Transcript Arvo2005poster

Role of Hepatocyte Growth Factor (HGF) in Expression of Proteases
and Cell Migration During Retinal Neovascularization.
E.S. ColomboB, E. MlnarikB, G. MenicucciB, P.G. McGuireB,,, and A. DasA.
ADept. of Surgery/ Division of Ophthalmology, BDept. of Cell Biology and Physiology Univ. Of New Mexico Health Sciences Center, Albuquerque,NM
Abstract
Purpose: Hepatocyte growth factor (HGF), acting through its receptor, c-met, is a cytokine whose role in promoting cell
migration has been extensively documented. HGF and c-met have been implicated in the progression of angiogenesis
in many cancer models and in the mouse model of retinal neovascularization. However, HGF mediation of protease
expression and of an endothelial cell migratory phenotype in the retina has not been explored. We determined if HGF
and/or hypoxia effects cell migration and protease levels in retinal microvascular endothelial cells and in a mouse
model of retinal neovascularization. Methods: Bovine retinal microvascular endothelial cells (BRMVEC) were grown to
near confluence and then stimulated with HGF (10ng/ml and 30 ng/ml) under normoxic and/or hypoxic conditions. At
varying time points, conditioned media and cell extracts were collected and analyzed by zymography or by reversetranscription PCR. Stable hypoxic conditions were achieved by adding 100M CoCl2 to the cell media. Migration
assays were performed by growing BRMVEC’s to confluence with or without HGF and wounding the monolayer. After 24
hours, the amount of cellular migration into the wound was analyzed. Retinal neovascularization was induced in
newborn mice by exposure to 75% oxygen followed by room air. Eyes were collected at day 17 following removal from
high oxygen, fixed in 10% formalin and embedded in paraffin. Sections were cut, and immunohistochemistry was
performed to localize retinal c-met expression. Results: After 24 hours of HGF stimulation there is a hypoxia-dependent,
and dose-dependent increase in the 54 kDa form of urokinase in endothelial cells. This expression pattern correlated
with the PCR analysis of uPA and uPAR mRNA. There is also a hypoxia-dependent and dose dependent increase in the
active form of MMP 2. The wound assay demonstrated that HGF significantly increased BRMVEC cell migration
compared to control samples. Immunohistochemical staining showed localized c-Met expression in the retina, which
was increased in experimental animals compared to control. Conclusion: HGF, acting through its receptor c-met, may
play an important role in the initial stages of retinal angiogenesis by stimulating a migratory phenotype and endothelial
cell protease production.
According to the National Diabetes Information Clearinghouse, diabetic retinopathy cases
between 12,000 to 14,000 new cases of blindness each year. Additionally, diabetic retinopathy is the
leading cause of blindness in individuals between 20 and 74 years of age in the United States.
Understanding the process by which vessels are stimulated to pathologically invade the retina may
provide important clues to future treatments of this devastating disease. It is believed that the
primary stimulus to induce new vessel growth in the retina is hypoxia. Many angiogenic factors have
been shown to be upregulated in response to hypoxic conditions. These factors include basic
fibroblast growth factor (bFGF), insulin-like growth factor (IGF) and vascular endothelial growth factor
(VEGF)3.
Studying cancer models of angiogenesis has proven useful in discerning new potential targets.
Hepatocyte Growth Factor, and its receptor, c-Met have been implicated in many cancer models1.
It is believed that HGF, when acting through c-Met causes the loss of cell-cell junctions1, the
increased expression of proteases4,5 VEGF1, and cell proliferation1 and/or anti-apoptosis6 of
endothelial cells.
The induction and activation of proteases by growth factors has been studied, however
examination of HGF’s effects of proteases in a hypoxia model of retinal angiogenesis has not been
published. We examined HGF’s effects on urokinase which when activated, converts plasminogen
to plasmin. Plasmin effectively degrades basement membranes and activates other proteases, such
as MMPs. These particular proteases are important because they have been well characterized in
hypoxia-induced retinal angiogenesis and their inhibition has resulted in significant decreases in
retinal angiogenesis in some models7,8.
Finally we examined weather HGF could induce an invasive phenotype in retinal microvascular
endothelial cells, and if this invasion was mediated by urokinase. In order to accomplish this, cells
were cultured on matrigel with HGF or with HGF and a urokinase-specific inhibitor, Ǻ6 which has
been previously established as a potent inhibitor of hypoxia induced retinal neovascularizaiton9.
BREC invasion +/- HGF and urokinase inhibitor
3.
4.
5.
6.
7.
8.
9.
The authors would like to thank the generous financial support from the American Diabetes Association and
NIH Grant EY12604-06 who made this work possible.
4000
4000
2000
30
# migratory cells
*
6000
*
3000
2000
1000
0
0
1
ITS
1
ITS
HGF 10
25
*
top its, bottom its
top its, bottom hgf
20
Its/its
Its+Ǻ6/hgf
top its+Ǻ6, bottom its
top its, bottom serum
15
*
10
HGF 10
5
0
1
Media + ITS
Media + ITS + 10ng/ml HGF
Media + ITS
Media + ITS + 10ng/ml HGF
Many studies indicate that HGF causes cells to increase protease expression. We hypothesized that HGF
stimulation results in an increase in urokinase expression. Here we show that HGF induces significant increases
in urokinase activity in both the conditioned media and in cell extracts after 24 hours of HGF stimulation.
Urokinase in Conditioned Media, n=3, p=0.0001. Urokinase in Cell Extracts, n=3, p=0.001
•We determined if HGF increased uPAR
gene expression by using RT-PCR.
Following 24 hours of HGF stimulation,
RNA was extracted, cDNA was
synthesized and uPAR gene expression
was determined relative to GAPDH.
uPAR RNA Expression
25000
20000
15000
•Because uPAR activates urokinase, its
increased expression offers a mechanism
for increased urokinase activation seen in
response to HGF.
10000
5000
0
ITS only control
HGF- 10 ng
BREC Migration Assay
Top=ITS, Bottom=ITS
*
40
Top=ITS, Bottom=Serum
30
20
Top: ITS
Bottom: ITS
Top: ITS
Bottom: MCDB 131 Complete
Top=ITS, Bottom=ITS+HGF
10
Top=ITS, Bottom=ITS+VEGF
0
1
Top: ITS
Bottom: ITS + 10ng/ml HGF
Its/complete media
Next, we hypothesized that the invasive phenoype induced by HGF was mediated by urokinase
activation. BRECs were cultured in Matrigel invasion chambers with 10ng/HGF as a chemoattractant.
The urokinase-specific inhibitor, Ǻ6 was used to determine if inhibition of urokinase activity decreased
BREC invasion into the matrigel. Our data indicates that HGF-stimulated invasion through basement
membrane can be completely inhibited by Ǻ6, this suggests that HGF mediates its invasive potential
through urokinase activation. ITS/ITS:ITS/HGF, n=4 p=0.003. ITS/HGF:ITS,Ǻ6/HGF, n=4 p=0.016.
Having established previously that HGF is upregulated in a mouse model of hypoxia-induced
angiogenesis, we decided to further elucidate what role HGF is playing in the retina. Because of many
previous reports of HGF mediating protease expression, we chose to examine the relationship between
HGF and urokinase. Using zymography assays, we determined that HGF increases urokinase activity in
vitro after 24 hours of stimulation. We further determined that this increased activity may be mediated
by incrased expression of urokinase receptor uPAR, because uPAR RNA level is nearly doubled in
response to HGF.
Next, we examined weather HGF induced a migratory phenotype in retinal microvascular
endothelial cells. Using a migration assay, and HGF as a chemoattractant, we determined that HGF
induces greater migration across a 0.8 um membrane than does VEGF, or complete media. Having
established this effect, we speculated that this migration was mediated by urokinase activity. To test
this, we used a urokinase-specific antagonist, Ǻ6 on microvascular cells grown on Matrigel, again using
HGF as a chemoattractant. The addition of Ǻ6 to the media reduced cell migration to background
levels, indicating that the migratory and invasive phenotype seen in response to HGF is mediated by
urokinase activity.
This study sheds new light on the role of HGF in mediating hypoxia-induced retinal angiogenesis. We
hope that this new information will lead to an increased understanding of hypoxia-induced retinal
angiogenesis, and thereby more pharmacologic targets for treatment of diabetic retinopathy and
retinopathy of prematurity.
Methods
Figure 3.
HGF Increases Cell Migration
50
Its/hgf
Conclusions
Figure 2.
HGF-induces UPAR Gene Expression
# of migratory cells
2.
Birchmeier C., Birchmeier W., Gherardi E., Vande Woude, Met, Metastasis, Motility and More. Nature Reviews, Mol. Cell Bio. 2003.
4:915-925
Kataoka H, Miyata S, Uchinokura S, Itoh, H. Roles of hepatocyte growth factor (HGF) activator and HGF activator inhibitor in the
pericellular activation of HGF/scatter factor. Cancer Metastasis Rev. 2003. 22(2-3):223-36.
Bergers G, Benjamin LE. Tumorigenesis and the angiogenic switch.
Nat Rev Cancer. 2003 Jun;3(6):401-10. Review.
Harvey P, Clark IM, Jaurand M-C, Warn RM, Edwards DR. Hepatocyte growth factor/scatter factor enhances the einvasion of
mesothelioma cell lines and the expression of matrix metalloproteinases. Br. J. Cancer. 2000. 83(9):1145-1153
Tacchini L, Matteucci E, De Ponti C, Desiderio, MA. Hepatocyte growth factor signaling regulates transactivation of genes belonging
to the plasminogen activation system via hypoxia inducible factor-1. Exp. Cell Res. 2003. 290:391-401.
Reveneau S, Paimell R, Peheuninck J, Leroy C, De Launoit Y, Fafeur V. Inhibition of JNK by HGF/SF prevents apoptosis induced by TNFalpha.
Ann N Y Acad Sci. 2003. 1010:100-3.
Majka, S., P. G. McGuire, et al. (2002). "Regulation of matrix metalloproteinase expression by tumor necrosis factor in a murine model
of retinal neovascularization." Invest Ophthalmol Vis Sci 43(1): 260-6.Wang
Le Gat, L., K. Gogat, et al. (2003). "In vivo adenovirus-mediated delivery of a uPA/uPAR antagonist reduces retinal
neovascularization in a mouse model of retinopathy." Gene Ther 10(25): 2098-103.
McGuire, P. G., T. R. Jones, et al. (2003). "The urokinase/urokinase receptor system in retinal neovascularization: inhibition by A6
suggests a new therapeutic target." Invest Ophthalmol Vis Sci 44(6): 2736-42
35
IDV Value
IDV Value
8000
References
1.
Urokinase Activity in Cell Extracts
Urokinase activity in
Conditioned Media
IDT Values
Background
Figure 4.
HGF Increases Cell Invasion
Figure 1.
HGF-Induces Urokinase Activity
Top: ITS
Bottom: ITS + 10ng/ml VEGF
Having established that HGF induces urokinase activation, we speculated that the increase in protease
activity would result in a migratory phenotype. BREC’s responded to HGF chemoattraction with a significant
increase in migration compared to Serum-Free and Complete-Media controls. This indicates that retinal
microvascular cells respond to HGF by increasing migration toward the HGF stimulus. n=4, p=0.003
Animal Model of Retinal Neovascularization. Retinal neovascularization was induced in newborn mice according to the protocol of Smith et al.
C57Bl/6 mice were exposed to 75% oxygen for 5 consecutive days beginning on post natal day 7. Mice were brought to room air on day 12. By day 18,
retinal neovascularization developed in 100% of the animals. Animas were sacrificed and the eyes collected for analysis. Animals exposed to room air
only served as age-matched controls. All experiments were consistent with the Association for Research in Vision and Ophthalmology (ARVO) statement
for the use of animals in ophthalmic and vision research
Cell culture. Migration and Invasion Assays: 0.8 μm pore size cell culture inserts (Falcon, BD Biosciences, Bedford, MD), coated with 50 μg/ml
fibronectin (Calbiochem, San Diego, CA) or Matrigel coated invasion assay chambers (BD Laware, Bedford MA), respectively, were plated with 2 X 104
Bovine Retinal Endothelial Cells (Vec Technologies, Rensselaer, NY). Cells were kept in MDCD-131 complete media (Vec Technologies, Inc., Rensselaer,
NY), until cultures were 70% confluent on the upper chamber, then transferred to serum free media, MCDB-131 with Insulin, Transferrin and Sodium
Selenite (ITS) (Sigma-Aldrich, St. Louis, MO) and 1X Pennicillin/Streptomycin overnight. The following day, 10 ng/ml HGF (R&D Systems Inc.,
Minneapolis, MN), 10 ng/ml VEGF, or MDCB-131 complete was added to the lower chambers, the upper chambers maintained in serum free media,
except where 100mM Ǻ6 (Angstrom Pharm., San Diego, CA) was added as described. 24 hours later, any cells remaining in the upper chamber were
removed with a cotton swab, the membranes were removed and coversliped with Vectashield (Vector Labs, Burlingame, CA) and photographs were
taken. Protease Assays: In a fibronectin-coated 24 well plate (Bectin Dickinson, Bedford, MA), Bovine Retinal Endothelial Cells were grown to 90%
confluence and transferred into serum free media overnight, media contained Insulin, Transferrin and Sodium Selenite (ITS) (Sigma-Aldrich, St. Louis,
MO). The following day, 10ng/ml of human recombinant HGF (Chemicon GF116, Temecula, CA) was added to the media. 24 hours later, conditioned
media was collected and either cell monolayer mRNA was extracted using Trizol Reagent (Invitrogen, Carlsbad, CA), or cell extracts were collected for
zymographic analysis (see below). cDNA was synthesized using (Superscript First Strand Synthesis System, Invitrogen, Carlsbad, CA). PCR primers
and probes were designed using MacVector 7.1.1 (Accelrys, San Diego, CA). Forward Primer: 5’CTAAAGGACTGAGGAACCCAAGC3’; Reverse
Primer: 5’GCCGCAACGCTATCAATAATCAC3’
Protease Activity. Zymographic analysis was performed using BRECs solubilized in 0.1% Triton X-100, 0.1 M phosphate (pH 7.3). Samples were
electrophoresed in 10% polyacrylamide minigels, into which casein and plasminogen (Sigma, C-7891 and P-5661, respectively) was cross-linked. After
electrophoresis, the enzymes were renatured by soaking the gels in a solution of 2.5% Triton X-100 followed by multiple rinses with distilled water. Next,
gels were incubated in 100mM Tris pH 8 at 37ºC for 3 hours with gentle rocking. The zones of proteolysis corresponding to the presence of urokinase
were visualized by staining the gel with 0.125% Coomassie blue and comparing the position of the bands with a molecular weight standard. The degree
of digestion related to the amount of enzyme present in the various samples was quantified using the image analysis software (Alpha Imager; Alpha
Innotech).