\set{final}

\def\Author{Kay}
\def\author{kay}
\def\vol{4}
\def\year{1998}
\def\page{22}
\def\txt_title{Fibroblast growth factor 2 uses PLC-[gamma] 1 for cell proliferation and PI3-kinase for alteration of cell shape and cell proliferation in corneal endothelial cells}
\def\txt_authors{EunDuck P. Kay, Sun Young Park, MinHee K. Ko, Sung Chul Lee}

\def\rcvd{1 July 1998}
\def\accept{8 October 1998}
\def\publ{27 October 1998}
\def\pdfsize{833}
\def\PMID{9808840}


\include{mvstyle.hsm}

\| External links

\| Internal defs

\| Key words: Corneal endothelial cells, fibroblast growth factor-2 (FGF-2), nuclear FGF-2, polymorphonuclear leukocytes, cell proliferation

\article{

\title{Fibroblast growth factor 2 uses PLC-\gamma 1 for cell
proliferation and PI3-kinase for alteration of cell shape and cell
proliferation in corneal endothelial cells}

\authors{\mailto{ekay@hsc.usc.edu}{EunDuck P. Kay},\sup{1,2} Sun
Young Park,\sup{1} MinHee K. Ko,\sup{1} Sung Chul Lee\sup{3}}

\institutions{\sup{1}Doheny Eye Institute and \sup{2}Department of
Ophthalmology, University of Southern California School of Medicine,
Los Angeles, CA, USA; \sup{3}Department of Ophthalmology, Yonsei
University College of Medicine, Seoul, Korea}

\correspondence{EunDuck P. Kay, DDS, PhD, Doheny Eye Institute, DVRC
#203, 1450 San Pablo Street, Los Angeles, CA, 90033; Phone: (323)
442-6625; FAX: (323) 442-6688; email:
\mailto{ekay@hsc.usc.edu}{ekay@hsc.usc.edu}}

\abstract

\abs_purpose{Fibroblast growth factor 2 (FGF-2) induces
endothelial-mesenchymal modulation in corneal endothelial cells,
including stimulation of cell proliferation and cell shape change
and induction of fibrillar collagen.  In the present study, we
investigated whether FGF-2 uses distinct signaling pathways for
individual biological activities.}

\abs_methods{Specific metabolic inhibitors were used to block cell
proliferation, while  reversion of cellular morphology (modulated
with FGF-2) was determined using specific antibodies and inhibitors. 
Immunocytochemical analysis was performed to identify any changes
observed in the cytoskeleton in relation to cell shape. Association
of cytoskeleton molecules with phosphatidylinositol 3-kinase was
determined using co-precipitation.  Cell proliferation was assayed
using a colorimetric method for determining the number of viable
cells.}

\abs_results{The fibroblastic morphology induced by FGF-2 reverted
to a polygonal shape in cells treated with anti-FGF-2 antibody,
anti-phosphatidylinositol 3-kinase antibody, LY294002, and
genistein, while anti-phospholipase C \gamma 1 antibody did not to
reverse the modulated cell morphology.  Cell proliferation mediated
by FGF-2 was blocked by metabolic inhibitors (genistein, LY294002
and wortmannin); genistein inhibited FGF-mediated cell proliferation
in a dose-response manner and had a maximum inhibition of 80% at 100
\mu M, while inhibitors of phosphatidylinositol 3-kinase had less
inhibitory effect than did genistein.  When cytoskeleton proteins
were examined, the characteristic punctated staining profiles of
vinculin observed in normal cells were maintained in fibroblastic
corneal endothelial cells treated with FGF-2.  The inhibitors that
cause reversion of cell shape also demonstrated the punctated
staining potential. Likewise, the staining profiles of
\alpha-actinin and smooth muscle \alpha-actin were not altered,
regardless of cell shape.  Filamentous actin and \alpha-actinin were
co-localized to the cytoskeleton and phosphatidylinositol 3-kinase
was associated with the cytoskeleton, regardless of cell shape.}

\abs_conclusions{These findings indicate that FGF-2 uses distinct
and/or dual signaling pathways for individual biological activities.}

\introduction

\p{The corneal endothelium is a monolayer of differentiated cells
located in the posterior portion of the cornea.  It is essential for 
maintaining corneal transparency, but
its capacity for regeneration after injury is severely limited in
humans, primates, and cats [1,2].  In response to certain
pathological conditions, corneal endothelial cells (CECs) in
vivo may respond by converting to fibroblast-like cells.  These
morphologically modulated cells then resume their proliferation
ability and start to produce fibrillar collagens, leading to the
formation of a fibrillar extracellular matrix.  A clinical example
of this process is the development of a retrocorneal fibrous
membrane [3,4], the presence of which blocks vision, thereby causing
blindness.  In our previous studies [5,6], we found that corneal endothelium
modulation factor (CEMF) secreted by polymorphonuclear leukocytes
(PMNs), fibroblast growth factor-2 (FGF-2), or a combination of the
two factors modulates phenotypes of CECs, leading to a
modulation similar to that observed in vivo (up-regulation of
cell proliferation, cell shape changes, and collagen phenotype
alteration).  We further found that CEMF could induce \i{de novo}
synthesis of FGF-2 and that the newly produced FGF-2 was the direct
mediator for the modulation of CECs [7].}

\p{FGF-2, a member of the fibroblast growth factor family, is a
multifunctional regulator of cell development, differentiation,
regeneration, senescence, proliferation, and migration [8-10].  In
normal cornea, FGF-2 is a component of Descemet's membrane that may
be necessary for wound repair [7,11-13].  The biological actions of
FGF-2 are mediated through transmembrane cell surface receptors that
possess tyrosine kinase activity [14,15].  One of the early
cellular events induced by the binding of FGF-2 to its receptor is
the stimulation of phospholipase C\gamma 1 (PLC-\gamma 1), which
hydrolyzes inositol phospholipid and generates the second messengers
diacylglycerol and inositol 1,4,5-phosphate [16,17].  PLC-\gamma 1
has been a major substrate of FGF receptor 1 (FGFR-1) [18] and is
known to be responsible for mitogenesis [19,20].  In our previous
studies, we demonstrated that PLC-\gamma 1 associated with the
cytoskeleton (vinculin and actin)
 plays a role in mitogenesis,
whereas the PLC-\gamma 1 complex has no effect on cell shape changes
mediated by FGF-2 [21].  Furthermore, PLC-\gamma 1 is
 involved in carcinogenesis or tumor progression [22-24]. 
Likewise, phosphatidylinositol (PI) 3-kinase, another intracellular
effector molecule, is implicated in a variety of cellular
processes, including mitogenesis, transformation, membrane ruffling,
actin polymerization, and vesicle transport [25-27].  Nevertheless,
limited information is available concerning the existence of a
specific signal transduction pathway for individual cellular
responses in a given growth factor-bound receptor system.}

\p{Although we demonstrated that PLC-\gamma 1 was responsible for
mitogenesis mediated by FGF-2 in CECs, the pathway leading to cell shape
change remains undefined.  It may be that different forms
of phosphoinositides are involved in differential cellular
responses [28].  In this respect, it is likely that PLC-\gamma 1 and
PI3-kinase are responsible for differential cellular responses
mediated by FGF-2.  In this report, we show that there are indeed
differential signaling pathways for mitogenesis and modulation of cell
shape mediated by FGF-2.  PI3-kinase is involved in the modulation of
cell shape and in mitogenesis, in part, whereas PLC-\gamma 1 is
responsible for mitogenesis, but is not involved in cell shape change.}

\methods

\subsection{Cell Cultures}

\p{Isolation and establishment in culture of rabbit CECs were
performed as previously described [5].  Briefly, Descemet's
membrane-corneal endothelium complex was treated with 0.2%
collagenase and 0.05% hyaluronidase for 90 min at 37 \deg C.
Cultures were maintained in Dulbecco's modified
Eagle's medium (DMEM) supplemented with 10% fetal bovine
serum and 50 \mu g/ml of gentamicin (DMEM-10) in a 5% CO\sub{2}
incubator.  This procedure has been shown to promote cell
proliferation during the early phase of culture and to maintain the
culture as a contact-inhibited monolayer when the cells reach
confluency.  First passaged CECs were used for all experiments.
Modulated CECs were prepared by treating the cells with FGF-2 (10
ng/ml) and heparin (10 \mu g/ml): the duration of the treatment of
cells with FGF-2 varied in experiments.}

\subsection{Treatment with Inhibitors and Neutralizing antibodies}

\p{When the modulated cells reached confluency with elongated cell
morphology, the cells were treated with genistein (Sigma, St. Louis,
MO), LY294002 (Sigma), anti-FGF-2 antibody (Upstate Biotechnology
Inc, UBI, Lake Placid, NY), anti-PI3-kinase antibody (Transduction
Laboratories, Lexington, KY), or anti-PLC-\gamma 1 antibody (UBI). 
For reversion of cellular morphology, the inhibitors were used at a
concentration of 10 \mu M for genistein and 20 \mu M for LY294002;
for the inhibitory action of the inhibitors on cell proliferation
mediated by FGF-2, a range of concentrations was used.  Neutralizing
antibody was added to the culture in the amount of 5 \mu g per 1 x
10\sup{6} cells.}

\subsection{Cell Proliferation Assay}

\p{Normal and modulated CECs (5 x 10\sup{3}) were plated in 96-well
tissue culture plates.  When the cells reached approximately 80%
confluency, they were treated with respective inhibitors for 24
h.  At the end of the incubation period, 100 \mu l of
CellTiter 96\sup{R}AQueous One Solution Cell Proliferation Assay reagent
(Promega, Madison, WI) was added to the wells.  The plates were
incubated for 1 h at 37\sup{o}C in a humidified 5% CO\sub{2}
atmosphere after which the absorbance was read at 490 nm using a 96 well
plate reader.  This step was repeated at the end of the 2- and
3-hour incubations.  Absorbance readings at the end of the
2-hour incubation were similar to those following the 3-hour
incubation, therefore, a 2-hour incubation was used.}

\subsection{Immunofluorescent Staining }

\p{Normal and modulated CECs (3 x 10\sup{4}/chamber) were seeded on
four-well chamber slides.  The cells were treated with inhibitors as
stated above.  Cells were then washed with phosphate buffered saline
(PBS) and fixed with 3% paraformaldehyde in PBS.  All washes and
incubations were carried out in PBS at room temperature.  Cells were
then permeabilized with 0.5% Triton X-100 in PBS for 5 min and
blocked with 2% bovine serum albumin (BSA).  The chamber slides were
incubated with primary antibody (1:50 dilution) for 2 h then
washed with PBS.  Cells were then incubated with the biotinylated
secondary antibody (Vector Laboratories, Inc., Burlingame, CA; 1:50
dilution) for 1 h followed by incubation with fluorescein
conjugated to avidin (Vector; 1:100 dilution) for 30 min. 
Following extensive washing, the slides were examined under a Leica
confocal microscope (Heidelberg, Germany).  For the co-localization
experiment of filamentous actin (F-actin) and \alpha-actinin, the
slides were simultaneously incubated with fluorescein phalloidin
(1:100 dilution) (Molecular Probes, Eugene, OR) and monoclonal
anti-\alpha-actinin antibody (1:100) (Sigma) for 2 h at room
temperature then washed with PBS.  Cells were then incubated with
biotinylated anti-mouse immunoglobulins (1:100) for 1 h, followed
by incubation with rhodamine conjugated to avidin (1:50) for 30
min at room temperature.  Following extensive washing, the
slides were examined under a Zeiss confocal microscope.}

\subsection{Immunoprecipitation}

\p{Cytoskeleton was prepared as previously described [21].  All
buffers were maintained at 4\sup{o}C during 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 \mu g/ml aprotinin, 10 \mu g/ml leupeptin, 0.2 mM sodium orthovanadate,
and 1 mM phenylmethylsulfonyl fluoride (PMSF).  Homogenates were
centrifuged and the pellet was dissolved in
sodium-dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE)
sample buffer followed by boiling.  Protein concentration was
determined using a BioRad Dc Protein Assay kit (Bio-Rad
Laboratories, Hercules, CA).}

\subsection{Immunoprecipitation}

\p{Samples of a constant amount (10-20 \mu g) were subjected to
immunoprecipitation.  Five micrograms of undiluted anti-PI3-kinase
antibody was added.  To this mixture, 50 \mu l of protein
G-sepharose resin (Sigma) was added, and incubation was carried out
at 4\sup{o}C for 1 h.  Following centrifugation at 10,000 rpm for
10 min, the resin was washed three times with PBS containing
protease inhibitors (PMSF, aprotinin and EDTA).  The protein was
eluted from the resin by boiling in SDS-PAGE sample buffer for 5
min.  After a brief spin, the supernatant was subjected to
SDS-PAGE.}

\subsection{SDS-Polyacrylamide Gel Electrophoresis}

\p{The conditions of electrophoresis were as described by Laemmli
[29].  Vinculin and \alpha-actinin were analyzed on a 5% gel under
reduced conditions and smooth muscle \alpha-actin was analyzed on an
11% gel under non-reducing conditions using a discontinuous
Tris-glycine buffer system (pH 8.3).}

\subsection{Immunoblot Analysis}

\p{Proteins separated by SDS-PAGE were electrophoretically
transferred to a polyvinylidene difluoride (PVDF) membrane at 0.22
ampere for 10 h in a semi-dry transfer system (Transfer buffer:
25 mM Tris, pH 8.3, 190 mM glycine, 20% MeOH).  Immunoblot analysis
was performed as described previously [21] using a commercial ABC
Vectastain kit (Vector Laboratories).  All washes and incubations
were carried out at room temperature in TTBS (0.9% NaCl, 100 mM
Tris-HCl, pH 7.5, 0.1% Tween 20).   Briefly, the PVDF membrane was
immediately placed into the blocking buffer (5% nonfat milk
containing TTBS) for 1 h.  The incubation with primary antibody
(1:5000 dilution) was carried out for 2 h; incubation with
biotinylated secondary antibody (1:2500 dilution) was carried out
for 1 h; and incubation with the ABC reagent was for 30 min. 
The membrane was treated with enhanced chemiluminescence (ECL)
reagent (Amersham Life Science, Buckinghamshire, England) for 1
min and the ECL-treated membrane was exposed further to ECL
film.}

\results

\subsection{Effect of Inhibitors on FGF-2 mediated Cell Shape Changes}

\p{The involvement of PI3-kinase in the signal transduction pathway
for cell shape change mediated by FGF-2 was investigated using
specific antibodies (\figref{1}).  The confluent CECs demonstrated a
polygonal monolayer, while the FGF-2-treated cells showed a loss of
polygonal shape and a subsequent change to an elongated morphology. 
When CECs were treated simultaneously with FGF-2 and anti-FGF-2
antibody, the neutralizing antibody was able to block the modulating
activity of FGF-2.  A similar reversion was shown when CECs were
treated simultaneously with FGF-2 and anti-PI3-kinase antibody.  On
the other hand, anti-PLC-\gamma 1 antibody was unable to block the
modulation activity of FGF-2.  This confirmed our previous
finding that PLC-\gamma 1 associated with vinculin and actin is not
involved in the modulation of cell shape  [21].  The effect of
inhibitors for PI3-kinase and receptor tyrosine kinase on the
fibroblastic CECs was determined (\figref{2}).  In the continuous
presence of FGF-2, polygonal cells were completely converted to
elongated fibroblastic cells.  The difference in the cellular
morphology of CECs treated with FGF-2 between \figref{1} and \figref{2}
was caused by differences in the duration of treatment with the
growth factor and when the treatment begins.  When the fibroblastic
CECs were treated with a low concentration of LY294002 (20 \mu M) to
specifically inhibit PI3-kinase, LY294002 was able to cause the cell
shape to revert completely to a polygonal morphology (\figref{2}{C}), while
wortmannin (a less specific inhibitor for PI3-kinase) or genistein
(an inhibitor for receptor tyrosine kinase), almost but not
completely, caused the modulated cellular morphology to revert to a
polygonal shape (\figref{2}{D}{E}).  It is of interest that a high concentration of
LY294002 (40 \mu M) demonstrated inhibitory activity on mitogenesis
as well (\figref{2}{F}).}

\subsection{Actions of Inhibitors on Cytoskeleton}

\p{To determine whether FGF-2 causes endothelial mesenchymal
transformation in CECs, the expression of smooth muscle
\alpha-actin, a mesenchymal marker, was determined in normal
polygonal CECs, modulated fibroblastic cells, and reversed polygonal
cells induced by inhibitors (\figref{3}).  Regardless of cell shape
(polygonal or fibroblastic) or cell state (endothelial versus
mesenchymal), all cells showed filamentous cytoplasmic distribution
of smooth muscle \alpha-actin. The staining of smooth muscle
alpha-actin appeared to be organized into bundles, similar to stress
fibers.  Neither LY294002 nor genistein altered the staining profile
of smooth muscle \alpha-actin, either qualitatively or
quantitatively (\figref{3}{C} and \figref{3}{D}).  When a similar
analysis was performed for \alpha-actinin, an actin-binding protein,
almost identical profiles to those seen with smooth muscle
\alpha-actin were observed: \alpha-actinin demonstrated filamentous
and punctate staining within the cells, regardless of cell shape
(\figref{4}).  Neither FGF-2 nor inhibitors altered the staining
profiles of \alpha-actinin.  When the staining profiles of vinculin,
another actin-binding protein, were analyzed, the characteristic
punctate staining was observed in normal CECs.  The cells that
reverted after treatment with inhibitors, as well as the modulated
fibroblastic cells, demonstrated a similar degree of punctate
staining profiles (\figref{5}): these cells showed far less staining
potential than the normal CECs.  Nevertheless, the characteristic
punctate staining potential was not altered by either FGF-2 or
inhibitors (\figref{5}).  When filamentous actin (F-actin) and
\alpha-actinin were simultaneously stained, CECs demonstrated
prominent F-actin bundles and marked colocalization profiles of
these proteins in normal and modulated CECs (\figref{6} and
\figref{7}).  These cells, regardless of cell shape, showed an
additional intracellular filamentous distribution of \alpha-actinin.
When the fibroblastic CECs were reverted by treatment with LY294002,
the induced polygonal cells demonstrated identical staining profiles
to those seen in normal CECs (\figref{8}): \alpha-actinin was
present along the cortical actin filament with occasional
intracellular stress fiber bundles and there was a marked
colocalization of F-actin and \alpha-actinin.}

\subsection{Association of PI3-kinase with Cytoskeleton}

\p{Our previous study showed that PLC-\gamma 1 is associated with
cytoskeleton [21]; PLC-\gamma 1 associated with vinculin and actin is
responsible for cell proliferation mediated by FGF-2, whereas the same
complex is not involved in the alteration of cellular morphology
mediated by FGF-2.  We, therefore, examined whether PI3-kinase also is 
associated with cytoskeleton proteins using immunoprecipitation with
anti-PI3-kinase antibody followed by immunoblotting with anti-smooth
muscle \alpha-actin antibody, anti-\alpha-actinin antibody or
anti-vinculin antibody (\figref{9}). The immune complex contained
\alpha-actinin (100 kDa), vinculin (116 kDa) and smooth muscle
\alpha-actin (45 kDa), regardless of cell shape (polygonal versus
fibroblastic).  The amounts of \alpha-actinin and smooth muscle
\alpha-actin were similar in both normal and fibroblastic CECs, whereas
the level of vinculin in the fibroblastic CECs appeared to be much lower
than that of the normal cells.  These findings are in agreement with the
immunofluorescent analysis.}

\subsection{Effect of Inhibitors on Cell Proliferation Mediated by FGF-2}

\p{In order to determine whether PI3-kinase is involved also in
mitogenic signaling pathway mediated by FGF-2, primary CECs were
treated with one of the following conditions: FGF-2, FGF-2 and
genistein in concentrations ranging from 10 to 100 \mu M, FGF-2 and
LY294002 in concentrations ranging from 5 to 50 \mu M, and
wortmannin in concentrations ranging from 10 to 500 nM (\figref{10}). 
Genistein inhibited the cell proliferation mediated by FGF-2 in a
dose-dependent manner, a mild inhibitory action was observed up to 50
\mu M, but there was a marked inhibition at 100 \mu M.  LY294002
demonstrated a moderate inhibitory action in a dose-dependent
manner: approximately 60% inhibition occured at 50 \mu M.  Unlike
LY294002, wortmannin, a less specific PI3-kinase inhibitor, reached
a maximal  inhibitory activity on cell proliferation mediated by
FGF-2 at low concentration of 20 nM.}

\discussion

\p{Our previous study demonstrated that in corneal endothelial
cells, FGF-2 is not merely a mitogen, it is also a potent modulator
of endothelial phenotypes [6,7,21].  Endothelial cells grown in the
continuous presence of FGF-2 not only proliferate excessively, they
also convert to fibroblast-like cells that begin to produce
fibrillar collagens (types I, III and V).  It is likely that several
distinct signal transduction pathways contribute to the
establishment of these cellular responses.  To understand how these
diverse actions (cell proliferation, cell shape change, collagen
phenotype switch) of FGF-2 on corneal endothelial cells are relayed,
we focused our attention on the earliest events, such as the direct
bindings of PLC-gamma 1 and PI3-kinase to FGF receptors. We
previously reported that PLC-\gamma 1 is the key molecule for the
mitogenic signaling pathway mediated by FGF-2 and that PLC-\gamma 1
is not involved in cell shape changes mediated by FGF-2 [21].  The
results presented here demonstrate that PI3-kinase is the mediator
for modulation of cellular morphology of CECs. LY294002 at a
concentration of 20 \mu M, which is low enough to block only
PI3-kinase, caused the fibroblastic morphology modulated by FGF-2 to
revert to a polygonal shape.  Likewise, anti-PI3-kinase antibody was
able to block the activity of FGF-2 on cell shape changes.  On the
other hand, anti-PLC-\gamma 1 antibody failed to block the
modulation activity of FGF-2.  These findings suggest that
PI3-kinase is the key molecule in dictating cell shape mediated by
FGF-2.  For mitogenic activity of FGF-2, the inhibitors for
PI3-kinase were able to moderately block cell proliferation.
LY294002 at low concentrations showed a moderate inhibitory action
on FGF-2-mediated cell proliferation. Wortmannin, a nonspecific
inhibitor of the enzyme, had a similar activity at low
concentrations.  These findings suggest that the initial steps of
the putative mitogenic pathways initiated by association with
PLC-\gamma 1 and PI3-kinase are non-overlapping.  It is not known
whether both pathways lead to entry into S phase or whether they
converge.}

\p{Our previous study also demonstrated that PLC-\gamma 1 is associated
with cytoskeleton, vinculin and actin [21].  Therefore, we further
investigated whether PI3-kinase is associated with cytoskeleton.  The
immunoprecipitation followed by immunoblotting analysis demonstrated
that PI3-kinase was associated with vinculin, \alpha-actinin, and smooth
muscle \alpha-actin.  The amount of \alpha-actinin and smooth muscle
\alpha-actin associated with PI3-kinase in the fibroblastic CECs
modulated with FGF-2 was similar to that of normal CECs, whereas
the level of vinculin associated with PI3-kinase in the modulated cells
was reduced when compared to that of normal cells.  Such a finding was
confirmed with the immunofluorescent staining profile of these proteins.
The staining profiles of \alpha-actinin and smooth muscle \alpha-actin
in the fibroblastic modulated cells were similar to those of polygonal
normal cells or cells that were reversed to a polygonal shape by
the inhibitors.  On the other hand, the characteristic punctate staining
of vinculin was less prominent in modulated CECs.}  

\p{The present findings suggest that the organization and
distribution of actin stress fibers (one major cytoskeleton system)
is not changed as a result of the action of FGF-2 or inhibitors.
These findings also confirm our previous observation that
cytochalasin B does not cause the FGF-2 mediated fibroblastic
morphology to revert to a polygonal shape [30], unlike colchicine,
which does cause the cellular morphology to revert from fibroblastic
to polygonal shape.  Taken together, it is likely that what dictates
cellular morphology in corneal endothelial cells is not a
microfilament system but a microtubule system.  Whether PI3-kinase
is associated with microtubule for its action remains to be
elucidated.}

\p{In summary, we conclude from the current study and our previous
study that FGF-2 may use PLC-\gamma 1 for mitogenic signaling and
PI3-kinase for modulation of cellular morphology and mitogenic
response.  Further investigation focusing on the mechanism by which
PI3-kinase and PLC-\gamma 1 direct the initiation of cell
proliferation should unravel the similarities and differences
between these two pathways.  Furthermore, investigations focusing on
the mechanism by which PI3-kinase directs modulation of cellular
morphology is expected to provide information on the specificity of
the individual cellular responses.}

\acknowledgements

\p{The authors thank Hae Won Kim (Yonsei University College of
Medicine, Seoul, Korea) for her technical service on confocal laser
microscopy and Yonsei University for the confocal analysis of the
present work performed at the facility.  The authors also thank Dr.
James Jester for his discussion of staining of cytoskeleton
proteins.  Support for this work was from NIH grants EY06431 and an
unrestricted grant from the Research to Prevent Blindness.}

\references

\p{1. Van Horn DL, Hyndiuk RA. Endothelial wound repair in primate
cornea. Exp Eye Res 1975; 21:113-124. \pubmed{809290}}

\p{2. Van Horn DL, Sendele DD, Seideman S, Buco PJ. Regenerative
capacity of the corneal endothelium in rabbit and cat. Invest Ophthalmol
Vis Sci 1977; 16:597-613. \pubmed{873721}}

\p{3. Michels RG, Kenyon KR, Maumence AE. Retrocorneal fibrous membrane.
Invest Ophthalmol 1972; 11:822-831. \pubmed{4116098}}

\p{4. Brown SI, Kitano S. Pathogenesis of the retrocorneal membrane.
Arch Ophthalmol 1966; 75:518-525. \pubmed{5324210}}

\p{5. Kay EP, Nimni ME, Smith RE. Modulation of endothelial cell
morphology and collagen synthesis by polymorphonuclear leukocytes.
Invest Ophthalmol Vis Sci 1984; 25:502-512. \pubmed{6715126}}

\p{6. 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 Sci 1993; 34:663-672. \pubmed{8449685}}

\p{7. Kay EP, Gu X, Smith RE. Corneal endothelial modulation: bFGF as
direct mediator and corneal endothelium modulation factor as inducer.
Invest Ophthalmol Vis Sci 1994; 35:2427-2435. \pubmed{8163333}}

\p{8. Mason IJ. The ins and outs of fibroblast growth factor. Cell 1994;
78:547-552. \pubmed{8069907}}

\p{9. Gospodarowicz D. Biological activities of fibroblast growth
factors. Ann N Y Acad Sci 1991; 638:1-8. \pubmed{1785796}}

\p{10. Burgess WH, Maciag T. The heparin-binding (fibroblast) growth
factor family of proteins. Annu Rev Biochem 1989; 58:575-606.
\pubmed{2549857}}

\p{11. Vlodavsky I, Folkman J, Sullivan R, Fridman R, Ishai-Michaeli R,
Sasse J, Klagsbrun M. Endothelial cell-derived basic fibroblast growth
factor: synthesis and deposition into subendothelial extracellular
matrix. Proc Natl Acad Sci U S A 1987; 84:2292-2296. \pubmed{3470794}}

\p{12. Landshman N, Belkin M, Ben-Hanan I, Ben-Chaim O, Assia E, Savion
N. Regeneration of cat corneal endothelium induced in vivo by fibroblast
growth factor. Exp Eye Res 1987; 45:805-811. \pubmed{3428400}}

\p{13. 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. \pubmed{2401214}}

\p{14. Ullrich A, Schlessinger J. Signal transduction by receptors with
tyrosine kinase activity. Cell 1990; 61:203-212. \pubmed{2158859}}

\p{15. Klagsbrun M, Baird A. A dual receptor system is required for
basic fibroblast growth factor activity. Cell 1991; 67:229-231.
\pubmed{1655276}}

\p{16. Suh PG, Ryu SH, Moon KH, Suh HW, Rhee SG. Cloning and sequence of
multiple forms of phospholipase C. Cell 1988; 54:161-169.
\pubmed{3390863}}

\p{17. Meisenhelder J, Suh PG, Rhee SG, Hunter T. Phospholipase C-gamma
is a substrate for the PDGF and EGF receptor protein-tyrosine kinases in
vivo and in vitro. Cell 1989; 57:1109-1122. \pubmed{2472219}}

\p{18. Burgess WH, Dionne CA, Kaplow J, Mudd R, Friesel R, Zilberstein
A, Schlessinger J, Jaye M. Characterization and cDNA cloning of
phospholipase C-gamma, a major substrate for heparin-binding growth
factor 1 (acidic fibroblast growth factor)-activated tyrosine kinase.
Mol Cell Biol 1990; 10:4770-4777. \pubmed{2167438}}

\p{19. Smith MR, Liu YL, Matthews NT, Rhee SG, Sung WK, Kung HF.
Phospholipase C-gamma 1 can induce DNA synthesis by a mechanism
independent of its lipase activity. Proc Natl Acad Sci U S A 1994;
91:6554-6558. \pubmed{8022819}}

\p{20. Smith MR, Ryu SH, Suh PG, Rhee SG, Kung HF. S-phase induction and
transformation of quiescent NIH 3T3 cells by microinjection of
phospholipase C. Proc Natl Acad Sci U S A 1989; 86:3659-3663.
\pubmed{2726744}}

\p{21. Gu X, Seong GJ, Lee YG, Kay EP. Fibroblast growth factor 2 uses
distinct signaling pathways for cell proliferation and cell shape
changes in corneal endothelial cells. Invest Ophthalmol Vis Sci 1996;
37:2326-2334. \pubmed{8843917}}

\p{22. Chang JS, Noh DY, Park IA, Kim MJ, Song H, Ryu SH, Suh PG.
Overexpression of phospholipase C-gamma1 in rat 3Y1 fibroblast cells
leads to malignant transformation. Cancer Res 1997; 57:5465-5468.
\pubmed{9407951}}

\p{23. Arteaga CL, Johnson MD, Todderud G, Coffey RJ, Carpenter G, Page
DL. Elevated content of the tyrosine kinase substrate phospholipase
C-gamma 1 in primary human breast carcinomas. Proc Natl Acad Sci U S A
1991; 88:10435-10439. \pubmed{1683701}}

\p{24. Lee SJ, Lee SD, Park JG, Kim CM, Ryu SH, Suh PG. Overexpression
of phospholipase C-gamma 1 in colorectal carcinomas is associated with
overexpression of factors that bind its promoter. J Biol Chem 1995;
270:16378-16384. \pubmed{7608208}}

\p{25. Kapeller R, Cantley LC. Phosphatidylinositol 3-kinase. Bioessays
1994; 16:565-576. \pubmed{8086005}}

\p{26. Wennstrom S, Hawkins P, Cooke F, Hara K, Yonezawa K, Kasuga M,
Jackson T, Claesson-Welsh L, Stephens L. Activation of phosphoinositide
3-kinase is required for PDGF-stimulated membrane ruffling. Curr Biol
1994; 4:385-393. \pubmed{7922352}}

\p{27. Stephens LR, Jackson TR, Hawkins PT. Agonist-stimulated synthesis
of phosphatidylinositol(3,4,5)-trisphosphate: a new intracellular
signaling system? Biochim Biophys Acta 1993; 1179:27-75.
\pubmed{8399352}}

\p{28. Cantley LC, Auger KR, Carpennter C, Duckworth B, Graziani A,
Kapeller R, Soltoff S. Oncogenes and signal transduction. Cell 1991;
64:281-302. \pubmed{1846320}}

\p{29. Laemmli UK. Cleavage of structural proteins during the assembly
of the head of bacteriophage T4. Nature 1970; 227:680-685.
\pubmed{5432063}}

\p{30. Kay EP, Oh S. Colchicine reverts cell shape but not collagen
phenotypes in corneal endothelial cells modulated by polymorphonuclear
leukocytes. Invest Ophthalmol Vis Sci 1987; 28:826-833.
\pubmed{3553059}}

}

\beginfigures

\figfile{1}{
\figtitle{1}{Effect of neutralizing antibodies on cell
morphology of corneal endothelial cells modulated with FGF-2 (10
ng/ml) supplemented with heparin (10 \mu g/ml)}

\p{The confluent first passage CECs were treated with different 
conditions for 24 h. (x150)}

\p{(A) Normal CECs}

\ctr{\jpgimage{1a}{191}{153}{24}}

\p{(B) FGF-2}

\ctr{\jpgimage{1b}{190}{151}{24}}

\p{(C) FGF-2 and anti-FGF-2 antibody}

\ctr{\jpgimage{1c}{189}{151}{25}}

\p{(D) FGF-2 and anti-PI3-kinase}

\ctr{\jpgimage{1d}{191}{152}{25}}

\p{(E) FGF-2 and anti-PLC-\gamma 1}

\ctr{\jpgimage{1e}{193}{152}{23}}
}

\figfile{2}{
\figtitle{2}{The effect of inhibitors on cell morphology of CECs
modulated with FGF-2}

\p{The first passage CECs were treated with
FGF-2 (10 ng/ml) supplemented with heparin (10 \mu g/ml) upon
plating.  On day 4, in which cellular morphology was changed to
elongated fibroblastic shape, the cells were exposed to inhibitors
for 48 h.  (x150)}

\p{(A) normal CECs}

\ctr{\jpgimage{2a}{186}{149}{25}}

\p{(B) CECs treated with FGF-2}

\ctr{\jpgimage{2b}{186}{144}{26}}

\p{(C) CECs treated with FGF-2 and 20 \mu M LY294002}

\ctr{\jpgimage{2c}{187}{148}{25}}

\p{(D) CECs treated with FGF-2 and 10 \mu M genistein}

\ctr{\jpgimage{2d}{186}{148}{25}}

\p{(E) CECs treated with FGF-2 and 100 nM wortmannin}

\ctr{\jpgimage{2e}{185}{147}{25}}

\p{(F) CECs treated with 40 \mu MLY294002}

\ctr{\jpgimage{2f}{188}{150}{22}}
}

\figfile{3}{
\figtitle{3}{Immunofluorescent analysis of smooth muscle \alpha-actin in CECs}

\p{CECs modulated with FGF-2 (10 ng/ml and 10 \mu g/ml
heparin) were treated with inhibitors in the presence of FGF-2 for
24 h.  Cells were stained with anti-smooth muscle \alpha-actin
antibody, processed and analyzed on confocal laser microscope as
described in the text. (X 400)}

\p{(A) normal CECs}

\ctr{\jpgimage{3a}{263}{212}{13}}

\p{(B) cells treated with FGF-2}

\ctr{\jpgimage{3b}{264}{208}{18}}

\p{(C) cells treated with FGF-2 and LY294002 (20 \mu M)}

\ctr{\jpgimage{3c}{264}{207}{20}}

\p{(D) cells treated with FGF-2 and genistein (10 \mu M)}

\ctr{\jpgimage{3d}{265}{209}{17}}
}

\figfile{4}{
\figtitle{4}{Immunofluorescent analysis of \alpha-actinin in CECs}

\p{Cells modulated with FGF-2 were treated with inhibitors in the
presence of FGF-2 for 24 h.  Cells were stained with anti
\alpha-actinin antibody, processed, and analyzed on confocal
microscopy as described in the text. (X 400)}

\p{(A) normal CECs}

\ctr{\jpgimage{4a}{263}{211}{19}}

\p{(B) cells treated with FGF-2 }

\ctr{\jpgimage{4b}{264}{209}{23}}

\p{(C) cells treated with FGF-2 and LY294002 (20 \mu M)}

\ctr{\jpgimage{4c}{264}{209}{19}}

\p{(D) cells treated with FGF-2 and genistein (10 \mu M)}

\ctr{\jpgimage{4d}{267}{208}{21}}
}

\figfile{5}{
\figtitle{5}{Immunofluorescent analysis of vinculin in CECs}

\p{Cells modulated with FGF-2 were treated with inhibitors in the
presence of FGF-2 for 24 h.  Cells were stained with
anti-vinculin antibody, processed and analyzed on confocal
microscopy as described in the text.  (X 400)}

\p{(A) normal CECs}

\ctr{\jpgimage{5a}{263}{209}{23}}

\p{(B) cells
treated with FGF-2}

\ctr{\jpgimage{5b}{260}{210}{17}}

\p{(C) cells treated with FGF-2 and LY294002 (20 \mu M)}

\ctr{\jpgimage{5c}{264}{208}{16}}

\p{(D) cells treated with FGF-2 and genistein (10 \mu M)}

\ctr{\jpgimage{5d}{260}{205}{14}}
}

\figfile{6}{
\figtitle{6}{Colocalization of F-actin and \alpha-actinin in
normal CECs}

\p{Cells were treated with Triton-X-100, BSA and double labeled for
F-actin and \alpha-actinin as described in the text. Fluorescein
signals (green) are F-actin positive, rhodamine signals (red) are
\alpha-actinin positive, and yellow signals show the colocalization
of F-actin and \alpha-actinin.  The last panel is a phase-contrast
image. (bar = 25 \mu m)}

\p{\ctr{\jpgimage{6a}{302}{217}{30}}}

\p{\ctr{\jpgimage{6c}{302}{218}{27}}}

\p{\ctr{\jpgimage{6b}{303}{217}{32}}}

\p{\ctr{\jpgimage{6d}{303}{213}{12}}}
}

\figfile{7}{
\figtitle{7}{Colocalization of F-actin and \alpha-actinin in
modulated CECs}

\p{Modulated cells were induced with treatment of FGF-2, and
staining procedures were described in \figref{6} and in the text.
Fluorescein signals (green) are F-actin positive, rhodamine signals
(red) are \alpha-actinin positive, and yellow signals show the
colocalization of F-actin and \alpha-actinin. The last panel is a
phase-contrast image. (bar = 25 \mu m)}


\p{\ctr{\jpgimage{7a}{342}{244}{29}}}

\p{\ctr{\jpgimage{7c}{341}{243}{42}}}

\p{\ctr{\jpgimage{7b}{342}{242}{50}}}

\p{\ctr{\jpgimage{7d}{339}{242}{30}}}

}

\figfile{8}{
\figtitle{8}{Colocalization of F-actin and \alpha-actinin in
modulated CECs treated with LY294002}

\p{Modulated cells induced by FGF-2 were treated with LY294002 (20
\mu M) and the staining procedures were described in \figref{6} and
in the text.  Fluorescein signals (green) are F-actin positive,
rhodamine signals (red) are \alpha-actinin positive, and yellow
signals show the colocalization of F-actin and \alpha-actinin. (bar
= 25 \mu m)}

\p{\ctr{\jpgimage{8a}{342}{241}{51}}}

\p{\ctr{\jpgimage{8c}{341}{238}{58}}}

\p{\ctr{\jpgimage{8b}{342}{245}{71}}}
}

\figfile{9}{
\figtitle{9}{Association of vinculin, smooth muscle \alpha-actin
and \alpha-actinin with PI3-kinase in the cytoskeleton}

\p{Cytoskeleton fractions of the first passage CECs and modulated
CECs were prepared, and proteins (30 \mu g/sample) associated with
PI3-kinase were detected by immunoprecipitation with PI3-kinase
antibody and immunoblotting with vinculin antibody, smooth muscle
\alpha-actin antibody or \alpha-actinin antibody.  1, normal CECs;
2, CECs modulated with FGF-2.  M, protein size marker.}

\p{\ctr{\jpgimage{9}{617}{272}{13}}}
}

\figfile{10}{
\figtitle{10}{The effect of inhibitors on cell proliferation
mediated by FGF-2 in CECs}

\p{The first passage CECs were plated into a 96-well plate and
treated with one of the following conditions for 24 h when cells
reached 80% confluency: FGF-2, FGF-2 and genistein in concentration
ranging from 10 to 100 \mu M (top panel); FGF-2 and LY294002 in
concentration ranging from 5 to 50 \mu M (middle panel); and
wortmannin in concentration ranging from 10 to 500 nM (bottom
panel).  The cell proliferation assay was performed using CellTiter
96\sup{R}AQueous One Solution Cell Proliferation Assay (Promega) as
described in the text. Inhibitory activity of the inhibitors were
presented as percent of FGF-2 mediated cell proliferation.}

\p{\ctr{\gifimage{10a}{364}{197}{3}}}

\p{\ctr{\gifimage{10b}{364}{197}{3}}}

\p{\ctr{\gifimage{10c}{363}{202}{3}}}
}