Fibrolast Growth Factor 2 Uses Distinct Signaling Pathways
for Cell Proliferation and Cell Shape Changes in Corneal
Endothelial Cells
Xin Gu,* Gongje Seong,^ Young Ghee Lee,\ and EunDuck P. Kay*-\%
Purpose. Fibroblast growth factor 2 (FGF-2) is not only a potent mitogen, it is a modulator
for corneal endothelial cells. To define how the modulation activities of FGF-2 are mediated,
we used pharmacologic inhibitors to examine the association of phospholipase C-yl (PLCy) with FGF receptor or with cytoskeleton.
Methods. Cell proliferation was determined either by the incorporation of 3H-thymidine into
DNA or by counting cell numbers in the absence or presence of the inhibitors. The protein
expression was analyzed by immunoprecipitation and immunoblot analysis. Cell shape change
was determined by phase-contrast microscopy.
Results. FGF-2 stimulated DNA synthesis, whereas genistein inhibited the FGF-2-mediated cell
proliferation in a dose-dependent manner, regardless of the concentration of FGF-2. The
PLC-yl specific antisense oligonucleotide primer was able to inhibit cell proliferation by 25%
in the absence of FGF-2; however, the antisense primer was not able to override the action
of FGF-2. Fibroblast growth factor receptor was phosphorylated on treatment of the cells with
FGF-2; however, 24-hour treatment with the growth factor significandy reduced phosphorylation of the receptor. Phospholipase Cyl appears to be abundant in cytoplasm in the absence
and presence of FGF-2, and a minor portion of the molecule is translocated to membrane
after treatment with FGF-2; genistein inhibited the translocation. When the cytoskeleton
fraction of the normal and the modulated corneal endothelial cells was immunoprecipitated
with PLC-7I antibodies, PLC-7I, actin, and vinculin were coprecipitated in both cell cultures.
Phospholipase C7I associated with cytoskeleton was phosphorylated on treatment of the cells
with FGF-2. In the presence of FGF-2 of the modulated cells, cytochalasin B, which did not
revert the modulated cell morphology, abolished the association of PLC-yl with actin and
vinculin; colchicine, which did revert the modulated cell shape to the polygonal shape, did
not block the association of these three molecules. Interestingly, colchicine slighdy enhanced
the stimulatory effect of FGF-2 on corneal endothelial proliferation in contrast to die effect
of cytochalasin B, which slighdy decreased die FGF-2 action on cell proliferation.
Conclusions. The association of PLC-yl widi cytoskeleton plays a role in cell proliferation,
whereas the association of PLC-yl with actin and vinculin has no effect on cell shape changes.
These findings indicate that FGF-2 appears to use distinct signaling pathways for cell proliferation and cell shape changes in corneal endothelial cells. Invest Ophthalmol Vis Sci.
1996;37:2326-2334.
X he corneal endothelium is essential for maintaining
corneal transparency, but the regeneration capacity
of corneal endothelium after in vivo injury is severely
limited in humans, cats, and primates.1'2 Under the
From the *Dolieny Eye Institute and the %Depattment of Ophthalmology, University
of Southern California School of Medicine, Los Angeles, California, and the
tDepartment of Ophthalmology, Yonsei University College of Medicine, Seoul, Korea.
Supported by National Institutes of Health grants EY06431 and EY03040, and by
an unrestricted grant from Research to Prevent Blindness, Inc., Neiu York, New
York.
Submitted for publication January 5, 1996; revised May 28, 1996; accepted June
24, 1996.
Proprietary interest category: N.
Reprint requests: EunDuck P. Kay, Doheny Eye Institute, 1450 San Pablo Street,
Us Angeles, CA 90033.
2326
conditions that accompany inflammation and atypical
wound repair, corneal endothelium in vivo responds
by converting to fibroblast-like cells.3"5 One such clinical example is the development of retrocorneal fibrous membrane, in which corneal endothelial cells
are modulated to fibroblast-like cells and ultimately
lose the characteristics of endothelial cells. Cell proliferation then resumes, resulting in layers of elongated
cells. These cells produce fibrillar collagen, which
leads to the formation of a fibrillar extracellular matrix. Development of retrocorneal fibrous membrane
can be induced in vitro by either of two distinct pro-
Investigative Ophthalmology & Visual Science, October 1996, Vol. 37, No. 11
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Signaling Pathways of Fibroblast Growth Factor 2 in Cornea! Endothelial Cells
tein factors, corneal endothelium modulation factor
(CEMF)"7 or basic fibroblast growth factor 2 (FGF-2),
or by a combination of the two.7'8
Fibroblast growth factor-2, a ubiquitous multifunctional growth factor, is one of nine heparin-binding
polypeptides known to be present in many tissues and
cell lines.9"12 In the eye, FGF-2 is a constituent of Descemet's membrane. There, it acts as an autocrine
growth factor to induce cell proliferation and migration of corneal endothelial cells, actions that indicate
FGF-2's involvement in wound repair.781314 The biologic actions of FGF-2 are mediated through transmembrane cell surface receptors that possess tyrosine
kinase activity.1'"17 One of the early cellular events
induced by the binding of FGF-2 to its receptor is
the stimulation of phosphatidylinositol-specific PLCy, which hydrolyzes inositol phospholipids and generates the second messengers, diacylglycerol and inositol
phosphates.18 '"' Phospholipase C-y has been shown to
be a major substrate of FGF receptor 1 (FGFR-1),20 so
activation of PLC-y stimulates the phosphatidylinositol turnover pathway, leading to increases in cytosolic
calcium and protein kinase C.21 This increase in cytosolic calcium eventually initiates cell proliferation.
Furthermore, our previous studies7'8 demonstrated that FGF-2 is the key molecule for the corneal
endothelial modulation that leads to fibrosis. To understand the molecular mechanism by which FGF-2
transduces the signals to cells for modulation, the current study investigated the early event of signal transduction pathways mediated by FGF-2 in the presence
of CEMF, which induces FGF-2 production in corneal
endothelial cells.8
METHODS
Cell Cultures
Isolation and establishment in culture of primary rabbit corneal endothelial cells were carried out as previously described.*' Cultures were maintained in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal calf serum and 50 //g/ml of
gentamicin in a humidified atmosphere of 5% CO2 in
air. Modulated endothelial cells were established as
previously described8 and were maintained in the presence of FGF-2 (10 ng/ml) and CEMF (0.25 //g/ml).
Animal experiments were performed in compliance
with the ARVO Statement for the Use of Animals in
Ophthalmic and Vision Research.
Preparation of Subcellular Fractions
To obtain subcellular fractions, the cultured cells were
washed three times with ice-cold phosphate buffered
saline (PBS), scraped off with homogenization buffer
(20 raM HEPES, pH 7.4, 1 mM EDTA, 10% glycerol,
50 mM NaCl, 20 mM /3-glycerophosphate, 2
2327
leupeptin, 1 mM sodium orthovanadate, and 1 mM
phenylmethylsulfonyl fluoride), and homogenized
with a glass homogenizer (20 strokes) on ice. The
volume of homogenate was adjusted to 1.5 ml with
homogenization buffer followed by ultracentrifugation in a Beckman TLS 55 rotor (Beckman Instruments, Fullerton, CA) at 40,000 rpm for 20 minutes
at 4°C. The supernatant was used as the cytosolic fraction; the pellet was homogenized further with 1 ml
of homogenization buffer supplemented with 2% Noctyl-/?-D-glucoside and was used as the membrane
fraction.
Cytoskeleton Isolation
Cytoskeleton was prepared by a modification of a previously described method.22 All buffers were maintained at 4°C during the cytoskeleton isolation. Cells
were washed with ice-cold microtubule stabilization
buffer containing 0.1 M Pipes, pH 6.9, 2 M glycerol,
1 mM EGTA, and 1 mM magnesium acetate. Cells
were homogenized with cold microtubule stabilization
buffer containing 0.2% Triton-X-100,10 //g/ml aprotinin, 10 fig/ml leupeptin, 200 //M sodium orthovanadate, and 1 mM phenylmethylsulfonyl fluoride. Homogenates were centrifuged, and the insoluble pellet
was dissolved further in sodium dodecyl sulfate
(SDS) —electrophoresis sample buffer followed by boiling. The protein fractions of the insoluble pellet were
analyzed and discarded from the second preparation
because the protein profiles were identical with the
soluble fractions.
DNA Synthesis
The first-passaged cells were maintained for 2 days in
DMEM containing 10% fetal calf serum and placed in
serum-free medium for 48 hours to deprive residual
growth activities of serum. The medium was then replaced with serum-free DMEM with or without FGF2, 5 //Ci of [3H]-thymidine (90 Ci/mmol; Amersham
Life Science, Buckinghamshire, UK), and with or without genistein. Labeling was terminated by washing the
cells with ice-cold PBS three times, followed by the
addition of 0.5 ml of dissolution buffer (25 mM
HEPES, pH 7.5, 0.1% Triton-X-100) and 5 fi\ of 2%
sodium deoxycholate. After cells were lysed, the lysate
was collected into a 1.5 ml microfuge tube, precipitated with trichloroacetic acid to a final concentration
of 10% and centrifuged at 14,000 rpm for 10 minutes
at 4°C. The pellet was then washed with 400 fi\ of
10% trichloroacetic acid, dissolved in 200 fi\ of 0.2 M
NaOH, and counted.
Irnmunoprecipitation
Portions of the cytoplasmic and membrane fractions
or the cytoskeleton fraction were subjected to irnmunoprecipitation: The samples were adjusted to 0.5 ml
with PBS, and 5 fi\ of undiluted primary monoclonal
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
2328
Investigative Ophthalmology 8c Visual Science, October 1996, Vol. 37, No. 11
antibodies (PLC-yl, FGF receptor, /?-actin, or phosphotyrosine) was added. To this mixture was added
50 (JA of protein G-Sepharose resin, and incubation
was carried out at 4°C for 1 hour or 17 hours with
constant shaking. After centrifugation at 10,000 rpm
for 10 minutes, the resin was washed three times with
PBS containing protease inhibitors (phenylmethylsulfonyl fluoride, aprotinin, and EDTA). The protein was
eluted from the resin by boiling in SDS-polyacrylamide sample buffer for 5 minutes. After a brief spin,
the supernatant was subjected to SDS-polyacrylamide
gel electrophoresis (PAGE). The proteins separated
by SDS-PAGE were transferred electrophoretically to
a polyvinylidene fluoride membrane. Immunoblot
analysis was performed with a commercial kit (Vectastain ABC kit; Vector Laboratories, Burlingame, GA)
All subsequent incubations were carried out at room
temperature in buffer I (0.9% NaCl, 100 mM Tris,
pH 7.5, 0.1% Tween 20). The polyvinylidene fluoride
membrane was washed, and the remaining accessible
sites were blocked with buffer I. Primary antibody incubations at a 1:5000 dilution were carried out for 1
hour. Membranes were washed extensively with buffer
I and incubated with biotinylated antibody (1:5000
dilution) for 1 hour. After the wash, the membranes
were incubated with Vectastain ABC reagent for 30
minutes. After extensive washes, the membranes were
incubated with either diaminobenzoic acid containing
0.03% NiCl2 and hydrogen peroxide or with the enhanced chemiluminescence (ECL) Western blot reagents (Amersham). The ECL-treated membranes
were exposed further to ECL film.
SDS-PAGE
Polypeptides were electrophoresed under the conditions described by Laemmli.23
Synthesis of Phospholipase C-yl Specific
Antisense Primer
A 32-base-pair antisense phosphorothioated oligonucleotide, 5'-AGC TGA GCA AAC TGC CCG TAG GTG
ATG TCC CC-3', was generated from the sequences
of prior X domain, which shares the least homology
with other phospholipase family members, and a thioated sense strand primer, 5'-GG GGA CAT CAC CTA
CGG GCA GTT TGC TCA GCT-3', was synthesized.
The first-passaged cells were plated in serum-supplemented medium (DMEM-10). When cells reached
50% confluency, they were washed with suspension
culture medium and transferred to one of the following: serum-free medium (SFM), SFM plus 50 fjM antisense primer, SFM plus 50 fjM sense primer, SFM plus
10 ng/ml FGF-2, or SFM plus 50 /JM antisense primer
and 10 ng/ml FGF-2 in the presence of lipofectin. Six
hours later, the medium was replaced with DMEM-10
for 48 hours, after which the cells were trypsinized and
12
16
Incubation (hours)
20
24
FIGURE l. Stimulatory effect of fibroblast growth factor 2
(FGF-2) on DNA synthesis in corneal endothelial cells. Firstpassaged cells at day 2 were starved of fetal calf serum for
48 hours, followed by labeling with 3H-thymidine with (•)
or without (O) FGF-2 at 10 ng/ml for 30 minutes and for
2, 4, 6, 8, 16, and 24 hours. Data are representative of three
separate experiments performed in triplicate.
proliferation was assayed by cell counting in triplicate
dishes.
Materials
The FGF-2 was purchased from Intergen (Purchase,
NY); antibodies directed against PLC-yl, phosphotyrosine, and FGFR-1 were purchased from Upstate Biotechnology Incorporation (Lake Placid, NY); antibodies directed against vinculin and /3-actin, as well as
protein G-Sepharose, were purchased from Sigma (St.
Louis, MO). Vectastain ABC kit was purchased from
Vector Laboratories, and ECL Western blot detection
reagents was purchased from Amersham.
RESULTS
Stimulated DNA Synthesis by Fibroblast Growth
Factor 2
It is known that some growth factors exert a downregulatory effect, whereby their continuous presence
causes a loss of receptors from the cell surface and
results in attenuation of the cellular response to the
ligands.24 Therefore, it was crucial to ascertain
whether FGF-2 caused downregulation of proliferation in corneal endothelial cells. Control cells showed
a linear and continuous incorporation of 3H-thymidine into DNA as a function of labeling time up to 24
hours (Fig. 1). Cells treated with FGF-2 had a profile
of DNA synthesis almost identical to the control cells
up to 6 hours of treatment; thereafter, DNA synthesis
was markedly enhanced as a function of labeling and
treatment time, indicating that FGF-2 stimulated DNA
synthesis in corneal endothelial cells. Genistein,25 a
specific inhibitor against receptor tyrosine kinases, was
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Signaling Pathways of Fibroblast Growth Factor 2 in Cornea! Endothelial Cells
2329
139 KD -
100
200
300
FIGURE 2. Inhibitory effect of genistein on cell proliferation
stimulated by fibroblast growth factor 2 (FGF-2) in corneal
endothelial cells. The first-passaged cells were starved of
serum on day 2 for 48 hours, followed by treatment with
genistein at 0, 10, 50, 100, and 250 //M, in the absence of
FGF-2 (O), with 2 ng/ml of FGF-2 (D), or with 10 ng/ml
of FGF-2 (•). After 16 hours, the cells were labeled with 5
pCi of HH-thymidine for 6 hours. Data are representative of
three separate experiments performed in triplicate.
used to inhibit cell proliferation mediated by FGF-2
(Fig. 2). The first-passaged cells were starved of serum
for 48 hours and treated with 10, 50, 100, or 250 ^M
of genistein in the presence of either 2 ng/ml or 10
ng/ml of FGF-2, whereas control cells received no
FGF-2. In the cells treated with FGF-2, 10 fiM of genistein had no significant inhibitory effect on DNA synthesis, regardless of the concentration of FGF-2; a 50
fiM or higher concentration of genistein demonstrated a dose-dependent inhibitory effect on DNA
synthesis, regardless of the concentration of FGF-2. It
is of interest to note that the control cells, which received no growth factors for 48 hours before treatment
with genistein, showed a dose-dependent inhibition
of DNA synthesis, albeit at low levels.
Fibroblast Growth Factor Receptor and
Phospholipase C-yl Expression
The biologic activities of FGF-2 are mediated by highaffinity cell surface receptors with an intrinsic protein
tyrosine kinase activity. Therefore, we determined the
time course of FGF-2-induced phosphorylation of
high-affinity FGFR. Membrane fractions initially were
immunoprecipitated with antibodies directed against
FGFR-1 and separated by SDS-PAGE. The proteins
were transferred to polyvinylidene fluoride membrane
and immunoblotted with anti-P-tyrosine antibodies,
followed by visualization of the proteins by ECL reagents. Of note, a phosphotyrosyl-containing band,
representing autophosphorylated FGFR-1, was detected in the absence of FGF-2 stimulation (Fig. 3).
The cytosolic fraction of quiescent NIH 3T3 cells contained autophosphorylated FGFR-1 (130 kDa) in the
3 4
1 2
Genistein (uM)
FIGURE 3. Time course of phosphorylation of FGF receptor
1 (FGFR-1) induced by fibroblast growth factor 2 (FGF-2)
in corneal endothelial cells. First-passaged cells on day 2
were starved of fetal calf serum for 48 hours followed by
treatment with FGF-2 at 10 ng/ml for 1 hour (lane 2), 6
hours (lane 3), or 24 hours (lane 4). The membrane fractions were immunoprecipitated with antibody directed
against FGFR-1, separated on 5% SDS-PAGE under the reduced conditions, and transferred to polyvinylidene fluoride
membrane followed by immunoblot analysis with anti-phosphotyrosine antibody using Vector ABC kit and the enhanced chemiluminescence system. 1 = control cells without FGF-2.
absence of FGF-2 treatment.2ti Results show that FGF2 increased tyrosine phosphorylation of FGFR with
increasing incubation times; a peak was reached after
6 hours of FGF-2 stimulation, after which autophosphorylation of FGFR-1 was markedly reduced. To gain
more insight into the consequences of the activated
FGF receptors, the levels of membrane-associated
PLC-yl were determined. There is a negligible
139kD -
mm
1 2 34
*«
«t«
5 6
FIGURE 4. Immunoblot analysis of membrane-associated
phospholipase C (PLC)-yl (lanes 1 to 4) and of cytoskeleton-associated PLC-yl (lanes 5 and 6) mediated by fibroblast
growth factor 2 (FGF-2). The first-passaged cells at day 2
were starved of serum for 48 hours, followed by treatment
with FGF-2 at 10 ng/ml for 1 hour (lane 2), 6 hours (lanes
3 and 5), or 24 hours (lanes 4 and 6). A portion of samples
was imniunoprecipitated with antibodies directed against
PLC-yl, subjected to a 5% SDS-PAGE under the reduced
conditions, and transferred to immobilon membrane, followed by immunoblot analysis with anti-PLC-yl antibody
and (lanes 1 to 4) and anti-P-tyrosine antibody (lanes 5 and
6). 1 = control cells without FGF-2 treatment.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Investigative Ophthalmology 8c Visual Science, October 1996, Vol. 37, No. 11
2330
Cytosol
Membrane
-PLC-71
1 2
3
4
12
3 4
B
-PLC-Yi
1 2
3 4
12
3
4
FIGURE 5. The inhibitory effect of genistein on translocation
ofPLOyl (A) and the effect of cycloheximide on phospholipase C (PLC)-yl level (B)). (A) On day 3, the first-passaged
cells were starved of fetal calf serum for 48 hours, followed
by treatment with fibroblast growth factor 2 (FGF-2) at 10
ng/ml, with or without genistein at 100 /xM for 24 hours.
(B) The first-passaged cells at day 2 were treated with cycloheximide (1 /ig/ml) for 3 days in the presence or absence
of FGF-2 (10 ng/ml). Cytosolic and membrane fractions
were subjected to 5% SDS-PAGE under the reduced conditions and transferred to immobilon membrane, followed by
immunoblot analysis with anti-PLOyl antibody. (A) 1 =
control cells; 2 = control cells treated widi genistein; 3 =
cells treated with FGF-2; 4 = cells treated with FGF-2 and
genistein. (B) 1 = control cells; 2 = control cells treated
with cycloheximide; 3 = cells treated with FGF-2; 4 = cells
treated with FGF-2 and cycloheximide.
amount of membrane-associated PLC-yl in the absence of FGF-2 stimulation (Fig. 4). When cells were
treated with FGF-2 for up to 24 hours, the amount of
PLC-yl, with an apparent molecular weight of 145
kDa associated with membrane fraction, increased
with the increasing incubation times of FGF-2; a peak
was reached after 6 hours, after which the level of the
membrane-associated PLC-yl appears to be maintained. When phosphorylation of the PLC-yl of these
samples was examined with anti-P-tyrosine antibodies,
to our surprise, no tyrosine phosphorylation of PLCyl was observed in the protein band corresponding
to PLC-yl that was identified by immunoblotting with
anti-PLC-yl antibodies in the presence of excess orthovanadate (data not shown). Recent studies22
showed that tyrosine phosphorylation of PLC-yl was
observed only in the cytoskeleton fraction in rat hepatocytes. Therefore, cytoskeleton fractions from the
FGF-2 treated corneal endothelial cells for either 6
hours or 24 hours were prepared and immunoprecipitated widi anti-PLC-yl antibodies, followed by immunoblotting with anti-P-tyrosine antibodies (Fig. 4, lanes
5 and 6). Results show uiat PLC-yl associated with
cytoskeleton is phosphorylated after FGF-2 treatment
of cells, regardless of the treatment time.
Subcellular Location of Phospholipase C-yl
Genistein is known to cause a subcellular translocation
of PLC activity from the membrane fraction to the
cytosolic fraction in rat 3Y1 fibroblast.^ To determine
whether genistein is able physically to translocate the
enzyme, the cells were treated with genistein in the
presence or absence of FGF-2, and membrane and
cytosolic fractions were analyzed (Fig. 5A). The results
show that FGF-2 significandy enhanced the amount
of PLC-yl in the membrane fraction. When cells were
treated with both FGF-2 and genistein, the inhibitor
reduced the amount of membrane-associated PLC-yl.
Although FGF-2 is only able to translocate a minor
fraction of cytosolic enzyme to membrane-associated
enzyme, a major portion of PLC-yl is present in the
cytosolic fraction, regardless of the presence or absence of FGF-2. To determine whether PLC-yl is indeed an abundant and stable cytosolic protein, cells
were treated with cycloheximide for up to 3 days, with
or without FGF-2, followed by immunoblot analysis.
There appear to be no quantitative changes in cytosolic PLC-yl, regardless of the presence or absence
of FGF-2 or the inhibitor, suggesting that PLC-yl is a
stable protein (Fig. 5B). In the absence of protein
synthesis, translocations of a minor fraction of cytosolic enzyme to membrane-associated fraction occurs,
suggesting that this event is a physical property.
To determine whether the abundant cytosolic
PLC-yl is biologically functional, PLC-yl-specific antisense phosphorothioated oligonucleotide primers
were used to prevent de novo translation of PLC-yl.
The primer was directed against the sequence before
X domain, the sequence that has the least homology
with other phospholipases. When cells reached approximately 50% confluency, cells were treated with
antisense or sense strand primers for 6 hours, after
which cells were maintained in growth-supporting medium, with or without FGF-2, for 48 hours. Phospholipase C-yl specific antisense oligonucleotide primer
was able to inhibit cell proliferation by 25% in the
absence of FGF-2 in corneal endothelial cells (Table
1), whereas sense strand had no inhibitory effect on
cell proliferation. Although antisense primer slightly
TABLE 1. The
Effect of PLC-yl-Specific
Antisense Oligonucleotide Primers on Cell
Proliferation of Corneal Endothelial Cell
Condition
Cell Number
(XI Of cells)
Control
AS primer
S primer
FGF-2
FGF-2 + AS
FGF-2 + S
6.8
5.1
8.1
12.3
10.5
12.9
±
±
±
±
±
±
0.02
0.04
0.71
1.09
0.81
0.70
% Control
100.0
75.0
119.1
180.0
154.4
189.7
AS = antisense primer; S = sense strand primer.
When thefirst-passagedcells reached approximately 50%
confluency, cells were treated and analyzed. Data are
representative of three separate experiments performed in
duplicate.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Signaling Pathways of Fibroblast Growth Factor 2 in Corneal Endothelial Cells
FIGURE 6. The effect of colchicine and cytochalasin B on cell
morphology of normal and of modulated corneal endothelial cells with fibroblast growth factor 2 (10 ng/ml), supplemented with heparin (10 ^.g/ml), and corneal endothelium
modulation factor (CEMF) 0.25 (/zg/ml). (A) Normal
cells. (B) Modulated cells. (C) Modulated cells treated with
0.1 fj,g/m\ colchicine. <D) Modulated cells treated with 1
/Ltg/ml cytochalasin B. Magnification, X50.
inhibited cell proliferation stimulated with subsequently added FGF-2, the PLC-yl specific antisense
primer was not able to override the action of FGF-2.
Association of Phospholipase C-yl With
Cytoskeleton
Little is known about the molecular events that occur
after the tyrosine phosphorylation of PLC-yl. Recently, it was reported22 that PLC-yl is associated with
actin and thatEGF induces translocation of PLC-yl to
the cytoskeleton. Therefore, the relationship between
PLC-yl and cytoskeleton was determined with an antimicro tubule agent (colchicine) and an anti-microfilament agent (cytochalasin B). To enhance modulation of corneal endothelial cells to fibroblast cells,
cells were modulated with FGF-2 and CEMF, after
which they were treated with either colchicine or cytochalasin B. At a concentration of 0.1 jug/ml, colchicine was effective in altering the fibroblastic morphology to polygonal shape, whereas cytochalasin B at 1
//g/ml did not alter the cell shape (Fig. 6). The solubilized cytoskeleton was prepared, and 0.5 mg of protein
of each sample was subjected to immunoprecipitation
with anti-PLC-yl antibodies followed by immunoblot
with anti-PLC-yl antibodies, anti-/?-actin antibodies,
or anti-vinculin antibodies (Fig. 7). The level of PLCyl increased in die modulated cells when compared
to the level in the control cells. Although colchicine
did not change the relative level of PLC-yl in the
modulated cells, cytochalasin B significantly decreased
the PLC-yl level in the modulated cells. Vinculin, with
an apparent molecular weight of 116 kDa, was determined in all samples—control cells and modulated
2331
cells, with or without inhibitor. However, the amount
of vinculin associated with PLC-yl specifically increased in the modulated cells; although colchicine
did not change the amount of vinculin, cytochalasin
B slightly reduced the amount of vinculin. When /?actin associated with PLC-yl was determined with anti/?-actin antibody, the modulated cells contained a
higher amount of /3-actin than did the control cells.
Cytochalasin B slightly reduced the amount of actin
in the modulated cells, whereas colchicine did not
change the level of actin. These findings indicate that
PLC-yl was associated with cytoskeleton proteins (vinculin and /3-actin) in corneal endothelial cells and
that the modulated fibroblastic corneal endothelial
cells contain relatively high amounts of PLC-yl, vinculin, and /?-actin complex compared to control cells.
The effect of colchicine and cytochalasin B on cell
proliferation mediated by FGF-2 was determined (Table 2). The modulated cells treated with cytochalasin
B showed a slight decrease in cell numbers and DNA
synthesis, whereas colchicine had a slightly stimulatory
effect on cell proliferation (Table 2).
DISCUSSION
Fibroblast growth factor, a multifunctional, autocrine
growth factor, is known to play a role in a variety of
biologic processes, such as embryonic development,
angiogenesis, transformation, and wound healing.'1""'27
Our previous study demonstrated diat in corneal endothelial cells, FGF-2 is not merely a mitogen, it is a
potent modulator of endothelial phenotypes.8 Endothelial cells grown in the continuous presence of FGF2 not only proliferate excessively but also convert to
205 140 -
^
83 45 kD
- PLC - 7 i
~ Vinculin
- j3"Actin
12
3 4
FIGURE 7. Association of actin and vinculin with phospholipase C (PLC)--yl in the cytoskeleton. The first-passaged
modulated cells, prepared as described in die legend of
Figure 6, were treated with either colchicine (0.1 //g/ml)
or cytochalasin B (1 /jg/ml) for 48 hours. Cytoskeleton fractions were prepared and proteins (0.5 mg/sample) associated with PLC-yl were detected by immunoprecipitation
with PLC-yl antibodies and immunoblotting widi PLC-yl
antibody, vinculin antibody, or /?-actin antibody. 1 = normal
cells; 2 = modulated cells with fibroblast growth factor 2
and corneal endothelium modulation factor (CEMF); 3 =
modulated cells treated with colchicine; 4 = modulated cells
treated with cytochalasin B.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Investigative Ophthalmology 8c Visual Science, October 1996, Vol. 37, No. 11
2332
TABLE 2.
The Effect of Colchicine and
Cytochalasin B on Cell Proliferation of
Corneal Endothelial Cells
When thefirst-passagedcells reached 90% confluency, the cells
were treated with FGF-2 (10 ng/ml), FGF-2 + colchicine (0.1
/^,g/ml), or FGF-2 + cytochalasin B (1.0 ^ig/ml) for 48 hours, at
which time one set of cells was tiypsinized and cell numbers were
counted. Another set of cells was labeled with 'H-thymidine for 6
hours. Data are representative of three separate experiments
performed in triplicate.
takes at least 6 hours of treatment with FGF-2 for the
cells to reach the highest level of membrane-associated PLC-yl. This time course coincides with the lag
period of DNA synthesis when cells are treated with
FGF-2. The requirement for long-term exposure to
FGF-2 for the initiation of maximal DNA synthesis in
corneal endothelial cells agrees with the reported evidence on the continuous exposure of Balb/c 3T3 cells
to FGF-1 for a minimum of 12 hours.31 This further
correlates with the maintenance of a low level of FGF
receptors on the cell surface during the entire Gl
phase of the cell cycle.31 Furthermore, the current
study shows that tyrosine-phosphorylated PLC-yl is
associated with cytoskeleton and that genistein inhibits not only the translocation of PLC-yl but also cell
proliferation mediated by FGF-2, suggesting that translocation of PLC-yl is essential for cell proliferation.
fibroblast-like cells that begin to produce fibrillar collagens (types I, III, and V). In vivo, such phenotypic
modulation leads to ectopic corneal fibrosis that
causes blindness by blocking light transmittance. To
understand how these diverse effects (cell proliferation, cell shape change, collagen phenotypic switch)
of FGF-2 on corneal endothelial cells are triggered,
we focused our attention on the earliest events that
occurred after binding FGF-2 to cells. Among the
known high-affinity FGF receptors, the expression of
FGFR-1 was observed in corneal endothelium in vivo.12
It is generally considered that activation of PLC-yl by
receptor tyrosine kinase is mediated directly by tyrosine phosphorylation of PLC-yl,l<)'28 and that this activity suggests translocation of cytosolic PLC-yl to the
membrane to interact with receptor tyrosine kinases.
Nonetheless, subcellular localization of biologically
functional PLC-yl has not been understood clearly.
Recently, EGF-stimulated PLC-yl has been reported
to be associated with the cytoskeleton, such as with
actin and an unidentified 110 kDa molecule.22 Studies
with A-431 cells in vitro show that the EGF receptor
can increase PLC-yl activity independently of tyrosine
phosphorylation.29 More recent studies show that
phosphatic acid acts as an allosteric activator in vitro
for both native and tyrosine phosphorylated PLC-yl.30
There appear, therefore, to be a number of mechanisms capable of activating PLC-yl. However, it is not
clear which ones are physiologically relevant and
might account for FGF-2-induced activation of PLCyl activity in specific cell types.
The current study proposes to elucidate the molecular event that occurs after the binding of FGF-2
to its receptors in corneal endothelial cells. Unlike
other systems, the addition of FGF-2 to corneal endothelial cells does not induce a rapid phosphorylation
of FGF receptor.1122 It takes at least 6 hours of treatment with FGF-2 for the cells to reach the maximal
state of phosphorylation of FGFR-1. Furthermore, it
Little is known about the molecular events that
occur after the tyrosine phosphorylation of PLC-yl.
It is possible that the SH2 domains of PLC-yl interact
intramolecularly with the phosphorylated tyrosine residues of the receptors. Such an intramolecular interaction may elicit a conformational change that allows
the SH3 domain to bind to the membrane cytoskeleton and position the catalytic X and Y domains at the
cytoplasmic side of the cell membrane. The truncated
PLC-yl containing the SH3 domain, with or without
the SH2 domains, localized to the actin cytoskeleton
when injected into the cell.32 Our study demonstrates
that PLC-yl is associated with vinculin and actin in
normal corneal endothelial cells. When the cells were
modulated to fibroblastic cells with simultaneous
treatment with FGF-2 and CEMF, the modulated cells
contained a higher level of PLC-yl in association with
the vinculin-actin complex. When the modulated fibroblastic endothelial cells were treated with either
colchicine or cytochalasin B, colchicine reverted the
cell shape to polygonal morphology, in contrast to
cytochalasin B, which was not able to revert the cell
shape. It is of interest to note that cytochalasin B appears to dissociate PLC-yl from the vinculin-actin
complex, whereas colchicine did not modulate the
association of these molecules. Thus, association of
PLC-yl to the vinculin-actin complex is not responsible for the modulation of cell morphology. On the
other hand, this complex appears to be involved in
cell proliferation; colchicine slighdy enhanced DNA
synthesis stimulated by FGF-2, whereas cytochalasin B
slightly reduced DNA synthesis in the FGF-2-treated
modulated cells. It has been reported that colchicine
demonstrates a synergistic effect with FGF-2 in the
initiation of DNA synthesis in aortic endothelial cell.33
However, there is a marked difference in the magnitude of the stimulatory effect of colchicine in aortic
or corneal endothelial cells. Recendy, it was reported3'1
that there is interaction between dynamin, a microtubule-associated protein, and the SH3 domain of PLC-
Cell
3
H-Thymidine
Cell Number Incorporation
(XlOf ± SD) (X105 cpm ± SD)
Control
3.70 ±
+FGF-2
10.30 ±
+FGF-2 + cytochalasin B 8.64 ±
+FGF-2 + colchicine
12.05 ±
0.21
0.19
0.56
0.28
3.10 ±
13.73 ±
10.29 ±
14.83 ±
0.05
0.36
0.03
0.11
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
Signaling Pathways of Fibroblast Growth Factor 2 in Corneal Endothelial Cells
y l and that such an interaction may mediate the
movement of PLC-yl along a network of microtubules. Thus, actin cytoskeleton and microtubules are
respectively proposed to be involved in signaling pathways through PLC-yl by translocation of cytosolic
PLC-yl to particulate fractions.
In summary, we can conclude from the current
study that FGF-2 may use the PLC-yl-vinculin-actin
complex for the signaling of cell proliferation and
that this complex is not responsible for the cell shape
change mediated by FGF-2. However, the exact sequence of molecular events in the signaling pathways
for these individual events must be further elucidated.
12.
13.
14.
Key Wards
15.
corneal endothelial cells, fibroblast growth factor receptor,
fibroblast growth factor 2, phospholipase C-yl, signal transduction
16.
Acknowledgment
The authors thank Dr. Sue Goo Rhee, National Institutes
of Health, for advice and discussion.
References
1. Van Horn DL, Hyndiuk RA. Endothelial wound repair
in primate cornea. Exp Eye Res. 1975;21:113-124.
2. Van Horn DL, Sendele DD, Seidman S, Buco PJ. Regenerative capacity of the corneal endothelium in rabbit and cat. Invest Ophthalmol Vis Sri. 1977; 16:597-613.
3. Michels RG, Kenyon KR, Maumenee AE. Retrocorneal
fibrous membrane. Invest Ophthalmol. 1972; 11:822831.
4. Brown SI, Kitano S. Pathogenesis of the retrocorneal
membrane. Arch Ophthalmol. 1966; 75:518-525.
5. Waring GO III. Posterior collagenous layer of the cornea. Ultrastructural classification of abnormal collagenous tissue posterior to Descemet's membrane in 30
cases. Arch Ophthalmol. 1982; 100:122-134.
6. Kay EP, Nimni ME, Smith RE. Modulation of endothelial cell morphology and collagen synthesis by polymorphonuclear leukocytes. Invest Ophthalmol Vis Sri.
1984;25:502-512.
7. Kay EP, Gu X, Ninomiya Y, Smith RE. Corneal endothelial modulation: A factor released by leukocytes
induces basic fibroblast growth factor that modulates
cell shape and collagen. Invest Ophthalmol Vis Sri.
1993; 34:663-672.
8. Kay EP, Gu X, Smith RE. Corneal endothelial modulation: bFGF as direct mediator and corneal endothelium modulation factor as inducer. Invest Ophthalmol
Vis Sri. 1994; 35:2427-2435.
9. Burgess WH, Maciag T. The heparin-binding (fibroblast) growth factor family of proteins. Annu Rev Biochem. 1989; 58:575-606.
10. Maries I, Adelaide J, Raybaud F, et al. Characterization
of the HST-related FGF 6 gene, a new member of the
fibroblast growth factor gene family. Oncogene. 1989;
4:335-340.
11. Crossley PH, Martin GR. The mouse FGF8 gene encodes a family of polypeptides and is expressed in
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
2333
regions that direct outgrowth and patterning in the
developing embryo. Development. 1995; 121:439-451.
Wilson SE, Walker JW, Chwang EL, He YG. Hepatocyte growth factor, keratinocyte growth factor, their
receptors, fibroblast growth factor receptor-2, and the
cells of the cornea. Invest Ophthalmol Vis Sri. 1993;
34:2544-2561.
Landshman N, Belkin M, Ben-Hanan I, Ben-Chaim
O, Assia E, Savion N. Regeneration of cat corneal endothelium induced in vivo byfibroblastgrowth factor.
Exp Eye Res. 1987; 45:805-811.
Hyldahl L, Schofield PN, Engstrom W. Stimulatory
effects of basic fibroblast growth factor on DNA synthesis in the human embryonic cornea. Development.
1990;109:605-611.
Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell. 1990;61:203212.
Klagsbrun M, Baird A. A dual receptor system is required for basic fibroblast growth factor activity. Cell.
1991:67:229-231.
Vainikka S. Partanen J, Bellosta P, et al. Fibroblast
growth factor receptor-4 shows novel features in genomic structure, ligand binding and signal transduction.
EMBOJ. 1992:11:4273-4280.
Suh PG, Ryu SH, Moon KH, Suh HW, Rhee SG. Cloning and sequence of multiple forms of phospholipase
C. Cell. 1988:54:161-169.
Meisenhelder J, Suh PG, Rhee SG, Hunter T. Phospholipase C-y is a substrate for the PDGF and EGF
receptor protein-tyrosine kinases in vivo and in vitro.
Cell. 1989:57:1109-1122.
Burgess WH, Dionne CA, Kaplow J, et al. Characterization and cDNA cloning of phospholipase C-y, a major
substrate for heparin-binding growth factor 1 (acidic
fibroblast growth factor)-activated tyrosine kinase. Mol
Cell Biol. 1990; 10:4770-4777.
Presta M, Maier JAM, Ragnotti G. The mitogenic signaling pathway but not the plasminogen activator-inducing pathway of basic fibroblast growth factor is
mediated through protein kinase C in fetal bovine
aortic endothelial cells. / Cell Biol. 1989; 109:18771884.
Yang LJ, Rhee SG, Williamson JR. Epidermal growth
factor-induced activation and translocation of phospholipase C-yl to the cytoskeleton in rat hepatocytes.
JBiol Chem. 1994;269:7156-7162.
Laemmli UK. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature.
1970;227:680-685.
Masui H, Castro L, Mendelsohn J. Consumption of
EGF by A431 cells: evidence for receptor recycling. J
Cell Biol. 1993;120:85-93.
Nakanishi O, Shibasaki F, Hidaka M, Homma Y, Takenawa T. Phospholipase C-yl associates with viral and
cellular sre kinases. / Biol Chem. 1993; 268:1075410759.
Prudovsky I, Savion N, Zhan X, et al. Intact and functional fibroblast growth factor (FGF) receptor-1 trafficks near the nucleus in response to FGF-1. / Biol
Chem. 1994;269: 31720-31724.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017
2334
Investigative Ophthalmology 8c Visual Science, October 1996, Vol. 37, No. 11
27. Mason IJ. The ins and outs offibroblastgrowth factors.
Cell. 1994; 78:547-552.
28. Noh D-Y, Shin SH, Rhee SG. Phosphoinositide-specific
phospholipase C and mitogenic signaling. Biochem Biophys Ada. 1996; in press.
29. Hernandez-Sotomayor SMT, Carpenter G. Non-catalytic activation of phospholipase G-yl in vitro by epidermal growth factor receptor. Biochem J. 1993;
293:507-511.
30. Jones GA, Carpenter G. The regulation of phospholipase C-yl by phosphatidic acid. Assessment of kinetic
parameters. J Biol Chem. 1993;268:20845-20850.
31. Zhan X, Hu X, Friesel R, Maciag T. Long term growth
factor exposure and differential tyrosine phosphorylation are required for DNA synthesis in BALB/c 3T3
cells. J Biol Chem. 1993; 268:9611-9620.
32. Bar-Sagi D, Rotin D, Batzer A, Mandiyan V, Schlessinger J. SH3 domains direct cellular localization of
signaling molecules. Cell. 1993; 74:83-91.
33. Liaw L, Schwartz SM. Microtubule disruption stimulates DNA synthesis in bovine endothelial cells and
potentiates cellular response to basicfibroblastgrowth
factor. AmJPathol. 1993; 143-937-948.
34. Seedorf K, Kostka G, Lammers R, et al. Dynamin binds
to SH3 domains of phospholipase Cy and GRB-2. /
Biol Chem. 1994;269:16009-16014.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933414/ on 06/18/2017