17 A combined approach identifies a limited number of new thyroid hormone target in post-natal mouse cerebellum

Laure Quignodon, Carmen Grijota-Martinez1, Emmanuel Compe2, Romain Guyot, Nathalie Allioli, David Laperrie`re4, Robert Walker3, Paul Meltzer3, Sylvie Mader4, Jacques Samarut and Fre´de´ric Flamant Institut de Ge´nomique Fonctionnelle de Lyon, Ecole Normale Supe´rieure de Lyon, Universite´ de Lyon, UMR INRA CNRS 5242, IFR128 46 alle´e d’Italie, 69364 Lyon Cedex 07, France

1Instituto de Investigaciones Biome´dicas Alberto Sols (CSIC-UAM) and Center for Biomedical Research on Rare Diseases (CIBERER), 28029 Madrid, Spain

2Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire, Illkirch, France

3Cancer Genetics Branch, NHGRI, Bethesda, Maryland, USA

4Institut de recherche en Immunologie et Cancer, Universite´ de Montre´al, Montre´al H3C3J7, Quebec, Canada

(Requests for offprints should be addressed to F Flamant; Email: frederic.fl[email protected])

Abstract

Thyroid hormones act directly on transcription in the post-natal developing cerebellum, controlling neuronal, and glial cell differentiation. We have combined three experimental approaches to identify the target genes that are underlying this phenomenon: 1) a microarray analysis of gene expression to identify hormone responsive genes in the cerebellum of Pax8K/K mice, a transgenic mouse model of congenital hypothyroidism; 2) a similar microarray analysis on primary culture of cerebellum neurons; and 3) a bioinformatics screen of conserved putative-binding sites in the mouse genome. This identifies surprisingly a small set of target genes, which, for some of them, might be key regulators of cerebellum development and neuronal differentiation. Journal of Molecular Endocrinology (2007) 39, 17–28

Introduction granular layer, where they undergo terminal differen- tiation. Hypothyroidism results in delayed proliferation, Thyroid hormones (THs, i.e. 3,5,30-tri-iodothyronine; reduced migration, and increased rate of apoptosis of T3, and its less active precursor thyroxine (T4)) are EGL cells, concomitant to a reduction in Purkinje cells essential regulators of brain development, congenital arborization. It is also responsible for delayed oligoden- hypothyroidism resulting in severe and irreversible drocytes differentiation (Rodriguez-Pena 1999)and mental retardation (Zoeller & Rovet 2004). Under- aberrant maturation of Bergmann processes of the standing TH action during brain development would Golgi epithelial cells, a cerebellum-specific astrocyte bring new insight to our knowledge of basic neurode- population (Morte et al. 2004). In vitro experiments velopment mechanisms, help clinical investigations of suggest that T3 acts in a cell autonomous manner on all TH-deficiency consequences and also allow us to address these cell types (Thompson 1996, Trentin et al. 1998, the growing concern that some chemicals present in Tang et al. 2000, Kimura-Kuroda et al. 2002, Heuer & food are dangerous for brain development (Zoeller & Mason 2003). As multiple cell–cell interactions underlie Rovet 2004). Although less common, TH excess is also cerebellum development, T3 also acts indirectly (Gomes detrimental to brain development (Kempers et al. 2003, et al. 1999, Martinez & Gomes 2005). 2005). Rodent cerebellum development is highly T3 directly activates gene expression by binding to TH sensitive to hypothyroidism and appears to be a suitable receptors (TRs) expressed from the two THRA and THRB experimental model to study congenital hypothyroidism genes (Yen 2001, Flamant et al. 2007). TRs act mainly as (Lauder 1977, Bernal et al. 2003, Bernal 2005). It takes heterodimers with retinoid X receptors (RXR), and place mainly during the first post-natal weeks, a remain bound to DNA in the absence of hormone developmental stage which coincides with a peak in binding. Well-characterized T3 response elements (TRE) the circulating level of T3 (Hadj-Sahraoui et al. 2000). At usually associate two half sites, related to the consensus this time, neuroblasts proliferate in the cerebellar 50AGGTCA30, arranged as a direct repeat separated by external granular layer (EGL), a transient structure four nucleotides (DR4), as an everted repeat separated by that disappears in mice 3 weeks after birth. The EGL cells six nucleotides (ER6) or as a palindrome (IR0; Desvergne are capable of self-renewal or migration toward the inner 1994). Whereas THRA expression is ubiquitous in

Journal of Molecular Endocrinology (2007) 39, 17–28 DOI: 10.1677/JME-06-0054 0952-5041/07/039–017 q 2007 Society for Endocrinology Printed in Great Britain Online version via http://www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access 18 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

cerebellum, THRB is expressed only in Purkinje cells. RNA extraction Unliganded TRa1, the receptor isotype encoded by RNAs were extracted from cerebellum using the THRA, exerts a negative influence on gene expression mammalian Genelute extraction kit (Sigma) and and is responsible for most of cellular alterations linked to further purified after RNase-free/DNaseI treatment hypothyroidism in cerebellar neurons (Morte et al. 2002). (Fermentas). RNAs were extracted from primary This explains why mice lacking THRA or both THRA and neuron cultures using the Nanoprep kit, also including THRB genes (Gothe et al. 1999, Gauthier et al. 2001)do DNaseI treatment (Stratagene, San Diego, CA, USA). not display the typical features of congenital hypothyr- oidism in cerebellum neurons (unpublished obser- vations). By contrast, THRA deletion is sufficient to Microarray analysis delay the differentiation of oligodendrocyte precursor RNAs were prepared for individual cerebellum either cells (Billon et al. 2002) and astrocytes (Morte et al. 2004). K K at P8 (three wild-type mice, four Pax8 / mice, Although the expression of a number of genes K K one Pax8 / mouse treated with TH for 6 h, has been shown to change in hypothyroid brain (Poguet K K two Pax8 / mice treated with TH for 48 h) or et al. 2003, Bernal 2005, Dong et al. 2005), very few direct K K P15 (four wild-type mice, three Pax8 / mice, target genes of T3 are known in brain. To our knowledge, K K one Pax8 / mice treated with TH for 6 h, three this includes only Hairless (Hr), Sygr1 (Potter et al. 2001), K K RC3/neurogranin (Guadano-Ferraz et al. 1997, Morte et al. Pax8 / mice treated with TH for 48 h). Each 1997), BTEB (Denver et al. 1999), and Rhes (Vargiu et al. individual RNA was compared with a pool of 2001). We combine here various in vivo and in vitro reference whole brain P15 (10 brains). Twenty-one approaches to identify several other genes, which are microarrays were hybridized to amino-allyled, oli- regulated by T3 in post-natal cerebellum. godTV primed, cDNA, prepared from 10 mg of these individual RNA and cross-linked to Cy3 or Cy5 dyes (Amersham). Microarrays for cerebellum RNA analysis were produced at NHGRI and contained Material and methods 14 000 spotted PCR products, prepared from a non- redundant cDNA library (http://research.nhgri.nih. Animals gov/microarray/downloadable_cdna.shtml). Image K K Pax8 / knockout mice (Mansouri et al. 1998), which analysis was performed using the ArraySuite software usually die within 3 weeks after birth, were produced package developed at NHGRI, and data with a from heterozygous parents with a mixed C57Bl6/129Sv quality, as defined in ArraySuite, inferior to 1 were genetic background. Wild-type littermates were used as discarded, leaving 12 800 spots analyzable in all control. TH treatments were performed by two i.p. microarrays. RNA (200 ng) prepared from cultured injections daily (Hegg et al. 1990) performed 48 and neurons was first submitted to two rounds of linear 24 h previous killing (0.2 mgT3,2mgT4 (Sigma) per amplification (MessageAmp aRNA Amplification Kit, gram of body-weight in 100 ml phosphate buffer saline). Ambion) before analysis with microarrays spotted All animal experimentations were performed under with 26 000 synthetic 50-mers oligonucleotides at animal care procedures were conducted in accordance Reseau National des Ge´nopoles (http://www.micro- with the guidelines set by the European Community array.fr). Two independent experiments were per- Council Directives (86/609/EEC). formed. Image analysis was performed using GenePix Pro 6.0 (Molecular Devices, Philadelphia, PA, USA). All data were filtered for low signal and the Primary neuron cultures threshold for TH induction was set to 2. All media were purchased from Invitrogen. Cerebellums were dissected from P2–P4 wild-type mice in Hank’s 2C 2C Quantitative reverse transcription-PCR analysis balanced sodium salt (HBSS, without Ca and Mg . (Q-RT-PCR) Supplemented with 1 mM NaPyruvate and 10 mM HEPES pH 7.4) rinsed in HBSS/Pyruvate/HEPES and cDNA were prepared from 100 ng (for cultured resuspended in serum-free culture medium (neurobasal neurons) or 1 mg (for cerebellum) RNA using AMV supplemented with B27 (2%) Glutamine (0.5 mM), reverse transcriptase (Promega) and random 6-mers penicillin (10 U/ml) and streptomycin (10 U/ml)), primers from an independent set of animals. After 1/50 before seeding on poly-L-lysine (Sigma) coated 24-well to 1/100 dilution, 5 ml cDNA were used for quantitative multiwells (Falcon; 3!105 cells/cm2). Whenever indi- PCR, using either sybrgreen (Invitrogen) or Taqman cated, cycloheximide (Sigma) was added for 6 h, together assay-on-demand assays (Applied Biosystems, Foster City, with T3 or not, to the culture medium at 100 mg/ml. T3 CA, USA) on a Opticon3 thermocycler (MJ Research). K (Sigma) was used at 10 7 M. Calibration was performed by 28S RNA, and

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access Thyroid hormone target genes in cerebellum . L QUIGNODON and others 19 quantitation was performed in duplicates or triplicates next performed as previously described (Compe et al. using the HPRT and TBP housekeeping genes as 2005). Coimmunoprecipitated DNA was quantified by K internal standards and the 2 DDCt method for analysis real-time quantitative PCR with Lightcycler apparatus (Livak & Schmittgen 2001). (Roche Diagnostic). The primers, whose sequences are available upon request, were designed in order to encompass the thyroid response elements found within Bioinformatics scanning of putative TH response the different promoters. The results are presented as elements percentages of immunoprecipitated DNA relative to the TRE matching exactly the consensus sequences (50- input from two independent experiments. We verified (A/G)G(G/T)TCA(N)4(A/G)G(G/T)TCA-30 for DR4, using several negative control fragments, chosen on non- 0 0 0 regulated genes, that background PCR amplification 5 -(A/G)G(G/T)TCATGA(C/A)C(C/T)-3 for IR0, 5 - ! . TGA(C/A)C(C/T)(N)6(A/G)G(G/T)TCA-30 for ER6) corresponds to 0 1% of input DNA. and located within 25 kb of annotated cap site were listed for the human and mouse genomes as described (Bourdeau et al. 2004). Hr genomic sequences were Results scanned with NUBISCAN (Podvinec et al. 2002) using a threshold of PZ0.05. Microarray analysis K K Pax8 / knockout mice suffer from congenital hypothyr- In vitro DNA– interaction oidism due to thyroid agenesis (Mansouri et al. 1998, K K Flamant et al. 2002). We harvested Pax8 / cerebellum A gel retardation assay was used to address the ability RNA at post-natal day 8 (P8) or day 15 (P15) either 6 or a a of TR /RXR heterodimers to bind double-stranded 48 h after TH treatment. When a stringent twofold 25–28 oligomers centered on consensus TRE sequence. change threshold was used, very limited changes in a a Recombinant human TR and RXR were prepared gene expression, both positive and negative, were a a from pSG5TR and pSG5RXR plasmids by coupled observed after TH treatment (Table 1). With the in vitro transcription/translation (TnT Coupled Reti- exception of A kinase (PRKA) anchor protein 1 culocyte Lysate Systems, Promega). TRE DNA probes 32 (Akap1), all genes responded to TH treatment only at were labeled with [g- P]ATP using T4 polynucleotide one developmental stage. Several genes, present on the kinase and then purified by acrylamide gel electro- microarrays, but absent in Table 1, are known to be ! 4 phoresis. DNA probe (2 10 c.p.m.) and 2 mlof sensitive to TH treatment. This included BTEB and programed lysate were incubated for 30 min at room cyclinD2 for which induction rate were close to the twofold temperature in 20 mM Tris (pH 8), 50 mM KCl, 10% threshold. We thus suspected that the statistical glycerol, 50 mM NaCl, 2 mM MgCl2, 2 mM dithiothrei- thresholds chosen for microarray analysis were too tol, 50 ng/ml poly(dI-dC)–poly(dI-dC) and 5 ng/ml stringent to identify some authentic TH target genes. salmon sperm DNA. Bound complexes were separated However, subsequent Q-RT-PCR experiments showed by low ionic strength acrylamide gel electrophoresis that, under this threshold, the rate of false positive was ! (0.5 Tris Borate EDTA buffer, 180 V for 90 min). Gels high (data not shown). were fixed in a 10% acetic acid/20% methanol solution To favor the detection of direct T3 target genes and and dried for autoradiography. For competition avoid that TH metabolism and transport might oppose to experiments, a 100-fold molar excess of unlabeled a rapid variation in cerebellum TH content (Galton double-stranded consensus DR4 oligonucleotide 2005), we performed another microarray analysis on 0 0 (5 CGATTTGAGGTCACAGGAGGTCACACAGT T3 ) primary neuron culture RNA. We compared untreated 0 K7 or aspecific oligonucleotide (5 GAGAGGAGATAAG- and T3 treated neurons (10 M, 6 h), in the presence of 0 CTGCCGCTAATGGCCGGGAAA3 ) were incubated cycloheximide, a translation inhibitor, which avoids the with prior to the addition of labeled double- activation of secondary target genes. This experiment stranded probe. identified a distinct, and larger set of genes, which were either activated or repressed by T3 treatment (Table 2). Chromatin immunoprecipitation (ChIP) assays Q-RT-PCR analysis of gene expression in whole Cerebella from six 20-day-old C57BL/6 animals were cerebellum pooled and fixed in 1% formaldehyde for 15 min at RT and this was followed by another incubation for 1 h at We performed Q-RT-PCR analysis to confirm the 4 8C. Cross-linking was stopped by addition of glycine to a regulation both in vivo and in cultured neurons, for final concentration of 0.125 M. ChIP experiments with candidate T3 target genes (Akap1, kin of IRRE like 3 antibodies raised against TRa1(Plateroti et al. 2001) were (Kirrel3), leucine rich repeat protein 1 (Lrrn1), www.endocrinology-journals.org Journal of Molecular Endocrinology (2007) 39, 17–28

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access 20 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

Table 1 Microarray analysis of thyroid hormone response in whole cerebellum

P8 (48 h) P15 (6 h) P15 (48 h) Locus ID Short name Full name

P8 (6 h) 2.70 2.80 2.33 1.93 11 640 Akap1 A kinase (PRKA) anchor protein 1 2.27 1.38 0.96 1.06 268 860 Abat 4-Aminobutyrate aminotransferase 1.22 2.13 0.50 0.94 25 262 Itpr1 Inositol 1,4,5-triphosphate receptor 1 1.08 0.97 2.19 1.48 103 712 RIKEN cDNA 6330403K07 gene 1.06 1.03 2.04 1.08 52 595 D15Ert- DNA segment, Chr 15, ERATO Doi 430, d430e expressed 1.05 1.03 2.02 1.59 11 972 Atp6v0d1 ATPase, HC transporting, lysosomal V0 subunit D1 0.99 0.90 2.11 1.41 12 864 Cox6c Cox6c: cytochrome c oxidase, subunit Vic 0.49 0.66 1.11 1.93 545 015 RIKEN cDNA 2610042L04 gene 0.71 0.48 0.88 0.52 17 536 Mrg1 Myeloid ecotropic viral integration site-related gene 1 0.79 1.42 0.12 0.27 66 674 RIKEN cDNA 6330409N04 gene 0.92 1.33 0.14 0.46 97 564 C86942 Expressed sequence C86942 (C86942) 0.94 1.19 0.16 0.33 15 387 Hnrpk Heterogeneous nuclear ribonucleoprotein K 0.72 1.04 0.18 0.37 20 877 Aurkb Aurora kinase B 0.96 1.34 0.20 0.33 12 550 Cdh1 Cadherin 1 0.71 1.13 0.23 0.63 18 676 Phf2 PHD finger protein 2 0.91 1.05 0.41 0.69 210 146 Irgq Immunity-related GTPase family, Q 0.77 0.88 0.41 0.43 14 646 Glns-ps1 Glutamine synthetase pseudogene 1 0.95 0.84 0.44 0.79 100 213 Rusc2 RUN and SH3 domain containing 2 0.78 0.97 0.48 0.95 70 834 Spag9 Sperm-associated antigen 9 0.90 0.92 0.50 0.66 20 729 Spin Spindlin

0 Table 2 Microarray analysis of 3,5,3 -tri-iodothyronine (T3)-mediated gene activation in cultured cerebellum neurons

Microarraya Q-PCR primary Q-PCR primary Full name Short name (CHXC6hT3) neurons (6 h T3) neurons (48 h T3)

Gene ID 225 872 Neuronal PAS domain protein 4 Npas4 7.99 NT NT 15 370 Nuclear receptor subfamily 4, group A, nur77 5.97 NT NT member 1 108 151 Semaphorin 3D Sema3d 4.97 1.27 1.92 20 750 Secreted phosphoprotein 1 Spp1 4.40 0.89 0.89 14 281 FBJ osteosarcoma oncogene Fos 3.79 NT NT 18 227 Nuclear receptor subfamily 4, group A, Nurr1 3.42 1.11 1.32 member 2 17 105 Lysozyme Lyzs 2.88 NT NT 11 840 ADP-ribosylation factor 1 ARF1 2.63 NT NT 70 273 Riken cDNA 2310051E17Rik gene 2.60 NT NT 58 243 Nucleosome assembly protein 1-like 5 Nap1l5 2.44 NT NT 71 307 Riken cDNA 1700018A04Rik gene 1700018A04Rik 2.21 NT NT 14 825 Chemokine (C-X-C motif) ligand 1 Cxcl1 2.11 0.83 1.32 19 252 Dual specificity phosphatase 1 Dusp1 2.10 NT NT 16 979 Leucine rich repeat protein 1 Lrrn1 2.09 1.19 2.22 20 983 IV Syt4 2.02 2.59 6.70 67 703 Kin of IRRE like 3 Kirrel 3 1.82 1.38 1.83 66 625 Riken cDNA 5730406M06Rik gene 0.50 NT NT 15 122 Hemoglobin a, adult chain 1 Hba-a1 0.50 1.75 2.92 67 527 Riken cDNA 1300007C21Rik 0.46 NT NT 231 798 Leucine-rich repeats and calponin Lrch4 0.45 NT NT homology domain containing 4 11 461 Actin b, cytoplasmic Actb 0.44 NT NT 21 770 Protein phosphatase 2, reg. subunit B, d PPP2R5D 0.43 NT NT

aAverage from two experiments, swapping cy3 and cy5 labeling.

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access Thyroid hormone target genes in cerebellum . L QUIGNODON and others 21

Pax8–/– Pax8–/–+TH WT * * * * * 2·00 * * * * * * 1·00 *

0·00 Hr NT3 HPRT PcP2 Lrrn1 Tgm1 Pdgfra Gabra6 Akap1 Kirrel3 Figure 1 Q-RT-PCR analysis of whole cerebellum RNA. RNAs were prepared from cerebellum at P8 K K K K from hypothyroid Pax8 / mice (nZ8), Pax8 / treated for 2 days with TH (nZ10), or wild-type K K littermates (nZ12). *Indicates mean values which significantly differ from Pax8 / mean value (Student t-test; P!0.05). Data are expressed as ratios compared with a reference sample (euthyroid whole brain at P15, except for Lrrn1, Kirrel3 and Tgm1 where a T3 treated P8 cerebellum was used, as expression was too low in the reference sample). transglutaminase 1 (Tgm1)) and a gene which was close to genes at P15, and the induction rate was generally found the twofold threshold in several experiments (Gabra6 to be reduced (data not shown). ((GABA-A) receptor, subunit a6) and D0H4S114/P311; Fig. 1). We also included control genes whose expression is known to be sensitive to TH deficiency. This last Q-RT-PCR analysis of gene expression in primary category included Hr, an authentic TH target gene neuronal cell cultures expressed in granular neurons, neurotrophin-3 (NT-3), which is indirectly regulated by TH (Poguet et al. 2003) Three modes of in vivo regulation by TH can be proposed. and PDGFRa, a marker of oligodendrocytes precursor First, liganded TR can exert direct transcriptional cells (Tekki-Kessaris et al. 2001) which differentiation is regulation on gene promoter. Alternatively, TH can strictly dependent on T3 (Tang et al. 2001)andPcp2,a first activate transcription factors or cofactors regulation, gene whose expression in Purkinje cells is known to be which in turn activate secondary targets in a cell reduced in case of hypothyroidism. Hr, Akap1, Gabra6, autonomous manner. Finally, non-cell autonomous Kirrel3, Lrrn1, Pcp2 and NT-3 were all activated at P8 when activation might occur, resulting from the activated TH treatment was performed in vivo, while PDGFRa was secretion of neurotrophic factors, like NT-3,byTH.To down-regulated, also not in a significant manner (Fig. 1). distinguish between these possibilities, we performed D0H4S114/P311 expression was not changed (data not Q-RT-PCR analysis of gene expression in primary cultures shown). The regulation was also addressed for some of cerebellar neurons, exposed to T3 (Fig. 2). We verified 3·00

2·50

2·00 Oh 6 h T3 1·50 24 h T3 48 h T3 CHX 1·00 CHX + 6 h T3

0·50

0·00 Hr NT3 PcP2 Akap1 Lrrn1 Gabra6 Kirrel3 Figure 2 Q-RT-PCR analysis of cultured neurons RNA. RNAs were prepared from cultured neurons 2 days after seeding. T3 was added 6, 24 or 48 h before. Cycloheximide was added (dark bars) to block RNA translation to identify genes, which are likely to be direct target genes. Only one representative experiment is shown. All presented activations were reproducible in three independent experiments. Data are expressed as ratios, compared with average level. www.endocrinology-journals.org Journal of Molecular Endocrinology (2007) 39, 17–28

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access 22 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

that serum-free culture conditions favor the survival of We also addressed the possibility that the list of genes neurons, mainly granular neurons, at the expense of glial with a consensus TRE in both the mouse and human cells. After 48 h of culture, more that 90% of the cells genomes contains other TH target genes, which are displayed the Tuj1 neuronal marker, whereas less that 2% regulated during cerebellum development, but were expressed the Glial fibrillary acidic protein, a marker for not identified in our microarray experiments. We first astrocytes (data not shown). When treated with T3,these monitored databases for expression patterns (GENSAT, cell cultures quickly reacted (Fig. 2) by a robust increase Genepaint, Allen Brain Atlas) and for published in Hr expression. The response amplitude appeared to be genetic evidence of possible involvement in neurode- less variable that the in vivo response. By contrast, NT-3 velopment and neural cells differentiation. We then was not activated in this system, confirming the previous selected 19 of these genes for further Q-RT-PCR conclusion that the in vivo activation of this gene by TH is expression analysis: Adam23, Chrna1, COUP-TFI, Crk, not a cell autonomous process (Poguet et al. 2003). c-ski, Dlgh1, a-laminin, Lin28, Midnolin, Prdx3, Prkca, A moderate augmentation of Akap1, Gabra6, Kirrel3, Lrrn1 Ptprj, SEMA4G, Sncb, Stathmin, Syngr, Tle6, Tgm1 and and Pcp2 also occurred, suggesting that T3 is acting in a Vav1. Among these, only Tgm1 was found to be highly cell autonomous manner to up-regulate these genes and induced in vivo in cerebellum (Fig. 1), but was not that this activation does not require previous NT-3 expressed in cultured neurons. activation. When T3 was added for 6 h in the presence Hr, which is directly regulated by T3 and possesses an of cycloheximide, a translation inhibitor, Hr, Akap1, Lrrn1 identified TRE 2345 upstream to its cap site and Kirrel3 activation was maintained indicating that (Thompson & Bottcher 1997), was not present in the these four genes are probably direct T3 target genes. list of genes with an evolutionary conserved TRE. This indicates that the choice of a stringent threshold precludes the identification of a fraction of the Bioinformatics scanning for thyroid response authentic TH target genes. To scan for response elements elements more loosely related to the consensus, we Z . To determine whether the direct target genes identified used NUBISCAN (P 0 05 threshold) to analyze in the present study were regulated by binding of TR on individual mouse genomic regions covering the candi- consensus TRE (DR4, IR0 and ER6), we performed a date TH target genes cap sites. This identified three bioinformatics screen of the whole mouse genome for the other putative TRE for Hr (Table 3). consensus elements located within 25 kb of annotated cap sites. As TRE consensus is frequent, even in random Identification of TRa/RXRa heterodimers bound to sequences, this method identified a large number of TRE in vitro and in vivo putative T3 target genes (5246 TRE with perfect match). Among the 43 up-regulated genes present in Tables 1 and To ascertain that we have identified bona fide direct 2, Akap1, ADP-ribosylation factor 1 (Arf1), ATPase, HC TRa1 target genes, the only isoform present in all transporting, lysosomal V0 subunit D1, cadherin 1 (CdhI), cerebellar cell types, we performed protein–DNA cytochrome c oxidase (Cox6c), hemoglobin a, adult chain interaction assays on some of the most likely candidate 1(Hba-a1), lysozyme, leucine-rich repeats and calponin genes. In vitro,TRa/RXRa heterodimers were able to homology domain containing 4, Lrrn1, protein phospha- bind to the TRE identified (Table 3) for Hr (K2345), tase 2, reg. subunit B, d, Myeloid ecotropic viral Akap1, CdhI, Hba-a1 and Tgm1 (Fig. 3). We finally used integration site-related gene 1 (Mrg1) and Spindlin ChIP assays to directly establish the actual occupancy of were found in this list. Although this is an indication promoters by TR (Fig. 4). In whole cerebellum extracts, that these genes might be direct TR target genes, the TRa was found to be present on fragments covering the observed frequency of TRE-containing genes (12 out of identified TRE for Hr, Akap1, Cdh1, Hba-a1, Mrg1, Tgm1. 43 regulated genes) is close to the frequency (30%) Taken together, these data strongly suggest that TRa expected for a random distribution of TRE elements in heterodimers, bound to identified DR4 elements, the genome. To focus our analysis on the TRE that are mediate positive regulation of Hr, Akap1, Hba-a1, more likely to be important for developmental Tgm1 and negative regulation of Cdh1 and Mrg1. regulation, and thus conserved during evolution, we crossed the list of the equivalent list obtained for the to identify TRE present in both Discussion homologous genes. This criterion has been shown to facilitate the recognition of true nuclear hormone target In the present study, we combined bioinformatics genes (Bourdeau et al. 2004). This allowed us to focus on prediction, primary neuronal culture and whole the subset of 157 putative target genes which possess a cerebellum RNA microarray analysis to identify genes, perfectly matched TRE in both species (113 DR4, 25 IR0 which are regulated by T3 in the developing and 19 ER6). This list included Akap1 and Mrg1. cerebellum. Previous attempts were either limited to

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access www.endocrinology-journals.org Table 3 Bioinformatics prediction of thyroid hormone response elements

Short Element Position/cap Mismatch in Mismatch in Full name name Mouse sequence type site mouse gene human gene T3 regulation

Gene ID 268 860 4-Aminobutyrate aminotransferase Abat AGAACTgaatccAGGTCA ER6 C18 413 1 1 CbC GGTGCAcagcAGGTCA DR4 K517 1 1 AGTTCGccctAGGTCA DR4 K12 838 1 1 11 461 Actin b, cytoplasmic Actb GGGTCAccagGGATCA DR4 K4666 1 1 NeK GGGTCActccAGGTAA DR4 K24 568 1 1 AGGTCAgtttAGTTTA DR4 C16 559 1 1 11 972 ATPase, HC transporting, lysoso- Atp6v0d1 AGTTCAgcaaAGTTCA DR4 K2746 0 1 CbC mal V0 subunit D1 11 640 A kinase (PRKA) anchor protein 1 Akap1 AGTTCAtacaGGTTCA DR4 K20 599 0 0 CbCNeC 11 840 ADP-ribosylation factor 1 ARF1 GGGTCAagaaAGTTCA DR4 C9508 0 1 NeC 12 550 Cadherin 1 Cdh1 TGTCCTcagggaAGTTCA ER6 C8445 1 1 CbK 14 281 FBJ osteosarcoma oncogene Fos AGGTCAgctcAGGACA DR4 K6100 1 1 NeC 14 399 (GABA-A) receptor, subunit a6 Gabra6 AGTTTAttgcAGTTCA DR4 K9252 1 1 CbCNeC 15 122 Hemoglobin a , adult chain 1 Hba-a1 AGTGCAgccaAGTTCA DR4 K2797 1 1 CbCNeC 231 798 Leucine-rich rep./calponin hom. Lrch4 AGGACAcagaAGTTCA DR4 K21 322 1 1 NeC domain containing 4 71 307 Myeloid ecotropic viral integration Mrg1 AGTTCAaggtGGGTCA DR4 K12 672 0 0 CbK

site-related gene 1 cerebellum in genes target hormone Thyroid 21 816 Nuclear receptor subfamily 4, group nur77 AGTTCAaaccGGTTCC DR4 C16 326 1 1 NeC A, member 1 TGTTCAcagtAGTTCA DR4 K10 369 1 1 AGGTCACGACCC IR0 K3340 1 1 15 460 Protein phosphatase 2, reg. subunit PPP2R5D TGACCCctctgaGGGTCA ER6 C871 0 1 NeC B, d 20 729 Spindlin Spin AGTTCAggctCGGTCA DR4 C19 359 1 1 CbK 21 816 Transglutaminase 1, K polypeptide Tgm1 GGGTCAaaaaAGTTCA DR4 K22 251 0 0 CbC 15 460 Hairless Hr AGGGCAtctgAGGACA DR4 K2345 2 CbCNeC AGGACAgcagAGGGCA DR4 K16 306 2 GGAGCAagaaAGGTCA DR4 C11 777 2

Downloaded fromBioscientifica.com at09/30/202111:32:40PM AGGTCAtgtcTGGGCA DR4 C21 564 2 ora fMlclrEndocrinology Molecular of Journal

CbC CbK, up- or down-regulated in cerebellum; NeCNeK, up- or down-regulated in primary neurons culture. . QUIGNODON L (2007) n others and 39, 17–28 via freeaccess 23 24 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

DR4 Hr Cdh 1 Akap–1 Tgm–1 Hbb-a

pSG5 + – – – – – – – – – – – – –

TRα/RXRα – ++++ + ++++++++

Labeled probe + + + + + + + + + + + + + +

DR4 control unlabled probe ––+ –– +–– ++++––

Aspecific unlabeled probe ––– + –– ––––– –– –

Figure 3 In vitro interaction between putative TRE and TRa/RXRa heterodimers. A double-stranded DR4 consensus sequence (50AGGTCAcaggAGGCTA30) was used as positive control and found to form a complex with recombinant TRa/RXRa. (2) but not with unprogramed reticulocytes lysate (1). Binding specificity was verified by competition with unlabeled specific (3) or non-specific (4) DNA. Similar complexes, destabilized by competition with excess of consensus DR4 TRE, formed with DR4 found in Hr, Akap1, Cdh1, Tgm1 and Hba-a1.

cultured neurons (Martel et al. 2002) comparison situation found in liver (Flores-Morales et al. 2002, Yen between hypothyroid and euthyroid animals (Dong et al. 2003) or skin fibroblasts (Moeller et al. 2005), et al. 2005), performed on whole brain RNA only where the same experimental approaches revealed that (Poguet et al. 2003) or at a post-developmental stage expression of many genes quickly changes after T3 (Haas et al. 2004). These studies identified few putative treatment. Although technical limitations might have TH target genes suggesting that, upon TH treatment, hampered the identification of the entire repertoire of the expression of a limited number of genes is changed TH target genes in cerebellum, the present study and the induction rates are limited. The numerous strongly suggests that the actual number of direct differences observed between hypothyroid and euthyr- transcriptional TH targets is much more limited than in oid animals (Dong et al. 2005) might thus mainly reflect other tissues. It seems therefore that gene regulation by a difference in cell composition. This contrasts with the TH in the developing cerebellum has to be tightly

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access Thyroid hormone target genes in cerebellum . L QUIGNODON and others 25

controlled. This also suggests that studying T3 action features of hypo- or hyper-thyroidism. Sema3d encodes will help to pinpoint few target genes involved in the a semaphorin expressed at high level in cerebellum basic molecular mechanisms underlying cell prolifer- granular cells (according to the GENSAT database), ation, differentiation, migration, and apoptosis in brain which might be important for EGL cells migration. Syt4 and bring a new insight into the genetic program which encodes Synaptotagmin IV a secretory vesicle protein controls brain development. thought to function as an inhibitor of neurotransmitter Parts of the TH-induced changes in gene expression release, and as a neuroprotective factor (Ferguson et al. that we have observed are indirectly related to TH 2004). This view has been recently challenged (Ting function, and only reflect the advancement of et al. 2006) in favor of an alternative hypothesis stating cerebellum maturation and cellular differentiation. that Synaptotagmin IV regulates glial glutamate release PDGFRa for example is a specific marker for oligoden- (Zhang et al. 2004). Itpr1, also called Pcp1, is enriched in C drocyte precursor cells in brain. The decrease of Purkinje cells, where it has a crucial role for Ca2 PDGFRa during post-natal development, its increased signaling (Matsumoto & Kato 2001). expression in hypothyroid mice or its slight decrease By combining expression and protein–DNA after TH in vivo treatment probably reflects only interaction analysis, we accumulated enough evidence oligodendrocytes precursor differentiation. Gabra6 is to ascertain that T3 directly up-regulates in the expressed only in inner granular cells and its up-regula- cerebellum the expression of the following six genes: tion by TH might also reflect progression of this cell Akap1, Cdh1, Hba-a1, Hr, Mrg1, Tgm1. Only Hr was type toward terminal differentiation. However, already known to be a TH target gene. In vitro neuronal although it is not a direct TR target gene, Gabra6 is cultures also suggests that Lrrn1 and Kirrel3 belong to activated by TH in cultured neurons, most probably in a this category of direct TH targets, although we did not cell autonomous manner. This gene encodes a identify evolutionary conserved TRE for these two cerebellum-specific subunit of GABA receptors whose genes. Three of these genes (Cdh1, Mrg1, Akap1) were specific function has not been clarified by knockout identified by whole cerebellum microarray analysis, analysis (Jones et al. 1997). As GABA is believed to exert, three (Lrrn1, Kirrel3, Hba-a1) by microarray analysis of beside its neurotransmitter function, a trophic effect cultured neurons, and the last gene (Tgm1) was found during brain development (Owens & Kriegstein 2002), by bioinformatics screening. Interestingly, T3 seems to it would be interesting to explore the possible directly mediate both positive (Akap1, Hba-a1, Hr, implication of Gabra6 in this poorly understood process Kirrel3, Lrrn1, Tgm1)andnegative(Cdh1, Mrg1) and its putative regulation by TH. Some other indirect transcriptional regulation. This reinforces the view TH-mediated gene regulation might explain some that the current model for TR-mediated transcription

In vivo TRa1 recruitment

2

% IP/Input 1

0 Hr Akap1 Hbb-a Tgm1 Cdh1 Mrg1 Figure 4 In vivo recruitment of TRa on identified TRE. The recruitments of TRa1 on the TRE found in the promoter of Hr, Akap1, Hba-a1, Tgm1, Cdh1 and Mrg1 were analyzed by ChIP assays in cerebella of 20-day-old C57BL/6 mice. Specific immunoprecipitated DNA was quantified by real-time quantitative PCR. The results are presented as percentages of immunoprecipitated DNA relative to the input, after background subtraction. www.endocrinology-journals.org Journal of Molecular Endocrinology (2007) 39, 17–28

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access 26 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

(Rosenfeld et al. 2006) is still incomplete (Nygard et al. European Network of Excellence (EU contract no. 2003, 2006). The expression of Akap1, Kirrel3, Lrrn1, FOOD-CT-2004-506319) the CRESCENDO EU inte- and Tgm1 was not significantly reduced in hypothyroid grated project. E C was supported by Agence Nationale K K Pax8 / mice when compared with wild type. These de la Recherche (ANR-06-BLAN-0141-01) and Associ- genes thus seem to belong to a distinct category of T3- ation pour la Recherche contre le Cancer (A06-2-3153). regulated genes that are mainly regulated by supra- L Q was supported by the Fondation pour la Recherche physiological levels of T3 (Yen et al. 2003). Such a Me´dicale. The authors declare that there is no conflict regulation might thus be more relevant to the of interest that would prejudice the impartiality of this developmental alterations resulting from hyperthyroid- scientific work. ism, rather than hypothyroidism. Lrrn1 function is unknown. It encodes a neuronal protein with an IgCAM domain which might indicate an intervention in neuronal migration. The poorly References studied Kirrel3 gene encodes a nephrin-like trans- involved in homophilic adhesion Almeida A, Bolanos JP & Moreno S 2005 Cdh1/Hct1-APC is essential (Serizawa et al. 2006) able to support stem cells for the survival of postmitotic neurons. Journal of Neuroscience 25 8115–8121. proliferation (Ueno et al. 2003). Its expression pattern Bernal J 2005 Thyroid hormones and brain development. Vitamins and also suggests a function in late differentiation processes, Hormones 71 95–122. especially synapses formation (Tamura et al. 2005). Bernal J, Guadano-Ferraz A & Morte B 2003 Perspectives in the study Tgm1 encodes a Tgm1 with an important function in of thyroid hormone action on brain development and function. skin (Matsuki et al. 1998). Its expression pattern and Thyroid 13 1005–1012. function in brain are unknown. Hba-a1, only known Billon N, Jolicoeur C, Tokumoto Y, Vennstrom B & Raff M 2002 Normal timing of oligodendrocyte development depends on function, is to encode the ‘a’ subunit of hemoglobin thyroid hormone receptor alpha 1 (TRa1). EMBO Journal 21 6452– and its expression in cerebellum is surprising at first 6460. glance. However, it has been shown that Hbb-b, which Bourdeau V, Deschenes J, Metivier R, Nagai Y, Nguyen D, Bretschnei- encodes the other subunit of hemoglobin, is present in der N, Gannon F, White JH & Mader S 2004 Genome-wide cultured oligodendrocyte precursor cells, and down- identification of high-affinity estrogen response elements in human regulated when their differentiation is triggered by NT-3 and mouse. Molecular Endocrinology 18 1411–1427. Compe E, Drane P, Laurent C, Diderich K, Braun C, Hoeijmakers JH & withdrawal and TH addition (Dugas et al. 2006). Egly JM 2005 Dysregulation of the peroxisome proliferator- Therefore, hemoglobin might exist in non-erythroid activated receptor target genes by XPD mutations. Molecular and cells and fulfill alternative function. Akap1 encodes an Cellular Biology 25 6065–6076. anchoring protein widely expressed in brain (McKee Denver RJ, Ouellet L, Furling D, Kobayashi A, Fujii-Kuriyama Y et al. 2005) and crucial for cAMP/PKA signaling & Puymirat J 1999 Basic transcription element-binding protein (BTEB) is a thyroid hormone-regulated gene in the developing (Newhall et al. 2006), a pathway which mediates the central nervous system. Evidence for a role in neurite outgrowth. anti-apoptotic activity of IGF1 on cerebellum granular Journal of Biological Chemistry 274 23128–23134. neurons (Subramaniam et al. 2005). The down- Desvergne B 1994 How do thyroid hormone receptors bind to regulation of cdh1, encoding E-cadherin, should also structurally diverse response elements? Molecular and Cellular influence granular cell cycling and differentiation Endocrinology 100 125–131. (Almeida et al. 2005). Dong H, Wade M, Williams A, Lee A, Douglas GR & Yauk C 2005 Molecular insight into the effects of hypothyroidism on the Together with the previously identified other T3 developing cerebellum. Biochemical and Biophysical Research Com- target genes (Sygr1, RC3/neurogranin, BTEB, and Rhes), munications 330 1182–1193. the new set of genes that we have identified offer Dugas J, YC T, Speed TP, Ngai J & Barres B 2006 Functional genomic interesting new entry points to the genetic program analysis of oligodendrocyte differentiation. Journal of Neuroscience 26 which control the complex network of cell–cell 10967–10983. Ferguson GD, Wang H, Herschman HR & Storm DR 2004 Altered interactions that coordinate the normal post-natal hippocampal short-term plasticity and associative memory in development of cerebellum. synaptotagmin IV (K/K) mice. Hippocampus 14 964–974. Flamant F, Poguet AL, Plateroti M, Chassande O, Gauthier K, Streichenberger N, Mansouri A & Samarut J 2002 Congenital hypothyroid Pax8(K/K) mutant mice can be rescued by Acknowledgements inactivating the TRa gene. Molecular Endocrinology 16 24–32. Flamant F, Gauthier K & Samarut J 2007 Thyroid hormones signaling We thank Jean-Marc Egly for his kind support, Nadine is getting more complex: STORMs are coming. Molecular Endo- Aguilera and the PBES staff for mouse breeding and crinology 21 321–333. Aure´lie Laugraud for statistical analysis. This work was Flores-Morales A, Gullberg H, Fernandez L, Stahlberg N, Lee NH, Vennstrom B & Norstedt G 2002 Patterns of liver gene expression supported by French Ministery of Research (ACI governed by TRb. Molecular Endocrinology 16 1257–1268. Biologie Cellulaire, Mole´culaire et Structurale) Ligue Galton VA 2005 The roles of the iodothyronine deiodinases in contre le Cancer (e´quipe labellise´e) the CASCADE mammalian development. Thyroid 15 823–834.

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access Thyroid hormone target genes in cerebellum . L QUIGNODON and others 27

Gauthier K, Plateroti M, Harvey CB, Williams GR, Weiss RE, Refetoff S, Matsumoto M & Kato K 2001 Altered calcium dynamics in cultured Willott JF, Sundin V, Roux JP, Malaval L et al. 2001 Genetic analysis cerebellar cells from IP3R1-deficient mice. Neuroreport 12 reveals different functions for the products of the thyroid hormone 3471–3474. receptor alpha locus. Molecular and Cellular Biology 21 4748–4760. McKee AE, Minet E, Stern C, Riahi S, Stiles CD & Silver PA 2005 A Gomes FC, Maia CG, de Menezes JR & Neto VM 1999 Cerebellar genome-wide in situ hybridization map of RNA-binding proteins astrocytes treated by thyroid hormone modulate neuronal reveals anatomically restricted expression in the developing mouse proliferation. Glia 25 247–255. brain. BMC Developmental Biology 5 14. Gothe S, Wang Z, Ng L, Kindblom JM, Barros AC, Ohlsson C, Moeller LC, Dumitrescu AM, Walker RL, Meltzer PS & Refetoff S 2005 Vennstrom B & Forrest D 1999 Mice devoid of all known thyroid Thyroid hormone responsive genes in cultured human fibroblasts. hormone receptors are viable but exhibit disorders of the pituitary– Journal of Clinical Endocrinology and Metabolism 90 936–943. thyroid axis, growth, and bone maturation. Genes and Development 13 Morte B, Iniguez MA, Lorenzo PI & Bernal J 1997 Thyroid 1329–1341. hormone-regulated expression of RC3/neurogranin in the Guadano-Ferraz A, Escamez MJ, Morte B, Vargiu P & Bernal J 1997 immortalized hypothalamic cell line GT1-7. Journal of Neuro- Transcriptional induction of RC3/neurogranin by thyroid hormone: chemistry 69 902–909. differential neuronal sensitivity is not correlated with thyroid Morte B, Manzano J, Scanlan T, Vennstrom B & Bernal J 2002 Deletion hormone receptor distribution in the brain. Brain Research. of the thyroid hormone receptor a1 prevents the structural Molecular Brain Research 49 37–44. alterations of the cerebellum induced by hypothyroidism. PNAS 99 Haas MJ, Mreyoud A, Fishman M & Mooradian AD 2004 Microarray 3985–3989. analysis of thyroid hormone-induced changes in mRNA expression Morte B, Manzano J, Scanlan TS, Vennstrom B & Bernal J 2004 in the adult rat brain. Neuroscience Letters 365 14–18. Aberrant maturation of astrocytes in thyroid hormone receptor a1 Hadj-Sahraoui N, Seugnet I, Ghorbel MT & Demeneix B 2000 knockout mice reveals an interplay between thyroid hormone Hypothyroidism prolongs mitotic activity in the post-natal mouse receptor isoforms. Endocrinology 145 1386–1391. brain. Neuroscience Letters 280 79–82. Newhall KJ, Criniti AR, Cheah CS, Smith KC, Kafer KE, Burkart AD Hegg E, Li S, Faure M & Ver O 1990 Keystone Symposia, Nuclear Hormone & McKnight GS 2006 Dynamic anchoring of PKA is essential during receptors Taos, NM, USA (abstract). oocyte maturation. Current Biology 16 321–327. Heuer H & Mason CA 2003 Thyroid hormone induces cerebellar Nygard M, Wahlstrom GM, Gustafsson MV, Tokumoto YM, Bondesson Purkinje cell dendritic development via the thyroid hormone M, Brooksbank C, Causton HC, Cavalieri D, Gaasterland T, receptor alpha1. Journal of Neuroscience 23 10604–10612. Hingamp P et al. 2003 Hormone-dependent repression of the E2F-1 Jones A, Korpi ER, McKernan RM, Pelz R, Nusser Z, Makela R, Mellor gene by thyroid hormone receptors. Molecular Endocrinology 17 JR, Pollard S, Bahn S, Stephenson FA et al. 1997 Ligand-gated ion 79–92. channel subunit partnerships: GABAA receptor a6 subunit gene Nygard M, Becker N, Demeneix B, Pettersson K & Bondesson M inactivation inhibits delta subunit expression. Journal of Neuroscience 2006 Thyroid hormone-mediated negative transcriptional 17 1350–1362. regulation of Necdin expression. Journal of Molecular Endocrinology Kempers MJ, van Tijn DA, van Trotsenburg AS, de Vijlder JJ, Wiedijk 36 517–530. BM & Vulsma T 2003 Central congenital hypothyroidism due to Owens DF & Kriegstein AR 2002 Is there more to GABA than synaptic gestational hyperthyroidism: detection where prevention failed. inhibition? Nature Reviews. Neuroscience 3 715–727. Journal of Clinical Endocrinology and Metabolism 88 5851–5857. Plateroti M, Gauthier K, Domon-Dell C, Freund JN, Samarut J Kempers MJ, van Trotsenburg AS, van Tijn DA, Bakker E, Wiedijk BM, & Chassande O 2001 Functional interference between thyroid Endert E, de Vijlder JJ & Vulsma T 2005 Disturbance of the fetal hormone receptor a (TRa) and natural truncated TRDa isoforms thyroid hormone state has long-term consequences for treatment of in the control of intestine development. Molecular and Cellular thyroidal and central congenital hypothyroidism. Journal of Clinical Biology 21 4761–4772. Endocrinology and Metabolism 90 4094–4100. Podvinec M, Kaufmann MR, Handschin C & Meyer UA 2002 Kimura-Kuroda J, Nagata I, Negishi-Kato M & Kuroda Y 2002 Thyroid NUBIScan, an in silico approach for prediction of nuclear receptor hormone-dependent development of mouse cerebellar Purkinje response elements. Molecular Endocrinology 16 1269–1279. cells in vitro. Brain Research. Developmental Brain Research 137 55–65. Poguet AL, Legrand C, Feng X, Yen PM, Meltzer P, Samarut J Lauder JM 1977 The effects of early hypo- and hyper-thyroidism on the & Flamant F 2003 Microarray analysis of knockout mice identifies development of rat cerebellar crotex. III. Kinetics of cell cyclin D2 as a possible mediator for the action of thyroid hormone proliferation in the external granular layer. Brain Research 126 during the postnatal development of the cerebellum. Developmental 31–51. Biology 254 188–199. Livak JL & Schmittgen TD 2001 Analysis of relative gene expression Potter GB, Beaudoin GM, III., DeRenzo CL, Zarach JM, Chen SH & data using real-time quantitative PCR and the 2-DDCT method. Thompson CC 2001 The hairless gene mutated in congenital hair Methods 25 402–408. loss disorders encodes a novel nuclear receptor corepressor. Genes Mansouri A, Chowdhury K & Gruss P 1998 Follicular cells of the and Development 15 2687–2701. thyroid gland require Pax8 gene function. Nature Genetics 19 87–90. Rodriguez-Pena A 1999 Oligodendrocyte development and thyroid Martel J, Cayrou C & Puymirat J 2002 Identification of new thyroid hormone. Journal of Neurobiology 40 497–512. hormone-regulated genes in rat brain neuronal cultures. Neuroreport Rosenfeld MG, Lunyak VV & Glass CK 2006 Sensors and signals: a 13 1849–1851. coactivator/corepressor/epigenetic code for integrating signal- Martinez R & Gomes FC 2005 Proliferation of cerebellar neurons dependent programs of transcriptional response. Genes and induced by astrocytes treated with thyroid hormone is mediated by Development 20 1405–1428. a cooperation between cell contact and soluble factors and involves Serizawa S, Miyamichi K, Takeuchi H, Yamagishi Y, Suzuki M & Sakano the epidermal growth factor-protein kinase a pathway. Journal of H 2006 A neuronal identity code for the odorant receptor-specific Neuroscience Research 80 341–349. and activity-dependent axon sorting. Cell 127 1057–1069. Matsuki M, Yamashita F, Ishida-Yamamoto A, Yamada K, Kinoshita C, Subramaniam S, Shahani N, Strelau J, Laliberte C, Brandt R, Kaplan D Fushiki S, Ueda E, Morishima Y, Tabata K, Yasuno H et al. 1998 & Unsicker K 2005 Insulin-like growth factor 1 inhibits extracellular Defective stratum corneum and early neonatal death in mice signal-regulated kinase to promote neuronal survival via the lacking the gene for transglutaminase 1 (keratinocyte transgluta- phosphatidylinositol 3-kinase/protein kinase A/c-Raf pathway. minase). PNAS 95 1044–1049. Journal of Neuroscience 25 2838–2852. www.endocrinology-journals.org Journal of Molecular Endocrinology (2007) 39, 17–28

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access 28 L QUIGNODON and others . Thyroid hormone target genes in cerebellum

Tamura S, Morikawa Y, Hisaoka T, Ueno H, Kitamura T & Senba E Ueno H, Sakita-Ishikawa M, Morikawa Y, Nakano T, Kitamura T & Saito 2005 Expression of mKirre, a mammalian homolog of Drosophila M 2003 A stromal cell-derived membrane protein that supports kirre, in the developing and adult mouse brain. Neuroscience 133 hematopoietic stem cells. Nature Immunology 4 457–463. 615–624. Vargiu P, Morte B, Manzano J, Perez J, de Abajo R, Gregor Sutcliffe J Tang DG, Tokumoto YM & Raff MC 2000 Long-term culture of purified & Bernal J 2001 Thyroid hormone regulation of rhes, a novel Ras postnatal oligodendrocyte precursor cells. Evidence for an intrinsic homolog gene expressed in the striatum. Brain Research. Molecular maturation program that plays out over months. Journal of Cell Brain Research 94 1–8. Biology 148 971–984. Yen PM 2001 Physiological and molecular basis of thyroid hormone Tang DG, Tokumoto YM, Apperly JA, Lloyd AC & Raff MC 2001 Lack of action. Physiological Reviews 81 1097–1142. replicative senescence in cultured rat oligodendrocyte precursor Yen PM, Feng X, Flamant F, Chen Y, Walker RL, Weiss RE, Chassande cells. Science 291 868–871. O, Samarut J, Refetoff S & Meltzer PS 2003 Effects of ligand and Tekki-Kessaris N, Woodruff R, Hall AC, Gaffield W, Kimura S, Stiles CD, thyroid hormone receptor isoforms on hepatic gene expression Rowitch DH & Richardson WD 2001 Hedgehog-dependent profiles of thyroid hormone receptor knockout mice. EMBO Reports oligodendrocyte lineage specification in the telencephalon. 4 581–587. Development 128 2545–2554. Zhang Q, Fukuda M, Van Bockstaele E, Pascual O & Haydon PG 2004 Thompson CC 1996 Thyroid hormone-responsive genes in developing Synaptotagmin IV regulates glial glutamate release. PNAS 101 cerebellum include a novel synaptotagmin and a hairless homolog. 9441–9446. Journal of Neuroscience 16 7832–7840. Zoeller RT & Rovet J 2004 Timing of thyroid hormone action in the Thompson CC & Bottcher MC 1997 The product of a thyroid developing brain: clinical observations and experimental findings. hormone-responsive gene interacts with thyroid hormone receptor. Journal of Neuroendocrinology 16 809–818. PNAS 94 8527–8532. Ting JT, Kelley BG & Sullivan JM 2006 Synaptotagmin IV does not alter excitatory fast synaptic transmission or fusion pore kinetics in mammalian CNS neurons. Journal of Neuroscience 26 372–380. Trentin AG, Gomes FC, Lima FR & Neto VM 1998 Thyroid hormone Received in final form 16 April 2007 acting on astrocytes in culture. In Vitro Cellular and Developmental Accepted 28 April 2007 Biology. Animal 34 280–282. Made available online as an Accepted Preprint 2 May 2007

Journal of Molecular Endocrinology (2007) 39, 17–28 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/30/2021 11:32:40PM via free access