V. Annexe : Identification de nouveaux gènes régulés par les hormones thyroïdiennes dans des cultures neuronales de cerveau de rat.

Table des matières

Dans une première pour étudier les mécanismes moléculaires responsables des effets des HT sur le développement du cerveau, nous avons utilisé la méthode des puces d’ADN permettant l’identification de gènes précoces régulés par la triiodothyronine (T3) dans des cultures primaires de neurones (traitement de 3 h). Cette méthode nous a permis d’identifier trois gènes, le facteur de transcription basique (BTEB), la glycoprotéine des pores nucléaires (P62) et la protéine de morphogénétique des os (BMP-4) régulés positivement par la T3 ainsi qu’un autre gène, le gène neuronal d’induction de l’apoptose (DP5), régulé négativement par cette même hormone. Nous avons aussi confirmé que tous ces gènes étaient régulés de la même façon par l'hormone thyroïdienne dans le cerveau en développement. Ces résultats contribuent à augmenter notre compréhension des voies de signalisation régulées par les hormones thyroïdiennes et ouvre de nouvelle voies pour l’étude des mécanismes moléculaires des hormones thyroïdiennes sur le cerveau en développement.

Cet article a été rédigé par le Dr Jack Puymirat et moi-même.

Accepté dans Neuroreport

Identification of New Thyroid Hormone-Regulated Genes in Rat Brain Neuronal Cultures.

Julie Martel, Christelle Cayrou and Jack Puymirat1

Laboratory of Human Genetics, Laval University Medical Research Center, CHUQ, Pavillon CHUL. 2705 Laurier Bd, Sainte-Foy, Quebec, Canada G1V 4G2

1Corresponding author:Dr. Jack Puymirat

Laboratory of Human Genetics. Laval University Medical Research Centre. CHUQ, Pavillon CHUL. Ste-Foy, Quebec, Canada, G1V 7P4.

Tel: 418-654-2186Fax: 418-654-2207

Email: jack.puymirat@crchul.ulaval.ca


As a first approach to study the molecular mechanisms that underlie the effects of thyroid hormones on the developing brain, we used a cDNA microarray technology to identify early thyroid hormone-regulated genes in brain neuronal cultures treated with Triiodothyronine (T3) for 3 hrs. We identified three genes that were up-regulated by T3 [basic transcription element-binding protein (BTEB), nuclear pore glycoprotein (P62) and bone morphogenetic protein-4 (BMP-4)] and one that was down-regulated, the neuronal apoptosis-inducing gene (DP5). We confirmed that these genes were also regulated by the thyroid state in the developing brain. Our findings enrich our knowledge of signaling pathways regulated by thyroid hormones and open new avenues for studying the molecular mechanisms of thyroid hormones in the developing brain.

Key Words: neurons, brain, thyroid hormones, gene expression, microarray, basic transcription element-binding protein (BTEB), nuclear pore glycoprotein (P62), bone morphogenetic protein-4 (BMP-4), neuronal apoptosis-inducing gene (DP5).


Thyroid hormones (TH) are major epigenetic factors for the normal development of the brain where the absence of TH at birth affects neuronal survival, proliferation and differentiation [1]. Their action in the developing brain are mainly mediated through two majors ligand-activated thyroid receptors (TRs) isoforms, TRa1 and TRb1 [2,3,4]. Although many TH-regulated genes have been identified [5], direct TH target genes that are crucial for normal development are not yet identified. The limited success in identifying early immediate TH-regulated genes may result from: i) the complexity of the rodent brain, which is composed of several different cell types, each of them maturing and becoming responsive to TH at different developmental stages; ii) the long period between induction of hypothyroidism and gene screening. In the few attempts that have been made, induction of neonatal hypothyroidism has extended from days to weeks before screening. Thus, early-immediate T3-response genes could not have been isolated in these screening attempts.

Primary neuronal cultures are powerful tools to study hormonal gene regulation. Primary neurons could be treated with TH for short periods, and therefore could be used to identify early-immediate TH-regulated genes.

We have used an oligonucleotide microarray to identify early TH-regulated genes in primary neuronal cultures treated with T3 for 3 hrs. We have identified four TH target genes, including three newly identified TH-regulated-genes. Three of them were up-regulated and one other was down-regulated.

Materials and Methods

Primary Cerebral Hemisphere Neuronal Cultures.

Cerebral hemispheres neuronal cultures were prepared from 15-day old rat embryos. Cells were dissociated and plated onto gelatin/L-poly-Lysine-coated coverslips at a density of 6 x 106 cells/ 100 mm diameter dish. Cells were grown in serum-free medium in an incubator at 37oC with 5% CO2 as previously described [6]. More than 80% of the cells were neurofilament-positive, indicating their neuronal phenotype. For experiments with TH, 30 nM T3 was added to the cultures 3 hrs before analysis. The dose of T3 used in this study was based on a previous dose-response study [6].


Pregnant Sprague-Dawley rats were obtained from Charles River (Wilmington, MA USA). Rat fetuses and neonates were rendered hypothyroid by continuous administration of goitrogen propylthiouracil (PTU; Sigma-Aldrich, Oakville, ON Canada) at a concentration of 0.005% in the mother’s drinking water from the 16th day of gestation and maintained continuing throughout the period of the study (postnatal day 17) as previously described [7]. Two groups of hypothyroid pups (n = 6 per treatment) received daily subcutaneous injections of either saline (PTU) or 0.5 mg of T4/100 g body weight (PTU + T4) since birth to the time of sacrifice (24 h after the last injection). To verify the efficacy of the treatments in altering thyroid state, plasma T3 levels were determined in euthyroid, hypothyroid, and T3-replaced hypothyroid rat pups using an Elisa Immuno Assay (EIA, Bayer Corporation, Tarrytown, NY USA). Animal care was in accordance with the institutional guidelines set by the Animal Care and Use Committee of the CHUL Research Center.

cDNA Microarray

Total RNA was isolated from 10-day old primary cultures using TRIZOL reagent (Life Technologies, Burlington, ON Canada). RNA concentrations were determined by measuring the A260 and by blotting the RNAs on a nylon membrane and hybridizing it a with cyclophilin probe, an unrelated gene constitutively expressed in cells [8]. The quality of the RNA was assessed by hybridization with a rat BTEB probe, a gene known to be up-regulated by T3 in primary neurons [9].

Twenty (20) mg of total RNA was converted to cDNA with biotin-labeled-dCTP and -dUTP (Enzo diagnostics). Labeled probes were hybridized to oligonucleotide microarray (Hu6800 GeneChip arrays) competitively at 45°C for 16 hrs. This array has 8000 probe sets corresponding to mouse cDNAs and ESTs. After several washing procedures, arrays were stained with streptavidin-phycoerythrin (Molecular Probes) and washed again. The chip was scanned with GeneArray Scanner (HP and Affymetrix). The average difference in expression on the chip was computed using Affymetrix GeneChip Analysis Suite (version 3.3) with default parameters.

Genes with ratio a number above 2-fold induction or repression over steady state expression in all experiments were selected and subjected to further analysis.

Northern blot analysis

Total RNA was prepared with TRIZOL reagent (Life Technologies, Burlington, ON Canada) from either 10-day old primary neuronal cultures grown in presence or absence of T3 (30 nM) during 3 hrs or from the brain of 17-day old postnatalanimals. Total RNA (20 mg/lane) was separated by electrophoresis in a 1% formaldehyde-agarose gel and transferred onto a nylon membrane. Blots were hybridized with [32P]-dCTP-labelled cDNA probes by random priming and washed with 0.1 x SSC, 0.5% SDS at 67oC. Normalization was done by hybridization with a 32P-labeled cyclophilin cDNA probe. Quantification of relative RNA levels from the autoradiograms were determined by densitometry on an AlphaImager 1200 version 3.3b (alpha Innotech Corporation, St Leandro, CA USA)


To investigate the effects of TH on gene expression in the developing brain, RNAs were prepared from untreated neuronal cells or cells treated with T3 for 3hrs, labeled with fluorescent dye and hybridized with the oligonucleotide microarray as described in materials and methods.

We sampled 8000 genes on the oligonucleotide microarray, which represents approximately 25% of the expressed genes in the brain, assuming that the brain express about 30,000 genes. We found that three genes [the basic transcription element-binding protein (BTEB), the bone morphogenetic protein-4 (BMP-4) and the nuclear pore glycoprotein (P62)] displayed greater than 2-fold increase in expression levels after T3 treatment whereas only one other gene, [the neuronal apoptosis-inducing gene (DP5)], displayed more than a 2-fold decrease in gene expression after signal enhancement (Table 1).

The changes in mRNA levels of novel T3-target genes on the microarray were confirmed by Northern blot analysis in primary neuronal cultures (Fig 1A). The levels of BTEB-, P62- and BMP-4-mRNA were increased significantly by 4.4-, 1.4- and 1.8-fold respectively after a 3 hrs treatment with T3. In contrast, the level of DP5 mRNA was decreased by 1.6-fold. To determine whether these genes were also regulated by the thyroid state in vivo, we next analyzed their levels of expression in 17-day old brain derived from euthyroid (Control, C), hypothyroid (PTU) and T4-treated hypothyroid (PTU + T4) animals (Fig 1B). Induction of hypothyroidism by PTU treatment reduced BTEB-, P62- and BMP-4-mRNA levels relative to the euthyroid controls (2.1-, 1.3- and 2-fold respectively) and increased DP5 mRNA by 2-fold. Conversely, treatment of hypothyroid animals with T4 significantly increased BTEB-, P62- and BMP4-mRNAs relative to untreated animals (1.5-, 1.3- and 1.9-fold respectively) and decrease DP5 mRNA 1.8-fold.


Using a cDNA microarray approach, we have identified three new TH-regulated genes. It has been previously described that BTEB is regulated by TH in the developing rat brain [7]. It is of note that we were unable to detect certain genes that have been previously shown to be regulated by TH (for rev., see ref. 5) : i) The myelin basic protein (MBP), proteolipid protein (PLP), myelin associated glycoprotein (MAG), 2', 3'-cyclic nucleotide 3'-phosphodiesterase (CNP) and metallothreonin genes are regulated by T3 in other nerve cell types (oligodendrocytes or astrocytes) which were not present in our cultures , ii) other genes like Purkinje cell protein-2 (PCP2) have been shown to be regulated in a specific brain area (cerebellum) which was not present in our cultures, iii) Finally, late TH-regulated genes could not have been isolated in this screening.

Among the newly identified T3 target genes, the neuronal apoptosis-inducing gene (DP5) has been shown to be involved in neuronal apoptosis in vivo. DP5 is induced during neuronal apoptosis observed in rat sympathetic neuron cultures deprived of NGF [10] or in cortical neurons exposed to Amyloid b protein [11]. Moreover, overexpression of DP5 in cultured neurons is sufficient to induce apoptosis [10]. One should note that deficiency of TH at birth induces cerebellar granule cell apoptosis in the internal granular layer of the cerebellum [12]. It is therefore possible that the up-regulation of DP5 in the hypothyroid state may participate in granule cell apoptosis. On the other hand, TH have been proposed to act as a timing clock by pushing neuroblasts out of their mitotic phase. The identification of BMP-4, which has been shown to push neuroblasts out of their mitotic phase [13], raised the hypothesis that BMP-4 may be a mediator in the control of the proliferative phase of neuroblasts by TH during brain development. Finally, P62 is a member of the nuclear pore complex glycoproteins that are thought to play an important role in the transport of regulatory proteins across the nuclear envelope [14]. Some data also suggests that TH may regulate the transport of proteins though the nuclear envelope as is the case for the B-cell translocation gene 1 (BTG1) [15], a member of an antiproliferative protein family or for the mitogen-activated protein kinase (MAPK) [16]. Thus, TH may also be involved on gene transport in the developing brain.


The identification of three new TH-regulated genes in the developing brain enhances our understanding of the TH-regulatory-cascades and opens new avenues for studying the molecular mechanisms of TH in respect to their effects on nerve cell proliferation, apoptosis and protein metabolism.

Acknowledgments. We thank Drs. Tom Hudson for cDNA microarray analysis, Dr. Imaizumi K for the DP5 cDNA probe, Dr. Miller for the P62 cDNA probe and Dr. Harris SE for the BMP-4 cDNA probe. This work was supported by a grant from the Canadian Institutes of Health Research No 11082 (JP). CC is a fellow of the Thyroid Foundation of Canada.


1. Legrand, J. J. Physio. (Paris) 78, 603-652 (1982).

2. Sap J, Munoz A , Damm K et al. Nature 324, 635-640 (1986).

3. Weinberger C, Thompson C , Ong ES et al. Nature 324, 641-646 (1986)

4. Murray MB, Zilz ND, McCreary NL, et al. J. Biol. Chem. 263, 12770-12777 (1988).

5. Thompson CC and Potter GB. Cerebral cortex 10, 939-945 (2000).

6. Garza R, Dussault JH and Puymirat J. Dev. Brain Res. 43, 287-298 (1988).

7. Denver J, Ouellet L, Furling D et al. J Biol Chem. 274, 23128-23134 (1999).

8. Danielson PE, Forss-Petter S, Brow MA et al, DNA 7, 261-267 (1988).

9. Puymirat J, Etongue-Mayer P and Dussault JH. J. Biol. Chem. 270, 30651-30656 (1995).

10. Imaizumi K, Tsuda M, Imai Y et al. J Biol Chem. 272, 18842-18848 (1997).

11. Imaizumi K, Morihara T, Mori Y et al. J Biol Chem. 274, 7975-7981 (1999).

12. Xiao Q, Nikodem VM. Front Biosci. 15, A52-57 (1998).

13. Li W, Cogswell CA and LoTurco JJ. J. Neurosci. 18, 8853-862 (1998).

14. D'Onofrio M, Lee MD, Starr CM, et al. J. Biol. Chem. 266,11980-11985 (1991).

15. Rodier A, Marchal-Victorion S, Rochard P, et al. Exp Cell Res , 249(2), 337-48 (1999).

16. Lin HY, Davis FB, Gordinier JK et al. Am J Physiol 276(5 Pt 1), C1014-24 (1999).

Figure legend

Figure 1. Effect of T3 on gene expression. A) Northern blot prepared with total RNA (20 mg/lane) from neuronal cultures grown in the presence or absence of T3 (30 nM) for 3 hrs. Blots were normalized with the cyclophilin probe. B) Northern blot prepared with total RNA from the brain of 17-day postnatal euthyroid (C), hypothyroid (PTU) and hypothyroid animals treated with T4 (PTU + T4) . Blots were normalized with a cyclophilin probe.

Table 1: Neuronal Genes Regulated by T3 Determined by oligonucleotide Microarray Analysis.

This table represents a comparison between RNA from single T3-treated primary neurons culture and the same untreated primary neurons culture. Results represent mean of four independent studies.