ANb

BSA

a

ARRAA

KOBGPN

1

dwm

A1

sms(t

L

0h

Animal Reproduction Science 135 (2012) 18– 24

Contents lists available at SciVerse ScienceDirect

Animal Reproduction Science

journa l h omepa g e: www.elsev ier .com/ locate /an i reprosc i

reporter promoter assay confirmed the role of a distal promoterOBOX binding element in enhancing expression of GDF9 gene inuffalo oocytes

haskar Roy1, Sandeep Rajput1, Sarvesh Raghav, Parveen Kumar2, Arpana Verma,andeep Kumar, Sachinandan De, Surender Lal Goswami, Tirtha Kumar Datta ∗

nimal Genomics Lab, ABTC, NDRI, Karnal 132001, India

r t i c l e i n f o

rticle history:eceived 5 June 2011eceived in revised form 28 August 2012ccepted 13 September 2012vailable online 29 September 2012

eywords:ocytesuffalo

a b s t r a c t

Growth differentiation factor 9 is primarily expressed in oocytes and plays a vital role inoocyte cumulus crosstalk. Earlier studies with buffalo oocytes revealed differential expres-sion of this gene under different media stimulation conditions which, in turn, are correlatedwith the blastocyst yield. In this study, different germ cell specific cis elements includinga NOBOX binding elements (NBE) and several E-boxes were identified at the 5′ upstreamregion of buffalo GDF9 gene and their potential role in GDF9 expression was investigated.Transfecting oocytes with GDF9 promoter deletion constructs harbouring the NBE reportergene revealed a 33% increase in GFP as well as the luciferase signal signifying its role in

DF9romoterOBOX element

stimulating the minimal promoter activity of GDF9 in buffalo oocytes. Site directed muta-tion of core binding nucleotides at NBE at 1.8 kb upstream to TSS further confirmed itsrole for enhancing the basal transcriptional activity of GDF9 promoter in buffalo oocytes.Current work will provide important leads for understanding the role of GDF9 in oocytescompetence and designing a more physiological IVF protocol in case of buffalo.

. Introduction

Indian river buffaloes (Bubalus bubalis) are the major

airy animals in India that produce about two-third of theorld’s buffalo milk and nearly half of the world’s buffaloeat (FAOSTAT, 2005). Regardless of their superior milk

∗ Corresponding author at: Animal Biotechnology, Genomics Lab,nimal Biotechnology Centre, National Dairy Research Institute, Karnal32001, Haryana, India. Tel.: +91 184 2259506; fax: +91 184 2250042.

E-mail addresses: bhaskar.biotech@gmail.com (B. Roy),rajputbt@gmail.com (S. Rajput), sarveshraghav@gmail.com (S. Raghav),eparveen@gmail.com (P. Kumar), arpanandri@gmail.com (A. Verma),

andeepndribiotech@gmail.com (S. Kumar), sachinandan@gmail.comS. De), surender.goswami@gmail.com (S.L. Goswami),irthadatta@gmail.com (T.K. Datta).

1 Authors have equal contribution.2 Present address: Department of Hematology, BMC, Lund University,

und 22184, Sweden.

378-4320/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.anireprosci.2012.09.006

© 2012 Elsevier B.V. All rights reserved.

producing ability, these animals however, suffer from sev-eral reproductive constraints making them low in repro-ductive efficiency (Nandi et al., 2002). As an aid to augmentreproduction rate in the buffalo, several assisted reproduc-tion technologies (ARTs) have been tried, but with limitedsuccess (Singh et al., 2009). Intrinsically, to understand thepeculiarities of buffalo reproduction and for standardiz-ing optimum ART protocols in this species, developmentpotential of oocytes is considered to be crucial as they ulti-mately contribute to the health of future embryos (Sirardet al., 2006). Oocyte to cumulus paracrine regulatory loopplay an important role in the acquisition of oocyte develop-mental competence and subsequent embryogenesis. GDF9plays a direct role in this bi-directional communication

process (Kidder and Vanderhyden, 2010). Previous stud-ies have demonstrated the ability of GDF9 to mimic oocytefunctions in oocytectomised COCs, including regulatingcumulus cell progesterone synthesis, suppression of LH

duction Science 135 (2012) 18– 24 19

Table 1Overlapping primer sets used for PCR amplification of buffalo GDF9 5′

upstream region.

Primer name Primer sequence 5′–3′

Set 1forward GATGAATGTGCAGACCCAAASet 1 reverse AGAGAGGGGAAAGGAGACCASet 2 forward CGCGAGGTTTTGTACTGTCASet 2 reverse GCCTTTATTTGGGTCTGCACSet 3 forward GTTCAACGGCATTCAGGACTSet 3 reverse CTACACCCTGCCAGTGGATTSet 4 forward CCACACTCCTTCCCAAACTC

B. Roy et al. / Animal Repro

receptor expression, promoting granulosa cell growth andregulating cumulus expansion (Yeo et al., 2008). Reduceddevelopmental competence of IVM oocytes have also beenattributed to inappropriate levels or composition of oocytesecreted GDF9 (Hussein et al., 2006). Recent work in ourlaboratory has established that the magnitude of GDF9expression varies in buffalo oocytes under different in vitromaturation conditions which eventually is correlated withblastocyst development rate (Jain et al., 2012). In spite ofthe growing importance of GDF9 however, information onits expression regulation in oocytes have been reportedonly recently. Considering the importance of buffalo andthe necessity of designing potent ART protocols for thisspecies, the current study aimed to examine the expressioncontrol mechanism of GDF9 by analysing the 5′ upstreamregion of buffalo GDF9 gene and characterizing its putativecis-acting regulatory elements.

2. Materials and methods

All media and chemicals were procured fromSigma–Aldrich, St. Louis, MO, USA unless otherwiseindicated. Disposable plastic wares used were fromThermo Fisher, Rochester, NY, USA.

2.1. In vitro maturation of oocytes

Buffalo ovaries were collected from an abattoir andtransported to the laboratory in pre-warmed normal saline(35–37 ◦C) supplemented with antibiotics (streptomycinsulphate 100 �g/ml and penicillin 100 U/ml). Cumulusoocyte complexes (COCs) were collected within 4–6 h afterthe slaughter of animals by aspiration of visible folliclesin oocyte collection medium (Dulbecco’s phosphate buffersaline supplemented with 0.01% l-glutamine, 0.4% BSA,100 �g/ml streptomycin sulphate, 100 U/ml penicillin andsodium pyruvate 36 �g/ml). The COCs were selected basedon their morphological characteristics and only excel-lent grade oocytes with more than 5 layers of cumulusmass were selected for in vitro maturation. A pool of 25COCs were placed for in vitro maturation in maturationmedium (TCM-199 with 10% FBS, 0.005% streptomycin,0.01% sodium pyruvate and 0.005% glutamine) supple-

mented with 0.5 mg/ml each of FSH and LH and 50 ng/mlEGF. Maturation drops (100 �l) were prepared in advance,overlaid with mineral oil and equilibrated for at least 2 hbefore putting the oocytes into them. Oocytes were then

ECR5 Enh ancer2

Con Reg

Con sv

Reg io

n 3

Con sv

Region

4

Con served

Region 5

Enh ancer3

5’

(-808 bp)(-1195 bp)(-1349 bp)(-1725 bp)(-2194 bp)

DelCon3 Forward Primer

SDM Primers

DelCon2 Forward Primer

DelCoP

Fig. 1. Putative map of buffalo GDF9 5′ upstream sequence and identified consdeletion constructs and site directed mutants. CR: conserved region, ECR: extende

Set 4 reverse CCCATCCACAGAGGGTATTGSet 5 forward AGCCCTGTGGCTTTGTCTTASet 5 reverse TTGCTGAACCATTTTGGCTA

allowed to mature for 24 h at 38.5 ◦C in an atmosphere of5% CO2 (v/v).

2.2. Isolation of buffalo GDF9 5′ upstream sequence

A standard phenol chloroform extraction method wasused to isolate genomic DNA from peripheral blood leuko-cytes (Sambrook and Russell, 2001). The primer walkingapproach was used for amplification of the 5′ flankingregion of GDF9 gene involving five sets of overlappingPCR primers (Table 1). The overall amplification strategyand primer scheme is described in Fig. 1. DNA sequencesfrom related species were analysed for homology align-ment and primers were designed at consensus sequencesaccordingly. PCR was performed using Pfx DNA polymerase(Invitrogen, USA) under the cycling parameters of firstcycle at 95 ◦C for 2 min, 31 cycles at 95 ◦C for 15 s, 58 ◦Cfor 30 s and 68 ◦C for 1 min and finally extension at 72 ◦Cfor 8 min using the thermal cycler (Eppendorf, USA). ThePCR products were purified using QIAquick gel extractionkit (Qiagen, USA) and sequenced. The targeted 4244 bpsequence upstream from TSS was reconstructed by align-ing the overlapping region of the five individual fragmentsamplified by PCR.

2.3. Analysis of sequence for transcription factor bindingsite prediction

The GDF9 5′ flanking sequences of cow (ENSB-

TAG00000009478), sheep (AF078545), goat (EF446168)and mouse (ENSMUSG00000018238) were retrieved fromEnsemble Genome Browser (http://www.ensembl.org/index.html) and NCBI database (http://www.ncbi.nlm.

Minimal Promoter ElementEnh ancer1

CR1

served

ion 2

TSS

(-246 bp)

ATG

n1 Forward rimer

DelCon Reverse Primer

EXON1

erved regions. Arrows indicate the primer locations for generating thed conserved region, MP: minimal promoter, SDM: site directed mutation.

20 B. Roy et al. / Animal Reproduction Science 135 (2012) 18– 24

Table 2Identified putative and common germ cell specific transcription factor binding sites on 5′ upstream sequence of buffalo GDF9 gene.

TFs Description Position Sequence Strand

LHXF Lim homeodomain factors −3638 to −3622 tagttgtaTAATttggg (+)HOMF Homeodomain transcription factors −3639 to −3623 ctagttgtaTAATttgg (+)PARF PAR/bZIP family −3620 to −3604 acaggttaaGTAAtttg (−)HOMF Homeodomain transcription factors −3480 to −3464 ctgacacagTAATtact (+)MAZF Myc associated zinc fingers −3294 to −3282 tgggGAGGggagt (+)EBOX E-box binding factors −2980 to −2968 aaggCACAtggat (+)RXRF RXR heterodimer binding sites −2621 to −2597 gaaagaagtgggtaaAGGTgacctc (+)PERO Peroxisome proliferator-activated receptor −2620 to −2598 aaagaagtgggtaaAGGTgacct (+)YBXF Y-box binding transcription factors −2565 to −2553 atgacTGGCtcaa (−)CAAT CCAAT binding factors −2564 to −2550 tgagCCAGtcatttc (+)GCNR Germ cell nuclear receptors −2573 to −2555 gactggcTCAAggtcactt (−)EBOX E-box binding factors −2265 to −2253 tgatcaCATGccg (+)NFKB Nuclear factor kappa B/c-rel −2168 to −2156 ccGGGActtgccg (+)CREB cAMP-responsive element binding proteins −2164 to −2144 tcgggtTGACggcggcaagtc (−)EBOX E-box binding factors −1884 to −1872 ccgccgCGTGacg (+)SP1F GC-Box factors SP1/GC −1869 to −1855 aagtgGGCGgtgcct (+)MAZF Myc associated zinc fingers −1850 to −1838 tggggtGGGGcca (+)EBOX E-box binding factors −1843 to −1831 gggccaAGCGcct (+)EV1 EVI1-myleoid transforming protein −1522 to −1506 ctgacAAGAgaagtctc (+)STAF Signal transducer and activator of transcription −1250 to −1232 gccattttcGGAAagtgct (+)SNAP snRNA-activating protein complex −1234 to −1216 gtatcCTTAagtagaaagc (−)LHXF Lim homeodomain factors −701 to −685 atcacattTAATcttga (−)OCT1 Octamer binding protein −696 to −682 gaCATCacatttaat (−)FAST FAST-1 SMAD interacting proteins −582 to −566 ccagtTGTGaatttctc (+)MYT1 MYT1 C2HC zinc finger protein −410 to −398 tggAAGTttcagc (+)EBOX E-box binding factors −423 to −411 tttcCACAtgcct (+)HOXF Factors with moderate activity to homeodomain consensus sequence −386 to −370 agattctTAAAatgttc (−)BRNF Brn POU domain factors −331 to −313 tttttttTAATcactcctc (+)HOMF Homeodomain transcription Factor −269 to −253 aagagatTTTAaataaa (+)SORY SOX/SRY-sex/testis determining and related HMG box factors −264 to −248 tggCATTtatttaaaat (−)NR2F Nuclear receptor subfamily 2 factors −254 to −230 aatgccAGGGgaaaggaaaaggaaa (+)HOMF Homeodomain transcription Factor −205 to −189 agtttgtccTAATtaga (−)GATA GATA binding factors −169 to −157 agctGATAaaaat (+)HOMF Homeodomain transcription factors −162 to −146 agctgaggaTAATtttt (−)BRNF Brn POU domain factors −139 to −121 ctgTAATcagaccttaagg (−)FKHD Fork head domain factors −134 to −118 cttctgtaATCAgacct (−)HOXF Factors with moderate activity to homeo domain consensus sequence −91 to −75 ggtctctTAAAtaaact (−)VTATA Vertebrate TATA binding protein factor −90 to −74 gtttattTAAGagacca (+)

TF: transcription factors, positions are relative to the start sequence ATG, capital letter in sequence indicate the core sequences responsible for interactingw

nsttmMmis

2p

dcomdpd

deletion constructs with restriction enzymes KpnI, NheIand KpnI, AgeI for their successive directional incorpo-ration into two reporter vectors viz. pGL4.17 [luc2/Neo]

Table 3Primers for GDF9 DelCon 1–3 amplification. Underlined sequences arerestriction enzyme sites, GCTAGC for NheI, GGTACC for KpnI.

Primer name Primer sequence 5′–3′ Productlength

DelCon constant GCTAGCGGCTTGGAAGAATTAGCAAGG

ith TF proteins.

ih.gov/), and compared with generated 5′ flankingequence of buffalo GDF9 gene using CLUSTALW2 mul-iple sequence alignment programme. For predicting theranscription factor binding sites, buffalo, cow, human and

ouse GDF9 5′ flanking sequences were analysed usingatinspector (solution parameters: core similarity 1.0;atrix-optimised) (Quandt et al., 1995). Similar but not

dentical germ cell specific transcription factor bindingites were screened manually and listed in Table 2.

.4. GDF9 promoter deletion constructs and reporterlasmids

Nested sets of primers were designed for preparation ofeletion constructs. For amplifying three sets of deletiononstructs, a constant reverse primer and progressive setsf nested forward primers (Table 3) were used with ter-

inal RE sites of KpnI and NheI/BshTI for the purpose of

irectional cloning. GFP deletion constructs (delcon 1–3)rimer sequences were the same as used for luciferaseeletion constructs except the terminal RE sites. The PCR

reactions for generating all the deletion constructs wereperformed in a total volume of 50 �l using Pfx proof readingtaq polymerase (Invitrogen, USA). The thermal parame-ters for amplifying the constructs are outlined in Table 4.Amplified PCR products were first ‘A’ tailed and subse-quently cloned into pGEM-T vector (Promega, Madison,WI, USA). Subcloning was carried out by restricting the

reverseDelCon1 forward GGTACCTGTGCAGACCCAAATAAAGG 743 bpDelCon2 forward GGTACCCATTCAGCAGTTTGGGTGTG 1113 bpDelCon3 forward GGTACCAAGACGGACGGAGGACAAG 2201 bp

B. Roy et al. / Animal Reproduction

Table 4Optimised PCR amplification conditions for amplification of deletion con-structs 1, 2 and 3.

Stage Cycles Temp Time

For DelCon1Initial denaturation 1 1 95 ◦C 2 minAmplification 2 34 95 ◦C 15 s

60 ◦C 30 s68 ◦C 1 min

For DelCon2Initial denaturation 1 1 95 ◦C 2 minAmplification 2 34 95 ◦C 15 s

65 ◦C 30 s68 ◦C 1.3 min

For DelCon3Initial denaturation 1 1 95 ◦C 2 minAmplification 2 39 95 ◦C 15 s

60 ◦C 30 s68 ◦C 2.3 min

(Promega, Madison, WI, USA) and pTurboGFP-PRL-dest1(Evrogen, Moscow, Russia), respectively. Orientation ofthe incorporated deletion constructs in the reporter vec-tors was confirmed by customized sequencing. Resultingdeletion constructs in the subsequent text are referred asDelcon1 (755 bp insert), Delcon2 (1125 bp insert) and Del-con3 (2213 bp insert).

2.5. Site directed mutation of NOBOX binding element

Site-directed mutagenesis was carried out at −1881 bpposition in putative NOBOX transcription factor-bindingsites (NBE) within the Delcon3 using Quick Change Light-ning SDM kit (Stratagene, Santa Clara, CA, USA). The NBEsite TAGTTG was changed to either TAGGCG or GGATCC.The two mutagenic primer pairs used for PCR amplifica-tion were 5′ TTCCGGGAGAACTAGGCGCCGACTGCCTCCAG3′ for NBE Delcon3-mut1 and 5′-AGGTCACTTCCGGGAGAA-CGGATCCCCGACTGCCTCCAGGTGAC-3′ for NBE Delcon3-mut2 along with complementary reverse primers (mutatednucleotides are underlined). The resultant mutation1 har-boured a RE site for EheI and mutation2 harboured a BamHIsite. Generated mutants were confirmed by restrictiondigestion analysis.

2.6. Reporter promoter assay

Transient transfection of denuded and zona permeabil-ized oocytes was performed 4 h after in vitro maturation,following the optimised electroporation protocol withsingle pulse of 500 V at pulse duration of 50 �s in ahypo-osmolar buffer (Eppendorf, Delhi, India) followedby culturing transfected oocytes for another 12 h. Theefficiency of transfection following electroporation wasoptimised using GFP signal data obtained after trans-fecting oocytes with a GFP reporter vector having CMVconstitutive promoter. The electroporation parameters as

described above yielded more than 75% transfection effi-ciency. GFP reporter promoter assay was performed with20 ng/�l of GDF9 promoter deletion constructs contain-ing GFP vector (pTurboGFP-PRL-dest1) where pAcGFP-N1

Science 135 (2012) 18– 24 21

(Clontech, Palo Alto) vector served as positive controlhaving a constitutive CMV promoter upstream of GFPreporter gene, while non-transfected oocyte group servedas negative control. Fluorescence images of transfectedoocytes were captured and processed for densitometryanalysis using ImageJ software. For the luciferase assaypGL4.13/SV40-promoter vector served as the positive con-trol and promoter less pGL4/Basic vector was used as thenegative control. Renilla luciferase was used as an internalcontrol for normalizing the transfection efficiency. Tripli-cate lysate samples (25 �l) were used for luciferase assayswith the Dual-Glo luciferase reporter assay kit (Promega,WI, USA). Light emission was quantified using a lumi-nometer (Turner Biosystems, Promega, WI, USA). Fireflyluciferase was normalized to renilla luciferase used as aninternal control for determining the transfection efficiencyand the results were presented as normalized relative lightunit (RLU).

2.7. Statistical analysis

Data for fluorescence intensity and RLU values for dif-ferent deletion constructs were analysed for the differencebetween means using two way ANOVA and significanceof difference between means was calculated at 5% level ofsignificance (P < 0.05).

3. Results

3.1. 5′ flanking sequence of buffalo GDF9 gene

The alignment of buffalo GDF9 5′ upstream sequencewith other mammalian species revealed the presence offive extensively conserved regions (CR1 to 5) up to astretch of 2 kb (Fig. 1). Three enhancer stretches weremapped on this 5′ upstream sequence depending onthe identified putative and common germ cell specifictranscription factor binding sites that were including sev-eral oocytes specific Ebox and NOBOX binding elements(Table 2). These identified regions in buffalo sequencerevealed extensive similarity when compared with thebovine, ovine, caprine and murine sequences. In buffalo,the putative minimal promoter (MP) that is located withinthe CR1 region, was found to be extended until −320 bpfrom ATG, and this CR1 region along with CR2 regionwas ascertained to harbour various germ cell specific cis-acting regulatory elements viz. NR2F, SORY, Foxh1, OCTetc. Three conserved Eboxes (CANNTG) and one NOBOXbinding Element (NBE) located within this region werefound to be conserved in buffalo, cow, sheep, goat andmouse. A potential stretch of enhancer region (ER1) wasobserved to be expanded over CR1 and CR2 region, whileanother stretch of enhancer region (ER2) was identifiedbeyond the CR2. These ER regions however, revealedunique buffalo specific sequence that was found to beless populated with oocyte specific TF binding sites. The

CR4, CR5 and an extended CR5 (ECR5) region were againfound to be harbouring potential enhancer elements andassigned to be the potential location for enhancer region 3(ER3).

22 B. Roy et al. / Animal Reproduction Science 135 (2012) 18– 24

0

10

20

30

40

50

60

70

80

Ctrl N Ctrl DelCon1 DelCon2 DelCon3

a

bb

c

a

% F

luore

scence Inte

nsity

Fig. 2. Differential fluorescence signals expressed by oocytes on trans-fection of GDF9 promoter deletion constructs. Control: positive controlvector (pAcGFP-N1). Negative control: non transfected oocytes. DelCon1,2, 3: GFP reporter-deletion constructs 1, 2 and 3. The negative controlvs

3s

wdsddctwoiiRaG

0

0.5

1

1.5

2

2.5

3

3.5

4

P Cntrl N Cntrl DelCon1 DelCon2 DelCon3

a

a

b

cc

Re

lative

Lu

cife

rase

Activity

Fig. 3. Luciferase signals obtained on transfection of reporter constructspGL4-deletion constructs 1, 2 and 3. RLU values were calculated as theratio of firefly: renilla relative light unit. Negative control: promoterless pGL4/basic vector, positive control: pGL4/SV40-promoter. Fireflyluciferase count was normalized to renilla luciferase as internal control.The negative control value has been taken as basal value. Bars indi-

F

alue has been taken as basal value. Bars indicate average ± SEM. Differentuperscripts indicate significant difference (p < 0.05).

.2. Functional analysis of GDF9 promoter by deletiontudy

Two alternate reporter assays viz. GFP and luciferaseere used to reconfirm the regulatory role of differenteletion constructs in the present study. Fig. 2 repre-ents the comparative fluorescence intensity values forifferent deletion constructs using GFP assay. Both of theelcon1 and 2 deletion constructs revealed stable fluores-ence signals after 12 h of transfection, whereas, in oocytesransfected with delcon3 (2213 bp), the signal intensityas increased by nearly 33%. Similarly, transfection of

ocytes with Luc-GDF9 deletion constructs revealed a 3 foldncrease in normalised RLU value with delcon3 in compar-son with the immediate smaller fragment delcon2 (Fig. 3).

esults indicate that the region from 2144 to −1056 bp has

significant effect on basal transcriptional activity of theDF9 delcon1 promoter (755 bp). Presence of a potential

ig. 4. Responsive cis-element map for deletion construct 3. The NOBOX element a

cate average ± SEM. Different superscripts indicate significant difference(p < 0.05).

NBE with a consensus sequence of TAGTTG at −1881 bp(Fig. 4) was prominent in this region, role of which wasfurther confirmed using site directed mutagenesis experi-ments.

3.3. Site directed mutation of NBE

Luciferase data obtained after transfection of mutateddelcon3 constructs (having mutated NBE sequence)revealed that delcon3-mut2 had a larger effect on reportergene expression than delcon3-mut1. RLU values for therespective mutants were 1.8 and 1.2, respectively, as com-pared to the non-mutated wild type whose value rested

as 2.75 (Fig. 5). The delcon3-mut1 with mutated sequenceof TAGGCG resulted in 30% reduction in promoter activitywhereas DelCon3-Mut2 with a mutated sequence GGATCC

t −1881 was found significant for buffalo oocyte specific GDF9 expression.

B. Roy et al. / Animal Reproduction Science 135 (2012) 18– 24 23

0

0.5

1

1.5

2

2.5

3

3.5

DelCon3Wild DelCon3Mut1 DelCon3Mut2

a

b

c

Re

lative

Lu

cife

rase

Activity

Fig. 5. Firefly luciferase activities resulted by GDF9 deletion construct 3promoter with mutations at NOBOX site. DelCon3Mut1 and DelCon3Mut2

Table 5Buffalo GDF9 5′ upstream sequence specific transcription factor bindingsites and their relative position to the TSS.

TF bindingelements

Consensus sequence Position(s) relative toTSS

TATA box TATTTAA −23

E-box CAGCTG −113; −1140; −1335;−2797; −3029; −4105

CACCTG −462; −725CAAGTG −2511CATGTG −1075; −2985CAGGTG −1864CAACTG −12CATCTG −660

NBE TAATTG NilTAATTA −140; −3408

are 2 mutants at 2 alternate core binding sequence sites. DelCon3 Wild

indicates constructs without mutation. Bars indicate average ± SEM. Dif-ferent superscripts indicate significant difference (p < 0.05).

resulted in 50% reduction in comparison to that of wild typesequence.

4. Discussion

The 5′ upstream sequence of GDF9 gene in buffalo wasfound to be conserved to a large extent in equivalencewith sequences from cow, sheep, goat and mouse. Analysisof the buffalo sequence further confirmed the occurrenceof conserved regions (CR) and the presence of function-ally important domains within 2 kb region upstream of itstranscription start site (Fig. 1). The extended comparativesequence analysis revealed that the degree of homologyhowever, reduced from CR1 to CR5 with presence of con-served oocyte specific transcription factor binding sites andminimal promoter element at CR1 and CR2 regions. Thissuggests the functional significance of this region and itsregulatory role in mammalian GDF9 gene as a whole. Theregions of CR3 to CR5 were meagrely populated with suchregulatory elements raising the possibility that individualcis elements at these regions were working as enhancersites (Tsunemoto et al., 2008). Results from earlier muta-tion studies of EBoxes and NBE for Ca2+ATPase and Oct4genes support such speculation (Baker et al., 1998; Choiand Rajkovic, 2006). The extended conserved region (ECR)of buffalo and the cow GDF9 gene were found to sharethe greatest homology among the species compared. TheNOBOX element (NBE) present on the buffalo ECR washowever, unique in its location and sequence conservationrepresenting striking deviation from the reported mousesequence. Marked species specificity has been reportedin GDF9 expression between rodent population and bovi-dae group along with human during the primordial follicledevelopment but this differential control of GDF9 expres-sion is yet to be examined at the molecular level. Based onavailable information, NBE seems to be the major control-ling factor for GDF9 expression in mouse ovary primordialfollicles (Suzumori et al., 2002). However, its relevance isyet to be determined in livestock species such as cow or

sheep.

The sequence of a 246 bp proximal region stretch of buf-falo GDF9 gene was compared for similarity with othermammalian species including cow, sheep, goat and mice.

TAGTTG −1881; −3575

DR0 element AGGTCAAATTCA −2688

It revealed notable homology and qualified this region tobe assigned as the minimum promoter element (MPE) forGDF9 (Fig. 1). A potential TATA element was detected at 23nucleotides upstream of the mapped TSS which further sig-nifies this stretch as a potential core promoter element forbuffalo GDF9 gene. Four putative NBEs (TAATTA/TAGTTG)were found at −140, −1881, −3408, and −3575 bp pos-itions (from TSS) in buffalo sequence analysed (Table 5).In mouse, NBE has been shown to be important for GDF9and Oct4 expression having a sequence motif of TAATTG(Choi and Rajkovic, 2006). These NBEs are identical to thecore binding sequences of homeoprotein, MSX1 (Catronet al., 1993), but having minor difference from the bind-ing sites of other homeoproteins such as CAAGTG for TITF1(Guazzi et al., 1990), TAAGTA for NKX3-1 (Steadman et al.,2000) and TAAGTG for BAPX1 (Kim et al., 2003). In viewof the consistent occurrence of TAATTG, Choi and Rajkovic(2006) experimented whether the NOBOX can also bind tothe consensus sequence of TAGTTG; TAATTA, thus, estab-lished their specific role as NOBOX binding cis elements.Subsequently, the same group reported that NBE motifcould also be responsible for regulating oocyte expressedgenes like Pad6 (Choi et al., 2010). However, Tsunemotoet al., 2008 reported that the NBE having role in oocytespecific gene expression is considered to have a consensussequence 5′TAATTG/A3′. They further revealed that morethan one NBE could have synergistic role in controllingthe basal level of transcription for oocyte specific genes asthey serve as the binding site of NOBOX transcription factorwhich, in turn, is recognized as the oocyte pro-survival fac-tor (Qin et al., 2007). Based on this, we speculated that the5′TAATTA3′ motifs in buffalo sequence could be potentiallyimportant for enhancing the basal transcriptional activityof GDF9 gene.

In mouse, a 3.3 kb 5′ upstream sequence of GDF9 genehas been reported sufficient to drive the reporter geneexpression in oocytes and inclusion of extra size of upto 10.7 kb in 5′ upstream region has resulted in the sig-

nificant increase of reporter gene activity in oocytes (Yanet al., 2006). This reflects the presence of enhancer activ-ity in the upstream region between −3.3 and −10.7 kb.In the present study, we designed the deletion construct

2 duction

1ewdo(mLb(wtoigiat(G

5

osttoatNssbaatispd

A

it

A

cj

4 B. Roy et al. / Animal Repro

(755 bp) to establish the strength of minimal promoterlement whereas the second deletion construct (1125 bp)as strategically planned for verifying the effect of pre-icted region (enhancer region 1) on enhancing the effectf minimal promoter located downstream. The 3rd one2213 bp) was designed to include the distal enhancer ele-

ent (predicted as enhancer region 2). On transfectinguc-GDF9 constructs in oocytes, the results revealed sta-le reporter signal with both 755 and 1125 bp constructsdelcon1 and 2) which abruptly increased by three foldsith the transfection of third construct. Our strategy to

arget a site directed mutagenesis at NBE site was basedn the available knowledge that it harbours a high affin-ty binding site for NOBOX, an oocyte-specific homeoboxene expressed in germ cell and in primordial and grow-ng oocytes during different stages of folliculogenesis (Choind Rajkovic, 2006). Thus based on the deletion and muta-ion results of the present study, we conclude that this NBEat −1881) could have a potential role for enhancing theDF9 transcriptional activity in buffalo oocytes.

. Conclusion

To the best of our knowledge, this is the first reportn the detailed GDF9 promoter analysis in any livestockpecies. The 5′ upstream region of buffalo GDF9 was foundo share a good level of homology with other species withhe presence of some unique sequence stretches. Importantocyte specific cis-acting regulatory elements like E-Boxesnd NBEs were identified. Functional analysis of the puta-ive promoter elements further highlights the role of theOBOX Binding Element at −1881 bp upstream of tran-

cription stat site in oocyte specific GDF9 expression. NBEerves as the binding site for homeodomain family mem-er transcription factor NOBOX which is reported to ben oocyte pro-survival factor. Thus the evidence gener-ted in the present study was realistic enough to interprethe functional role of mutated NBE on GDF9 expressionn buffalo oocytes. Further information will help in under-tanding buffalo oocyte biology and designing appropriaterotocols for ART protocol for this important species ofairy animals.

cknowledgements

Fund received from NAIP C1056 of ICAR and usefulnputs of Ms. Rupinder Kaur in improving the text arehankfully acknowledged.

ppendix A. Supplementary data

Supplementary data associated with this articlean be found, in the online version, at doi:10.1016/.anireprosci.2012.09.006.

Science 135 (2012) 18– 24

References

Baker, D., Dave, V., Reed, T., Misra, S., Periasamy, M., 1998. A novelEbox/ATrich element is required for the muscle specific expression ofthe sarcoplasmic reticulum Ca2+ ATPase (SERCA2) gene. Nucleic AcidsRes. 26, 1092–1098.

Catron, K.M., Iler, N., Abate, C., 1993. Nucleotides flanking a conservedTAAT core dictate the DNA binding specificity of three murine home-odomain proteins. Mol. Cell Biol. 13, 2354–2365.

Choi, M., Lee, O.H., Jeon, S., Park, M., Lee, D.R., Ko, J.J., Yoon, T.K., Rajkovic,A., Choi, Y., 2010. The oocyte-specific transcription factor, Nobox, reg-ulates the expression of Pad6, a peptidylarginine deiminase in theoocyte. FEBS Lett. 584, 3629–3634.

Choi, Y., Rajkovic, A., 2006. Characterization of NOBOX DNA binding speci-ficity and its regulation of Gdf9 and Pou5f1 promoters. J. Biol. Chem.281, 35747–35756.

FAOSTAT, 2005. FAO Statistical Databases (Online). FAO, Rome, Availablefrom: http://faostat.fao.org/ (verified 18.09.11).

Guazzi, S., Price, M., De Felice, M., Damante, G., Mattei, M.G., Di Lauro, R.,1990. Thyroid nuclear factor 1 (TTF-1) contains a homeodomain anddisplays a novel DNA binding specificity. EMBO J. 9, 3631–3639.

Hussein, T.S., Thompson, J.G., Gilchrist, R.B., 2006. Oocyte-secreted factorsenhance oocyte developmental competence. Dev. Biol. 296, 514–521.

Jain, T., Jain, A., Kumar, P., Goswami, S.L., De, S., Singh, D., Datta, T.K., 2012.Kinetics of GDF9 expression in buffalo oocytes during in vitro matura-tion and their associated development ability. Gen. Comp. Endocrinol.178, 477–484.

Kidder, G.M., Vanderhyden, B.C., 2010. Bidirectional communicationbetween oocytes and follicle cells: ensuring oocyte developmentalcompetence. Can. J. Physiol. Pharmacol. 88, 399–413.

Kim, D.W., Kempf, H., Chen, R.E., Lassar, A.B., 2003. Characterization ofNkx3.2 DNA binding specificity and its requirement for somatic chon-drogenesis. J. Biol. Chem. 278, 27532–27539.

Nandi, S., Raghu, H.M., Ravindranatha, B.M., Chauhan, M.S., 2002. Pro-duction of buffalo (Bubalus bubalis) embryos in vitro: premises andpromises. Reprod. Domest. Anim. 37, 65–74.

Qin, Y., Choi, Y., Zhao, H., Simpson, J.L., Chen, Z.J., Rajkovic, A., 2007. NOBOXhomeobox mutation causes premature ovarian failure. Am. J. Hum.Genet. 81, 576–581.

Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T., 1995. Matlndand matinspector: new fast and versatile tools for detection of con-sensus matches in nucleotide sequence data. Nucleic Acids Res. 23,4878–4884.

Sambrook, J., Russell, D., 2001. Molecular Cloning. A Laboratory Manual,3rd ed. Cold Spring Harbor Press, Cold Spring Harbor, NY.

Singh, B., Chauhan, M.S., Singla, S.K., Gautam, S.K., Verma, V., Manik,R.S., Singh, A.K., Sodhi, M., Mukesh, M., 2009. Reproductive biotech-niques in buffaloes (Bubalus bubalis): status, prospects and challenges.Reprod. Fertil. Dev. 21, 499–510.

Sirard, M.A., Richard, F., Blondin, P., Robert, C., 2006. Contribution of theoocyte to embryo quality. Theriogenology 65, 126–136.

Steadman, D.J., Giuffrida, D., Gelmann, E.P., 2000. DNA-binding sequenceof the human prostate-specific homeodomain protein NKX3.1. NucleicAcids Res. 28, 2389–2395.

Suzumori, N., Yan, C., Matzuk, M.M., Rajkovic, A., 2002. Nobox is ahomeobox encoding gene preferentially expressed in primordial andgrowing oocytes. Mech. Dev. 111, 137–141.

Tsunemoto, K., Anzai, M., Matsuoka, T., Tokoro, M., Shin, S.W., Amano,T., Mitani, T., Kato, H., Hosoi, Y., Saeki, K., 2008. Cis-acting elements(E-box and NBE) in the promoter region of three maternal genes (his-tone H1oo, nucleoplasmin 2, and zygote arrest 1) are required foroocyte-specific gene expression in the mouse. Mol. Reprod. Dev. 75,1104–1108.

Yan, C., Elvin, J.A., Lin, Y.N., Hadsell, L.A., Wang, J., DeMayo, F.J., Matzuk,M.M., 2006. Regulation of growth differentiation factor 9 expression

in oocytes in vivo: a key role of the E-box. Biol. Reprod. 74, 999–1006.

Yeo, C.X., Gilchrist, R.B., Thompson, J.G., Lane, M., 2008. Exogenous growthdifferentiation factor 9 in oocyte maturation media enhances subse-quent embryo development and fetal viability in mice. Hum. Reprod.23, 67–73.