5
Regulation of the Human SIX3 Gene Promoter Johannes Lengler and Jochen Graw 1 Institute of Mammalian Genetics, GSF-National Research Center for Environment and Health, D-85764 Neuherberg, Germany Received August 13, 2001 A 2-kb promoter fragment of SIX3, a human tran- scription factor essential for vertebrate eye develop- ment, has been characterized in a gene reporter assay system. The peak of activity implies the 2-kb sequence of SIX3, whereas 5*-deletion constructs of the pro- moter decreases successively to 60% of the activity starting from the entire promoter. In contrast, cutting off 300 bp of the 3* promoter extinguishes its activity completely. Coexpression experiments of different other transcription factors illuminate the regulation of SIX3 during eye development: Pax6 activates the 2703/2349 SIX3 promoter threefold, and PROX1 even eightfold. In contrast, Msx2 represses the entire SIX3 promoter. Furthermore, Six3 is regulated by its own negative feedback loop. In conclusion, SIX3 expres- sion underlies a complex regulation, which is an im- portant part to understand the network of transcrip- tion factors during eye development. © 2001 Academic Press Key Words: SIX3; promoter; gene reporter assay; transcription factor; eye development; mouse; man. The SIX3 gene is a member of the Six-gene family of vertebrates, which are homologous to the Drosophila sine oculis gene. The Six proteins are characterized by a highly conserved homeodomain and a specific Six domain, which is supposed to participate in transcrip- tional activation (1). Vertebrate Six3 genes are in- volved in head midline and eye formation in several organisms (2). Comparison of Six3 genes of chicken (3), Xenopus laevis (4), medaka (5) and zebrafish (2) indi- cate, that they are more closely related to the recently described Drosophila gene optix than sine oculis (6). In the mouse, Six3 is early expressed during devel- opment of the neural plate (E 6.5) and later on during eye development (7). Its expression takes place in the lens vesicle (E 9.5) and afterwards in the lens (E 12.5). One day later, it retrieves from the lens fibre cells to the epithelium and there it remains strongly ex- pressed. Together with Six3, transcription factors like Prox1 (8), Msx2 (9) and Pax6 (10) are coexpressed dur- ing several stages of eye development. Recent publication of the human genome enables analysing sequences upstream of the SIX3 coding re- gion. In this report, we describe the amplification and characterization of the promoter region, which is re- sponsible for SIX3 expression in a cell culture system. Furthermore, we illuminate the effect of several tran- scription factors being part of a complex network reg- ulating SIX3. MATERIALS AND METHODS Expression systems. Pax6 cDNA subcloned into pBlue- scriptKS(1) vector was digested with HindIII and BamHI and sub- cloned into eukaryotic expression vector pcDNA3.1 (Invitrogen, Karlsruhe, Germany). For expression of PROX1 and Six3, the corre- sponding pcDNA3 expression constructs are already described (11). Msx2 coding sequence was cloned into expression vector pcDNA3.1. PCR amplification of promoter sequences. Genomic DNA was pre- pared from human blood cells using the Puregene kit following exactly the instructions of the manufacturer (www.gentra.com). For- ward primers contain a XhoI site for cloning, whereas the reverse PCR primer has a HindIII restriction site (Table 1). The PCR condi- tions were 67 mM Tris–HCl (pH 8.8), 16.6 mM (NH 4 ) 2 SO 4 , 6.7 mM MgSO 4 , 0.07% b-mercaptoethanol (vol/vol), 200 mM each of dNTP, 0.1 mM each of oligonucleotide primers, 100 ng template DNA, 0.5 units Taq DNA polymerase and 0.5 units Pfx DNA polymerase (both Invitrogen, Karlsruhe, Germany). During the PCR of 40 cycles, the annealing temperature of 52°C was reduced to 44°C, 1°C per five cycles, in a “touchdown” PCR (12). The PCR products were cloned into the cloning site of the pPLLucII reporter vector (13) using SalI/BamHI and HindIII restriction enzymes and then transformed into DH5a bacteria. DNA was prepared using a plasmid NucleoSpin column (Macherey Nagel, Du ¨ ren, Germany). Transfection and reporter gene assay. For luciferase (Luc) re- porter assays, 293 fibroblast cells were cultivated in 96-well plates for 24 h and transfected by PolyFect Transfection Reagent (Qiagen, Hilden, Germany) using 0.7 mg plasmid-DNA. The DNA mix in transfection reaction is compounded by (i) 0.5 mg reporter vector, containing a sequence 59 of SIX3 fused to the Luc reporter gene, (ii) 0.1 mg effector (either Pax6-, Six3-, Msx2-pcDNA3.1 or Prox1- pcDNA3, or the parental plasmid pcDNA3.1 as negative control), and (iii) 0.1 mg pRL-SV40 for transfection control. Cells were harvested 48 h after transfection and cellular extracts were assayed with the Dual-Luciferase Reporter Assay System (Pro- mega, Heidelberg, Germany) in quadruplicate and triplicate, and the standard deviation was calculated ( n 5 7). 1 To whom correspondence should be addressed. Fax: 149(89)3187-2210. E-mail: [email protected]. Biochemical and Biophysical Research Communications 287, 372–376 (2001) doi:10.1006/bbrc.2001.5605, available online at http://www.idealibrary.com on 372 0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

Regulation of the Human SIX3 Gene Promoter

Embed Size (px)

Citation preview

Page 1: Regulation of the Human SIX3 Gene Promoter

R

JID

R

smsomsocoo2epnspt

t

vsadtvoXcd

oelOtp

1

Biochemical and Biophysical Research Communications 287, 372–376 (2001)

doi:10.1006/bbrc.2001.5605, available online at http://www.idealibrary.com on

0CA

egulation of the Human SIX3 Gene Promoter

ohannes Lengler and Jochen Graw1

nstitute of Mammalian Genetics, GSF-National Research Center for Environment and Health,-85764 Neuherberg, Germany

eceived August 13, 2001

Prox1 (8), Msx2 (9) and Pax6 (10) are coexpressed dur-i

agcsFsu

M

scKsM

pewPtM0uIaciSic

pfHtc0p(

wms

A 2-kb promoter fragment of SIX3, a human tran-cription factor essential for vertebrate eye develop-ent, has been characterized in a gene reporter assay

ystem. The peak of activity implies the 2-kb sequencef SIX3, whereas 5*-deletion constructs of the pro-oter decreases successively to 60% of the activity

tarting from the entire promoter. In contrast, cuttingff 300 bp of the 3* promoter extinguishes its activityompletely. Coexpression experiments of differentther transcription factors illuminate the regulationf SIX3 during eye development: Pax6 activates the703/2349 SIX3 promoter threefold, and PROX1 evenightfold. In contrast, Msx2 represses the entire SIX3romoter. Furthermore, Six3 is regulated by its ownegative feedback loop. In conclusion, SIX3 expres-ion underlies a complex regulation, which is an im-ortant part to understand the network of transcrip-ion factors during eye development. © 2001 Academic Press

Key Words: SIX3; promoter; gene reporter assay;ranscription factor; eye development; mouse; man.

The SIX3 gene is a member of the Six-gene family ofertebrates, which are homologous to the Drosophilaine oculis gene. The Six proteins are characterized by

highly conserved homeodomain and a specific Sixomain, which is supposed to participate in transcrip-ional activation (1). Vertebrate Six3 genes are in-olved in head midline and eye formation in severalrganisms (2). Comparison of Six3 genes of chicken (3),enopus laevis (4), medaka (5) and zebrafish (2) indi-ate, that they are more closely related to the recentlyescribed Drosophila gene optix than sine oculis (6).In the mouse, Six3 is early expressed during devel-

pment of the neural plate (E 6.5) and later on duringye development (7). Its expression takes place in theens vesicle (E 9.5) and afterwards in the lens (E 12.5).ne day later, it retrieves from the lens fibre cells to

he epithelium and there it remains strongly ex-ressed. Together with Six3, transcription factors like

1 To whom correspondence should be addressed. Fax:49(89)3187-2210. E-mail: [email protected].

372006-291X/01 $35.00opyright © 2001 by Academic Pressll rights of reproduction in any form reserved.

ng several stages of eye development.Recent publication of the human genome enables

nalysing sequences upstream of the SIX3 coding re-ion. In this report, we describe the amplification andharacterization of the promoter region, which is re-ponsible for SIX3 expression in a cell culture system.urthermore, we illuminate the effect of several tran-cription factors being part of a complex network reg-lating SIX3.

ATERIALS AND METHODS

Expression systems. Pax6 cDNA subcloned into pBlue-criptKS(1) vector was digested with HindIII and BamHI and sub-loned into eukaryotic expression vector pcDNA3.1 (Invitrogen,arlsruhe, Germany). For expression of PROX1 and Six3, the corre-

ponding pcDNA3 expression constructs are already described (11).sx2 coding sequence was cloned into expression vector pcDNA3.1.

PCR amplification of promoter sequences. Genomic DNA was pre-ared from human blood cells using the Puregene kit followingxactly the instructions of the manufacturer (www.gentra.com). For-ard primers contain a XhoI site for cloning, whereas the reverseCR primer has a HindIII restriction site (Table 1). The PCR condi-ions were 67 mM Tris–HCl (pH 8.8), 16.6 mM (NH4)2SO4, 6.7 mMgSO4, 0.07% b-mercaptoethanol (vol/vol), 200 mM each of dNTP,

.1 mM each of oligonucleotide primers, 100 ng template DNA, 0.5nits Taq DNA polymerase and 0.5 units Pfx DNA polymerase (bothnvitrogen, Karlsruhe, Germany). During the PCR of 40 cycles, thennealing temperature of 52°C was reduced to 44°C, 1°C per fiveycles, in a “touchdown” PCR (12). The PCR products were clonednto the cloning site of the pPLLucII reporter vector (13) usingalI/BamHI and HindIII restriction enzymes and then transformed

nto DH5a bacteria. DNA was prepared using a plasmid NucleoSpinolumn (Macherey Nagel, Duren, Germany).

Transfection and reporter gene assay. For luciferase (Luc) re-orter assays, 293 fibroblast cells were cultivated in 96-well platesor 24 h and transfected by PolyFect Transfection Reagent (Qiagen,ilden, Germany) using 0.7 mg plasmid-DNA. The DNA mix in

ransfection reaction is compounded by (i) 0.5 mg reporter vector,ontaining a sequence 59 of SIX3 fused to the Luc reporter gene, (ii).1 mg effector (either Pax6-, Six3-, Msx2-pcDNA3.1 or Prox1-cDNA3, or the parental plasmid pcDNA3.1 as negative control), andiii) 0.1 mg pRL-SV40 for transfection control.

Cells were harvested 48 h after transfection and cellular extractsere assayed with the Dual-Luciferase Reporter Assay System (Pro-ega, Heidelberg, Germany) in quadruplicate and triplicate, and the

tandard deviation was calculated (n 5 7).

Page 2: Regulation of the Human SIX3 Gene Promoter

R

pd2ccctplbdw

tmetdecirfio

sotccap

Safs2gpct

tttcospa

sii2ftdopi

gr41sm

ASS5wawm7mpglb

pq

TABLE 1

Vol. 287, No. 2, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ESULTS AND DISCUSSION

Characterization of the SIX3 promoter. The computerrogram PromoterInspector (http://genomatix.gsf.de) pre-icts a promoter, which lies close to the start codon (bp278 to 23) of the human SIX3 gene. For that reason, we

loned a 2-kb fragment directly upstream of the SIX3oding region, as well as several corresponding deletiononstructs, in front of a Luc reporter gene. For transfec-ion experiments, the kidney 293 epithelium cell line wasreferred the differentiated cell lines like for example theens epithelial cells. This 293 cell line can be used as aasic transfection system (14). It is supposed to lack en-ogenous eye specific factors, which are in focus of thisork.Transfecting the different promoter constructs into

he 293 fibroblast cell line, the entire 2-kb SIX3 pro-oter shows 6 times higher activity than the promot-

rless reporter vector (Figs. 1A and 1B). 59 Deletions ofhe 2-kb to a 0.7-kb fragment result in a successiveecrease of the promoter activity, which is in the short-st fragment 60% of the entire promoter. However,utting the promoter sequence only 300 bp of its 39 end,ts activity drops down to the level of the promoter-lesseporter plasmid. The role of the 2348/144 promoterragment is in line with the database analysis, reveal-ng a promoter element between position 2278 and 23f the initiation codon.Therefore, it might be concluded, that the proximal

equence is essential, but not sufficient for the functionf the SIX3 promoter. Our data suggest sequences be-ween 2703 to 21015 and 21016 to 21937, whichause each an activating effect in the 293 fibroblast cellulture. The length of the fragment from 21937/144 ist least necessary to comprise the entire activity of theromoter.

Function of Pax6 and PROX1 on the SIX3 promoter.ince Pax6 binding sites are known in detail (15), wenalysed the SIX3 promoter using MatInspector Pro-essional (http://genomatix.gsf.de). Eight Pax6 bindingites are predicted at positions 21730, 21701, 21546,1451, 21450, 21305, 21134 and 2416. We investi-

ated the ability of Pax6 to stimulate the entire SIX3romoter in co-transfection experiments into the 293ell line. Pax6-pcDNA3.1 expression vector activatedhe 0.4-kb fragment threefold as compared to the co-

Primers Used to Amplify the H

Forward 21937/21912 59-TTA GC2703/2680 59-TTA GC

Reverse 2329/2383 59-CCA GC144/118 59-TAT CG

Note. Restriction sites XhoI in the forward and HindIII in the rev

373

ransfection of parental pcDNA3.1 vector used as con-rol (Fig. 2A). This 400-bp sequence is characterized byhe missing 39 end of the SIX3 promoter (Fig. 1A). Inontrast, Pax6 was not able to activate significantlyther parts of the SIX3 promoters, including the 39equence 2383/144. In line, a Pax6 binding site wasredicted in the 2703/2349 SIX3 promoter fragmentt position 2416.Interestingly, PROX1 exhibits a similar effect de-

cribed above for Pax6. PROX1 does not show anynfluence on the entire 2-kb promoter of SIX3, however,t activates the 0.4-kb fragment even eightfold (Fig.A). Therefore, the 2704/2349 fragment was selectedor a cotransfection experiment in a different cell line,he human lens epithelial cell line CD5A lacking en-ogenous PROX1 expression (11). Increasing amountsf PROX1-pcDNA3 were cotransfected with SIX3 re-orter plasmid and activated the 2704/2349 fragmentn a dose-response manner (Fig. 2B).

Taking all observations on the SIX3 activation to-ether, there is a good evidence for Pax6 and PROX1esponsive sequences between bp 2703 and 2349 (Fig.). However, since all fragments containing the 2348/44 sequence at the 39-end of the promoter demon-

trate no stimulation neither by Pax6 nor by PROX1, itight have the function of a silencer.

Function of Six3 and Msx2 at the SIX3 promoter.s described for Pax6, we investigated the influence ofix3 itself on its own promoter. It is noteworthy thatix3 represses the entire SIX3 promoter as well as its9 deletion constructs (Fig. 3). The repression by Six3as observed to be less than 50% using the 21016/144s well as the entire 2-kb promoter. Six3 demonstratedeaker repression upon the 2703/144 promoter frag-ent, reducing the reporter activity to approximately

0%. However, deleting the 39-end of the SIX3 pro-oter, 2703/2349 obliterates the Six3 repression com-

letely. Therefore, it might be concluded, that the SIX3ene can be regulated by its own negative feed backoop, and the SIX3 responsive element is suggestedetween 2348 and 144.Msx2 affects the SIX3 promoter very similar as com-

ared to Six3. It reduces the entire promoter below auarter of its basal activity. Therefore, Msx2 is a re-

an SIX3 Promoter Sequences

CG AGT TTC AGC TTT ATT GAG GGC AGA TTA G-39CG AGA AGA AGA AGT CTC AGA GAC AGC GG-39

AGC TTA ATG ACA ACA GCC TCA TAA TAT CTC C-39GC TTC AAG AAG TGG GAG GAA TAG AGG TCT AG-39

e primers are underlined.

um

C TC TAA A

ers

Page 3: Regulation of the Human SIX3 Gene Promoter

pt

Sptw1qwafM

apedespa

asc

Apprms

Vol. 287, No. 2, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ressor of SIX3 expression, which is even strongerhan Six3 itself.

We demonstrated Six3 and Msx2 repressing theIX3 gene. However, since activity of the 2703/2349romoter cannot be reduced, this fragment is supposedo be unimportant down-regulating SIX3 expression,hereas Six3 and Msx2 are able to repress the 2703/44 fragment. Moreover, the 2703/2349 promoter se-uence misses the proximal 2348/144 fragment,hich all fragments being able to be repressed by Six3nd Msx2 have in common. Therefore, the 2348/144ragment is considered to contain both, a Six3 and a

sx2 responsive element (Fig. 4).

FIG. 1. (A) Deletion constructs of the human SIX3 promoter.ccession No. AF083891. PCR primers used for amplification of therimer positions. 2278/23 sequence indicated in black contains a prromoter fragments. Transfection experiments (n 5 7) were perforeporter construct was cotransfected with pRL-SV40 vector for interneasured for Luc expression by the Dual Luciferase assay system. A

et as 1.

374

In conclusion, during early eye development, Pax6 islready expressed (10), while Six3 begins to be ex-ressed shortly before the lens is formed (7). Then,xpression of both genes is mainly colocated in theeveloping eye. Six3 withdraws from the lens to itspithelium around E 14.5, similar when Pax6 expres-ion decreases in the lens fibre cells. These gene ex-ression patterns support our observation that Pax6ctivates Six3 expression.Expression of Prox1 begins in the lens placode E 9.5

nd is observed until at least ten day old mice (8). Itupports Six3 expression. Possibly, Six3 expression isontrolled by the co-located Msx2 repressor (16), which

e underlying sequence appears in the GenBank/EMBO with thenomic sequence are given in Table 1. Numbers above indicate theoter sequence predicted by PromoterInspector. (B) Function of SIX3d with kidney 293 fibroblast cells. In all transfections, a luciferasestandardisation of efficiency of the gene transfer. Cell extracts wereata were normalised to the promoter-less plasmid (0/0), which was

Thge

ommealll d

Page 4: Regulation of the Human SIX3 Gene Promoter

mml

an

DPcPctaiwmPPSaf

tTfmmip

Vol. 287, No. 2, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

75

ight function as counterpart of Prox1 and, further-ore, Six3 is controlled by its own negative feedback

oop.Since Six3 is involved in important genetic pathways

s development of brain (17) and eye (18), a complexetwork of transcription factors, of which Six3 itself is

FIG. 2. (A) Pax6 and PROX1 activate the human SIX3 promoter.eletion constructs of the SIX3 promoter were cotransfected withax6 or PROX1. The relative luciferase activity of each promoteronstruct cotransfected with the expression vector containing eitherax6 or PROX1 was measured and compared to the activity of theorresponding promoter fragment alone (as shown in Fig. 1A). In allransfection experiments, we observed a stimulating effect of Pax6nd PROX1 only upon the 2703/2349 promoter, whereas the activ-ties of all fragments containing the 2349/144 39-promoter elementere not changed. (B) PROX1 activates the 2703/2349 SIX3 pro-oter element in lens epithelial cells. To affirm the activating role ofROX1 on the 2703/2349 promoter element, increasing amounts ofROX1 were cotransfected into CD5A cells. PROX1 stimulates theIX3 promoter element in a dose-dependent manner above its basalctivity. For reference, the pPLLucII vector without 2703/2349ragment is shown on the left.

FIG. 3. Six3 and Msx2 repress the human SIX3 promoter. Dele-ions of the SIX3 promoter were cotransfected with Six3 or Msx2.he relative luciferase activity of each promoter construct cotrans-

ected with the expression vector containing either Six3 or Msx2 waseasured and compared to the activity of the corresponding pro-oter fragment alone (as shown in Fig. 1A). All fragments contain-

ng the 2349/144 39-promoter sequence were observed to be re-ressed.

3

Page 5: Regulation of the Human SIX3 Gene Promoter

aa

A

(Uas

R

1

1

1

1

1

1

1

1

1

bc

Vol. 287, No. 2, 2001 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

part, controls the temporal and spatial expressionnd fine tuning of Six3.

CKNOWLEDGMENTS

We gratefully acknowledge the receipt of clones from R. BallingGBF, Braunschweig, Germany), D. Davidson (Edinburgh, Scotland/K), P. Gruss (Gottingen, Germany), Y. Kamachi (Nagoya, Japan),nd S. I. Tomarev (Bethesda, MD). Oligonucleotides were synthe-ised by Utz Linzner (GSF, Institute of Experimental Genetics).

EFERENCES

1. Seo, H. C., Drivenes, Ellingsen, S., and Fjose, A. (1998) Expres-sion of two zebrafish homologues of the murine Six3 gene demar-cates the initial eye primordia. Mech. Dev. 73, 45–57.

2. Kawakami, K., Ohto, H., Takizawa, T., and Saito, T. (1996)Identification and expression of six family genes in mouse retina.FEBS Lett. 393, 259–263.

3. Bovolenta, P., Mallamaci, A., and Boncinelli, E. (1996) Cloningand characterisation of two chick homeobox genes, members ofthe six/sine oculis family, expressed during eye development. Int.J. Dev. Biol. 1(Suppl.), 73S–74S.

4. Ghanbari, H., Seo, H. C., Fjose, A., and Brandli, A. W. (2001)Molecular cloning and embryonic expression of Xenopus Six ho-meobox genes. Mech. Dev. 101, 271–277.

5. Loosli, F., Koster, R. W., Carl, M., Krone, A., and Wittbordt, J.(1998) Six3, a medaka homologue of the Drosophila homeoboxgene sine oculis is expressed in the anterior embryonic shield andthe developing eye. Mech. Dev. 74, 159–164.

6. Gallardo, M. E., Lopez-Rios, J., Fernaud-Espinosa, I., Gra-nadino, B., Sanz, R., Ramos, C., Ayuso, C., Seller, M. J., Brunner,H. G., Bovolenta, P., and Rodriguez, d. C. (1999) Genomic cloningand characterization of the human homeobox gene SIX6 revealsa cluster of SIX genes in chromosome 14 and associates SIX6hemizygosity with bilateral anophthalmia and pituitary anoma-lies. Genomics 61, 82–91.

7. Oliver, G., Mailhos, A., Wehr, R., Copeland, N. G., Jenkins, N. A.,and Gruss, P. (1995) Six3, a murine homologue of the sine oculisgene, demarcates the most anterior border of the developing

FIG. 4. Regulation of the human SIX3 promoter. PAX6 and PROetween 2703 and 2349 (arrows). In contrast, SIX3 activation isorresponding responsive elements are considered between 2348 an

376

neural plate and is expressed during eye development. Develop-ment 121, 4045–4055.

8. Tomarev, S. I., Zinovieva, R. D., Chang, B., and Hawes, N. L.(1998) Characterization of the mouse Prox1 gene. Biochem. Bio-phys. Res. Commun. 248, 684–689.

9. Holme, R. H., Thomson, S. J., and Davidson, D. R. (2000) Ectopicexpression of Msx2 in chick retinal pigmented epithelium cul-tures suggests a role in patterning the optic vesicle. Mech. Dev.91, 175–187.

0. Walther, C., and Gruss, P. (1991) Pax-6, a murine paired boxgene, is expressed in the developing CNS. Development 113,1435–1449.

1. Lengler, J., Krausz, E., Tomarev, S., Prescott, A., Quinlan, R. A.,and Graw, J. (2001) Antagonistic action of Six3 and Prox1 at thegamma-crystallin promoter. Nucleic Acids Res. 29, 515–526.

2. Don, R. H., Cox, P. T., Wainwright, B. J., Baker, K., and Mattick,J. S. (1991) ‘Touchdown’ PCR to circumvent spurious primingduring gene amplification. Nucleic Acids Res. 19, 4008.

3. Kamachi, Y., and Kondoh, H. (1993) Overlapping positive andnegative regulatory elements determine lens-specific activity ofthe delta 1-crystallin enhancer. Mol. Cell. Biol. 13, 5206–5215.

4. Tamura, K., Tanimoto, K., Murakami, K., and Fukamizu, A.(1992) A combination of upstream and proximal elements isrequired for efficient expression of the mouse renin promoter incultured cells. Nucleic Acids Res. 20, 3617–3623.

5. Epstein, J., Cai, J., Glaser, T., Jepeal, L., and Maas, R. (1994)Identification of a Pax paired domain recognition sequence andevidence for DNA-dependent conformational changes. J. Biol.Chem. 269, 8355–8361.

6. Liu, Y. H., Ma, L., Wu, L. Y., Luo, W., Kundu, R., Sangiorgi, F.,Sneard, M. L., and Maxson, R. (1994) Regulation of the Msx2homeobox gene during mouse embryogenesis: A transgene with439 bp of 59 flanking sequence is expressed exclusively in theapical ectodermal ridge of the developing limb. Mech. Dev. 48,187–197.

7. Wallis, D., and Muenke, M. (2000) Mutations in holoprosen-cephaly. Hum. Mutat. 16, 99–108.

8. Ohto, H., Kamada, S., Tago, K., Tominaga, S. I., Ozaki, H., Sato,S., and Kawakami, K. (1999) Cooperation of six and eya inactivation of their target genes through nuclear translocation ofEya. Mol. Cell. Biol. 19, 6815–6824.

activates SIX3 expression; the responsive elements are most likelytrolled by MSX2 as well as its own negative feedback loop. The44 (bars).

X1cond 1