Conserved transcription factor binding sites suggest an activator basal promoter and a distal inhibitor in the galanin gene promoter in mouse ES cells

1

Q1

2

345

678910111213

14

15

16

17

18

33

34

35

36

37

38

39

40

41

42

43Q2

44

Gene xxx (2014) xxx–xxx

GENE-39429; No. of pages: 7; 4C:

Contents lists available at ScienceDirect

Gene

j ourna l homepage: www.e lsev ie r .com/ locate /gene

Conserved transcription factor binding sites suggest an activator basalpromoter and a distal inhibitor in the galanin gene promoter in mouseES cells

OF

Sayonara Gonzalez a,b,⁎, Renata Binato a,⁎, Letícia Guida b, André Luiz Mencalha a, Eliana Abdelhay a

a Instituto Nacional de Câncer, Centro de Transplante de Medula Óssea, RJ, Brazilb Departamento de Genética Médica, Instituto Fernandes Figueira, Fundação Oswaldo Cruz, RJ, Brazil

UAbbreviations: Bp, Base pair; ChIP, Chromatin immunoEDTA, Ethylenediaminetetraacetic acid; EMSA, ElectrophEmbryonic stem cells; HEPES, 4-(2-Hydroxyethyl)piperaKilobase; KCl, Potassium chloride; MEF, Mouse embryoMgCl2,Magnesium chloride;ml,Mililiter;mM,Milimolar;PCR, Polymerase chain reaction; PR, Proximal region; TF,borate/EDTA.⁎ Corresponding authors at: Praça da Cruz Vermelha

Laboratórios do CEMO, INCA, Rio de Janeiro, RJ CEP 203207 1874.

E-mail addresses: [email protected] (S. [email protected] (R. Binato).

0378-1119/$ – see front matter © 2014 Published by Elsehttp://dx.doi.org/10.1016/j.gene.2014.01.059

Please cite this article as: Gonzalez, S., et al., Cin the galanin gene promoter in mouse ES ce

O

a b s t r a c t
a r t i c l e i n f o
19

20

21

22

23

24

Article history:Accepted 21 January 2014Available online xxxx

Keywords:ES cellsGalanin promoterCREBHOXPAXSP1Transcriptional activation

25

26

27

28

CTED P

R

Galanin and its receptors have been shown to be expressed in undifferentiatedmouse embryonic stem (ES) cellsthrough transcriptome and proteomic analyses. Although transcriptional regulation of galanin has beenextensively studied, the regulatory proteins that mediate galanin expression in mouse ES cells have not yetbeen determined. Through sequence alignments, we have found a high degree of similarity between mouseand human galanin upstream sequences at −146 bp/+69 bp (proximal region) and −2408 bp/−2186 bp(distal region). These regions could be recognized by ES cell nuclear proteins, and EMSA analysis suggests aspecific functionality. Analysis of the proximal region (PR) using EMSA and ChIP assays showed that the CREBprotein interacts with the galanin promoter both in vitro and in vivo. Additional EMSA analysis revealed that anSP1 consensus site mediated protein–DNA complex formation. Reporter assays showed that CREB is an activatorof galanin expression and works cooperatively with SP1. Furthermore, analysis of the distal region (DR) usingEMSA assays demonstrated that both HOX-F and PAX 4/6 consensus sites mediated protein–DNA complexformation, and both sites inhibited luciferase activity in reporter assays. These data together suggest that CREand SP1 act as activators at the basal promoter, while HOX-F and PAX 4/6 act as silencers of transcription. Theinterplay of these transcription factors (TF) may drive regulated galanin expression in mouse ES cells.

© 2014 Published by Elsevier B.V.

293032

31

E

45

46

47

48

49Q3

50

51

52

53

54

55

56

57

NCO

RR1. Introduction

Galanin is a highly conserved 29 (30 in human) amino acid peptidethat plays a neuromodulatory role, in addition to an important trophicrole, in neuronal tissues after injury and disease (Lang et al., 2007).Galanin has also been suggested to have a biological activity inprogenitor or stem cells from both mesodermal and ectodermal origin(Louridas et al., 2009), and has been found through transcriptome andproteomic analyses in several lineages of undifferentiated human andmouse ES cells (Anisimov et al., 2002; Ramalho-Santos et al., 2002;Sato et al., 2003). To accomplish these functions, galanin signals throughthree receptors: GalR1 (Habert-Ortoli et al., 1994), GalR2 (Howard et al.,1997; Smith et al., 1997), and GalR3 (Wang et al., 1997).

58

59

60

61

62

63

64

65

66

67

68

precipitation; DR, Distal region;oretic mobility shift assay; ES,zine-1-ethanesulfonic acid; kb,nic fibroblasts; mg, Miligram;PBS, Phosphate-buffered saline;Transcription factor; TBE, Tris/

23, 6°andar ALA C, Divisão de.230-130, Brazil. Tel.: +55 21

), [email protected],

vier B.V.

onserved transcription factorlls, Gene (2014), http://dx.do

Transcriptional in vivo and in vitro studies of the galanin gene fromdifferent organisms have made it possible to identify a plethora ofregulatory elements at the basal promoter, such as NRE, AP1, STAT,CRE, SP1 and ERE, that activate galanin expression in several cell typesor tissues (Kofler et al., 1996; Rökaeus et al., 1998). Corness et al.(1997) found a CRE element at the proximal promoter and a repressorelement localized at −2.2 kb and −1.4 kb of the rat galanin gene thatis active in primary sensory neurons in culture. In another study,Bacon et al. (2007) found ETS, STAT and Bicoid consensus sites localizedat −1.9 kb of the mouse galanin gene that direct galanin expressionafter axotomy. Additionally, bovine galanin promoter sequencesspanning from 5 kb or 131 bp were studied in human neuroblastomaSH-SY5Y cells and in transgenic mice, and the presence of silencer andenhancer sequences was revealed (Rökaeus et al., 1998). However,despite the large amount of data addressing galanin activation,functional transcriptional regulators of mouse galanin expression in EScells have not yet been determined.

In the present study, we have found, through sequence alignment, ahigh degree of conservation between mouse and human galaninupstream sequences, located at −146 bp/+69 bp of the proximalregion (PR) and −2408 bp/−2186 bp of the distal region (DR). Inboth regions, therewas conservation of transcription factor (TF) bindingsites including HOX-F and PAX 4/6 in the DR, and SP1 and CRE in the PR.By analyzing the proximal region, we showed through EMSA and ChIP

binding sites suggest an activator basal promoter and a distal inhibitori.org/10.1016/j.gene.2014.01.059

http://dx.doi.org/10.1016/j.gene.2014.01.059

mailto:[email protected]




http://www.sciencedirect.com/science/journal/03781119


T

O

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

2 S. Gonzalez et al. / Gene xxx (2014) xxx–xxx

UNCO

RREC

assays that CREB proteins interact with the galanin gene promoterin vitro and in vivo. Further EMSA analysis using SP1, HOX-F and PAX4/6 consensus sites showed that all of these sequences also have theability to bind the nuclear proteins of ES cells. When the cooperativerole of these sites was investigated through transfection assays, weobserved that SP1 and CRE, using proximal constructs, activatedluciferase signal cooperatively, while using distal constructs, HOX-Fand PAX 4/6 exerted an inhibitory action on the luciferase signal.

These data together suggest that CRE and SP1 at the basal promoteract as activators, while HOX-F and PAX 4/6 act as silencers of galaninexpression in mouse ES cells.

2. Materials and methods

2.1. Computer analysis of the promoter region

Approximately 3 kb of the upstream sequence of the mouse(L11144.1) and human (L38575.1) galanin genes was obtained fromGenBank (www.ncbi.nlm.nih.gov.br). Conservation between bothspecies was assessed through the sequence alignment program FASTAVIRGINIA (http://fasta.bioch.virginia.edu/fasta_www2/fasta_list2.shtml),and conservation of TF binding sites was assessed using online databanksMatInspector (www.genomatix.de), rVista 2.0 (http://rvista.dcode.org)and IFTI (www.ifti.org).

2.2. Cells and culture conditions

USP-1, a murine ES cell line (Sukoyan et al., 2002) was kindlyprovided by Dr. Lygia Pereira (Universidade de São Paulo) and culturedas follows: ES cells were maintained on a feeder layer of irradiatedMEFs (murine embryonic fibroblast) in DMEM high glucose media(Invitrogen) supplemented with 15% fetal bovine serum (Hyclone),100 mM nonessential amino acids (Invitrogen), 2 mM glutamine(Invitrogen), 100 U of penicillin/ml (Invitrogen), 100 mg/ml of strepto-mycin (Invitrogen), 0.55 μM β-mercaptoethanol (Sigma) and 1000 U/mlof leukemia inhibitory factor (LIF; Chemicon) on 0.2% gelatin-coatedplates, at 37 °C in a humidified chamber with 5% CO2. The mediumwas changed 3 h before splitting, and the cells were rinsed twice with1X phosphate-buffered saline (PBS), treated with 0.25% trypsin/0.5mM EDTA for 5 min and split 1:3 to be maintained in culture or usedfor preparation of protein nuclear extracts or transfection assays afterremoval of feeder cells. Feeders were removed by replating into a newgelatin-coated dish for propagation. These ES cells without MEF cellswere expanded 1–2 passages in 50% MEF conditioned media and 50%fresh ES cell media.

2.3. Preparation of nuclear extract

For preparation of nuclear protein extracts, 1 × 107 ES cells withoutMEF cellswere rinsed twicewith PBS 1X and treatedwith 0.25% trypsin/0.5 mM EDTA. The material was centrifuged and the nuclear proteinswere extracted as described by Dignam et al. (1983). Protein concentra-tions were determined by the Bradford method (Bradford, 1976).

2.4. Electrophoretic mobility shift assays (EMSA)

For EMSA assays, we used probes from PCR synthesized fragmentsand double-stranded (ds) DNA oligonucleotides, both end-labeledwith [γ-32P] ATP (Amersham, Pharmacia) using T4 polynucleotidekinase (Invitrogen). The primer sequences used to generate PCR frag-ment probes corresponding to the PR and DR were as follows: PRsense5′-AGACTGTGGGTGATC CTCTC-3′, PRantisense 5′-CTGGATGGTCGCTTACTG-3′, DR sense 5′-GCTTTGTGTGCTGTGTCCATTACT-3′, DRantisense5′-CCCATA TCACTGACAGATTCGCTCC-3′. The following primers (andtheir complements) were annealed to generate dsDNA oligonucleotideprobes: SP1-sense 5′-CAGGAGGC GGCGCTGAGCGG-3′, CREB-sense 5′-

Please cite this article as: Gonzalez, S., et al., Conserved transcription factorin the galanin gene promoter in mouse ES cells, Gene (2014), http://dx.do

F

AGCCGGTGACGCGGCAGCTCC-3′, CREB -Mut-sense 5′-AGCCGGTGAAGCGGCAGCTCC-3′, HOX-F-sense 5′-ATTACTCTA ATGGTGACCCATTTTG-3′, PAX-4/6-sense 5′-CTCCAGGCTGGGCTGCCTT-3′. Gel shift reactionswere performed in a binding buffer [25 mM HEPES (pH-7.6), 30 mMKCl, 5 mMMgCl2, 5% glycerol], containing 1 μg of poly (dIdC), 7–10 μgof nuclear extract proteins and 80,000 cpm of [γ-32P]-labeled PCRdsDNAprobes. The reactions occurred at 25 °C for 40 min andwere sub-sequently resolved on a 4% or 4.5% native polyacrylamide gel when PCRfragments or oligonucleotides were utilized as probes, respectively; in0.5 × TBE buffer (22.5 mM Tris/Borate/1 mM EDTA) for 4 h at 140 V.For competition reactions, a 100 molar excess of unlabeled competitoroligonucleotide or a mutated version of the specific consensus bindingsite was added 20 min before addition of the probe. For supershiftexperiments, 2 μg of anti-CREB1 (C-21) antibody (Santa Cruz Biotech-nology Inc.) was incubatedwith a nuclear protein extract 60 min beforeaddition of the specific probe. After addition of the specific probe, theincubation was continued for 40 min at 25 °C. The dried gels wereexposed on Kodak MS films on intensifying screens at −70 °C for 48 h.

ED P

RO2.5. Chromatin immunoprecipitation (ChIP) assay

The ChIP assays with anti-CREB antibody (Santa Cruz technologies)were performed according to the online protocol provided byAbcam Company (http://www.abcam.com/ps/pdf/protocols/N-ChIP.pdf). DNA extractions from bound fractions were performed followingthe Abcam (www.abcam.com) protocol. The immunoprecipitatedDNA was amplified for sequences containing binding sites by usingthe following CRE element sequence primers present in the mousegalanin gene promoter: ChIP-sense 5′-GGTCTGAGAC TGTGGGTGAT-3′;ChIP-antisense 5′-CTGCTGCCGCTATTTATG-3′. Quantification was eval-uated by RT-qPCR analysis. Immunoprecipitation of anti-tubulinantibody (Santa Cruz Technologies) was used as a negative control.

2.6. Reporter vector design

The galanin firefly luciferase reporter constructs were made bycloning fragments corresponding to the PR and DR of themouse galaninpromoter into the firefly PGL3-Basic vector (Promega) upstream ofthe luciferase reporter gene. The constructs p80, p60 and p60mutwere obtained by cloning PCR-amplified fragments corresponding to−78 bp/+357 bp and −60 bp/+357 bp into XhoI and HindIII restric-tion sites of pGL3-Basic. The primer sequences used for the PCR reac-tions were: p80-sense: 5′-ATATCTCGAGTCGCAGGAGGCGGCGCT-3′;p60-sense: 5′ATC-TCTCGAGAGCCGGTGACGCGGCAG-3′. The p60mutconstruct contained, in the forward primer, a C to A base substitution(underlined) at the CRE binding site to generate a mutant version ofthis element: p60mut-sense: 5′-ATCTCTCGAGAGCCGGTGAAGCGGCAG-3′. Reverse primer pPR-antisense 5′-GCCCGAAGCTTCATCTGGAAGGAAAAGTGG-3′ was used to generate all the products. Forward primershave a XhoI restriction site and reverse primers have aHindIII restrictionsite. To generate the p2400 and the p2300 constructs corresponding tothe DR, we first digested the p80 luciferase reporter construct with SacIand XhoI restriction enzymes. This digested construct was used for sub-sequent cloning of−2408 bp/−2186 bp and−2365 bp/−2186 bp PCRfragments from the mouse galanin gene separately. The primersequences used for the PCR reactions were: p2400 sense: 5′-GGAATGAGCTCCTTTGTGTGCTGTGTCC-3′ and p2300 sense: 5′-TGAATGAGC TCTAGCTCCACGCTGGGCTG-3′. Reverse primer pDR-antisense: 5′-GGCTGCTC GAGCCCATATCACTGACAGAT-3′ was usedto generate both products. Forward and reverse primers have SacIand XhoI restriction sites, respectively. All reporter pGL3 vectorsequences were confirmed by DNA sequencing at the AutomatedDNA Sequencing Facility of Instituto de Biofísica Carlos ChagasFilho-UFRJ.


http://www.ncbi.nlm.nih.gov.br)

http://fasta.bioch.virginia.edu/fasta_www2/fasta_list2.shtml)

http://www.genomatix.de)

http://rvista.dcode.org)

http://www.ifti.org)

http://www.abcam.com/ps/pdf/protocols/N-ChIP.pdf)

http://www.abcam.com/ps/pdf/protocols/N-ChIP.pdf)

http://www.abcam.com)


186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

3S. Gonzalez et al. / Gene xxx (2014) xxx–xxx

2.7. Transient transfection and luciferase reporter assay

ES cells (5 × 104) were plated into 24-well plates pretreated with0.2% gelatin and incubated for 4 h in ES cell media as describedpreviously. Then, the cells were transfected with 1.0 μg of either pGL3reporter vectors containing galanin promoter inserts or pGL3-basicvector. To normalize transfection efficiency, we co-transfected 0.01 μgof Renilla luciferase phRL-TK plasmid into each well. Transfection wascarried out using Lipofectamine LTX 2000 (Invitrogen) according tothe manufacturer's protocol. Firefly and Renilla luciferase activitieswere measured in cell lysates 48 h after transfection using the DualGloLuciferase Assay System (Promega) on a Veritas TM MicroplateLuminometer (Turner Biosystems), following the manufacturer'sprotocol. All experiments were performed in triplicate. Ratios of Renillaluciferase readings to firefly luciferase readings weremeasured for eachexperiment, and triplicates were averaged. The average values ofthe tested constructs were normalized to the activity of the emptypGL3-basic vector, which was arbitrarily set at a value of 1.

UNCO

RRECT

Fig. 1. Alignment of the mouse (M) and human (H) galanin promoter. (A) and (B) Distal regionfactor binding sites. (C) Schematic representation depicting the conserved binding sites of thescription start site. The dashed line indicates a 1.0 kb sequence insertion in the human promoter(E) EMSA assay of binding reactions using PCR probes (P), corresponding to the PR and DR, andof unlabeled wild type (Cwt) oligonucleotides. The arrows indicate the resulting DNA–protein


F

3. Results

3.1. In silico analysis of conserved regions within the galanin gene promoter

Evolutionary conservation of sequences can be analyzed throughsequence alignments of distantly related species, such as human andmouse. Because noncoding DNA has diverged between species, aconserved region in a noncoding sequence suggests the biologicalactivity of regulatory elements that have been under selective pressure.This computational approach, supplemented with experimentalanalysis, is required for the elucidation of cis-acting transcriptionalregulatory elements.

Initially, we performed a computational comparative analysis of 3 kb(upstream from the translation codon) of the human (GenBank accessionnumber L38575.1) and mouse (GenBank accession number L11144.1)galanin genes. The alignment of both sequences demonstrated a highdegree of similarity and conservation at position −146 bp/+69 bp andat −2408 bp/−2186 bp (Figs. 1A and B). Subsequent in silico analysis

ED P

RO

O

(DR) and proximal region (PR), respectively. Boxes demonstrate conserved transcriptionpromoter region of the human and mouse galanin gene. Positions are relative to the tran-. The arrows indicate the primer positions that generated the PCR fragment probe. (D) andES cell nuclear extracts (NE). Competition reactionswere performed using 100-fold excesscomplexes. DR—distal region, PR—proximal region.



219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282


of these two overlapping sequences revealed the conservation of putativebinding sites at the proximal region (PR) as reported previously(Kofler et al., 1996; Rökaeus et al., 1998). A CRE site, the cAMP responseelement that is recognized by CREB (CRE-binding protein), was found at−55 bp/−47 bp. CREB is a cellular transcription factor (TF) that is highlyexpressed in ES cells and regulates the transcription of numerous targetgenes (Bleckmann et al., 2002). We also found a binding site forSP1 at−75 bp/−66 bp, which is a TF implicated in the activation of a va-riety of genes involved in proliferation, differentiation and apoptosis(Black et al., 2001). At the distal region (DR), our analysis found HOX-Fconsensus sequence at−2389 bp/−2366 bp and PAX 4/6 consensus se-quence at−2360 bp/−2342 bp that are recognized by the products ofdevelopmental genes belonging to PAX and HOX family members(Gehring and Hiromi, 1986; Krumlauf, 1994; Wehr and Gruss, 1996).

It is interesting to note that between the PR and the DR, there is a1.0 kb sequence in the human promoter with no similarity to themouse gene, suggesting that there was a sequence insertion in thehuman genome during evolution. With the exception of this sequenceinsertion, the binding sites were conserved at the PR and DR of bothspecies, suggesting a selective pressure on these regulatory sequencesand possibly a functional role (Fig. 1C).

3.2. Binding of transcription factors to the galanin promoter

To investigate the functionality of the conserved binding sites of theproximal and distal regions of the galanin promoter, we carried outEMSA assays using PCR-amplified fragments covering the PR and DR,containing conserved putative TF binding sites and nuclear proteinsfrom ES cells. We observed the formation of three specific complexes(a, b, and c) and one nonspecific complex (d) on the PR (Fig. 1D). Com-petition analysis using 100-fold excess of unlabeled oligonucleotide as acompetitor showed that specific binding was abolished (Fig. 1D). EMSAassays of the DR showed the formation of one specific complex (Fig. 1E),

UNCO

RRECT

Fig. 2. Characterization of CREB binding to the galanin promoter. (A) EMSA analysis of binding rpetition was performed using 100-fold excess of unlabeled wild type (Cwt) or mutated (CMuassays were performed using CREB-1 antibody. Control reactions were prepared using nonspewith the galanin promoter in vivo. Q-PCR quantification of the CRE consensus sequence in ES cewith protein A nonspecific amplification. Normalized fractions were used to calculate the boun


ED P

RO

OF

and this binding was abolished when we used 100-fold excess of theunlabeled oligonucleotide as a competitor. These results reinforce theassumption of a functional role for conserved regions.

Within the PR, the CRE binding site has been found to be conservedthrough evolution in the rat, bovine, human and mouse galanin pro-moters. The bovine CRE binding site has been shown to be functionalin human SH-SY5Y cells and rat pheochromocytoma-derived PC12cells (Rökaeus et al., 1998). In addition, CREB protein has been shownto be highly expressed at the inner cell mass (ICM), a region of theblastocyst from which ES cells are derived (Bleckmann et al., 2002).

Because most of the findings addressing the regulation of galaninexpression were generated in other species, it is not possible to predictwhether the regulatory network of galanin transcription, as currentlydefined, also applies to mouse ES cells. To address this issue, weexamined the consensus sequences located at the PR. EMSA assaysusing an oligonucleotide containing a CRE consensus binding site andES cell nuclear proteins resulted in the formation of complexes. Whenwe used 100-fold molar excess of the unlabeled DNA-oligo containingthe CRE consensus binding site, the complex was completely abolished.Conversely, the complex was maintained when we used an unlabeledmutated analog as a competitor, confirming its specificity. Supershiftassays were performed using anti-CREB antibody and this assaydemonstrated that CREB was present in the protein complex formedwith the CRE consensus sequence (Fig. 2A).

To confirm the EMSA results, we performed ChIP assays. The ChIPassay allows in vivo analysis of nuclear protein–DNA interactions.Chromatin fractions bound to the CREB antibody in ES cells werequantified by RT-qPCR using primers to amplify the promoter regionthat contains the CRE consensus site. The results revealed that CREBwas bound to this site in vivo (Fig. 2B).

We next investigated whether the SP1 site was also shifted in EMSAanalysis. This SP1 site, similar to the CRE site, is also evolutionarilyconserved (Rökaeus et al., 1998). In our assays, nuclear proteins from

eactions using a CREB oligonucleotide as probe (P) and ES cell nuclear extracts (NE). Com-t) oligonucleotides. The arrows indicate the resulting DNA–protein complexes. Supershiftcific antibodies, anti c-Myc and anti cyclin-D. (B) ChIP analysis confirms CREB interactionlls. DNA amplification was quantified in bound and unbound fractions after normalizationd/input ratio.



283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325


ES cells were able to bind to the SP1 consensus sequence (Fig. 3). Com-petition analysis using 100-fold excess of unlabeled oligonucleotidesconfirmed the specificity of binding.

To investigate the consensus sequences located at the DR, we usedtwo [γ-32P]-labeled double-stranded DNA oligonucleotides, one con-taining the consensus sequence for HOX-F and the other containingthe consensus sequence for PAX 4/6, and ES cell nuclear extracts.When the HOX-F consensus sequence was used, we observed sixspecific complexes. (Fig. 3B, complexes a, b, c, d, e and f). These seemedto display different binding affinities because shifted bands wereof different intensities. After competition with 100-fold excess ofunlabeled oligonucleotide, complexes a, b, c, d and e disappearedcompletely, while complex f showed a significant reduction in intensity.

When PAX 4/6 consensus siteswere subjected to EMSA analysis, twocomplexes were formed, as indicated in a and b (Fig. 3C). Only complexa was completely abolished by the competition reaction with 100-foldexcess of the unlabeled oligonucleotide.

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

3.3. Transcriptional activity of the proximal region of the galanin genepromoter

To clarify the regulatory role of TF binding sites found by EMSAassays on mouse galanin gene expression, we created reporterconstructs and performed transfection assays in ES cells.

Initially, we created a P80 construct that contained CRE and SP1consensus sequences and cloned it into a pGL3 reporter vectorupstream of the firefly luciferase gene. This construct demonstratedhigh transcriptional activity in the reporter assay when comparedwith pGL3 (15-fold induction). When we deleted the SP1 bindingsite (P60 construct), we observed a 3.5-fold reduction of luciferaseexpression, suggesting a stimulatory role for this sequence on the

UNCO

RRECT

Fig. 3. Interactions of nuclear proteins from ES cells with the galanin gene promoter. EMSA anprobes (P) and ES cell nuclear extracts (NE). Competition reactionswere performed using 100-fDNA–protein complexes.


galanin promoter. When the plasmid containing a sequence alterationin the CRE consensus site (P60mut construct) was transfected into EScells, we observed a greater reduction of luciferase expression, similarto the control levels (Fig. 4A).

To evaluate the regulatory role of DR on basal galanin transcription,we cloned a PCR fragment containing HOX-F and PAX4/6 binding sitesinto the P80 construct upstream of the CRE and SP1 sites. When theresulting construct (P2400) was transfected into ES cells, the promoteractivity was reduced by 1.5× compared to basal promoter activity(P80) When we removed the HOX-F binding site and retained onlythe PAX4-6 upstream PR (P2300 construct), we observed a slightreduction in luciferase activity (Fig. 4B).

ED P

RO

OF4. Discussion

The galanin gene has been isolated from several species of verte-brates, including mammals, birds, reptiles and fish, and has been de-scribed to have wide-ranging biological effects, particularly within theneuroendocrine and central or peripheral nervous system. Studies ofthe bovine galanin promoter in human neuroblastoma SH-SY5Y cellsdemonstrated that 0.1 kb of the promoter conferred high basal expres-sion of the gene. In this region, a highly conserved cyclic AMP responseelement (CRE)-like sequence between−66 and−44 bp of the galaninpromoter was found (Rökaeus et al., 1998). The same group showedthat 131 bp of the bovine galanin promoter, including the CRE consen-sus binding site, is sufficient to induce high basal activity in SH-SY5Ycells. However, the presence of upstream silencers was detected in theregion between 451bp and5 kb. The samewas observed in human cho-riocarcinoma cells and rat pheochromocytoma cells, but not in humanbreast cancer cells, indicating that the repressor element present inthe galanin gene promoter could act in a tissue-specific manner

alysis of binding reactions using (A) SP1, (B) HOX-F and (C) PAX4-6 oligonucleotides asold excess of unlabeledwild type (Cwt) oligonucleotides. The arrows indicate the resulting



ECTED P

RO

OF

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

Fig. 4. Functional analysis of the mouse galanin promoter. (A) Reporter activity of P80, P60 and P60Mut constructs in transiently transfected ES cells. The histogram shows induction ofluciferase activity when we used the construct p80 (with SP1 and CRE consensus binding sites) and the levels of luciferase activity decreased when we used the construction withoutthe SP1 consensus binding site, indicating cooperative activity among the 2 binding sites to increase transcriptional activation. (B) Reporter activity of p2400 and p2300 constructs intransiently transfected ES cells. The histogram shows reduction of luciferase activity when we used both constructions p2400 and p2300 (the first containing PAX4-6 and HOXconsensus binding sites and the second with only the PAX4-6 consensus binding site) in the presence of the basal promoter. Luciferase activity was measured and normalized to Renillalevels. P80—construct containing CRE and SP1 consensus sequences. P60—construct containing the CRE consensus sequence. P60Mut—construct containing an alteration of the CREconsensus sequence. P2400—construct containing HOX and PAX4-6 consensus sequences. P2300—construct containing PAX4-6 consensus sequence. LUC—empty vector.


UNCO

RR(Rökaeus et al., 1998). Although described as a neuropeptide, galanin

expression was also identified in mouse embryonic stem (ES) cellsthrough transcriptome analysis as one of themost abundant transcripts,indicating that galanin may play an important role in these cells.Tarasov et al. have suggested that galanin could affect pathwaysinvolved in regulating cell number, and this effect may be modulatedby LIF (Anisimov et al., 2002; Tarasov et al., 2002).

In light of this, understanding how galanin is regulated in ES cells be-comes important, as this gene has an effect on the growth of these cells.

To identify the putative cis-acting elements responsible for mousegalanin gene expression,we compared approximately 3 kb of upstreamsequences from mouse and human galanin genes. This comparisonshowed two regions of similarities between these two species, locatedat 146 bp/+69 bp (proximal region) and at −2408 bp/−2186 bp(distal region). Further in silico analysis revealed the conservation ofseveral transcription factor binding sites, including SP1 and CRE, at theproximal region, and HOX-F and PAX4/6 at the distal region.

The CRE binding site has been found to be evolutionarily con-served. The bovine CRE binding site has been shown to bind PMA(phorbol-12-myristate-13-acetate)-inducible human proteins fromSH-SY5Y cells and also proteins from bovine chromaffin cells.Additionally, in unstimulated conditions, without PMA administra-tion, it has been demonstrated that CRE on the bovine promoterconfers high basal galanin expression in SH-SY5Y cells and ratpheochromocytoma-derived PC12 cells (Rökaeus et al., 1998).


As the galanin gene has been described to play an important role inthe biological activity of stem cells, we investigated whether the con-served regions of the galanin gene were functional in mouse ES cells.First, we explored whether nuclear proteins could bind to the proximalregion through EMSA analysis. We found that this region was bound byES cell nuclear proteins in a specific manner, reinforcing the possibilityof their functionality. To further dissect the proximal region, weexplored protein binding to CRE consensus sites in EMSA assays. Wefound that nuclear proteins isolated from ES cells could bind to thissequence, and this binding disappeared when using an unlabeledprobe containing CRE sequence. Conversely, when we used a mutatedCRE site in competition assays, the protein binding could not be compet-ed away, confirming the specificity of the complex. The presence ofCREB protein in the complex was verified by supershift assays. Anantibody against CREB-1 protein inhibited the formation of thesecomplexes, indicating that CREB-1 binds to the mouse galaninpromoter. Correspondingly, when we performed a ChIP assay using EScells, immunoprecipitation with anti-CREB antibody, followed by theamplification of genomic DNA surrounding the CRE site in the galaninproximal promoter, revealed that CREB was also bound to the galaninpromoter in vivo.

Additional EMSA assays showed that an SP1 site was also shifted inEMSA analysis, and this binding disappeared when we competed withunlabeled probe containing SP1 sequence. The SP1 site, similar tothe CRE site, is also evolutionarily conserved and is arranged in an



T

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457Q4

458

459

460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529

530


UNCO

RREC

overlapping manner with the NGF responsive element (NRE) at thegalanin promoter. It has been shown that administration of NGF caninduce the activation of the bovine galanin gene in PC12 cells throughthis NRE site. Interestingly, this sequence can also bind nuclear proteinsin unstimulated basal conditions (Rökaeus et al., 1998).

The study of the distal region identified protein binding to an HOX-Fconsensus sequence, resulting in six specific complexes that seemed todisplay different binding affinities because shifted bands showed differ-ent intensities. After competition with the unlabeled oligonucleotide, 5of the 6 complexes disappeared completely, while one complex showeda significant reduction in intensity but did not disappear. Differences inbinding activities may be a result of diverse complexes forming at theHOX-F site. This site is the general recognition site for HOX genes, aset of homeodomain TFs that in the vertebrate genome are organizedin four clusters composed of 13 subgroups, called paralog groups(Sharkey et al., 1997). The HOX-F consensus site is recognized by HOXparalogs 1–8, belonging to the four HOX clusters (matrix family assign-ment—www.genomatix.de). When the PAX 4/6 consensus sites weresubjected to EMSA analysis, a specific complex was formed.

Tounderstand the regulatory role of elements in the proximal regionon galanin expression, deletional and mutational analysis of galanin re-porter constructs was performed in ES cells. The reporter vector bearingCRE and SP1 consensus sequences demonstrated high transcriptionalactivity. When we removed the SP1 binding site, we observed a 3.5-fold reduction of luciferase expression, suggesting a stimulatory rolefor this sequence on the galanin promoter. When the plasmid contain-ing a sequence alteration to the CRE consensus site was transfectedinto ES cells, we observed a greater reduction in luciferase expression,similar to the control levels (Fig. 4). The importance of SP1 and CREbinding sites to galanin gene expression has been shown for the bovinegalanin gene in induced and non-induced processes, as discussedearlier. Their relevance is also supported by the observation thatinteractions among SP1 and CRE binding sites have been reported inthe literature for other promoters and may be a common signaturethat drives the expression of several genes (Piera-Velazquez et al.,2007; Xia et al., 2008). It is possible that the same mechanism mayoperate to regulate mouse galanin in ES cells.

To evaluate the regulatory role of DR on basal galanin transcription,we cloned a PCR fragment containing HOX-F and PAX4/6 binding sitesinto the construct containing CRE and SP1 sites. When the resultingconstruct was transfected into ES cells, the promoter activity wasreduced by 1.5×. When we removed the HOX-F binding site andretained only the PAX4-6 site, we observed a slight reduction inluciferase activity. Pax and Hox genes are developmentally regulatedand can activate or repress several genes, mainly regulating celldifferentiation. It has also been postulated that Hox genes are crucialto tissue-specific stem cells, specifically for self-renewal, tissue specific-ity and quiescence (Shah and Sukumar, 2010). Expression profiles of EScells have been rigorously investigated because of the therapeuticpossibilities of these cells. Expression of Pax and Hox genes has beendocumented (Baumann et al., 2003; Ramalho-Santos et al., 2002;Sharkey et al., 1997), but their roles have not yet been elucidated.Because galanin has been suggested to be involved in the maintenanceof ES cells in an undifferentiated and proliferative state (Tarasov et al.,2002), we postulate that Pax and Hox genes may suppress galaninexpression in mouse ES cells and, as a result, these cells can follow theappropriate path of differentiation.

In the present work, our in silico analysis, together with themolecular studies of the galanin gene in mouse ES cells, suggested thatboth SP1 and CREB proteins may serve as important transcriptionfactors to activate galanin basal promoter activity. However, HOX-Fand PAX4-6 binding sites, which are conserved in the distal region,may interact with homeobox and paired box family members and


ED P

RO

OF

may function to downregulate galanin gene expression. Binding of spe-cific combinations of these trans-activating factors is likely necessary forappropriate galanin regulation in mouse ES cells.

Acknowledgment

This work was supported by grants from Convênio INCA/FIOCRUZ.

References

Anisimov, S.V., Tarasov, K.V., Tweedie, D., Stern, M.D., Wobus, A.M., Boheler, K.R., 2002.SAGE identification of gene transcripts with profiles unique to pluripotent mouseR1 embryonic stem cells. Genomics 79, 169–176.

Bacon, A., Kerr, N.C., Holmes, F.E., Gaston, K., Wynick, D., 2007. Characterization of an en-hancer region of the galanin gene that directs expression to the dorsal root ganglionand confers responsiveness to axotomy. J. Neurosci. 27, 6573–6580.

Baumann, S., et al., 2003. Expression of regulatory genes Oct-4, Pax-6, Prox-1, Ptx-2 at theinitial stages of differentiation of embryonic stem cells in vitro. Ontogenez 34,174–182.

Black, A.R., Black, J.D., Azizkhan-Clifford, J., 2001. Sp1 and krüppel-like factor family oftranscription factors in cell growth regulation and cancer. J. Cell. Physiol. 188,143–160.

Bleckmann, S.C., Blendy, J.A., Rudolph, D., Monaghan, A.P., Schmid, W., Schütz, G., 2002.Activating transcription factor 1 and CREB are important for cell survival duringearly mouse development. Mol. Cell. Biol. 22, 1919–1925.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgramquantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.72, 248–254.

Corness, J.D., Burbach, J.P., Hökfelt, T., 1997. The rat galanin-gene promoter: response tomembers of the nuclear hormone receptor family, phorbol ester and forskolin.Brain Res. Mol. Brain Res. 47, 11–23.

Dignam, J.D., Lebovitz, R.M., Roeder, R.G., 1983. Accurate transcription initiation by RNApolymerase II in a soluble extract from isolated mammalian nuclei. Nucleic AcidsRes. 11, 1475–1489.

Gehring, W.J., Hiromi, Y., 1986. Homeotic genes and the homeobox. Annu. Rev. Genet. 20,147–173.

Habert-Ortoli, E., Amiranoff, B., Loquet, I., Laburthe, M., Mayaux, J.F., 1994. Molecular clon-ing of a functional human galanin receptor. Proc. Natl. Acad. Sci. 91, 9780–9783.

Howard, A.D., et al., 1997. Molecular cloning and characterization of a new receptor forgalanin. FEBS Lett. 405, 285–290.

Kofler, B., Liu, M.L., Jacoby, A.S., Shine, J., Iismaa, T.P., 1996. Molecular cloning and charac-terisation of the mouse preprogalanin gene. Gene 182, 71–75.

Krumlauf, R., 1994. Hox genes in vertebrate development. Cell 78, 191–201.Lang, R., Gundlach, A.L., Kofler, B., 2007. The galanin peptide family: receptor pharmacol-

ogy, pleiotropic biological actions, and implications in health and disease. Pharmacol.Ther. 115, 177–207.

Louridas, M., Letourneau, S., Lautatzis, M.E., Vrontakis, M., 2009. Galanin is highlyexpressed in bone marrow mesenchymal stem cells and facilitates migration ofcells both in vitro and in vivo. Biochem. Biophys. Res. Commun. 390, 867–871.

Piera-Velazquez, S., Hawkins, D.F., Whitecavage, M.K., Colter, D.C., Stokes, D.G., Jimenez,S.A., 2007. Regulation of the human SOX9 promoter by Sp1 and CREB. Exp. Cell Res.313, 1069–1079.

Ramalho-Santos, M., Yoon, S., Matsuzaki, Y., Mulligan, R.C., Melton, D.A., 2002."Stemness": transcriptional profiling of embryonic and adult stem cells. Science298, 597–600.

Rökaeus, A., Jiang, K., Spyrou, G., Waschek, J.A., 1998. Transcriptional control of the galaningene. Tissue-specific expression and induction by NGF, protein kinase C, and estro-gen. Ann. N. Y. Acad. Sci. 863, 1–13.

Sato, N., Sanjuan, I.M., Heke, M., Uchida, M., Naef, F., Brivanlou, A.H., 2003. Molecular sig-nature of human embryonic stem cells and its comparison with the mouse. Dev. Biol.260, 404–413.

Shah, N., Sukumar, S., 2010. The Hox genes and their roles in oncogenesis. Nat. Rev. Can-cer 10, 361–371.

Sharkey, M., Graba, Y., Scott, M.P., 1997. Hox genes in evolution: protein surfaces andparalog groups. Trends Genet. 13, 145–151.

Smith, K.E., et al., 1997. Expression cloning of a rat hypothalamic galanin receptor coupledto phosphoinositide turnover. J. Biol. Chem. 272, 24612–24616.

Sukoyan, M.A., Kerkis, A.Y., Kerkis, I.E., Mello, M.R.B., Visintin, J.A., Pereira, L.V., 2002. Es-tablishment of new murine embryonic stem (ES) cell lines for the generation ofmouse models for human genetic diseases. Braz. J. Med. Biol. Res. 35, 535–542.

Tarasov, K.V., Tarasova, Y.S., Crider, D.G., Anisimov, S.V., Wobus, A.M., Boheler, K.R., 2002.Galanin and galanin receptors in embryonic stem cells: accidental or essential? Neu-ropeptides 36, 239–245.

Wang, S., He, C., Hashemi, T., Bayne, M., 1997. Cloning and expressional characteriza-tion of a novel galanin receptor: identification of different pharmacophoreswithin galanin for the three galanin receptor subtypes. J. Biol. Chem. 272,31949–31952.

Wehr, R., Gruss, P., 1996. Pax and vertebrate development. Int. J. Dev. Biol. 40, 369–377.Xia, Y., et al., 2008. Sp1 and CREB regulate basal transcription of the human SNF2L gene.

Biochem. Biophys. Res. Commun. 368, 438–444.


http://www.genomatix.de)


Documents

Conserved transcription factor binding sites suggest an activator basal promoter and a distal inhibitor in the galanin gene promoter in mouse ES cells