Structural and functional characterization of the humanPAX7 5 0-flanking regulatory region

Yana V. Syagailoa,1, Olga Okladnovaa,2, Ella Reimera, Marcus Graßlea, Rainald Mossnera,Stefan Gattenlohnerb, Alexander Marxb, Jobst Meyera, Klaus-Peter Lescha,*

aDepartment of Psychiatry and Psychotherapy, University of Wurzburg, Fuchsleinstrasse 15, 97080 Wurzburg, GermanybInstitute of Pathology, University of Wurzburg, Wurzburg, Germany

Received 28 July 2001; received in revised form 11 June 2002; accepted 19 June 2002

Received by W.-H. Li

Abstract

The human PAX7 gene is a member of the paired box containing gene family of transcription factors implicated in development of the

skeletal muscle of the trunk and limbs as well as elements of the central nervous system. To understand the molecular mechanisms involved

in its expression, we have localized the transcription start sites in adult skeletal muscle and functionally characterized the 5 0-flanking

regulatory region responsible for PAX7 expression in this tissue. The major transcription start was identified 664 bp upstream from the

ATG codon using primer extension and 5 0-rapid amplification of cDNA ends (5 0-RACE). Analysis of the 5 0-flanking sequence revealed the

absence of a TATA-box and the presence of an inverted CCAAT-box. Several consensus sites for common transcriptional regulators

including Oct-1, NF1, AP2, AP4, CREB, Sp1, Nkx2.5, and MyoD are present in the promoter region. To determine the sites critical for

the function of the PAX7 promoter, a series of deletion fragments of the 5 0-flanking region were cloned adjacent to luciferase reporter gene

and expressed in RD, Cos-7 and JAR cell lines. The maximal promoter activity was achieved by a fragment extending from the position

2403 to 1373. No strong positive or negative regulatory elements were discovered by adding of further sequences (up to 2.97 kb). A

polymorphic (CCT)n repeat sequence was found 107 bp upstream of the transcription initiation site. PCR-based systematic screening for

length variations in 227 unrelated individuals of a Caucasian population showed a bimodal distribution of three alleles containing 8, 10 or 11

repeat units. When different variants of this PAX7 gene-linked polymorphic region (PAX7-LPR) were fused to a luciferase reporter gene and

transfected into RD cells, the variant with 11 repeat units revealed higher transcriptional efficiency compared to the 8 or 10 repeat alleles.

q 2002 Elsevier Science B.V. All rights reserved.

Keywords: Transcription regulation; Promoter; Paired box genes; Trinucleotide repeat polymorphism

1. Introduction

Nine members of the Pax gene family, Pax1 to Pax9,

have been identified in mice and humans (Gruss and

Walther, 1992; Tremblay and Gruss, 1994). These genes

exhibit spatial and temporal regulation during development

and encode proteins acting as transcriptional regulators. The

common feature of these proteins is a conserved 128-amino

acid DNA-binding domain, the paired domain, initially

found in some Drosophila segmentation genes (Bopp et

al., 1986; Walther et al., 1991). Mutations in several Pax

genes have been found in both congenital mouse and human

diseases, and have been shown to disrupt the developmental

processes in tissues in which they are expressed (Gruss and

Walther, 1992; Chalepakis et al., 1993; Strachan and Read,

1994).

In the developing mouse embryo, Pax7 is expressed from

Gene 294 (2002) 259–268

0378-1119/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved.

PII: S0378-1119(02)00798-9

www.elsevier.com/locate/gene

Abbreviations: ANOVA, analysis of variance; bp, base pair(s); cDNA,

DNA complementary to RNA; cpm, counts per minute; dNTP, deoxyribo-

nucleoside triphosphate; d.p.c., day post coitum; DTT, dithiothreitol;

EDTA, ethylenediaminetetraacetic acid; ENase, restriction endonuclease;

EtdBr, ethidium bromide; FCS, fetal calf serum; FKHR, fork head domain

in rhabdomyosarcoma gene; kb, kilobase(s) or 1000 bp; nt, nucleotide(s);

Pax, paired box-containing gene; PAX7-LPR, PAX7 gene-linked poly-

morphic region; PBS, phosphate-buffered saline; PCR, polymerase chain

reaction; PMA, phorbol 12-myristate 13-acetate; RACE, rapid amplifica-

tion of cDNA ends; RNase, ribonuclease; RT-PCR, reverse transcription

PCR; SDS, sodium dodecyl sulfate; UTR, untranslated region

* Corresponding author. Tel.: 149-931-201-77-600; fax: 149-931-201-

77-620.

E-mail address: kplesch@mail.uni-wuerzburg.de (K.-P. Lesch).1 Present address: German Cancer Research Center; Heidelberg,

Germany.2 Present address: Institute of Oncological Chemistry, University of

Dusseldorf, Dusseldorf, Germany.

day 9 post coitum (d.p.c.) in the ventricular zone of the

neural tube, which generates a stem cell population for the

entire central nervous system, and dermomyotomal parts of

the somites, from which muscle cells derive (Jostes et al.,

1991). At early stages Pax7 expression is also detected in all

brain vesicles, but later (11 d.p.c.) this expression is

retracted to the mesencephalon (Jostes et al., 1991; Stoy-

kova and Gruss, 1994). No spontaneous mutations in Pax7/

PAX7 were found. Mice with a targeted inactivation of Pax7

have been generated (Mansouri et al., 1996). Homozygous

knockout animals die within the first three postnatal weeks

and show malformations in facial skeletal structures. Cell

culture and electron microscopic analysis revealed a

complete absence of satellite cells in Pax72/2 skeletal

muscle, whereas the proportion of muscle-derived stem

cells was unaffected. In contrast to its paralogue, Pax3,

which plays an essential role in regulating the developmen-

tal program of embryonic myoblasts (Maroto et al., 1997;

Tajbakhsh et al., 1997), Pax7 is expressed in satellite cells

residing in the adult muscle (Seale et al., 2000). Interest-

ingly, differential expression of alternately spliced Pax7

transcripts correlates with muscle regenerative efficiency

in different strains of mice (Kay et al., 1995, 1998).

A specific chromosome translocation, t(1;13)(p36;q14),

resulting in the fusion of PAX7 to a fork head domain

gene (FKHR), has been implicated in the genesis of alveolar

rhabdomyosarcoma (Davis et al., 1994). The resulting

chimeric protein is a more potent transcriptional activator

compared to the wild-type Pax7 protein and possibly acts by

suppressing the apoptotic program normally responsible for

elimination of aberrant cells (Bernasconi et al., 1996). Cells

of other tumors, embryonal rhabdomyosarcoma and

Ewing’s sarcoma, were found to express wild-type PAX7

at elevated levels in comparison with primary human

myoblasts (Bernasconi et al., 1996; Barr et al., 1999).

Thus, PAX7 gene expression, which is rigorously controlled

during normal tissue development, is abnormally regulated

in certain pathological states. Therefore, in order to eluci-

date the molecular mechanisms controlling the expression

of the human PAX7 gene in normal and pathological tissues,

we have determined the transcription start site(s) and carried

out the structural and functional analysis of the regulatory

region responsible for its expression in adult skeletal

muscle. Finally, a polymorphic repetitive sequence located

within the basal promoter and its functional assessment are

reported.

2. Materials and methods

2.1. Isolation and sequencing of the 5 0-flanking regulatory

region

The human genomic PAC library no. 704 (Resource

Center/Primary Database of the German Human Genome

Project, Berlin), constructed by ligation of DNA from a

human male fibroblast cell line into pCYPAC-2 vector,

was screened under high-stringency hybridization condi-

tions with a 386 bp 32P-labeled EcoRI-SspI DNA fragment

encompassing a part of the human PAX7 promoter sequence

deposited in the EMBL/GenBank database (Accession

number: X96744). PAC DNA preparation from a single

positive clone (RPCIP704B248Q25), restriction endonu-

clease (ENase) mapping, and subcloning of restriction frag-

ments were carried out according to established protocols.

Fragments containing exon 1 and the 5 0-flanking sequences

were identified by Southern analysis and subcloned into

pBluescript II KS(1) vector (Stratagene). The 3 0 end of

the PAC clone was identified by Southern analysis using

four different short cDNA fragments containing the exons

1, 2, 3 and 4 as probes. Double-stranded DNA was

sequenced by the dideoxy chain termination method with

flanking reverse and forward primers as well as by primer

walking. Sequence analysis of the 5 0-regulatory region was

performed using on-line program MatInspector V2.2 (http://

transfac.gbf.de/TRANSFAC/index.html). The sequence

was deposited in the EMBL/GenBank database (Accession

number: AJ130875).

2.2. Northern blot

Human Multiple Tissue Northern (MTN) Blot membrane

(Clontech) with equal amounts of poly(A)1 RNA (2 mg)

from various peripheral tissues was hybridized with a 573

bp 32P-labeled fragment of human PAX7 cDNA using the

protocol provided by the manufacturer. The cDNA probe

was amplified by PCR with primers P7-1 (5 0-GCGCAAG-

CAGCGACGCAGTCG) and P7-2 (5 0-GAGAAGTCAGC-

CTGTGGCTGCG). Hybridized membrane was exposed to

X-ray film (Kodak XR) at 2708C.

2.3. RT-PCR amplification

For RT-PCR analysis, 5 mg of total RNA isolated from

frontal, temporal, occipital and entorhinal cortex, cerebel-

lum, thalamus, subthalamic nuclei, substantia nigra, hippo-

campus, putamen, amygdala and caudate nucleus were

reverse transcribed with 200 units of Superscript II reverse

transcriptase (Gibco BRL) in a 20 ml reaction volume

following the manufacturer’s recommended protocol. 3 ml

of this reaction was used as template for PCR with the

paired-box-specific primers P7-3 (5 0-TTGCCG-CTAC-

CAGGAGACCG) and P7-4 (5 0-CCAGGATGCC-

GTCGATGCTG) and b-actin specific primers (5 0-CCT-

CGCCTTTGCCGATCC; 5 0-GGATCTTCATGAGGTAG-

TCAGTC). PCR (40 s at 948C, 30 s at 588C, 60 s at 728C

for 35 cycles) was performed in a final volume of 50 ml

containing 20 pmol of each primer, 200 mM of each

dNTP in 1.5 mM MgCl2, 20 mM (NH4)2SO4, 75 mM

Tris–HCl (pH 9.0), 0.01% Tween-20, and 0.5 units of Taq

DNA polymerase (Eurogentec). Fragments were separated

on a 2% agarose gel stained with EtdBr.

Y.V. Syagailo et al. / Gene 294 (2002) 259–268260

2.4. Primer extension analysis

Antisense oligonucleotides P7-5 (5 0-AGCCGGCCGCA-

TCATTCTCGGTAC) and P7-6 (5 0-GACCACACTCGCT-

TCTCCCTCCTC) were end-labeled with [g-32P]ATP (3000

Ci/mmol) using T4 polynucleotide kinase (Promega). 0.5

mg of poly(A)1 RNA from human adult skeletal muscle

or 30 mg of yeast tRNA were mixed with the labeled primer

(1.0 £ 105 cpm) in 20 ml of hybridization buffer (225 mM

KCl and 10 mM Tris–HCl (pH 8.3)), denatured at 858C for

10 min, and incubated at 658C for 30 min. The hybridization

mix was slowly cooled to 378C for 60 min, and 40 ml of the

reverse transcription mixture (50 mM Tris–HCl (pH 8.3), 75

mM KCl, 3 mM MgCl2, 7.5 mM DTT, 0.75 mM dNTPs, and

30 units of RNasin) was then added. Primer was extended by

200 units of Superscript II reverse transcriptase (Gibco

BRL) at 428C for 60 min, and reactions were terminated

by addition of 5 ml of 0.5 M EDTA (pH 8.0) and 0.5 ml

of 10 mg/ml RNase A1. After incubation at 378C for 30 min,

extended fragments were purified by phenol-chloroform

extraction, analyzed on 6% polyacrylamide/8 M urea gels,

and sized against sequencing reactions.

2.5. 5 0-Rapid amplification of cDNA ends (5 0-RACE)

5 0-RACE was performed with 1 mg of poly(A)1 RNA

using the Marathon RACE kit (Clontech) according to the

manufacturer’s instructions. Initial amplification was

carried out with gene-specific primer P7-7 (5 0-GCCTTCT-

TTCTCCGGACCACACTC), followed by nested PCR

using gene-specific primer P7-6. PCR products were cloned

in pGEM-T vector (Promega) and automatically sequenced.

2.6. Trinucleotide repeat genotyping

Genomic DNA from 227 unrelated healthy volunteers

(133 males, mean age 28.7 ^ 7.9 years; 94 females,

32.3 ^ 11.6 years) was extracted from whole blood. The

study was approved by the ethics committee of the Univer-

sity of Wurzburg, and informed consent was obtained from

all subjects. Primers P7-8 (5 0-CGTTATTGGTCCTCCG-

CTC) and P7-9 (5 0-CGCGAGTGATCAGCTGGGTA)

flanking the PAX7 gene-linked polymorphic region

(PAX7-LPR) located directly upstream of the transcription

start site were used to generate fragments containing the

(CCT)n trinucleotide repeat. PCR (40 s at 948C, 30 s at

598C, 60 s at 728C for 38 cycles) was performed in a final

volume of 25 ml containing 60 ng genomic DNA, 10 pmol

of each primer, 200 mM of each dNTP in 1.5 mM MgCl2, 20

mM (NH4)2SO4, 75 mM Tris–HCl (pH 9.0), 0.01% Tween-

20, 2 mCi of [a-32P]dCTP (10 mCi/ml, 3000 Ci/mmol), and

0.2 units of Taq DNA polymerase (Eurogentec). Products

were fractionated by electrophoresis through 6% sequen-

cing gels.

2.7. Construction of reporter gene plasmids

The 2.97 kb XbaI-NaeI fragment including part of exon 1

and its upstream region was subcloned into the promoter-

less, luciferase reporter gene vector pGL3 basic (Promega)

in sense orientation and designated pXN. Serial deletion

fragments of pXN, HindIII-NaeI (21605 to 1373, pHN),

BglII-NaeI (21388 to 1373, pBgN), PstI-NaeI (21159 to

1373, pPN), EcoRI-NaeI (2633 to 1373, pRIN), BamHI-

NaeI (2403 to 1373, pBN), PvuII-NaeI (270 to 1373,

pPvN), and SmaI-NaeI (142 to 1373, pSN) were also

studied. To obtain construct pXND, pXN was digested

with PstI and re-ligated resulting in a plasmid containing

the 2.97 kb XbaI-NaeI fragment lacking 1.18 kb. Fragments

BamHI-NaeI (pNB), and PvuII-NaeI (pNPv) were also

fused to the luciferase gene in an antisense orientation.

The 1.0 kb EcoRI-NaeI (pRIN) fragment contains 10 CCT

repeats (allele 10, (CCT)10). For generation of constructs

with different length of repetitive sequence, genomic

DNAs with alleles containing length variants of the

(CCT)n repeat were amplified with primers P7-10 (5 0-

GCTGCAGGAATTCCTCCCTTC) and P7-11 (5 0-GCGA-

GTGATCAGCTGGGTAAG). EcoRI and PvuII-digested

products (alleles 8 and 11) were then ligated into pRIN.

All constructs were verified by cycle sequencing.

2.8. Cell culture

Cos-7 (African green monkey kidney, DSMZ ACC60)

cells were grown in Dulbecco’s modified Eagle’s medium

supplemented with 10% fetal calf serum (FCS), 50 mg/ml

streptomycin and 50 units/ml penicillin. JAR cells (human

placental choriocarcinoma, ATCC HTB-144) were grown in

RPMI 1640 medium supplemented with 10% FCS, 50 mg/

ml streptomycin and 50 units/ml penicillin. RD cells

(human embryonal rhabdomyosarcoma, ECACC 85111

502) were maintained in EMEM medium supplemented

with 2 mM glutamine, 2% NEAA, 2% vitamins, 10%

FCS, 50 mg/ml streptomycin and 50 units/ml penicillin.

All cells were cultured at 378C in a humidified 5% CO2

atmosphere.

2.9. Transfection and luciferase assays

Transfections were performed by the lipofection method

(Roche) according to standard procedures. Briefly, approxi-

mately 6 £ 105 cells on 60 mm plates were cotransfected

with 2.5 mg of luciferase constructs and 0.5 mg of

pRSV40-lacZ plasmid to control for transfection efficiency.

Cells were harvested in 250 ml lysis buffer (Roche) 40 h

after transfection. Luciferase assays were conducted with

100 ml of cell extracts using luciferin reagent (Promega).

Chemoluminescence was counted for 10 s in a liquid scin-

tillation spectrophotometer. The extracts (10 ml) were also

tested for b-galactosidase activity, and luciferase activities

were normalized to b-galactosidase activities using equal

Y.V. Syagailo et al. / Gene 294 (2002) 259–268 261

amounts of total protein determined by the method of Brad-

ford.

3. Results

3.1. Cloning of the 5 0-flanking regulatory region of the

human PAX7 gene

Screening of the human PAC library with a probe target-

ing a part of the human PAX7 5 0-flanking region yielded a

single ,70 kb clone covering the first three exons of PAX7

including its 5 0-flanking regulatory region. The partial

physical map of this clone is depicted in Fig. 1B. The

XhoI-BamHI (,4 kb) and the EcoRI-XhoI (,2.7 kb) frag-

ments were further characterized by restriction mapping and

sequencing (Fig. 1B).

3.2. Expression of the PAX7 gene

To determine the expression pattern of PAX7 and the size

of its mRNA, Northern blot from various human adult

tissues was hybridized with a cDNA fragment consisting

of exons 5–8. The probe detected a transcript of approxi-

mately 6.3 kb exclusively in adult skeletal muscle (Fig. 2).

No expression was observed in heart, brain, placenta, lung,

liver, kidney and pancreas. However, the existence of alter-

nate or truncated splice variants in tissues other than skeletal

muscle cannot be excluded. RT-PCR analysis of different

Y.V. Syagailo et al. / Gene 294 (2002) 259–268262

Fig. 1. Genomic structure of the 5 0-flanking regulatory region of the human PAX7 gene. (A) Schematic representation of the PAX7 gene physical map based on

the Homo sapiens DNA sequence from PAC 394P21 (Accession number: AL021528). Exons are shown as black boxes. (B) Partial restriction map of the

human PAC genomic clone RPCIP704B248Q25. X, XhoI; Xb, XbaI; H, HindIII; E, EcoRI; B, BamHI; N, NotI; S, SalI restriction sites. The transcription start

site is indicated. (C) Nucleotide sequence of the PAX7 gene-linked polymorphic region (PAX7-LPR; underlined sequences represent PCR primers P7-8 and

P7-9). The (CCT)n repeat is shown in boldface type.

Fig. 2. Northern analysis of PAX7 expression in peripheral tissues. A 6.3 kb

transcript was detected only in adult skeletal muscle.

regions of human adult brain using paired-box-specific

primers revealed transcripts in cerebellum and subthalamic

nuclei, as well as weak bands in temporal and frontal cortex,

thalamus, amygdala and putamen (Fig. 3).

3.3. Mapping of transcription start sites

Murmann et al. (2000) determined a major transcription

start site 70 bp upstream of the first ATG in the C2C12 mouse

myoblasts and tumor A673 cell lines, while the longest 5 0-

UTR (599 bp) was described for RNA isolated from alveolar

rhabdomyosarcoma tumor (Vorobyov et al., 1997). There-

fore, in order to define the transcription initiation site(s) of

PAX7 in adult skeletal muscle, primer extension analysis was

performed with primers (P7-5 and P7-6) from two putative

candidate regions. The primer extension analysis with distal

P7-5 primer revealed no extension product within the region

presented by Murmann et al. (2000), but showed an extension

product of a higher molecular weight, out of range of the

sequencing ladder, suggesting the presence of a possible

start site in the region described for alveolar rhabdomyosar-

coma (Vorobyov et al., 1997) (data not shown). The primer

extension analysis with primer P7-6 from the proximal

region yielded two extended products corresponding to size

of 76 and 100 nt (Fig. 4, lane 1). No specific signals were

observed in the negative control using yeast tRNA (Fig. 4,

lane 2). In order to confirm the results obtained with primer

extension, 5 0-RACE analysis using antisense primers speci-

fic for the proximal region was performed. 5 0-RACE consis-

tently produced a 91 bp fragment in two independent

experiments. The 5 0 end of this fragment revealed by

sequence analysis of ten clones and two primer extension

products are indicated in Fig. 5. The most upstream nucleo-

tide position found for the 5 0 end of the PAX7 cDNA initia-

tion site is labeled 11 in subsequent figures.

3.4. PAX7 promoter activity

Analysis of the 5 0-flanking sequence revealed a TATA-

like motif localized at position 2177 nt relative to the major

transcription initiation site (11). No obvious CCAAT

consensus was found within 100 bp 5 0 to the start site, but

one CCAAT box in reverse complementary orientation was

found 222 bp upstream of the 11 site (Fig. 5). Several

consensus sites for common transcription regulators includ-

ing Oct-1 (consensus binding motif NNGAATATKCANN-

NN), NF1 (core string TGGC), AP2 (MKCCCSCNGGCG),

AP4 (WGARYCAGCTGYGGNCNK), CREB (NNGNT-

GACGYNN), Sp1 (KRGGCGKRRY), Nkx2.5 (TYAAG-

TG), and MyoD (NNCANTGNY) (Quandt et al., 1995)

were mapped to the promoter region. The second consensus

site for a TATA box (ataTATAaataaata) motif was found at

position 1153 relative to the initiation start site. To identify

the DNA regions responsible for the control of PAX7 gene

transcription we analyzed the ability of serially truncated

fragments from the 5 0 end of PAX7 to drive the expression

of a luciferase reporter gene. Previous data (Schafer et al.,

1994) prompted us to choose the rhabdomyosarcoma cell

line (RD), which displays robust constitutive PAX7 expres-

sion, as an in vitro model for reporter gene experiments.

Additionally, JAR cells which are devoid of endogenous

PAX7 and Cos-7 cells expressing low levels of PAX7

mRNA (data not shown) were used for transient transfection

assays. Absolute activities of promoter constructs were up to

six-fold lower in JAR cells and up to three-fold lower in

Cos-7 compared to RD cells. The deletion constructs and

Y.V. Syagailo et al. / Gene 294 (2002) 259–268 263

Fig. 3. Expression of the PAX7 gene in human adult brain. Total RNA from temporal (1), occipital (9), frontal (11) and entorhinal (12) cortex, hippocampus (2),

thalamus (3), cerebellum (4), amygdala (5), substantia nigra (6), caudate nucleus (7), subthalamic nuclei (8) and putamen (10) was reverse-transcribed and

subjected to cDNA amplification using primers for PAX7 and b-actin. PCR products were separated on 2% agarose gel stained with EtdBr. DNA ladder

(Fermentas), which contains pUC19 plasmid DNA cut with MspI, was used as a size marker (M).

their promoter activities are depicted in Fig. 6. The pGL3

control vector was used as the reference construct. The

BamHI-NaeI fragment (pBN, bases 2403 to 1373) showed

the highest luciferase activity in RD cells, which was about

6% of pGL3 control vector activity (for comparison, the

activity of pBN in JAR or Cos-7 cells relative to pGL3

control vector was ,1% and 2%, respectively). Insertion

of the BamHI-NaeI fragment in a reverse orientation into

luciferase vector (pNB) revealed ,50% activity of the frag-

ment in sense orientation. Orientation-independent activity

of the basal promoter was previously reported for many

other promoters (Giger et al., 1995; Kozlov et al., 1995;

Okladnova et al., 1999). Further deletion of a fragment to

the PvuII site (270), which contains the TATA-like motif at

position 2177, still exhibits a relatively high level of

promoter activity. These data indicate that the TATA-like

motif and the sequence from the position 2177 to 270 are

not critical for PAX7 promoter activity in vitro. However,

the deletion of a further 111 nucleotides up to the SmaI site

(142) significantly decreased the promoter activity in RD

cells, indicating that the sequence around the initiation sites

is required for normal PAX7 promoter activity. Addition of

further sequences (up to 2.97 kb, construct pXN) had no

significant effect on transcription activity in RD cells (Fig.

6). No differences in PAX7 promoter activity were detected

between the various fragments used for Cos-7 transfection

(Fig. 6). In JAR cells (Fig. 6), the construct pBgN (21388 to

1373) showed increased activity in comparison to smaller

deletion fragments. Additionally, the fragment between

BglII and HindIII (position 21605) sites had a slightly

negative effect on the luciferase activity.

3.5. Allele frequencies and functional analysis of the

promoter-associated polymorphism

Sequencing of the PAX7 5 0-flanking region revealed a

(CCT)10 repeat located at positions 2136 to 2107 relative

to the position 11 of the major transcription initiation site.

The position and sequence of the repeated region are shown

in Fig. 1C. To determine whether this trinucleotide sequence

is polymorphic, flanking primers were used to PCR-amplify

genomic DNA from 227 unrelated individuals of Caucasian

population. Analysis of the reaction products revealed three

differentially sized fragments, which were subcloned and

sequenced. Allele 8 (eight repeat units; frequency 0.38)

and allele 10 (ten repeat units; frequency 0.59) were the

most abundant forms in the population study, whereas allele

11 (11 repeat units; frequency 0.03) was rare. The PAX7-

LPR allele frequencies and genotype distribution are

depicted in Fig. 7. The heterozygosity rate was 0.54.

The location of the PAX7-LPR suggested that it might

have an effect on transcription initiation driven by the basal

PAX7 promoter. To evaluate the impact of the PAX7-LPR

on transcriptional efficiency, RD cells were transiently

transfected with the luciferase reporter gene fused to 1.0

kb of the 5 0-flanking regulatory sequence (construct pRIN)

containing three length variants of the repeat. The (CCT)11

variant conferred a significantly higher transcriptional effi-

ciency on the PAX7 promoter in RD cells in comparison to

alleles with eight or ten repeat units (P , 0:05, one-way

ANOVA followed by Fisher’s PLSD test) (Fig. 8).

4. Discussion

The regulatory mechanism of PAX7 gene expression was

studied by isolation and functional characterization of the

promoter region initiating the transcription of the PAX7

mRNA in adult skeletal muscle. A transcript of ,6.3 kb

in length was detected in adult skeletal muscle by Northern

analysis of human peripheral tissues, indicating that the

human PAX7 mRNA is larger than the 4.9 kb mouse Pax7

transcript found in total embryonic tissue (Jostes et al.,

1991). A transcript of the same length was observed for

wild-type PAX7 in RD (embryonal rhabdomyosarcoma)

and A673 (Ewing’s sarcoma) cell lines by Barr et al.

Y.V. Syagailo et al. / Gene 294 (2002) 259–268264

Fig. 4. Determination of PAX7 transcription initiation sites by primer exten-

sion. Extension products were run against a sequencing reaction of the

PAX7 promoter region with primer P7-7 also used for primer extension

analysis. Lanes represent primer extension experiments with 0.5 mg

mRNA of human adult skeletal muscle (lane 1) and 30 mg of yeast tRNA

as negative control (lane 2). The protected bands are indicated by arrows.

Primer extension of yeast tRNA showed an unspecific extension product.

(1999). However, the longest wild-type PAX7 cDNA (2.2

kb), that was previously reported to contain the complete

coding region and 5 0 sequence, is considerably shorter

(Vorobyov et al., 1997). Therefore, in order to determine

the full length of the 5 0-UTR, we performed primer exten-

sion and 5 0-RACE. Primer extension analysis identified two

transcription start sites 640 bp and 664 bp upstream of the

ATG codon in adult skeletal muscle mRNA. The smaller

primer-extended product may represent an additional start

site or indicates mRNA secondary structure/degradation. A

transcriptional start site 655 bp upstream of the translation

initiation codon was mapped by 5 0-RACE with a slight

discrepancy (9 bp) in comparison to primer extension

results. The presence of GC-rich stretches in mRNA

sequences could result in the premature termination of the

reverse transcription of the long mRNA from poly(dT)

primer by 5 0-RACE procedure. Thus, PAX7 mRNA

contains a long 5 0-UTR encompassing 664 nucleotides.

Cellular mRNAs containing complex 5 0-UTR may be

post-transcriptionally regulated by their 5 0-UTR and are

Y.V. Syagailo et al. / Gene 294 (2002) 259–268 265

Fig. 5. Nucleotide sequence of the first exon in uppercase and a partial 5 0-flanking region of the human PAX7 gene (EMBL/GenBank accession number:

AJ130875). Restriction endonuclease sites used to generate deletion constructs are indicated. The transcription start sites detected by primer extension and 5 0-

RACE assays are denoted by arrows. Numbering of the sequence is relative to the major upstream transcription start site (11). Consensus sequences for DNA-

binding proteins (TATA-like motif; AP-2; AP-4; NF-1; MyoD; Oct1; Sp1; CCAAT; Nkx2.5) and the (CCT)n repeat are underlined.

Y.V. Syagailo et al. / Gene 294 (2002) 259–268266

Fig. 6. Transcriptional activity of the human PAX7 gene 5 0-flanking region. Transient transfection experiments were carried out in RD, Cos-7, and JAR cells

using serial 5 0 deletions of promoter-exon 1/luciferase reporter gene (Luc) fusion constructs. Luciferase transcription efficiency was studied in comparison to

pGL3 basic vector (pGL3b). Relative promoter-induced activities were calculated relative to pGL3 control vector (100%, data not shown). Results from three

to four independent experiments performed in duplicates and normalized for transfection efficiency are displayed as means ^ SD.

Fig. 7. (A) Allele frequencies of the PAX7-LPR in unrelated individuals (N ¼ 227). (B) Genotype frequencies of the PAX7-LPR.

often found to encode regulatory proteins, proto-oncogenes,

growth factors and their receptors, and homeodomain tran-

scription factors (van der Velden and Thomas, 1999).

In most eukaryotic genes, the TATA box is located 25–30

bp upstream of the mRNA initiation sites (Johnson and

McKnight, 1989) and directs accurate transcription initiation

through a mechanism that depends on its specific interaction

with a multiprotein complex called TFIID (Kaufmann and

Smale, 1994; Ghosh and Van Duyne, 1996). The A/T-rich

short TATA-like sequence (gaATAAAAgaaaa) found in the

PAX7 5 0-flanking region and located 177 bp upstream of the

transcription initiation site does not completely match the

TATA-box consensus sequence (STATAAAWRNNNN-

NNN) and is not critical for expression of the PAX7 gene

in vitro as was shown by functional analysis. Thus, in contrast

to the promoter of the human PAX3 gene, that contains

TATA and two CCAAT boxes upstream of the start site at

appropriate positions for classical promoter structure

(Okladnova et al., 1999), the PAX7 promoter lacks a canoni-

cal TATA box and resembles previously described TATA-

less human PAX2 and mouse Pax4 gene regulatory regions

(Stayner et al., 1998; Brink et al., 2001). Several potential

binding sites for Sp1, AP-2, AP-4, NF-1, MyoD, Oct1,

Nkx2.5, CREB, and CCAAT-binding (in negative orienta-

tion) transcription factors were identified upstream and

downstream of the transcription start site.

Functional mapping of the gene regulatory region

revealed the sequence between 270 and 1373 bp essential

for minimal basal activity of the PAX7 promoter. The maxi-

mal promoter activity was detected for pBN construct

(2403 to 1373 bp). Progressive addition of further

sequences (up to 2.97 kb, construct pXN) does not effec-

tively change the transcriptional activity detected with this

construct. Only the fragment between BamHI and EcoRI

sites (pRIN construct) has some regulatory elements,

which have a slightly negative effect on the promoter activ-

ity in RD cells. Our results suggest that the large region of

the human PAX7 promoter between XbaI and BamHI does

not contain any strong positive or negative regulatory

elements affecting the PAX7 control of transcription,

however, it is not unusual for tissue-specific regulatory

sequences to be located at a long distance from the tran-

scription start (Hardison et al., 1997; Jaakkola et al., 1997;

Maul et al., 1998). Further studies are required to define

factors necessary for PAX7 activation and regulation.

Our findings provide the first evidence for the expression

of the PAX7 gene in several regions of adult human brain.

The Pax7 gene is implicated in neurogenesis during devel-

opment and its expression was detected in the specific

regions of embryonic brain (Jostes et al., 1991; Stoykova

and Gruss, 1994) as well as in the adult brain of the mouse

(Ziman et al., 1997). Though Northern analysis with a speci-

fic probe from the 3 0 region of the gene showed no expres-

sion of PAX7 in adult human brain, the existence of alternate

or truncated transcripts could not be excluded. Indeed, RT-

PCR analysis with paired-box-specific primers indicated

PAX7 transcripts in cerebellar, temporal and frontal cortex,

subthalamic nuclei, thalamus, amygdala and putamen. Inter-

estingly, transcripts containing only the paired domain

encoding region of the PAX3 gene were found in many

human adult tissues including different brain regions

(Tsukamoto et al., 1994).

Results of our study have also demonstrated that the poly-

morphic (CCT)n repeat located 107 bp upstream of the tran-

scription start site within the core regulatory region shows

length variation in a western European population. Tandem

arrays of trinucleotides (triplet repeats) are abundant in

eukaryotic genomes and a subset of these repeats is asso-

ciated with the triplet expansion disorders (Mitas, 1997).

Intrastrand or interstrand folded structures, such as hairpins,

tetraplexes and triplexes have been described for the triplet

expansion disease-associated repeats (Kashi et al., 1997).

Such structures may promote triplet expansion by favoring

strand-slippage during DNA replication (Sinden and Wells,

1992). The presence of these structures on the template strand

may block the forward progress of DNA polymerase, thus

promoting dissociation of the polymerase and increasing the

likelihood that the nascent strand and the template strand will

slip relative to one another. Structures promoting the triplet

expansion may also block the forward progression of RNA

polymerase during transcription. Therefore, the PAX7

(CCT)n repetitive element described here may influence the

transcriptional efficiency of the PAX7 promoter. Population

and family-based association studies are currently under way

to determine the role of PAX7-LPR genotypes in neurode-

velopmental and neuromuscular disorders.

Acknowledgements

We thank G. Ortega and M. Schad for technical expertise,

G. Stober for sharing DNA samples, and J. Nicol for secre-

Y.V. Syagailo et al. / Gene 294 (2002) 259–268 267

Fig. 8. Transcriptional activity of PAX7-LPR alleles. The 1.0 kb fragments

EcoRI-NaeI containing 8, 10, or 11 repeat units had been fused to the

luciferase expression vector and transfected into RD cells. Luciferase activ-

ities of allelic constructs are calculated relatively to allele 10 (construct

pRIN) and displayed as means ^ SD. Allele 11 vs. allele 8 or allele 10,

**P , 0:05, one-way ANOVA followed by Fisher’s PLSD test.

tarial assistance. This work was supported by the Bundes-

ministerium fur Bildung, Wissenschaft, Forschung und

Technologie (BMBF, 512-4001-01KS9603 and

01GA9802/5).

References

Barr, F.G., Fitzgerald, J.C., Ginsberg, J.P., Vanella, M.L., Davis, R.J.,

Bennicelli, J.L., 1999. Predominant expression of alternative PAX3

and PAX7 forms in myogenic and neural tumor cell lines. Cancer

Res. 59, 5443–5448.

Bernasconi, M., Remppis, A., Fredericks, W.J., Rauscher, F.J., Schafer,

B.W., 1996. Induction of apoptosis in rhabdomyosarcoma cells through

down-regulation of PAX proteins. Proc. Natl. Acad. Sci. USA 93,

13164–13169.

Bopp, D., Burri, M., Baumgartner, S., Frigerio, G., Noll, M., 1986. Conser-

vation of a large protein domain in the segmentation gene paired and in

functionally related genes of Drosophila. Cell 47, 1033–1040.

Brink, C., Chowdhury, K., Gruss, P., 2001. Pax4 regulatory elements

mediate beta cell specific expression in the pancreas. Mech. Dev.

100, 37–43.

Chalepakis, G., Stoykova, A., Wijnholds, J., Tremblay, P., Gruss, P., 1993.

Pax: gene regulators in the developing nervous system. J. Neurobiol. 24,

1367–1384.

Davis, R.J., D’Cruz, C.M., Lovell, M.A., Biegel, J.A., Barr, F.G., 1994.

Fusion of PAX7 to FKHR by the variant t(1;13)(p36;q14) translocation

in alveolar rhabdomyosarcoma. Cancer Res. 54, 2869–2872.

Ghosh, G., Van Duyne, G.D., 1996. Pieces of the puzzle: assembling the

preinitiation complex of Pol II. Structure 4, 891–895.

Giger, R.J., Vogt, L., Zuellig, R.A., Rader, C., Henehen-Beatty, A., Wolfer,

D.P., Sonderegger, P., 1995. The gene of chicken axonin: complete

structure and analysis of the promoter. Eur. J. Biochem. 227, 617–628.

Gruss, P., Walther, C., 1992. Pax in development. Cell 69, 719–722.

Hardison, R., Slightom, J.L., Gumucio, D.L., Goodman, M., Stojanovic, N.,

Miller, W., 1997. Locus control regions of mammalian beta-globin gene

clusters: combining phylogenetic analyses and experimental results to

gain functional insights. Gene 205, 73–94.

Jaakkola, P., Vihinen, T., Maatta, A., Jalkanen, M., 1997. Activation of an

enhancer on the syndecan-1 gene is restricted to fibroblast growth factor

family members in mesenchymal cells. Mol. Cell. Biol. 17, 3210–3219.

Johnson, P.F., McKnight, S.L., 1989. Eukaryotic transcriptional regulatory

proteins. Annu. Rev. Biochem. 58, 799–839.

Jostes, B., Walther, C., Gruss, P., 1991. The murine paired box gene, Pax7,

is expressed specifically during the development of the nervous and

muscular system. Mech. Dev. 33, 27–37.

Kashi, Y., King, D., Soller, M., 1997. Simple sequence repeats as a source

of quantitative genetic variation. Trends Genet. 13, 74–78.

Kaufmann, J., Smale, S.T., 1994. Direct recognition of initiator elements by

a component of the transcription factor IID complex. Genes Dev. 8,

821–829.

Kay, P.H., Mitchell, C.A., Akkari, A., Papadimitriou, J.M., 1995. Associa-

tion of an unusual form of a Pax7-like gene with increased efficiency of

skeletal muscle regeneration. Gene 163, 171–177.

Kay, P.H., Harmon, D., Fletcher, S., Robertson, T., Ziman, M., Papadimi-

triou, J.M., 1998. Pax7 includes two polymorphic homeoboxes which

contain rearrangements associated with differences in the ability to

regenerate damaged skeletal muscle in adult mice. Int. J. Biochem.

Cell Biol. 30, 261–269.

Kozlov, S.V., Giger, R.J., Hasler, T., Korvatska, E., Schiorderet, D.F.,

Sonderegger, P., 1995. The human TAX1 gene encoding the axon-

associated cell adhesion molecule TAG-1/Axonin-1: genomic structure

and basic promoter. Genomics 30, 141–148.

Mansouri, A., Stoykova, A., Torres, M., Gruss, P., 1996. Dysgenesis of

cephalic neural crest derivatives in Pax7 2 /2 mutant mice. Develop-

ment 122, 831–838.

Maroto, M., Reshef, R., Munsterberg, A.E., Koester, S., Goulding, M.,

Lassar, A.B., 1997. Ectopic Pax-3 activates MyoD and Myf-5 expres-

sion in embryonic mesoderm and neural tissue. Cell 89, 139–148.

Maul, R.S., Zhang, H., Reid, J.D.T., Pedigo, N.G., Kaetzel, D.M., 1998.

Identification of a cell type-specific enhancer in the distal 5 0-region of

the platelet-derived growth factor A-chain gene. J. Biol. Chem. 273,

33239–33246.

Mitas, M., 1997. Trinucleotide repeats associated with human disease.

Nucleic Acids Res. 25, 2245–2254.

Murmann, O.V., Niggli, F., Schafer, B.W., 2000. Cloning and character-

ization of the human PAX7 promoter. Biol. Chem. 381, 331–335.

Okladnova, O., Syagailo, Y.V., Tranitz, M., Riederer, P., Stober, G., Moss-

ner, R., Lesch, K.P., 1999. Functional characterization of the human

PAX3 gene regulatory region. Genomics 57, 110–119.

Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T., 1995. MatInd

and MatInspector: new fast and versatile tools for detection of consen-

sus matches in nucleotide sequence data. Nucleic Acids Res. 23, 4878–

4884.

Schafer, B.W., Czerny, T., Bernasconi, M., Genini, M., Busslinger, M.,

1994. Molecular cloning and characterization of a human PAX-7

cDNA expressed in normal and neoplastic myocytes. Nucleic Acids

Res. 22, 4574–4582.

Seale, P., Sabourin, L.A., Girgis-Gabardo, A., Mansouri, A., Gruss, P.,

Rudnicki, M.A., 2000. Pax7 is required for the specification of

myogenic satellite cells. Cell 102, 777–786.

Sinden, R.R., Wells, R.D., 1992. DNA structure, mutations, and human

genetic disease. Curr. Opin. Biotechnol. 3, 612–622.

Stayner, C.K., Cunliffe, H.E., Ward, T.A., Eccles, M.R., 1998. Cloning and

characterization of the human PAX2 promoter. J. Biol. Chem. 273,

25472–25479.

Stoykova, A., Gruss, P., 1994. Roles of Pax-genes in developing and adult

brain as suggested by expression patterns. J. Neurosci. 14, 1395–1412.

Strachan, T., Read, A.P., 1994. PAX genes. Curr. Opin. Genet. Dev. 4, 427–

438.

Tajbakhsh, S., Rocancourt, D., Cossu, G., Buckingham, M., 1997. Redefin-

ing the genetic hierarchies controlling skeletal myogenesis: Pax-3 and

Myf-5 act upstream of MyoD. Cell 89, 127–138.

Tremblay, P., Gruss, P., 1994. Pax: genes for mice and men. Pharmacol.

Ther. 61, 205–226.

Tsukamoto, K., Nakamura, Y., Niikawa, N., 1994. Isolation of two

isoforms of the PAX3 gene transcripts and their tissue-specific alter-

native expression in human adult tissues. Hum. Genet. 93, 270–274.

van der Velden, A.W., Thomas, A.A., 1999. The role of the 5 0 untranslated

region of an mRNA in translation regulation during development. Int. J.

Biochem. Cell Biol. 31, 87–106.

Vorobyov, E., Mertsalov, I., Dockhorn-Dworniczak, B., Dworniczak, B.,

Horst, J., 1997. The genomic organization and the full coding region of

the human PAX7 gene. Genomics 45, 168–174.

Walther, C., Guenet, J.L., Simon, D., Deutsch, U., Jostes, B., Goulding,

M.D., Plachov, D., Balling, R., Gruss, P., 1991. Pax: a murine multi-

gene family of paired box-containing genes. Genomics 11, 424–434.

Ziman, M.R., Fletcher, S., Kay, P.H., 1997. Alternate Pax7 transcripts are

expressed specifically in skeletal muscle, brain and other organs of adult

mice. Int. J. Biochem. Cell Biol. 29, 1029–1036.

Y.V. Syagailo et al. / Gene 294 (2002) 259–268268