Tài liệu Báo cáo khoa học: Cell surface heparan sulfate proteoglycans Target and partners of the basic ®broblast growth factor in rat Sertoli cells pptx

Cell surface heparan sulfate proteoglycans
Target and partners of the basic ®broblast growth factor in rat Sertoli cells
Sylvie Brucato, Jean Bocquet and Corinne Villers
Laboratoire de Biochimie, IRBA, Universite
Â
de Caen, France
Basic ®broblast growth factor (bFGF) regulates diversi®ed
biological functions in rat Sertoli cells. This report demon-
strates t hat bFGF inhibits steroidogenesis in developing rat
Sertoli cells. Follicle stimulating horm one (FSH)-stimulated
estradiol p roduction was reduced by bFGF. M oreover, the
amount of cytochrome P450 aromatase, responsible for the
irreversible transformation of androgens into estrogens, is
decreased by b FGF at the transcriptional l evel. The bFGF
inhibitory eect was also observed in the presence of dibu-
tyryl-cAMP, ch olera toxin or R O-20-1724, all inducing high
levels of cAMP, the second messenger of FSH.
Heparan sulfate proteoglycans ( HSPGs) were shown t o be
required as cofactors for bFGF signaling. Indeed, sodium
chlorate, d escribed to drastically decrease prote oglycan sul-
fation, abolishes the bFGF do wnregulation of FSH-stimu-
lated estradiol synthesis p reviously observed. Glypican-1,
syndecan-1 and -4, potential bFGF coreceptors, are m ainly
regulated at the transcriptional level. This report s hows that
the b FGF r egulation of their expression speci®cally depends
on the nature of HSPG and of the Sertoli cell developmental
stage.
In conclusion, HSPG are partners and the target of bFGF
in rat Sertoli cells.
Keywords: bFGF; aromatase;heparan sulfate proteoglycans;
RT-PCR; Sertoli cells.

The basic ®broblast growth factor (bFGF or FGF-2)
belongs t o a large F GF fam ily of 21 structurally related
members [1]. This growth factor is produced by many cell
types and tissues, i ncluding testis [2]. Its biological activity is
pleiotropic [3] as it in¯uences aspects of both cellular
growth, differentiation but also angiogenesis, tissue repair
and cell migration. In rat testis, bFGF affects, for instance,
Leydig and Sertoli cell s teroidogenesis [4,5], Sertoli cell
transferrin production [6] and plasminogen activator activity
[7] but also c-fos [8] and FGFR-1 [9] mRNA expression.
The biological activity of bFGF is mediated by interac-
tion with high af®nity cell surface bFGF receptors (FGFR-1
to FGFR-4) [ 10]. In addition, bFGF b inds to h eparan
sulfate proteoglycans (HSPG) on the cell surface [11].
Oligosaccharidic sequences of HS chains are de®ned for the
bFGF binding and for the recognition of the speci®c bFGF
receptor, leading to the formation of a ternary complex
comprising HSPG±bFGF±FGFR. These oligosaccharidic
motifs are differently sulfated related to the synthesis
pathway itself and depending on the cell type. The r esulting
structural microheterogeneity modulates bFGF af®nity for
its coreceptor and, as a consequence, the growth factor
activity. Studies indicated that bFGF binding to HSPG
facilitates bFGF receptor binding and activation. bFGF
receptor binding to cells that do not express HSPG is
signi®cantly reduced when compared to cells expressing
HSPG [4,11±15].
Sertoli cells are the principal source of estradiol produc-
tion in the immature testis [16,17]. Signi®cant estrogen
synthesis is present in Sertoli cells of early postnatal rats,

with a sharp reduction during subsequent maturation
[18,19].
The p resent work ®rstly aims to evaluate the effect of
bFGF on follicle stimuling hormone (FSH)-estradiol syn-
thesis and cytochrome P 450 aromatase mRNA expression
in 20 days old-rat Sertoli cells. The involvement of t he
cAMP pathway w as evaluated using three approaches, all
inducing d ifferently high levels of cAMP: (a) dibutyryl cyclic
AMP (dbcAMP), a structural analogue of cAMP; (b)
cholera t oxin, a protein G s a ctivator; and (c) R O-20±1724, a
speci®c phosphodiesterase inhibitor.
Then, w e i nvestigated bFGF effect on FSH-estradiol
synthesis in t he absence of H SPG in 20-day-old-rat S ertoli
cells. These cells were treated with sodium chlorate to
completely inhibit sulfatation of proteoglycans and, in
consequence, abolish bFGF binding to HSPG.
Our previous studies indicated that in immature rat Sertoli
cells, cell surface proteoglycans are mainly represented by
HSPG [20,21] and among these, at least glypican-1, syn-
decan-1 and syndecan-4 mRNAs are e xpressed [22]. M ore-
over, syndecan-1 [23], syndecan-4 [24] and glypican-1 [25] are
potential coreceptors of bFGF, and are e ssentially regulated
at the transcriptional level [26]. Thus, using a semi-quanti-
tative RT-PCR, we had demonstrated in immature Sertoli
cells that glypican-1 and syndecan-1 mRNA expression was
Correspondence to S. Brucato, Laboratoire de Biochimie, IRBA,
Universite
Â
de Caen, Esplanade de la Paix, 14032 Caen cedex, France.
Fax:+33231954940,Tel.:+33231566576,

E-mail:
Abbreviations: bFGF, basic ®broblast growth factor; FSH, follicle
stimulating hormone; HSPG, heparan sulfate proteoglycan; PAPS,
phosphoadenosine phosphosulfate; FiRE, FGF-inducible response
element; FIN-1, FGF-inducible nuclear protein-1; DMEM, Dul-
becco's modi®ed Eagle's medium; RhFGF, recombinant human basic
FGF; PA PS, phosphoadenosine phosph osulfate; AMV, avian myelo-
blastosis virus.
(Received 30 May 2001, revised 1 October 2001, accepted 1 4
November 2 001)
Eur. J. Biochem. 269, 502±511 (2002) Ó FEBS 2002
speci®cally upregulated by PKC-activatio n i n c ontrast to
syndecan-4 transcription [22]. Until now, nothing has been
known about the bFGF regulation of glypican-1 and
syndecan-1 and -4 mRNA expression in developing Sertoli
cells. Our p resent study demonstrates that bFGF in¯uences
this expression in 20-day-old-rat Sertoli cells.
During testicular development, the physiology of Sertoli
cells i s m odi®ed. The cell proliferation decreases and ceases
allowing the establishment of the hematotesticular barrier
around the 20th day postpartum. In addition, some
enzymatic activities are modulated, such as the aromatase
activity which decreases upon ontogenesis. T herefore, the
study was extended and considered in 10- and 30-day-old-
rat Sertoli cells.
This report shows that HSPG are necessary for bFGF
signal transduction in acting as coreceptors and that HSPG
mRNAs expression is modulated by bFGF itself in
developing Sertoli cells.
MATERIALS AND METHODS

Materials
Ovine F SH (oFSH -21) was kindly p rovided b y t he National
Institute of A rthritis, Metabolic and Digestive Diseases
(Pituitary Hormone Distribution program, Bethesda, MD,
USA). Dulbecco's modi®ed Eagle's medium (DMEM),
Ham's F12 medium, Trypsin (USP Grade), trizol reagent
and DNA mass ladder were from Gibco-BRL (Cergy-
Pontoise, France). Collagenase-d ispase was from Boehrin-
ger-Mannheim (Meylan, France). Ultroser SF (steroid-free
serum substitute) was purchased from IBF-Biotechnics
(Villeneuve-La-Garenne, France). Bovine pancreas deoxy-
ribonuclease (DNase type I), hyaluronidase (type I-S),
testosterone, estradiol 17-b,(Bu)
2
cAMP (N6, 2¢-O-dibutyryl-
adenosine 3¢:5¢cyclic monophosphate), cholera toxin, Ro
20±1724, sodium chlorate, Hoescht 33258, calf thymus
DNA and agarose were purchased from Sigma (Saint-
Quentin Fallavier, France). Avian myeloblastosis virus
(AMV) reaction buffe r 5 ´, oligo d(T) 15, dNTPs, RNasin,
AMV-re verse transcr iptas e, Thermus aquaticus (Taq) DNA
polymerase reaction b uffer 1 0 ´, Taq DNA polymerase a nd
MgCl
2
were from Promega (Charbonnie
Á
re-les-bains,
France). The oligonucleotide primers were synthesized and
puri®ed by Eurobio (Les Ulis, France). Recombinant
human basic ®broblast growth factor (RhFGF) were from

R & D Systems (Abingdon, UK). 2,4,6,7[
3
H]-17b estradiol
(3.77 TBq ámmol
)1
) was from NEN (les Ulis, France). All
reagents were of analytical or molecular biology grade.
Cell culture
Ten-, 20- and 30-day-old Sprague±Dawley r ats f rom our
own colony were killed by cervical dislocation. Sertoli cells
were obtained b y sequential enzymatic digestion including
trypsin, collagenase and hyaluronidase as described previ-
ously [27].
Sertoli cells were seeded at the concentration of
250 000 cellsácm
)2
in 24-well dishes or in 75-cm
2
plastic
¯asks and cultured for48 h in Ham's F12/DMEM (1 : 1,v/v)
supplemented with 2% Ultroser SF i n order to attach the
Sertoli cells in a humidi®ed atmosphere of 5% CO
2
in air at
32 °C. Culture medium was rene wed after 48 h . Three days
after plating, residual germinal cells were removed by brief
hypotonic treatment using 20 m
M
Tris/HCl (pH 7.4) [28].
Sertoli cells were then cultured for two days in culture

medium devoid of Ultroser before being u sed on d ay 5 a fter
plating.
For the aromatase assay, Sertoli cells were incubated f or
24 h w ith testosterone (200 ngámL
)1
), oFSH (100 ngámL
)1
)
and/or bFGF (5 ngámL
)1
) and/or sodium chlorate (10 m
M
).
Before RNA extraction, cells were incubated for 24 h
either in the absence or in prese nce of FSH, dbcAMP,
cholera toxin or bFGF, either in combination with FSH or
dbcAMP or cholera toxin and bFGF.
Extraction of total RNA
Total RNA was extracted from rat Sertoli cells by single
step method of Chomczynski & Sacchi [29] using T rizol
reagent. T he integrity and quality o f puri®ed RNA were
controlled b y 1% agarose gel electrophoresis and measure
of the absorbance at 260 and 280 nm.
Semi-quantitative RT-PCR
Heat denatured total RNA (500 ng; 55±60 °C, 5 min) was
added to a reverse transcription r eaction m ixture containing
the reaction buffer (50 m
M
Tris/HCl, pH 8.3, 50 m
M

KCl;
10 m
M
MgCl
2
,0.5m
M
Spermidine, dithiothreito l 10 m
M
),
1 l
M
oligo d(T)
15
,500l
M
dNTPs, 20 UI RNasin, 18 UI
AMV-reve rse transcriptase in 20 lL ®nal volume. The
reaction was carried out at 37 °C for 60 min and followed
by 5 min denaturation at 95 °C.
Two microliters of the ®rst strand synthesis product
(0.1 lg) wa s used as template to amplify e ach cDNA. PCR
was performed with 250 l
M
dNTPs, Taq DNA polymerase
reaction buffer (50 m
M
KCl, 10 m
M
Tris/HCl, pH 9; 0 .1%

Triton X-100), 2.5 UI Taq DNA polymerase, MgCl
2
1.5 m
M
,10pmolofeachprimer(Table1)ina20-lL
reaction volume.
The PCR was started at 94 °C1minandfollowedbyup
to 27 cycles of am pli®cation for t he three proteoglycans and
20 cycles for the internal control, b-actin as described
previously [23], which consisted of a denaturating step (at
94 °C f or 1 min), an annealing s tep (at 55 °C f or 1 min) a nd
an extension step (at 72 °C for 2 min) then a ®nal
elongation step (at 72 °C for 10 min) in ROBOCYCLERÒ
Gradient 40 (Stratagene).
The cytochrome P450 aromatase c DNA was ampli®ed at
94 °C for 1 min for 3 0 cycles a s described previously [30],
which consisted of a denaturating step (at 94 °Cfor1min),
an annealing step (at 6 0 °C for 30 s) and an extension step
(at 72 °C for 1 min) then a ® nal elongation step (at 72 °C
for 10 min) in ROBOCYCLERÒ Gradient 40 (St ratagene).
To check for contaminating genomic DNA, a RT-PCR
was performed on RNA without AM V reverse tran scriptase
(data not shown). In a ll negative P CR control r eactions,
cDNA templates were replaced with sterile wate r to check
the absence of contaminants.
Aliquots ( 10 lL) of the P CR reaction w ere size-
separated on a 4% agarose gel equilibrated in Tris/
acetate/EDTA (40 m
M
Tris/acetate, 1 m

M
EDTA). Gels
were stained w ith e thidium b romide (1 lgámL
)1
), photo-
graphed using Polaroid ®lm under UV light and ana-
lysed using a AGFA SnapScan 1200
P
Scanner
Ò
, Adobe
Ó FEBS 2002 HSPG, target and partners of bFGF (Eur. J. Biochem. 269) 503
PHOTOSHOP
Ò
software and the
NIH IMAGE
computer
program ( />Radio immuno assay of estradiol 17-b
Culture medium was extracted with 5 vol. of diethylether
and estradiol was quanti®ed by radioimmunoassay using a
speci®c antibody purchased from Biosys (Compie
Á
gne,
France). The only signi®cant cross reactions were for
2-methoxy-estradiol ( 5%), e stra diol 17 a (0.28%), estrone
and estriol (0.45%). The sensitivity of the assay w as 6 pg per
tube. Intra- and interassay coef®cients of variation were less
than 10%. The analysis o f the radioimmunoassay data was
performed using the
SECURIA

program from the Packard
Instrument Company (Meriden, CT, USA).
DNA quanti®cation
The DNA content of the cell layer at the end of incubation
was quanti®ed by the method of West et al .[31].After
solubilization of the cell layer in 1
M
NaOH and s ubsequent
neutralization by 1
M
KH
2
PO
4
, DNA was quanti®ed in a
Kontron spectro¯uorimeter u sing Hoescht 33258 as ¯uo -
rescent probe and calf thymus as standard.
Statistical analysis
All experimental datawere presented as the meanof duplicate
(estradiol) determinations of three wells in, at l east, three
different cultures within e ach treatment grou p. Results were
normalized in pg (estradiol) per 10
6
cells. Statistical signi®-
cance between groups was determinated by Student's paired
t-test. Differe nces were c onsidered signi®cant at p < 0.05.
RESULTS
Cell surface HSPG are bFGF partners. bFGF inhibits
the FSH-stimulated estradiol synthesis in Sertoli cells
Sertoli cells from 20-days-old-rats were i ncubated for 24 h

with FSH (100 ngámL
)1
) and increasing concentrations of
bFGF (0.1±10 ngámL
)1
). FSH-stimulated estradiol synthe -
sis was inhibited and appeared to be dose-dependent
(Fig. 1 ). The maximal bFGF effect ()49%) on FSH-
induced estradiol production was r eached for 5 ngámL
)1
.In
contrast, estradiol synthesis was not regulated b y bFGF in
the absence of FSH (data not shown).
Direct implic ation o f cAMP increase, the second mes-
senger of FSH, in the bFGF regulation was evaluated by
addition to the culture medium of either 1 m
M
dbcAMP or
10 lgámL
)1
cholera toxin. Their addition increased e stradiol
production by a factor 9 and 7 , respectively (Table 2) as
FSH did (factor 9 ) (Fig. 1). In the presence of dbcAMP or
cholera toxin, bFGF addition induces a similar inhibition
()49% and )40%, r espectively) on estradiol synthesis as the
one described in Fig. 1 (Table 2).
The c AMP level elevation was also aprehended in the
presence of 100 ngámL
)1
FSH and 250 l

M
RO 20±1724, a
speci®c inhibitor o f cAMP-speci®c phosphodiesterase [32].
In these conditions, a signi®cant increase (about +65%) of
estradiol production was obtained (Fig. 2) as already
described. We observed that concomittant treatment with
5ngámL
)1
bFGF also induced a decrease (about )30%) of
FSH-stimulated estradiol synthesis. Nevertheless, this
decrease was lesser by comparison to the one observed in
the presence of FSH and bFGF ()49%). This result
suggested that bFGF action could induce, in part, a
Table 1 . Primers for PCR ampli®cation.
Target cDNA Localization and sequence of primer Size of PCR products (bp)
Syndecan-1 5¢-370 AGGTGCTTTGCCAGATATGACT-3¢ 432
5¢-802 CTCTTTGATGACAGAAGTGCCT-3¢
Syndecan-4 5¢-85 GAGTCGATTCGAGAGACTGA-3¢ 365
5¢-450 AAAAATGTTGCTGCCCTG-3¢
Glypican-1 5¢-566 GAATGACTCGGAGCGTACACTG-3¢ 488
5¢-1054 CCTTTGAGCACATTTCGGCAA-3¢
P450 aromatase 5¢-1555GCTTCTCATCGCAGAGTATCCGG-3¢ 289
5¢-1821CAAGGGTAAATTCATTGGGCTTGG-3¢
b-actin 5¢-2350 ACAGACTACCTCATGAAGAT-3¢ 665
5¢-3222 AGCCATGCCAAATGTCTCAT-3¢
Fig. 1. Dose-related eect of bFGF on FSH-stimulated estradiol syn-
thesis in immature cultured rat Sertoli cells. Sertoli cells were incubated
for 24 h with testosterone substrat e (200 ngáml
)1
) in the presence or

not (c ontrol) of FSH (100 ngáml
)1
) and of bFGF increasing conc en-
trations. Determination of estradiol production was pe rformed by
radioimmunoassay. Values are expressed in p g per 10
6
cells and are
representative o f three experiments (mean  SEM). **, Signi®cantly
dierent at P< 0.01; ***, s igni®cantly dierent at P <0.001from
FSH values. NS, not signi®cant.
504 S. Brucato et al. (Eur. J. Biochem. 269) Ó FEBS 2002
decrease of FSH-stimulated estradiol s ynthesis by stimulat-
ing cAMP-speci®c phosphodiesterase activity.
bFGF inhibits the FSH-stimulated cytochrome P450
aromatase mRNA expression
The relative expression of cytochrome P450 aromatase
mRNA was evaluated using semi-quantitative RT-PCR. In
the presence of 100 ngámL
)1
FSH, cytochrome P450
aromatase mRNA expression was highly increased as
described previously [30]. Sertoli cells from 20-day-old-rats
were then incubated for 24 h w ith 100 ngámL
)1
FSH and
5ngámL
)1
bFGF. In these conditions, cytochrome P450
aromatase mRNA e xpression was inhibited ()41%) by
comparison to FSH taken as control (Fig. 3). A similar

bFGF inhibitory effect on P450 aromatase mRNA e xpres-
sion was observed in the presence of dbcAMP ()41%) or
cholera toxin ()45%) (data not shown).
bFGF effect on FSH stimulated steroidogenesis
requires the presence of HSPG
We examined in what extend sodium chlorate treatment
could m odify inhibitory effect of exogenous bFGF on FSH-
stimulated estradiol synthesis. Indeed, Sertoli cells are
bFGF pr oducing c ells [2,33] and are the t arget of this
growth factor as bFGF inhibits FSH-induced estradiol
synthesis ([5], and our results). HSPG and especially
glypicans and synde cans have been described as coreceptors
for this growth factor via a highly sulfated sequence of their
heparan sulfate chains [34].
Sodium chlorate is an inhibitor of ATP sulfurylase and
hence of the production of phosphoadenosine phospho-
sulfate (PAPS), the active s ulfate donor for s ulfotrans-
ferases. It has been shown to abolish sulfation on proteins
and carbohydrate residues in intact cells without inhibiting
cell growth or protein synthesis [35±37], and proteoglycan
sulfation in cultured Sertoli cells from 20-day-old rats [38].
When Sertoli cells were incubated with 1 0 m
M
sodium
chlorate for 24 h, an increase of FSH-stimulated e stradiol
production (+42.5%) w as observed a s d escribed previously
[38]. A ddition of 10 m
M
NaCl, u sed as negative control,
did not induce any modi®cation of FSH-stimulated estradiol

synthesis (data not shown). H owever, c oncomitant treat-
ment of bFGF with 10 m
M
sodium chlorate totally abolished
the in hibitory b FGF e ffect previously observed on Sertoli cell
estradiol synthesis (Fig. 4). Similar results were obtained
when Sertoli cells were incubated with 1 m
M
dbcAMP
instead of F SH (data not shown). Thus, abolition b y sodium
chlorate of bFGF effe ct on FSH-stimulated steroidogenesis
could implicate HSPG in t he bFGF signaling.
bFGF effect on cell attachment
Addition of 5 ngámL
)1
bFGF did not promote any
signi®cant difference in cell attachment to substratum (data
not shown) as the DNA content o f the cell layer at the end
of the 24 h incubation period was i dentical in untreated
and bFGF-treated Sertoli cell cultures (2525  257 and
2538  246 ng per well in three different cell cultures).
Cell surface HSPG are bFGF partners during Sertoli cell
postnatal development
Developing Sertoli cells undergo structural, b iochemical and
functional modi®cations as previously mentioned. Thus, the
relationship between bFGF and H SPG was evaluated in
Sertoli cells from 10 to 30-days-old-rats.
bFGF effect on Sertoli cell estradiol synthesis
When Sertoli cells were incubated for 24 h with
100 ng ámL

)1
FSH, estradiol production decreased with
developing Sertoli cells (Table 3).
When Sertoli cells from 10-day-old-rats were incubated
for 24 h with 100 ng ámL
)1
FSH and 5 ngámL
)1
bFGF,
FSH-stimulated estradiol production decreased ()40%).
This inhibition was less important than in Sert oli cells from
20- and 30-day-old rats ()49 and )53%, respectively)
(Table 3). Thus, bFGF inhibitory e ffect was m ore pro-
nounced on FSH-stimulated estradiol production with
Table 2. bFGF eect on cAMP-stimulated estradiol synthesis in
immature cultured ra t Sertoli cells. Sertoli cells were incubat ed for 24 h
with testosterone substrate (200 ngámL
)1
) (control) in the presence of
1m
M
dbcAMP or 10 lgámL
)1
cholera toxin and/or 5 ngámL
)1
of
bFGF. Estradiol production was determined by radioimmunoassay.
Values are e xpressed in pg per 1 0
6
cells an d a re rep r esentative of three

experiments (mean  SEM).
Estradiol
(pg per 10
6
cells)
Control 53  4
dbcAMP (1 m
M
) 480  2
dbcAMP (1 m
M
) + bFGF (5 ng mL
)1
) 244  16
Cholera toxin (10 lgá mL
)1
) 379  29
Cholera toxin (10 lgá mL
)1
) + bFGF
(5 ngámL
)1
)
227  18
Fig. 2. Inhibitory eect of bFGF o n FSH-stimulated estradiol synthesis
by Sertoli cells cultured in the presence of Ro-20-1724, a cAMP phos-
phodiesterase inhibitor. Sertoli cells were incubated for 24 h with
testosterone substrate (200 ngáml
)1
), FSH ( 100 ngáml

)1
)and/orRo-20-
1724 (250 lM) and /or bFGF (5 ngáml
)1
). Estradiol production was
determined by radioimmunoassay. Valu es are expressed in pg p er 10
6
cells and are representative of three e xperiments (mean  SEM).
*, signi®cantly diere nt at p < 0 .05 from FSH or F SH + Ro-20- 1724
values.
Ó FEBS 2002 HSPG, target and partners of bFGF (Eur. J. Biochem. 269) 505
Sertoli cells aging. Moreover, cytochrome P450 aromatase
mRNA expression was regulated similarly by bFGF upon
development (data not shown). We suggest that bFGF
could participate in the Sertoli cell steroidogenesis decrease
by inhibiting cytochrome P450 aromatase mRNA expres-
sion and FSH-stimulated estradiol production.
bFGF effect on steroidogenesis in the absence
of cell surface HSPG
When Sertoli cells from 10- to 30-day-old-rats were
incubated with 100 ngámL
)1
FSH, 1 0 m
M
sodium chlorate
and 5 ngámL
)1
bFGF for 24 h , bFGF inhibitory e ffect on
Fig. 3. Dose-related eect o f bFGF on F SH-stimulated P450 aromatase mRNA in immature cultured rat S ertoli cells. Sertoli cells were incubated for
24 h in the absence of 100 ngáml

)1
FSH ( lane 1) or in the presence o f 100 ngáml
)1
FSH and increasing concentrations of bFGF (lanes 2 to 6). Total
RNA was extracted as described in Materials a nd m ethods. Then, 500 ng RNA was reverse-transcribed and am pli®ed by relative quantitative
RT-PCR as previously described. (A) Agarose gel of one representative experiment. (B1) The densitometry data are representative of three
experiments (mean  SEM). Aromatase mRNA level under treatment is expressed versus control which is arbitrarily set t o 100%. (B2) The
densitometry data are representative of three experiments (mean  SEM ). Aromatase mRNA level under treatment is expressed versus FSH which
is arbitrarily set to 100%.
506 S. Brucato et al. (Eur. J. Biochem. 269) Ó FEBS 2002
FSH-stimulated estradiol synthesis was not observed
(Table 3). In conclusion, bFGF requires c ell surface HSPG
for t he inhibition of steroidogenesis in developing Sertoli
cells.
Cell surface HSPG are bFGF targets during Sertoli cell
postnatal development
As previously mentioned, glypican-1, syndecan-1 and
syndecan-4 are coreceptors for bFGF in some cellular
models. As bFGF requires cell surface HSPG to regulate
FSH-stimulated estradiol p roduction, we evaluated if bFGF
itself could in¯uence s yndecan-1, syndecan-4 and glypican-1
mRNAs expression.
bFGF effect on glypican-1, syndecan-1 and syndecan-4
mRNAs expression
TherelativemRNAexpressionoftheseHSPGwas
evaluated using semi-quantitative RT-PCR a s d escribed
previously [22]. Figure 5 indicated that, when Sertoli cells
from 10-, 20- and 30-day-old-rats were incubated for 24 h
without any treatment, glypican-1 mRNA expression was
signi®cantly increased between 10- and 20-days old, t hen

unchanged between 20- and 30-days old. Syndecan-1 mRNA
expression was the same whatever rat age. Syndecan-4
mRNA expression increased highly between 10- a nd 20-days
old, then decreased between 20- and 30-days-old but
however, w as higher than at 10 days old ( Fig. 5).
When Sertoli cells from 10-day-old-rats were incubated for
24 h in the presence of 10 ngámL
)1
bFGF, b FGF inhibited
glypican-1 mRNA expression ()35%) but had no e ffect on
syndecan-1 mRNA expression. On the other hand, bFGF
stimulated syndecan-4 mRNA expression (+ 41%) (Fig. 6).
When Sertoli cells from 20-day-old rats were incubated
for 24 h in the presence of 10 ngámL
)1
bFGF, glypican-1
mRNA expression was inhibited ()37%) similarly to
10-day-old cells whereas syndecan-1 and syndecan-4
mRNAs e xpression was not modi®ed c ompared to t he
control (Fig. 6).
In Sertoli cells from 30-day-old rats, bFGF had no
inhibitory effect on glypican-1 mRNA expression, but
increased syndecan-1 and syndecan-4 mRNA expression
(+36% and +42%, respectively) (Fig. 6).
DISCUSSION
This report shows for the ®rst time in developing rat Sertoli
cells that HSPGs are partners for b FGF signal t ransduction
as coreceptors, a nd that HSPGs are a target of this growth
factor as their mRNA expression is modulated by bFGF
itself.

Under our cell culture conditions, 20-day-old-rat Sertoli
cells did not proliferate. Thus, bFGF effect was evaluated
by estradiol synthesis instead of cell proliferation test. We
demonstrated that bFGF regulates steroidogenesis upon
cell development. Thus, bFGF d ecreased c ytochrome P 450
aromatase mRNA e xpression but also inhibited the FSH-
stimulated estradiol synthesis. Therefore, the mechanism by
which bFGF inhibits Sertoli cell steroidogenesis is still
unknown in developing rat Sertoli cells. However, it seems
that bFGF mainly regulates steroidogenesis at the tran-
scriptional l evel and could also, in part, stimulate phospho-
diesterase activity as TGF-b does [ 39].
Thus, bFGF, among other testicular agents, could
participate in the decrease of this Sertoli cell FSH-stimulated
estradiol production during t estis development [40,41]. In
this way, the i nhibitory effect of bFGF on estradiol
production may represent signals destined to shut down
aromatase activity because recent observations made by
Sharpe et al. [42] have led to the conclusion that prolonged
exposure o f Sertoli cells to estrogens impairs or delays their
functional maturation.
Fig. 4. Eect of bFGF in the presence of sodium chlorate on FSH-
stimulated estradiol synthesis in immature rat Sertoli cells. Sertoli cells
were incubate d with t estoste rone su bstrate ( 200 ngáml
)1
)(control)in
the p resence of FSH (100 ngámL
)1
)and/orbFGF(5ngámL
)1

)and/or
sodium chlorate (10 m
M
) during 2 4 h . Estradiol pro duction was
determined by radioimmunoassay. Valu es are expressed in pg per 10
6
cells and are represen tative of t hree experiments (mean  SEM).
**, Signi®cantly di erent at P < 0.01 f ro m FSH values. NS, n ot
signi®cant.
Table 3. bFGF eect on FSH-stimulated estradiol synthesis in the presence or not of c ell surface HSPG during S ertoli cell development. Sertoli c ells
were incubated for 24 h with testosterone substrate (200 ngámL
)1
) in the p resence or not (control) of F SH (100 ngámL
)1
)and/orbFGF(5ngámL
)1
)
and/or sodium chlorate ( 10 m
M
). Estradiol production was determined by radioimmunoassay. Values are expressed i n pg per 10
6
cells and are
representative of three experiments (mean  S EM) for each studied age.
Sertoli cells
(days old)
Estradiol synthesis (pg per 10
6
cells)
Control FSH FSH + bFGF FSH + chlorate FSH + chlorate + bFGF
10 760  64 1220  27 830  50 1490  70 1380  100

20 53  4 458  28 234  3 653  12 578  69
30 25  0.6 111  9 52  5 170  22 147  15
Ó FEBS 2002 HSPG, target and partners of bFGF (Eur. J. Biochem. 269) 507
The bFGF binding requires speci®c motifs on highly
sulfated HS chains. Sodium chlorate, in inducing structural
alteration of HS chains, prevents the bFGF binding. Our
results indicate that the presence of sodium chlorate
abolishes the bFGF biological effects in d eveloping Sertoli
cells. Thus, inhibitory effect of bFGF on FSH-stimulated
estradiol production requires t he presence of cell surface
HSPG with correctly sulfated HS chains in addition to
bFGF receptor as shown in previous studies [4,11±15].
Among these HSPG, glypican-1, syndecan-1 and
syndecan-4 are potential coreceptors o f bFGF. We indicated
that the HSPG mRNA pattern is not similar in Sertoli cells
from 10, 20 a nd 30 days old. In addition, glypican-1 mRNA
expression was inhib ited ()35% and )37%, respectively) in
Sertoli cells f rom 10- and 20- day-old rats as d escribed in
oligodendrocytes [43] or in lung ®broblasts [44]. I n contrast,
this inhibition was abolished in 30-day-old rat Sertoli cells.
Syndecan-1 mRNA expression was not modi®ed by bFGF
in Sertoli cells from 10 and 20-day-old rats whereas it was
stimulated in Sertoli cells from 30-day-old rats. If bFGF
stimulated syndecan-4 mRNA expression in Sertoli cells
from 10 and 30-day-old rats, no effect was observed in
Sertoli cells from 20-day-o ld rats. Thus, bFGF effect seems
to be HSPG-, developmental stage- and cell type-speci®c.
Indeed, bFGF does not regulate syndecan-1 mRNA
expression in MCA3D keratinocytes [45,46] or in endothelial
cells [47]. However, this growth f actor increases syndecan-1

and syndecan-4 m RNAs expression in ®broblasts [48] and in
vascular smooth muscle cells [49], respectively.
At the transcriptional level, t he 5 ¢ ¯anking r egion o f
syndecan-1 contains an FGF-inducible response element
(FiRE) [50]. In this study, bFGF increases syndecan-1
mRNA expression in Sertoli cells from 30-day-old r ats. This
observation suggests that these cells might express all
transcription facto rs comp onents o f F iRE, namely USF,
the uncharacterized p46 nuclear proteins, A P-1 (Jun/Fos)
complexes and a putatively novel FGF-inducible nuclear
protein-1 (FIN-1) [50]. T hese transcriptional elements are
differentially regulated d epending on cell type and activating
growth factor [50]. In NIH 3T3 ®broblasts, FiRE w as
shown to be selectively induced by bFGF whereas in
keratinocytes, FiRE was not induced by this growth factor.
Whether FiRE is really expresse d in Sertoli cells or not,
some post-translational modi®cations, phosphorylation or
dephosphorylation of FiRE components could contribute
to the speci®city [51±54]. Inhibitory transcription factors
that bind to AP-1 or FIN-1 o r that inactive b inding or
transactivation capacity of, for example, FIN-1 and USF-1,
could explain activation or inhibition of FiRE. From our
results, it seems that bFGF r egulates differently HSPG
expression upon Sertoli cell maturation suggesting a func-
tional selectivity.
Assuming that levels of glypican-1 and syndecan protein
synthesis c orrelate well with mRNA s l evels, it is likely that,
in the p resence of b FGF, the plasma membrane will be
enriched with syndecans and a decrease of glypican-1.
Further experiments will be needed to understand the

biological signi®cance of t he different regulation o f their
expression. As a ®rst element step t owards understanding, a
recent demonstration indicated that the synthesis of cell
surface HSPG and FSH-stimulated estradiol synthesis are
inversely correlated. This suggests a potential role for these
Fig. 5. Evolution o f glypican-1, syndecan-1 and
syndecan-4 mRNAs expression during Ser toli
cells development. Sertoli cells from 10-, 20-
and 30-day-old rats w ere incubated for 24 h
without treatment. T otal RNA was extracted
as described in M aterials and methods. Then
RNA (500 ng) was reverse transcribed a nd
ampli®ed by relative q uantitative RT-PCR as
described previously [ 22]. Glyp-1, glypican-1;
Synd-1, syndecan-1; Synd-4, syndecan-4. (A)
Agarose gels of one representative experimen t.
(B) Densitometry data are representative of
®ve experiments (mean  SE) for each age.
508 S. Brucato et al. (Eur. J. Biochem. 269) Ó FEBS 2002
HSPG in the decrease of estradiol production [55]. The
aromatase activity d ecreases by modulation o f cytoskeleton
occuring during Sertoli cell development [56]. In this way,
syndecan-1 and syndecan-4 could participate to this event in
reorganizing actin ®laments via their cytoplasmic domain
[57] but also in the presentation and delivery of bFGF to its
receptors. Moreover, during Sertoli cell development,
phosphodiesterase activity increases [58]. Phamanthu et al .
[38] suggest a possible involvement of cell HSPG in the age-
related i ncrease in Sertoli cell phosphodiesterase activity and
in the concomitent loss of s teroidogenic response to FSH.

These data and our results suggest that bFGF could
modulate, in part, the decrease of FSH-stimulated estradiol
synthesis via HSPG.
REFERENCES
1. Nishimura, T., Nakatake, Y., Konishi, M. & Itoh, N. (2000)
Identi®cation of a novel FGF, FGF-21, preferentially e xpressed in
the liver. Bio chim. Biophys. Acta 1492, 203±206.
2. Han, I.S., Sylvester, S.R., Kim, K.H., Schelling, M.E.,
Venkateswaran, S., Blanckaert, V.D., McGu inness, M.P. &
Griswold, M .D. (1993) Basic ®broblast growth factor is a testi-
cular germ cell product which may regulate Sertoli cell function.
Mol. Endocrinol. 7, 889±897.
3. Bikfalvi, A., Klein, S., P intucci, G. & Rifkin, D.B. (1997)
Biological roles of ®broblast growth factor-2. Endrocrin. Rev. 1,
26±45.
4. Laslett, A.L., McFarlane, J .R., Hearn, M.T.W. & Risbridger,
G.P. (1995) Requirement for heparan s ulphate prote oglycans to
mediate b asic ®broblast growth factor (FGF-2) -induced stimu-
lation of Leydig cell steroidogenesis. J. Steroid. Biochem. Molec.
Biol. 54, 245±250.
5. Schteingart, H.F., Meroni, S.B., Capena, D.F., Pellizzari, E.H. &
Cigorraga, S.B. (1999) Eects of basic ®broblast growth factor
and nerve growth factor on lactate production, c-glut amyl trans-
peptidase and ar omatase activities in cultured S ertoli cells. Eur. J.
Endocrinol. 14 1, 539±545.
6. Boockfor, F.R. & Schwarz, L.K. (1990) Fibroblast growth factor
modulates the release of transferrin from cultured Sertoli cells.
Mol. Cell. Endocrinol. 73, 187 ±194.
Fig. 6. Action of bFGF on Glypican-1, syndecan-1 a nd syndecan-4 mRNAs expression during Sertoli cells development. Sertoli cells from 10-, 20- and
30-day-old rats were incubated for 24 h in the p resence (+) or in the absence (±) of 10 n gáml

)1
bFGF. Total RNA was extracted as described in
Materials and methods. Then, RNA (500 n g) was reverse transcribed and ampli®ed by re lative quantitative RT-PCR as described previously [22].
Glyp-1, glypican-1; Synd-1, syndecan-1; Synd-4, syndecan-4. (A) Agarose gel of one representative experiment for e ach studied age. (B) Densi-
tometry data a re representative of t hree, ®ve and three dierent experiments (mean  SE) for 10-, 20- and 3 0-d ay-old-rat Sertoli cells, respectively .
Ó FEBS 2002 HSPG, target and partners of bFGF (Eur. J. Biochem. 269) 509
7. Jaillard, C., Chatelain, P.G. & Saez, J.M. (1987) In vitro r egula-
tion of pig Sertoli cell growth and function: eects of ®broblast
growth factor and somatomedin-C. Biol. Reprod. 37, 665±674.
8. Smith, E.P., Hall, S.H., Monaco, L., French, S.H., Wilson, E.M.
& Conti, M. (1989) A rat Sertoli cell f actor similar to basic ®-
broblast growth fact or increases c-fos messenger ribon ucleic acid
in cultured Sertoli cells. Mol. Endocrinol. 3, 954±961.
9. Le Magueresse-Battistoni, B., Wol, J., Morera, A.M. &
Benahmed, M. (1994) Fibroblast growth factor receptor type 1
expression during rat testicular development and its regulation in
cultured Sertoli cells. Endocrinology 135, 2404±2411.
10. Nugent, M.A. & Iozzo, R.V. (2000) Fibroblast growth factor-2.
Int. J. Biochem. Cell Biol. 32 , 115±120.
11. Yayon, A., Klagsbrun, M., Esko, J.D., Leder, P. & Ornitz, D.M.
(1991) Cell surface, heparin-like molecules are required for binding
of basic ®broblast growth factor to its high anity receptor. Cell
64, 841±848.
12. Rapraeger, A.C., Krufka, A. & Olwin, B.B. (1991) Requirement of
heparan sulfate for bFGF-mediated ®b roblast growth and myo-
blast dierenciation . Science 252, 1705±1708.
13. Heath, W.F., Cantrell, A.S., M ayne, H.G. & R ichard, J.S. (1991)
Mutations in the heparin binding domains of h uman basic ®bro-
blast growth factor alter its biological activity. Biochemistry 30,
5608±5615.

14. Savona, C., Chambaz, E.M. & Feige, J.J. (1991) Proteoheparan
sulfate contribute to the binding of basic ®broblast growth factor
to its high anity receptors on bovine adrenocortical cells. Growth
Factor 5, 273±282.
15. Ornitz, D.M., Yayon, A., Flanagan, J.G., Svahn, C.M., Levi, E. &
Leder, P. (1992) He parin is required for cell-free binding of basic
®broblast growth factor to a soluble receptor and for mitogenesis
in whole ce ll s. MolCellBiol.12, 240±247.
16. Dorrington, J.H. & Fritz, I.B. (1975) Androgen synthesis and
metabolism by preparations from the sem iniferous tubule of the
rat testis. In Hormonal Regulation of Spermatogenesis (French,
F.S., Hansson, V., Ritzen, E .M. & Nayfeh, S.N., eds.), pp. 37± 52.
Plenum Press, New York.
17. Welsh, M.J. & W iebe, J.P. (1976) Sertoli ce lls from immature rats:
in v i tr o stimulation of s teroid met abolism. Biochem. Biophys. Res.
Commun. 69, 936±941.
18. Suarez-Quian, C.A., Dym, M., Makris, A., Brumbaugh, J., Ryan,
K.J. & Canick, J.A. (1983) Estrogen synthesis by immature r at
Sertoli cells in vitro. J. Androl. 4, 203±209.
19. Papadopoulos, V., Carreau, S ., Szerman, J.E., Drosdowsky,
M.A., Dehennin, L. & Scholler, R. (1986) R at testis 17b-estradiol:
identi®cation b y gas chromatography-mass spectrometry and
age related cellular distribution. J. Steroid. Biochem. 24, 1211±
1216.
20. Mounis,A.,Barbey,P.,Langris,M.&Bocquet,J.(1991)Deter-
gent-solubilized proteoglycans in rat testicular Sertoli cells. Bio-
chim. Biophys. Acta. 1074, 424±432.
21. Brucato,S.,Fagnen,G.,Villers,C.,Bonnamy,P.J.,Langris,M.&
Bocquet, J. (2001) Biochemical characterization of integral mem-
brane heparan sulfate proteoglycans in Sertoli cells from immature

rat testis. Biochim. Biophys. A cta. 1510, 474 ±487.
22. Brucato, S., Harduin-Lepers, A., Godard, F., Bocquet, J. &
Villers, C. (2000) Expression of glypican-1, syndecan-1 a nd
syndecan-4 mRNAs protein kinase C-regulated in rat immature
Sertoli cells b y semi-quantitative R T-PCR analysis. Biochim.
Biophys. Acta. 1474, 31±40.
23. Filla, M.S., D am, P. & Jalkanen, M. (1998) The cell surface
proteoglycan syndecan-1 med iates ®b roblast growth fac tor-2
binding and activity. J. Cell Physiol. 147, 310±321.
24. Steinfeld, R., Van den Berghe, H. & D avid, G. (1996) Stimulation
of ®broblast g rowth f actor receptor-1 o ccupancy and signaling by
cell surface-associated syndecans a nd glypican. J. Cell. Biol. 133 ,
405±416.
25. Bonneh-Barkay, D ., Shlissel, M., Berman, B ., Shaoul, E., Admon,
A.,Vlodavsky,I.,Carey,D.J.,Asundi,V.K.,ReichSlotky,R.&
Ron, D. (1997) Identi®cation of glypican as a modulator of the
activity of ®broblast growth factors. J. Biol. Chem. 27 2 , 12415±
12421.
26. Carey, D.J. (1997) Syndecans: multifonctional cell-surface
co-receptors. Biochem. J. 327, 1±16.
27. Tung, P.S., Skinner, M.K. & F ritz, I.B. (1984) Fibronectin syn-
thesis is a marker for peritubular contaminants in Sertoli cell-
enriched cultures. Biol. Reprod. 30, 199±211.
28. Galdieri, M., Ziparo, E., Palombi, F., Russo, M.A. & Stefanini,
M. (1981) Pure Sertoli cell cultures: a new model for the study of
somatic±germ cell interactions. J. Androl. 2, 249±254.
29. Chomczynski, P. & Sacchi, N. (1987) Single -step method of RNA
isolation by a cid g u anidinium thiocyanate-phenol-chlorofo rm
extraction. Anal. Biochem. 162, 156±159.
30. Carreau, S . & Levallet, J. (1997) Cytochrome P 450 aromatase in

male germ cells. Folia Histochem. Cytobiol. 35 , 195±202.
31. West, D.C., Sattar, A. & Kumar, S. (1985) A simpli®ed in situ
solubilization procedure for the determination of DNA and cell
numberintissueculturedmammaliancells.Anal. B iochem. 14 7 ,
289±295.
32. Conti, M., Nemoz, G., Sette, C. & Vicini, E. (1995) Recent pro-
gress in u nderstandin g the h ormonal regulation of p hosphod ies-
terases. Endocrine Rev. 16 , 370±389.
33. Mullaney, B.P. & Skinner, M.K. (1992) Basic ®broblast growth
factor (bFGF) gene expression and protein production during
pubertal dev elopment of the seminiferous t ubule: follicle-stimu-
lating hormone-induced Sertoli cell bFGF expression. Endocri-
nology 131, 2928±2934.
34.Aviezer,D.,Levy,E.,Safran,M.,Svahn,C.,Buddecke,E.,
Schmidt, A., David, G., Vlodavsky, I. & Yayon, A. (1994) Dif-
ferential structural requirements of heparin and heparan sulfate
proteoglycans that promote binding of basic ®broblast growth
factor to its receptor. J. Biol. Chem. 269, 114±121.
35. Brauer, P.R., Keller, K.M. & Keller, J.M. (1990 ) Concurrent
reduction in the sulfation of heparan sulfate and basement
membrane assemb ly in a c ell m odel s ystem. Development 110,805±
813.
36. Greeve, H ., Cully, Z., Blumberg, P . & Kresse, H . ( 1988) In¯ue nce
of chlorate on proteoglycan biosynthesis by cultured human
®broblasts. J. Biol. Chem. 263 , 12888±12891.
37. Humphries, D.E. & Silbert, J.E. (1988) Chlor ate: a reversible
inhibitor of proteoglycan sulfation. Biochem. Biophys. Res. Comm.
145, 365±371.
38. Phamantu, N.T., Fagnen, G., Godard, F., Bocquet, J. &
Bonnamy, P.J. (1999) Sodium chlorate induces undersulfation

of cellular proteoglycans and increases in FSH -stimulated
estradiol production in immature rat S ertoli cells. J. Androl. 20,
241±250.
39. Morera, A.M., Esposito, G., Ghiglieri, C., Chauvin, M.A.,
Hartmann, D.J. & Benhamed, M. (1992) Transforming growth
factor b1 inhibits gonadotropin action i n cultured porcine Sertoli
cells. Endocrinol. 130, 831±836.
40. Ackland, J.F., Schwartz, N.B., Mayo, K.E. & D odson, R.E.
(1992) Nonsteroidal signals originating in the gonads. Physiol.
Rev. 72, 731±765.
41. Rosselli, M . & Skinner, M.K. (1992) Developmental regulation of
Sertoli c ell aromatase ac tivity and plasminogen activator pro-
duction by hormones, retinoids and the testicular paracrine f actor,
PModS. Biol. Reprod. 46, 586±594.
42. Sharpe, R.M., Atanassova, N., McKinnell, C., Parte, P., Tu rner,
K.J., Fisher, J.S., Kerr, J .B., Groome, N.P., Macpherson, S.,
Millar, M .R. & Saunders, P.T. (1998) Abnormalities i n functional
development of t he Sertoli cells in rat treated ne onatally with
diethylstillbestrol: a poss ible role for estrogens in Sertoli cells
development. Biol. Repr od. 59, 1084±1094.
510 S. Brucato et al. (Eur. J. Biochem. 269) Ó FEBS 2002
43. Bansal, R., Kumar, M., Murray, K. & Pfeier, S.E. (1996) De-
velopmental and FGF-2-mediated regulation of syndecans (1±4)
andglypicaninoligodendrocytes.Mol. Cell Neurosc. 7, 276±288.
44. Romaris, M., Bassols, A. & David, G. (1995) Eect of trans-
forming growth factor-b1 and basic ®broblast growth factor on
the expression of c ell surface proteoglycans in human lung ®bro-
blasts. B i och em. J. 310, 73±81.
45. Tsuboi, T., Sato, C., Kurita, Y., Ron , D., Rubin, J.S. & Ogawa,
H. (1993) Keratinocyte growth factor (FGF-7) stimulates migra-

tion and plasminogen activator activity of normal human kerati-
nocytes. J. Invest. Dermatol. 101 , 49±53.
46. Jaakkola, P., Maatta, A. & Jalkanen, M. (1998) The activation
and composition of FiRE (an F GF-inducible response element)
dier in a cell type- and growth factor-speci®c manner. Onc ogene
17, 1279±1286.
47.Kainulainen,V.,Nelimarkka,L.,Jarvelainen,H.,Laato,M.,
Jalkanen, M. & Elenius, K. (1996) Suppr ession of syndecan-1
expression in endothelial c ells by tumor necrosis factor-alpha.
J. Biol. Chem. 271, 18759±18766.
48. Jaakkola, P., Vihinen, T ., Maatta, A . & Jalkanen, M. ( 1997)
Activation of an enhancer on the syndecan-1 gene is restricted to
®broblast growth factor fam ily members in mesenchy mal cells.
Mol. Cell. Biol. 17, 3210±3219.
49. Cizmeci-Smith, G., Langan, E., Youkey, J., Showalter, L.T. &
Carey, D.J. (1997) Syndecan-4 is a primary-response gene induced
by basic ® broblast growth factor an d arterial injury in vascular
smooth muscle cells. Artrioscler. Thromb. Vasc. Biol. 17 , 172±180.
50. Jaakkola, P. & Jalkanen, M. (2000) Transcritptional regulation of
syndecan-1 expression by growth factors. Prog. Nucle ic Acid R es.
Mol Biol. 63 , 109±138.
51. Smeal, T., Binetruy, B., M ercola, D.A., Birrer, M. & Karin, M .
(1991) Oncogenic and transcriptional cooperation with Ha-Ras
requires phosphorylation of c-J un on serines 63 and 73. Nature
354, 494±496.
52. Baker, S.J., Kerppola, T.K., Luk, D., Vandenberg, M.T.,
Marshak, D.R., Curran, T. & Abate, C. ( 1992) Jun is phospho-
rylated by several protein kinases at the same sites that are
modi®ed in serum-stimulated ®broblasts. MolCellBiol.12,
4694±4705.

53. Franklin, C .C., Sanchez, V., W agner, F., Woodgett, J.R. & Kraft,
A.S. (1992) Ph orbol ester-induced amino-terminal phosphoryl-
ation of human JUN but not J UNB regulates transcriptional
activation. Proc. Natl Acad. Sci. USA 89, 7247±7251.
54. Hill, C.S. & T reisman, R. (1995) Transc riptional regulation
by extracellular signals: mechanisms and speci®city. Cell. 80,
199±211.
55. Phamantu, N.T., Bonnamy, P.J., Bouakka, M. & Bocquet, J.
(1995) Inhibition of proteoglycan synthesis induces an increase in
follicle stimulating hormone (FSH) -stimulated es tradiol produc-
tion by immature rat sertoli cells. Mol. Cell Endo. 109, 37±45.
56. Meroni, S.B., Steingart, H.F., Pellizzari, E.H. & Cigorraga, S.B.
(1995) Possible involvement of micro®laments in the regulation
of Sertoli cell aromatase activity. Mol. Cell. Endocrinol. 112,
69±75.
57. Carey, D.J., Bendt, K.M. & Stahl, R.C. (1996) The cytoplasmic
domain of syndecan-1 i s required for cytoskeleton asso ciation but
not detergent insolubility. J. Biol. Chem. 271, 15253±15260.
58. Griswold, M.D. (1993) Postnatal S ertoli cell deve lopment. In The
Sertoli Cell (par Russel, L.D. & Griswold, M.D., eds), pp. 493±
508. Cache River Press, USA.
Ó FEBS 2002 HSPG, target and partners of bFGF (Eur. J. Biochem. 269) 511