscope. Magnification, x250.
The differentiation markers for neuron progenitor (nestin) and glial cells (glial fibrillary acidic protein, GFAP) were analyzed. Nestin mRNA was induced 10-fold in ES cells but only 2.5-fold in EC cells (Fig. 2A), consistent with the morphology shown (Fig. 1). GFAP mRNA was not expressed in T3-treated ES cells but induced 5-fold in T3-treated EC cells (Fig. 2B). This pattern of gene expression is consistent with the morphological observations and demonstrates T3-induced astroglial differentiation in EC cells and neuron progenitors in ES cells.
FIG. 2. Nestin and GFAP mRNA detection in differentiated ES and EC cells. The differentiation conditions were the same as those described in Fig. 1. Total RNA was isolated from differentiated ES and EC cells and reverse transcribed. The mRNA levels of nestin (A) and GFAP (B) were PCR amplified with 33P-labled primers, resolved in a 5% polyacrylamide gel, and quantified by phosphor-imager. The data are shown as mean value ± SE of three separate experiments. *, P < 0.001 compared with control.
We analyzed the endogenous expression of the previously described T3 target genes, RC3 and CamKIV in ES and EC cells. In T3-treated ES cells, RC3 was induced 21-fold and CamKIV 4.7-fold (Fig. 3). In contrast, T3 treatment of EC cells did not result in significant induction of RC3 or CamKIV expression compared with control cells.
FIG. 3. Differential mRNA expression of neurogranin (RC3) and CaMKIV in ES and EC cells. ES and EC cells were differentiated for 3 d with T3 treatment. Total RNA was isolated and used for RT-PCR to determine the mRNA levels of RC3 (A) and CaMKIV mRNA (B). PCR amplification was done in 25 cycles with 2 μl cDNA. The expression level of ?-actin mRNA was quantified using diluted cDNA (1:5). Quantification was performed by phosphor-imager and controlled for the level of ?-actin mRNA. The data shown are from three separate experiments. **, P < 0.001 compared with control; *, P < 0.05 compared with control.
EC cells are derived from a carcinoma, which may have reduced TR1 expression, as has been reported in PC12 cells (33). To test this hypothesis, TR1 mRNA content was determined by RT-PCR and was reduced 5.7-fold in EC cells compared with ES cells (Fig. 4). We directly analyzed functional TR protein by T3-binding activity in nuclear extracts from EC and ES cells. The maximum binding capacity of EC cells was reduced 35% compared with ES cells. The affinity for T3 was 1.3 nM in nuclear extracts from EC cells and 1.8 nM in nuclear extracts from ES cells (Fig. 5).
FIG. 4. TR1 mRNA expression in ES and EC cells. Cell differentiation and PCR analysis were the same as those described in Fig. 3.
FIG. 5. [125I]T3 binding to nuclear extracts of ES and EC cells and Scatchard analysis. Nuclear extracts of ES cells (15 μg) (A) and EC cells (15 μg) (B) were incubated with various concentrations of [125I]T3 either in the presence or absence of a 500-fold molar excess of cold T3 for 4 h at 4 C. The bound and free T3 was separated and radioactivity counted (as described in Materials and Methods). The assays were done in triplicates at each concentration, and the mean of three determinations with SE is shown.
A number of factors, including overexpression of COUPTF1, have been shown to modulate T3 responsiveness in developing neurons (37). To determine whether the reduced TR content in EC cells was limiting for the T3 response, we performed a transient transfection assay. A CamKIV promoter reporter construct (37, 38) was transfected into ES and EC cells. In ES cells, 5-fold T3 induction was observed; however, in EC cells, no T3 induction was detected. Cotransfection of a TR1 expression vector in EC cells restored 4.3-fold T3 induction (Fig. 6). These data indicate that the reduced TR content in EC cells is likely to be responsible for reduced T3-induced gene expression. EC cells also provide a neural differentiation model that is not T3 responsive to compare with wild-type ES cells and identify T3 target genes.
FIG. 6. T3 induction of rat CaMKIV-1060 reporter construct in ES and EC cells. The CaMKIV-1060 CAT reporter construct (0.2 μg) was transfected into ES cells and EC cells with T3 treatment and with and without cotransfection of a TR expression vector (0.2 μg). CAT activity was determined 48 h after transfection and normalized for protein concentration used in the assay. The assays were performed in duplicate. *, P < 0.05.
Effects of wild-type TR and T3 treatment on gene expression
Despite the profound effects of T3 on development and the clear requirement for expression of TR, relatively few T3 gene targets have been identified. We previously developed an ES cell line by introducing a mutation (P398H) into the TR gene. The morphology of neural differentiation of TR P398H ES cells was indistinguishable from neural differentiated wild-type ES cells. The T3 target genes (RC3 and CaMKIV), however, were significantly down-regulated in TR P398H ES cells (7).
We used differentiated TR P398H ES cells to compare with differentiated wild-type ES cells, profile gene expression, and identify T3 target genes. These cell lines differ only by whether functional TR was present. EC cells, although different in origin from ES cells, are functionally TR deficient and provide another comparison with wild-type ES cells. We treated cells with T3 for 3 d under differentiation conditions. The gene expression profiles of the three cell models (ES, P398H ES, and EC) and undifferentiated control cells were analyzed using DNA microarray (MU74Av2 gene chip, Affymitrx). The gene chip (MU74Av2) contains 12,489 mouse genes. The comparison analysis used Microarray Suite, and wild-type ES cells (undifferentiated) were used as a baseline to filter out the genes that were not present and unchanged when compared with T3-treated wild-type ES cells. Similarly, T3-treated wild-type ES cells were used as the baseline when compared with P398H ES cells. Data analysis showed that approximately 45% of genes in the gene chip were not present. In the present genes, in wild-type ES cells, T3-induced neural differentiation resulted in altered mRNA expression of 610 known genes (26.2% of those tested). Using a threshold of greater than 5-fold stimulation or repression, 88 genes were identified with 5- to 26-fold up-regulation or 5- to 18-fold down-regulation. The gene with the greatest induction was myosin light chain (MLC), a known T3 target gene, with 26-fold induction.
Because differentiation alone alters gene expression, many of the genes identified are likely to be only indirectly regulated by T3. We, therefore, compared gene expression in wild-type ES cells with expression in TR P398H ES cells, a cell line with a dominant-negative TR P398H mutation known to impair T3-dependent gene expression (7, 18), and EC cells with low nuclear TR content. The results from DNA microarray analysis of two mutant ES cell lines derived from two individual TR P398H ES clones showed that 75–79% of known genes had no detectable change compared with wild-type ES cells; 20–24% of known genes had modest changes and 0.7–1% of known genes (45–60 genes) were significantly up- or down-regulated (>5-fold). The expression of most unknown genes, 85–90%, was unchanged, and 0.2–0.3% (12–18 genes) had equal or greater than 5-fold change in mRNA level. These results indicate that T3 treatment influenced a wide range of genes in a stem cell differentiation model (>20% of known genes) either directly or indirectly. However, the number of genes directly targeted by T3 was small. These findings are similar to studies of T3-dependent gene expression in other T3 target tissues, such as liver (35, 36). The T3 treatment with 3-d duration was sufficient for detecting the changes in early differentiation events.
Identification of T3 target genes
Based on the results of gene profiles in wild-type ES, compared with the TR P398H ES and EC cells, we focused on 40 putative T3-responsive genes to characterize endogenous expression. We characterized mRNA expression by quantitative real-time PCR in wild-type ES cells, TR P398H ES cells, and EC cells. Consistent endogenous T3 regulation was confirmed in five genes, with response in wild-type ES, but not TR P398H ES cells, and EC cells. The regulated genes included neurexophilin 2, SPNR, synatotagmin II, KBP, and PSMA. PSMA mRNA was negatively regulated by T3, and the other genes were induced. In wild-type ES cells, T3 stimulated neurexophilin 2 mRNA 22-fold, SPNR 20-fold, synaptotagmin II 13-fold, and KBP 5.3-fold compared with baseline (untreated ES cells). In TR P398H ES cells, T3 also stimulated mRNA expression of these genes, but the extent was significantly lower compared with wild-type ES cells (Table 2). In P398H ES cells, neurexophilin 2 mRNA was stimulated 8.3-fold, SPNR 6.7-fold, synaptotagmin II 7.2-fold, and KBP 3-fold. PSMA expression in response to T3 was reduced 90% in wild-type ES cells and 41% in P398H ES cells, compared with baseline. The target genes were not detected, or they had less than 2-fold induction, in T3-treated EC cells compared with control EC cells (Table 2).
TABLE 2. mRNA expression of potential T3-responsive genes (arbitrary units)
Endogenous gene expression in both in vivo and in vitro models
The T3-responsive genes identified in stem cells are endogenously expressed in a range of tissues, including brain, liver, and testes. T3 can regulate genes at the transcriptional and posttranscriptional levels. T1-tubulin, for example, is T3 inducible but does not have a thyroid hormone response element in the gene. It is thought to be regulated by the influence of T3 on mRNA stability (39). We used cell lines from the tissues relevant to the identified genes and tissue samples from hypothyroid mice and mice with a TR P398H mutation. Neurexophilin 2, a gene dominantly expressed in mouse liver (40, 41), was examined in liver-derived HepG2 cells. The cells were grown in serum-free medium and treated with 3 nM T3 for 24 h. The mRNA level of neurexophilin 2 was stimulated 3.6-fold (P < 0.05) (Fig. 7A). Both SPNR and KBP are expressed in testis and were examined in a testis cell line, TM4. After 4 h of T3 treatment, the mRNA levels of SPNR and KBP were stimulated 2.3-fold and 1.2-fold compared with the levels without T3 (Fig. 7B).
FIG. 7. T3 responsiveness of neurexophilin 2, SPNR, and KBP in cell culture. HepG2 and TM4 cells were gown in serum-free media for 24 h before T3 treatment. T3 was added to culture at final concentration 3 nM. The treatment for HepG2 cells was 24 h and for TM4 cells was 4 h. mRNA levels were determined by quantitative real-time PCR and normalized for ?-actin mRNA levels. Nerexophilin 2 (A) was determined using HepG2 cells and SPNR (B) and KBP (C) using TM4 cells. The assays were performed in triplicate. *, P < 0.05.
We further examined mRNA expression in euthyroid, hypothyroid, and TR P398H mutant mice. Mice were made hypothyroid by feeding a PTU-containing iodine-deficient diet for 8.5 wk. We confirmed hypothyroidism by analyzing mRNA levels of well-known T3-targeted genes, MLC, and CamKIV. The mRNA level of MLC in heart was reduced 11-fold and CaMKIV expression in testes reduced 3.1-fold (Fig. 8). The mRNA expression of the T3 target genes was determined using tissues isolated from three different animal models: wild-type, hypothyroid, and TR P398H mutant mice. In liver, neurexophilin 2 was reduced 3.7-fold in P398H mice and 2.7-fold in hypothyroid mice compared with euthyroid mice (Fig. 9A). In testis, SPNR was reduced 5.4-fold in the P398H mice and 2.2-fold in hypothyroid mice (Fig. 9B); KBP was reduced 2.3-fold in testes of the P398H mice and 3.3-fold in hypothyroid mice compared with wild-type mice (Fig. 9C). Levels of PSMA mRNA were increased 2-fold in P398H mice and 3-fold in hypothyroid mice compared with wild-type mice (Fig. 9D). Synaptotagmin II mRNA had only modest reduction in the brain of hypothyroid or mutant TR mice compared with wild-type mice (data not shown).
FIG. 8. mRNA levels of known T3-responsive genes in hypothyroid mice. Hypothyroidism in mice was induced by feeding iodine-deficient diet supplement with 0.15% PTU for 8.5 wk. Total RNA was isolated from heart and testis of wild-type (n = 4), P398H (n = 4), and hypothyroid male mice (n = 3). RNA was treated with DNase I before RT. The level of mRNA expression was determined using quantitative real-time PCR and normalized for ?-actin mRNA levels of MLC (A) in heart and CaMKIV (B) in testis. The data shown are the mean values ± SE. Statistical comparison is between the wild-type and P398H mice or wild-type and hypothyroid mice: *, P < 0.05; **, P < 0.01.
FIG. 9. In vivo T3 response of identified genes using animal models (described in Fig. 8). Total RNA was isolated from liver and testis. The conditions for quantitative real-time PCR were as described (Fig. 8). Neurexophilin 2 (A) was analyzed using RNA from liver; SPNR (B), KBP (C), and PSMA (D) were from testis. The assays were done in triplicate. *, P < 0.05; **, P < 0.01.
We tested the response of these genes to T3 treatment in EC cells. SPNR, PSMA, and KBP were not detected by quantitative real-time PCR at baseline or with T3 treatment. Neurexophilin 2 and synaptotagmin II mRNA were detected at baseline but did not significantly change with T3 treatment, consistent with reduced TR content (data not shown).
These data confirm that the five genes, including four stimulated by T3 and one repressed, are regulated by T3 and require wild-type TR for the response. T3 regulation was confirmed for all genes in ES or other in vitro cell models and for four of the five genes in in vivo mouse models. In the case of synaptotagmin II, it is likely that the T3 response is present developmentally but not in the adult brain. This will require further investigation.
Discussion
We have demonstrated that ES and EC cells have a differential response to T3-induced differentiation with respect to cell morphology, endogenous gene expression, and T3-dependent gene expression. The restoration of the T3 response to EC cells with the addition of TR indicates that reduced nuclear TR content contributes to the observed reduced T3-induced gene expression. In addition to the reduction in nuclear TR, however, altered expression of TR coactivators and corepressors may also play a role in reduced T3-dependent gene expression in EC cells. Reduced TR expression and expression of mutant TRs that do not bind T3 have been reported in a variety of tumors and tumor cell lines (33, 42, 43). In PC12 neural stem cells, TR1 and T3 are critical for maintaining nerve growth factor-dependent neural differentiation, expression of neuronal genes, and normal neural morphology (39). This supports the role of TR in maintaining differentiated function in stem cells. The morphological phenotype of EC cell differentiation and relative sensitivity to T3 and RA is likely because of a number of factors in addition to TR content, related to the neoplastic tissue origin of these cells.
The observation that expression of 26% of the mRNAs from the genes studied are altered as a consequence of T3-induced differentiation of ES cells indicates a large fraction of genes required for this developmental program. The value of comparison with the expression profile in the TR mutant ES cells and EC cells was to determine those genes most likely to be directly mediated by TR. These comparative models, and a relatively high threshold for significant induction, narrowed the number of genes studied.
We identified five genes (neurexophilin 2, SPNR, KBP, PSMA, and synaptotagmin II) that meet the criteria for a T3/TR-mediated response that have not previously been shown to be regulated by T3. We demonstrated that the expression of neurexophilin 2, SPNR, KBP, and PSMA was reduced in TR P398H mice (14) and hypothyroid mice but induced in euthyroid mice. The synaptotagmin II gene was T3 responsive in ES cells but not in brain from animal models. It is possible that synaptotagmin II transiently responds to T3 in an early developmental stage but not in adult tissue.
The physiological function of these T3-responsive genes is not fully understood. SPNR is a microtube-associated RNA-binding protein, and its mRNA is expressed at high levels in the testis, exclusively in postmeiotic germ cells (44, 45). Low-level expression of SPNR was also seen in ovarian and brain tissue. Mice deficient in SPNR have a high rate of mortality and abnormalities in neurological development, spermatogenesis, and sperm morphology (45). Indeed, SPNR was reduced 5.4-fold in testis of the P398H mice, and the male mice had a significant reduction in fertility (18). Other TR mutant mice (PV and R384C) have been reported to have an increased mortality rate and reduced fertility (20, 21). KBP, a serine proteinase inhibitor, binds to and inhibits tissue kallikrein activity and is found in endothelial and smooth muscle cells of blood vessels (46, 47). Neurexophilin 2 is a member of the neurexophilin family, the ligand for neurexins, and is exclusively expressed in mouse liver (41).
PSMA gene expression has been linked to recurrence of prostate cancer (48). Down-regulation of PSMA gene expression by thyroid hormone may be associated with an effect of differentiating prostate cancer. A number of studies have looked at the influence of thyroid hormone on prostate cancer growth and differentiation. Growth and differentiation of the LNCaP prostate cancer cell line has been associated with T3 treatment (49). Thyroid hormone directly stimulates the expression of the prostate-specific antigen, and a response element has been identified (50).
Thyroid hormone regulation of development requires stimulation of a number of genes in a time- and location-specific manner. The majority of T3-regulated genes that have been identified and studied are in adult tissues, which are much more accessible for study. Models of thyroid hormone action in mammalian development have been difficult to develop, because thyroid hormone has such a broad range of actions. The models we have used permit the identification of those genes that require T3 and TR mediation for their effects. The genes identified in this way were all regulated by T3 in vitro and, in all but one gene, in the in vivo system. This approach holds promise for further understanding of T3 regulation of development.
References
Oppenheimer JH, Schwartz HL 1997 Molecular basis of thyroid hormone-dependent brain development. Endocr Rev 18:462–475
Lazarus JH 1999 Thyroid hormones and neurodevelopment. Clin Endocrinol (Oxf) 50:147–148
Lim W, Nguyen NH, Yang HY, Scanlan TS, Furlow JD 2002 A thyroid hormone antagonist that inhibits thyroid hormone action in vivo. J Biol Chem 277:35664–35670
Furlow JD, Yang HY, Hsu M, Lim W, Ermio DJ, Chiellini G, Scanlan TS 2004 Induction of larval tissue resorption in Xenopus laevis tadpoles by the thyroid hormone receptor agonist GC-1. J Biol Chem 279:26555–26562
Lee LR, Mortensen RM, Larson CA, Brent GA 1994 Thyroid hormone receptor- inhibits retinoic acid-responsive gene expression and modulates retinoic acid-stimulated neural differentiation in mouse embryonic stem cells. Mol Endocrinol 8:746–756
Münoz A, Wrighton C, Seliger B, Bernal J, Beug H 1993 Thyroid hormone receptor/c-erbA: control of commitment and differentiation in the neuronal/chromaffin progenitor line PC12. J Cell Biol 121:23–38
Liu YY, Tachiki KH, Brent GA 2002 Targeted thyroid hormone receptor gene dominant negative mutation (P398H) selectively impairs gene expression in differentiated embryonic stem cells. Endocrinology 143:2664–2672
Forrest D, Hallbook F, Persson H, Vennstrom B 1991 Distinct functions for thyroid hormone receptors and ? in brain development indicated by differential expression of receptor genes. EMBO J 10:269–275
Porterfield SP, Hendrich CE 1993 The role of thyroid hormones in prenatal and neonatal neurological development: current perspectives. Endocr Rev 14:94–106
Dowling AL, Zoeller RT 2000 Thyroid hormone of maternal origin regulates the expression of RC3/neurogranin mRNA in the fetal rat brain. Brain Res Mol Brain Res 82:126–132
Bernal J, Guadano-Ferraz A, Morte B 2003 Perspectives in the study of thyroid hormone action on brain development and function. Thyroid 13:1005–1012
Forres D, Erway LC, Ng L, Altschuler R, Curran T 1996 Thyroid hormone receptor ? is essential for development of auditory function. Nat Genet 13:354–357
Forrest D, Hanebuth E, Smeyne RJ, Everds N, Stewart CL, Wehner JM, Curran T 1996 Recessive resistance to thyroid hormone in mice lacking thyroid hormone receptor ?: evidence for tissue-specific modulation of receptor function. EMBO J 15:3006–3015
Brent GA 2000 Tissue-specific actions of thyroid hormone: insights from animal models. Rev Endocr Metab Disord 1:27–33
Fraichard A, Chassande O, Plateroti M, Roux JP, Trouillas J, Dehay C, Legrand C, Gauthier K, Kedinger M, Malaval L, Rousset B, Samarut J 1997 The T3R gene encoding a thyroid hormone receptor is essential for post-natal development and thyroid hormone production. EMBO J 16:4412–4420
Gaüthier K, Plateroti M, Harvey CB, Williams GR, Weiss RE, Refetoff S, Willott JF, Sundin V, Roux JP, Malaval L, Hara M, Samarut J, Chassande O 2001 Genetic analysis reveals different functions for the products of the thyroid hormone receptor locus. Mol Cell Biol 21:4748–4760
Wikstrom L, Johansson C, Salto C, Barlow C, Barros AC, Baas F, Forrest D, Thoren P, Vennstrom B 1998 Abnormal heart rate and body temperature in mice lacking thyroid hormone receptor . EMBO J 17:455–461
Liu Y-Y, Schultz JJ, Brent GA 2003 A thyroid hormone receptor gene mutation (P398H) is associated with visceral adiposity and impaired catecholamine-stimulated lipolysis in mice. J Biol Chem 278:38913–38920
Itoh Y, Esaki T, Kaneshige M, Suzuki H, Cook M, Sokoloff L, Cheng SY, Nunez J 2001 Brain glucose utilization in mice with a targeted mutation in the thyroid hormone or ? receptor gene. Proc Natl Acad Sci USA 98:9913–9918
Kaneshige M, Suzuki H, Kaneshige K, Cheng J, Wimbrow H, Barlow C, Willingham MC, Cheng S 2001 A targeted dominant negative mutation of the thyroid hormone 1 receptor causes increased mortality, infertility, and dwarfism in mice. Proc Natl Acad Sci USA 98:15095–15100
Tinnikov A, Nordstrom K, Thoren P, Kindblom JM, Malin S, Rozell B, Adams M, Rajanayagam O, Pettersson S, Ohlsson C, Chatterjee K, Vennstrom B 2002 Retardation of post-natal development caused by a negatively acting thyroid hormone receptor 1. EMBO J 21:5079–5087
Keller GM 1995 In vitro differentiation of embryonic stem cells. Curr Opin Cell Boil 7:862–869
Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M, Gepstein A, Livne E, Binah O, Itsk-vitz-Eldor J, Gesptein L 2001 Human embryonic stem cells can differentiate into myocytes with structural and functional properties of cardiomyocytes. J Clin Invest 108:407–414
Balasubramaniyan V, de Haas AH, Bakels R, Koper A, Boddeke HW, Copray JC 2004 Functionally deficient neuronal differentiation of mouse embryonic neural stem cells in vitro. Neurosci Res 49:261–265
Pera MF, Filipczyk AA, Hawes SM, Laslett AL 2003 Isolation, characterization, and differentiation of human embryonic stem cells. Methods Enzymol 365:429–446
Heins N, Englund MC, Sjoblom C, Dahl U, Tonning A, Bergh C, Lindahl A, Hanson C, Semb H 2004 Derivation, characterization, and differentiation of human embryonic stem cells. Stem Cells 22:367–376
Kania G, Blyszczuk P, Czyz J, Navarrete-Santos A, Wobus AM 2003 Differentiation of mouse embryonic stem cells into pancreatic and hepatic cells. Methods Enzymol 365:287–303
Roberston EJ 1997 Derivation and maintenance of embryonic stem cell cultures. Methods Mol Biol 75:173–184
Strickland S, Smith KK, Marotti KR 1980 Hormonal induction of differentiation in teratocarcinoma stem cells: generation of parietal endoderm by retinoic acid and dibutyryl cAMP. Cell 21:347–355
Clifford J, Chiba H, Sobieszczuk D, Metzger D, Chambon P 1996 RXR-null F9 embryonal carcinoma cells are resistant to the differentiation, anti-proliferative and apoptotic effects of retinoids. EMBO J 15:4142–4155
Greene LA, Tischler AS 1976 Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc Natl Acad Sci USA 73:424–2428
Weber E, Jilling T, Kirk KL 1996 Distinct functional properties of Rab3A and Rab3B in PC12 neuroendocrine cells. J Biol Chem 271:6963–6971
Munoz A, Wrighton C, Seliger B, Bernal J, Beug H 1993 Thyroid hormone receptor/c-erbA: control of commitment and differentiation in the neuronal/chromaffin progenitor line PC12. J Cell Biol 121:423–438
Bain G, Ray WJ, Yao M, Gottlieb DI 1994 From embryonal carcinoma cells to neurons: the P19 pathway. Bioessays 6:343–348
Flores-Morales A, Gullberg H, Fernandez L, Stahlberg N, Lee NH, Vennstrom B, Norstedt G 2002 Patterns of liver gene expression governed by TR?. Mol Endocrinol 16:1257–1268
Yen PM, Feng X, Flamant F, Chen Y, Walker RL, Weiss RE, Chassande O, Samarut J, Refetoff S, Meltzer PS 2003 Effects of ligand and thyroid hormone receptor isoforms on hepatic gene expression profiles of thyroid hormone receptor knockout mice. EMBO Rep 4:581–587
Liu YY, Brent GA 2002 A complex deoxyribonucleic acid response element in the rat Ca2+/calmodulin-dependent protein kinase IV gene 5'-flanking region mediates thyroid hormone induction and chicken ovalbumin upstream promoter transcription factor 1 repression. Mol Endocrinol 16:2439–2451
Means AR, Ribar TJ, Kane CD, Hook SS, Anderson KA 1997 Regulation and properties of the rat Ca2+/calmodulin-dependent protein kinase IV gene and its protein products. Recent Prog Horm Res 52:389–407
Lorenao PI, Menard C, Miller FD, Bernal J 2002 Thyroid hormone-dependent regulation of T1 -tubulin during brain development. Mol Cell Neurosci 19:333–343
Missler M, Sudhof TC 1998 Neurexophilins form a conserved family of neuropeptide-like glycoproteins. J Neurosci 18:3630–3638
Petrenko AG, Ullrich B, Missler M, Krasnoperov V, Rosahl TW, Sudhof TC 1996 Structure and evolution of neurexophilin. J Neurosci 16:4360–4369
Zhang XY, Kaneshige M, Kamiya Y, Kaneshige K, McPhie P, Cheng SY 2002 Differential expression of thyroid hormone receptor isoforms dictates the dominant negative activity of mutant ? receptor. Mol Endocrinol 16:2077–2092
Gittoes NJ, McCabe CJ, Verhaeg J, Sheppard MC, Franklyn JA 1998 An abnormality of thyroid hormone receptor expression may explain abnormal thyrotropin production in thyrotropin-secreting pituitary tumors. Thyroid 8:9–14
Schumacher JM, Lee K, Edelhoff S, Braun RE 1995 Spnr, a murine RNA-binding protein that is localized to cytoplasmic microtubules. J Cell Biol 129:1023–1032
Pires-daSilva A, Nayernia K, Engel W, Torres M, Stoykova A, Chowdhury K, Gruss P 2001 Mice deficient for spermatid perinuclear RNA-binding protein show neurologic, spermatogenic, and sperm morphological abnormalities. Dev Biol 233:319–328
Chao J, Miao RQ, Chen V, Chen LM, Chao L 2001 Novel roles of kallistatin, a specific tissue kallikrein inhibitor, in vascular remodeling. Biol Chem 382:15–21
Evans BA, Drinkwater CC, Richards RI 1987 Mouse glandular kallikrein genes. Structure and partial sequence analysis of the kallikrein gene locus. J Biol Chem 262:8027–8034
Ross JS, Sheehan CE, Fisher HA, Kaufman Jr RP, Kaur P, Gray K, Webb I, Gray GS, Mosher R, Kallakury BV 2003 Correlation of primary tumor prostate-specific membrane antigen expression with disease recurrence in prostate cancer. Clin Cancer Res 9:6357–6362
Esquenet M, Swinnen JV, Heyns W, Verhoeven G 1995 Triiodothyronine modulates growth, secretory function and androgen receptor concentration in the prostatic carcinoma cell line LNCaP. Mol Cell Endocrinol 109:105–111
Zhu W, Young CY 2001 Androgen-dependent transcriptional regulation of the prostate-specific antigen gene by thyroid hormone 3,5,3'-L-triiodothyronine. J Androl 22:136–141(Yan-Yun Liu and Gregory A)