Skip to main content

Full text of "USPTO Patents Application 10555896"

See other formats


PCT 



WORLD INTELLECTUAL PROPERTY ORGANIZATION 
International Bureau 




INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) 



(51) International Patent Classification 7 : 
A61K 31700 



A2 



(11) International Publication Number: WO 00/67737 

(43) International Publication Date: 16 November 2000 (16.1 1.00) 



(21) International Application Number: PCT/US00/ 12309 

(22) International FHing Date: 5 May 2000 (05.05.00) 



(30) Priority Data: 

60/132,964 



7 May 1999 (07.05.99) 



US 



(63) Related by Continuation (CON) or Contmuation-in-Part 
(CIP) to Earlier Application 

US 60/132,964 (CIP) 

Filed on 7 May 1999 (07.05.99) 



(71) Applicant (for all designated States except US): THE 

BRIGHAM AND WOMEN'S HOSPITAL, INC. [US/US]; 
75 Francis Street, Boston, MA 02115 (US). 

(72) Inventors; and 

(75) Inventors/Applicants (for US only): GALPER, Jonas, B. 
[US/US]; 186 Gardner Street, Brookline, MA 02445 (US). 
KONG, Dequan [CA/US]; 16 Palmer Street #1, Waltham, 
MA 02451 (US). 

(74) Agent: ELRIF1, Ivor, R.; Mintz, Levin, Cohn, Ferris, Glovksy 
and Popeo, P.C, ., One Financial Center, Boston, MA 021 1 1 
(US). 



(81) Designated States: AE, AG, AL, AM, AT, AU, AZ, BA, BB, 
BG, BR, BY, CA, CH, CN, CR, CU, CZ, DE, DK, DM, 
DZ, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, 
IN, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, 
LV, MA, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, 
RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, TZ, 
UA, UG, US, UZ, VN, YU, ZA, ZW, ARIPO patent (GH, 
GM, KE, LS, MW, SD, SL, SZ, TZ, UG, ZW), Eurasian 
patent (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), European 
patent (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, 
IE, IT, LU, MC, NL, PT, SE), OAPI patent (BF, BJ, CF, 
CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG). 



Published 

Without international search report and to be republished 
upon receipt of that report. 



(54) Title: USE OF HMGCoA REDUCTASE INHIBITORS IN THE PREVENTION OF DISEASES WHOSE PATHOGENESIS IS 
DEPENDENT ON NEOVASCULARIZATION 



(57) Abstract 

HMGCoA reductase inhibitors have a 
well-known mechanism in controlling cholesterol 
metabolism. HMGCoA reductase inhibitors also 
have a less well-known effect on gene expression. 
This invention provides a new use for HMGCoA 
reductase inhibitors in the treatment of diseases whose 
pathogenesis is dependent on neovascularization. 
HMGCoA reductase inhibitors are administered at 
anti-angiogenic therapeutic doses for the treatment of 
primary and metastatic tumors, inflammatory processes 
involving new vessel formation, diabetic retinopathy, 
rheumatoid arthritis, and atherosclerosis. HMGCoA 
reductase inhibitors affect the expression of genes 
through interference with the function of small GTP 
binding proteins (such as Rho). Because of the low 
incidence of side effects with these agents, HMGCoA 
reductase inhibitors could also be taken prophylactically 
to prevent the development of diseases in which the 
pathogenesis is caused by neovascularization. 




Pj^fiosphate ^ Pyrophosphate 



PyrophosphHtl 

aearvt 



QUUtueuq \ i ■ 

Protoln Protein GftranyJ- 
Furncsylallon Geranytatioti 



Cholesterol 



FOR THE PURPOSES OF INFORMATION ONLY 



Codes used to identify States party to the PCT on the front pages of pamphlets publishing international applications under the PCT. 



AL 


Albania 


ES 


Spam 


LS 


Lesotho 


SI 


Slovenia 


AM 


Armenia 


FI 


Finland 


LT 


Lithuania 


SK 


Slovakia 


AT 


Austria 


FR 


France 


LU 


Luxembourg 


SN 


Senegal 


AU 


Australia 


GA 


Gabon 


LV 


Latvia 


sz 


Swaziland 


AZ 


Azerbaijan 


GB 


United Kingdom 


MC 


Monaco 


TD 


Chad 


BA 


Bosnia and Herzegovina 


GE 


Georgia 


MD 


Republic of Moldova 


TG 


Togo 


BB 


Barbados 


GH 


Ghana 


MG 


Madagascar 


TJ 


Tajikistan 


BE 


Belgium 


GN 


Guinea 


MK 


The former Yugoslav 


TM 


Turkmenistan 


BF 


Burkina Faso 


GR 


Greece 




Republic of Macedonia 


TR 


Turkey 


BG 


Bulgaria 


HU 


Hungary 


ML 


Mali 


TT 


Trinidad and Tobago 


BJ 


Benin 


IE 


Ireland 


MN 


Mongolia 


DA 


Ukraine 


BR 


Brazil 


IL 


Israel 


MR 


Mauritania 


UG 


Uganda 


BY 


Belarus 


IS 


Iceland 


MW 


Malawi 


US 


United States of America 


CA 


Canada 


IT 


Italy 


MX 


Mexico 


uz 


Uzbekistan 


CF 


Central African Republic 


JP 


Japan 


NE 


Niger 


VN 


Viet Nam 


CG 


Congo 


KE 


Kenya 


NL 


Netherlands 


YU 


Yugoslavia 


CH 


Switzerland 


KG 


Kyrgyzstan 


NO 


Norway 


zw 


Zimbabwe 


a 


Cote d'lvoire 


KP 


Democratic People's 


NZ 


New Zealand 






CM 


Cameroon 




Republic of Korea 


PL 


Poland 






CN 


China 


KR 


Republic of Korea 


FT 


Portugal 






CU 


Cuba 


KZ 


Kazakstan 


RO 


Romania 






CZ 


Czech Republic 


LC 


Samt Lucia 


RU 


Russian Federation 






DE 


Germany 


U 


Liechtenstein 


SD 


Sudan 






DK 


Denmark 


LK 


Sri Lanka 


SE 


Sweden 






EE 


Estonia 


LR 


Liberia 


SG 


Singapore 







WO 00/67737 PCT/US00/1 2309 

USE OF HMGCoA REDUCTASE INHIBITORS 
IN THE PREVENTION OF DISEASES WHOSE PATHOGENESIS 
IS DEPENDENT ON NEOVASCULARIZATION 



TECHNICAL FIELD OF THE INVENTION 
This invention relates to methods of treating or preventing diseases whose pathogenesis 
is dependent on neovascularization. 



BACKGROUND OF THE INVENTION 
Angiogenesis (the formation of new blood vessels from a preexisting vasculature) 
involves the proliferation, migration, and differentiation of endothelial cells. Growth factors 
such as basic fibroblast growth factor and vascular endothelial growth factor (VEGF) are 
potent stimulators of angiogenesis. However, the balance between these pro- angiogenic 
stimulatory factors and other anti-angiogenic inhibitory factors regulates angiogenesis in the 
human body. In normal adults, angiogenesis plays a role in the female reproductive system, in 
the hair cycle, and in wound healing. 

Angiogenesis normally occurs in only a few adult human tissues under normal 
physiological conditions. In the adult, pro-angiogenic stimuli affect the pathogenesis of several 
disease states, including the growth and development of tumors. New blood vessels might 
facilitate the inflammation process by bringing in white blood cells and nutrients, and might 
result in the enhancement of tumor growth. For tumors, the repression or limitation of 
angiogenic activity could interfere with the development of new tumors and cause the 
regression of pre-existing tumors. Prevention of angiogenesis could avert the damage caused 
by the invasion of a new microvascular system. Therapies directed at control of the angiogenic 
processes could lead to the abrogation or mitigation of these diseases. 

For this reason, a considerable interest has arisen in the angiogenic mechanisms of 
disease and in the discovery of agents which might interfere with angiogenesis {see, Folkman, 
1 Nature Medicine 27-31 (1995); and Barinaga, 275 Science 482-4 (1997)). A number of 
anti-angiogenic agents have been developed and agents are undergoing clinical trials. These 
include (a) antibodies to angiogenic proteins, such as vascular endothelial growth factor 
(VEGF); (b) molecules which block the function of integrins that are found on the surface of 
endothelial cells, interact with the extracellular matrix, and are involved in the differentiation 
and migration of endothelial cells; (c) molecules which block the activity of 



WO 00/67737 PCT/USOO/12309 
metalloproteinases which breakdown the extracellular matrix and permit the migration of 
endothelial cells during new vessel formation (such as BB-94 (batimastat; British Biotech 
Pharmaceuticals, Oxford, UK)); and (d) agents such as angiostatin and endostatin which are 
secreted by tumors which interfere with the development of metastases by inhibiting new 
vessel formation (see, United States patents 5,885,795 and 5,854,205, both to O'Reilly et al, 
both incorporated herein by reference). Other anti-angiogenic agents are thalidomide, 
interleukin 12 (IL-12), TIE-2, anti-tumor necrosis factor a (TNF-a) antibodies, minocycline, a 
interferon, and the specific angiogenesis inhibitor AGM-1470 (Takeda- Abbott 
Pharmaceuticals). Anti-angiogenic agents might cause the regression and disappearance of 
tumors and the stabilization of atherosclerotic plaques {see, Moulton et al, 99 Circulation 
1726-1732 (1999); Bergers et al, 284(5415) Science 808-812 (1999)). Many of these agents 
are the subjects of clinical trials, but none have yet been approved for clinical use and their 
efficacy in human disease is unknown. 

What is needed is a method that is known to be safe and which can effectively inhibit 
the unwanted growth of blood vessels, especially growth of blood vessels into tumors. The 
method should be able to overcome the activity of endogenous growth factors. The method 
should also be able to modulate the formation of capillaries in other angiogenic disease states 
in which angiogenesis plays a role. The method for inhibiting angiogenesis should preferably 
produce few side effects. 

SUMMARY OF THE INVENTION 

The invention provides a new use for 3-hydroxy-3-methylglutaryl CoA (HMGCoA) 
reductase inhibitors (statins) in the treatment of diseases whose pathogenesis is dependent on 
neovascularization (angiogenesis). The methods are effective for modulating angiogenesis, and 
inhibiting unwanted angiogenesis, especially angiogenesis related to tumor growth. Among the 
new uses of HMGCoA reductase inhibitors are for the treatment and prevention of primary and 
metastatic tumors, for the treatment and prevention of the inflammatory process involving new 
vessel formation, for the treatment and prevention of diabetic retinopathy, for the treatment 
and prevention of rheumatoid arthritis, and for the treatment and prevention of atherosclerosis, 
by causing the regression of atherosclerotic lesions. 

The invention uses HMGCoA reductase inhibitors at therapeutic or prophylactic doses 
for the treatment or prevention of these diseases. HMGCoA reductase inhibitors are currently 



WO 00/67737 PCT/US00/12309 
in wide use in the treatment and prevention of coronary artery disease and stroke by reducing 
the level of lipids in the blood: HMGCoA reductase inhibitors are known to have a low 
incidence of side effects. Unexpectedly, however, HMGCoA reductase inhibitors can also be 
used to provide medically important anti-angiogenic effects, through a newly discovered 
mechanism by which the administration of HMGCoA reductase inhibitors is used to modulate 
the activity of small GTP-binding proteins, such as Rho. Among the HMGCoA reductase 
inhibitors that can be used are simvastatin (Zocor®; Merck), pravastatin (Pravachol®; Bristol 
Myers Squibb), lovastatin (Mevacor®; Merck), atorvastatin (Lipitor;® Park-Davis), fluvastatin 
(Lescol®; Sandoz) and cerevastatin (Bayer). 

The invention also provides a birth control method, in which an effective amount of an 
HMGCoA reductase inhibitor prevents uterine neovascularization. 

BRIEF DESCRIPTION OF THE DRAWINGS 

FIG. 1 shows the effect of simvastatin on the organization of capillary-like structures 
by human umbilical vein endothelial cells (HUVECs) grown on Matrigel®. FIG. 1 A shows a 
control. FIG. IB shows the effect of 0.1 jiM simvastatin. FIG. 1C shows the effect of 1 ^iM 
simvastatin. FIG. ID shows the effect of 5 ^M, simvastatin. 

FIG. 2 is a bar graph showing the effect of simvastatin on the proliferation (FIG. 2A) 
and migration (FIG. 2B) of endothelial cells. Cells were incubated with various concentrations 
of simvastatin for three days and cells harvested and counted. 

FIG. 3 is a bar graph showing the effect of HMGCoA reductase inhibitors in 
VEGF-mediated angiogenesis in a chorioallantoic membrane (CAM) model. VEGF with and 
without simvastatin was introduced onto the chorioallantoic membrane in a collagen 
containing gel sandwiched between a nylon mesh. Placed on the surface of the chorioallantoic 
membrane. Angiogenesis was quantified by counting the percentage of squares in the top mesh 
containing blood vessels. Chorioallantoic membranes were incubated with either vehicle, 250 
ng VEGF, simvastatin alone, or 250 ng VEGF plus various concentrations of simvastatin. 

FIG. 4 shows the effects of simvastatin on FGF-2 stimulated angiogenesis in a mouse 
corneal pocket model. P denotes the position of polymer implantation, arrows indicate the 
presence of blood vessels. FIG. 4A shows angiogenesis stimulated by a polymer containing 10 
ng FGF-2. FIG. 4B shows angiogenesis stimulated by a polymer containing 10 ng FGF-2 plus 
5 ^iM simvastatin. FIG. 4C - FIG. 4F are photomicrographs of sagittal sections of mouse 



WO 00/67737 PCT/US00/12309 
corneas. FIG. 4C shows a 24 hr incubation with the polymer alone. FIG. 4D shows a 24 hr 
incubation with polymer containing 10 ng of FGF-2. FIG. 4E shows a 24 hr incubation with 10 
ng of FGF-2 plus 5 \xM simvastatin. FIG. 4F shows a 24 hr incubation with 10 ng of FGF-2 
plus 10 nM simvastatin. 

FIG. 5 is a set of micrographs showing the effects of GGPP, GGTI-287 and C3 
exo-toxin on HUVECs cultured on Matrigel, thus demonstrating the involvement of a 
geranylgeranylated Rho GTPase in the formation of capillary-like structures. (FIG. 5A) 
control. (FIG. 5B) 5 (iM simvastatin plus 10 jiM FPP. (FIG. 5C) 5 |iM simvastatin plus 10 
GGPP. (FIG. 5D) 10 nM FTI-277. (FIG. 5E) 10 \jM GGTI-287. (FIG. 5F) 5 jig/ml C3 
exo-toxin. 

FIG. 6 is a schematic representation of the cholesterol biosynthetic pathway, including 
several cholesterol by-products, such as dolicholphosphate and ubiquinone. FIG. 6 shows the 
sites of action of BZA, TMD, and HMGCoA reductase inhibitors, such as mevinolin 
(lovastatin). 

DETAILED DESCRIPTION OF THE INVENTION 

Introduction. 

The invention provides for the use of HMGCoA reductase inhibitors in the treatment 
and prevention of diseases in whose pathogenesis involves angiogenesis. The mechanism by 
which HMGCoA reductase inhibitors regulatecholesterol metabolism is well understood. 
HMGCoA reductase inhibitors also have a less well-known effect on gene expression. But 
HMGCoA reductase inhibitors also have an effect independent of cholesterol lowering. The 
non-cholesterol lowering effects of HMGCoA reductase inhibitors are due to the interference 
of agents with the function of small GTP-binding proteins such as Rho and Ras, which play a 
role in gene expression. The interference of HMGCoA reductase inhibitors with the function of 
the small GTP-binding proteins effects the expression of genes coding for growth factor 
receptors and cytokines. The expression of these genes affect the inflammatory processes, cell 
migration, and cell cycle regulation involved in atherogenesis and tumor development. 
Furthermore, these drugs interfere with angiogenesis which is dependent on Rho. 
Angiogenesis plays an important role in atherogenesis and tumor development. Since these 
effects involve interference in the farnesylation of Ras or geranylgeranylationof proteins such 
as Rho or Rho family members, the effects are independent of cholesterol lowering. 



WO 00/67737 PCT/US00/12309 
This invention thus provides a new use for HMGCoA reductase inhibitors. HMGCoA 
reductase inhibitors can still be used for the treatment of hypercholesterolemia and secondary 
prevention in coronary artery disease. Unexpectedly, HMGCoA reductase inhibitors can now 
be administered to achieve results independent of cholesterol lowering. Based upon this 
invention, HMGCoA reductase inhibitors can not only achieve plaque reduction, decreased 
plaque growth, increased plaque stability, and the decreased the likelihood of plaque rupture 
due to effects on cholesterol lowering, but also by anti-angiogenic effects. The new use of 
HMGCoA reductase inhibitors is for the treatment of patients with rheumatoid arthritis, 
diabetes, psoriasis and other inflammatory diseases and both primary and metastatic cancer in 
which angiogenesis is necessary for the development of the disease. Hence, HMGCoA 
reductase inhibitors can also prophylactically prevent the development of tumors and the 
complications of diabetes and the vascularization or atherosclerotic lesions. 

The advantages of this invention over existing technological developments are that the 
prevention of new vessel formation is considered a novel, benign, and curative approach to the 
treatment of disease. Although anti-TNFa antibody and other antiproliferative agents have 
been tested for treatment of rheumatoid arthritis, the HMGCoA reductase inhibitors have far 
fewer side effects and could be more efficacious than these agents. Furthermore, the method of 
the invention could, in some cases, replace the chemotherapeutic agents currently used to 
relieve patients of the devastating side effects of many of these chemotherapeutic agents. Also, 
the use of antibodies is expensive and often can lead to a reverse immunologic response, thus 
limiting their use. In the case of diabetic retinopathy, the method of the invention could 
prevent the development of complications long before the need for laser therapy became 
necessary. The invention provides a rationale for testing the therapeutic or prophylactic 
dosage. 

HMGCoA reductase inhibitors. 

HMGCoA reductase inhibitors exert effects independent of cholesterol lowering. 
Abnormalities of lipid metabolism are known to importantly affect cardiovascular disease 
including atherosclerosis and heart failure. 3-Hydroxy-3-methylglutaryl coenzyme A 
(HMGCoA) reductase inhibitors, commonly referred to as "statins", are a group of 
cholesterol-lowering drugs, which decrease LDL cholesterol by inhibiting the rate-limiting 
enzyme in cholesterol biosynthesis (Goldstein & Brown. 343 Nature 425-30 (1990), Grundy, 
97 Circulation 1436-9 (1998)). Statins are widely used in the treatment and prevention of 



WO 00/67737 PCT/USOO/12309 
coronary artery and other forms of vascular disease, including hypercholesterolemia and 
atherosclerotic vascular disease. Many thousand of patients are currently being treated with 
these agents. The mechanism of action of these agents in the treatment and prevention of 
coronary artery disease was thought to be due to effects on cholesterol lowering. As a result of 
clinical trials, which have demonstrated that HMGCoA reductase inhibitors safely reduce 
cardiovascular morbidity and mortality, HMGCoA reductase inhibitors are now in wide use 
for the treatment of hypercholesterolemia and atherosclerotic cardiovascular disease 
(Scandinavian Simvastatin Survival Study Group, 344 Lancet 1383 (1994); Sacks et ai, 335 
N. Engl. J. Med. 1001-9 (1996), Shepherd et ai, 333 N. Engl. J. Med. 1301-7 (1995)). 
Recently, attention has been focused on non-cholesterol lowering effects of these agents (West 
of Scotland Coronary Prevention Study Group, 97 Circulation 1440-5 (1998); Sacks et ai, 97 
Circulation 1446-52 (1998)). 

Analysis of clinical data has demonstrated that cholesterol lowering alone does not 
account for the therapeutic effects of HMGCoA reductase inhibitors (Sacks et ai, 97 
Circulation 1446-52 (1998), Vaughan et ai, 348 Lancet 1079-82 (1996)). Inhibition of the 
cholesterol metabolic pathway by HMGCoA reductase inhibitors interferes with the synthesis 
of famesylpyrophosphate, which is not only a precursor to cholesterol, but is also required for 
four other pathways (see, FIG. 6). These pathways include the biosynthesis of ubiquinone, a 
component of the mitochondrial oxidative chain; and dolichol phosphate, which is required for 
the glycosylation of cell surface receptors. Two of these pathways include the 
famesylpyrophosphate (FPP) dependent, postradiational lipidation of small GTP-binding 
proteins, such as Ras and the GPP-dependent (FIG. 6). Interference with the processes that 
depend on any one of these four pathways, could be responsible for cholesterol independent 
effects of HMGCoA reductase inhibitors on atherogenesis and cardiovascular disease (Brown 
& Goldstein, 21 J. Lipid Res. 505-17 (1980)). Thus, HMGCoA reductase inhibitors might 
exert effects on the progression of coronary artery disease not only by cholesterol lowering, 
but also by cholesterol independent mechanisms involving interference with any of these 
pathways. 

Posttranslational lipidation of small GTP binding proteins like Ras and Rho is required 
for their membrane localization and function. The farnesylation of Ras may be a regulatable 
process. Induction of the cholesterol metabolic pathway was shown to increase the level of 



WO 00/67737 PCT/US00/12309 

farnesylation and membrane localization of Ras and the stimulation of Ras dependent gene 
expression (Gadbut et al, 16 EMBO J. 7250-60 (1997)). 

Furthermore, HMGCoA reductase inhibitors interfere with the farnesylation of 
proteins such as Ras and the geranylgeranylation of proteins such as Rho. Several ligands, 
receptors, and enzymes involved in cell signaling are either positively or negatively controlled 
by Rho. Ras dependent TGFp signaling can be mediated via an effect on the farnesylation of 
Ras. HMGCoA reductase inhibitors have been shown to increase the Rho-dependent 
expression of ecNOS, production of NO and inhibition of vascular smooth muscle cell 
proliferation (Laufs et a/., 97 Circ. 1 129 (1998); Guijarro et al, 83 Circ.Res. 490 (1998); 
Laufs & Liao, 273 J. Biol. Chem. 24266 (1998)), effects which might reverse endothelial cell 
dysfunction and interfere with atherogenesis. 

We have shown that TGFp signaling and the expression of TGFp, and the type II 
TGFp receptor are under the negative control of a Rho GTPase. Induction of the cholesterol 
metabolic pathway decreased the expression of TGFP! and the type II TGFP receptor, but 
inhibition of the cholesterol metabolic pathway by HMGCoA reductase inhibitors or the 
inhibition of the geranylgeranylation of Rho by the geranylgeranyltransferase inhibitor GGTI 
induced the expression of TGFP, and the type II TGFp receptor and increased TGF-psignaling 
(Park & Galper, 96 Proc. Natl. Acad. Sci. USA 1 1525-30 (1999)). 

Moreover, in the presence of TNFa, the HMGCoA reductase inhibitor lovastatin 
synergistically decreased the angiogenic response to the intradermal injection of Ras 
transformed NIH-3T3 cells (Feleszko et al, 81 Int. J. Cancer 560 (1999)). 

Angiogenesis affects the pathogenesis of atherosclerosis. Angiogenesis, the formation 
of new blood vessels from a preexisting vasculature, is physiologically involved in the female 
reproductive system, in wound healing and in the normal hair cycle (Folkman & Klagsbrun, 
235 Science 442-7 (1987)). Angiogenesis also affects the pathogenesis and development of 
tumors, psoriasis, rheumatoid arthritis and diabetic retinopathy and atherosclerosis (Folkman, 
1 Nature Medicine 27-31 (1995)). 

Clinical studies have shown an increase in neo-vascularization in atherosclerotic 
plaques which rupture or develop mural hemorrhage (Paterson, 25 Arch. Pathol. 474-487 
(1938)). In normal blood vessels, the microvascular network of vasa vasorum is confined to 
the adventitia and outer media. In vessels with atherosclerotic plaques, these adventitial vessels 
increase in number and extend into the intima of the atherosclerotic lesions (Barger et al, 310 



PCT/USOO/12309 

WO 00/67737 

N. Engl. J. Med. 175-177 (1984)). Casting studies have shown that these intimal vessels are 
branches of the native adventitial vasa vasorum (Zhang et al, 143 Am. J. Pathol. 164-72 
(1993)). Plaque vessels are often found in areas containing large numbers of macrophages, 
T-cells and mast cells, which can activate angiogenesis (Kaartinen et al, 123 Atherosclerosis 
123-31 (1996)). Their close proximity to inflammatory infiltrates means that these vessels may 
recruit inflammatory cells into the plaques. Furthermore, these vessels may be required for the 
supply of oxygen and nutrients necessary for the growth of the plaque beyond a certain stage. 
A recent study using anti-angiogenic agents, endostatin, and TNP-470 (which have no effect 
on cholesterol levels) shown that they inhibited plaque growth during treatment of cholesterol 
fed Apo-E -/- mice by 85% and 70%, respectively. Hence, angiogenesis can promote plaque 
development and inhibition of angiogenesis can suppress plaque growth (Moulton et al, 99 
Circulation 1726-32 (1999)).' 

Others have suggested that HMGCoA reductase inhibitors have anti-angiogenic effects. 
Treatment of cholesterol-fed monkeys with pravastatin (an HMGCoA reductase inhibitor) 
resulted in a decrease in both cellularity and neovascularization of atherosclerotic plaques 
(Williams et al, 31 J. Am. Coll. Cardiol. 684-91 (1998)). 

Furthermore, treatment of patients with diabetic retinopathy with pravastatin resulted in 
regression in the vascular lesions (Gordon et al, 1 12 Am. J. Ophthalmol. 385-91 (1991)). 

Vascular endothelial growth factor (VEGF) stimulates angiogenesis during vascular 
development and in response to pathological stimuli. VEGF affects not only in the 
development of the vascular system, but also appears to be involved in the pathogenesis of 
diseases in which angiogenesis has a role. Patients with proliferative diabetic retinopathy 
contain significantly higher levels of VEGF in their vitreous than those of control patients. 
These levels exceeded the concentration required for stimulation of proliferation of vascular 
endothelial cells in vitro (Adamis et al, 1 18 Am. J. Ophthalmol. 445-50 (1994)). 

Rheumatoid arthritis is characterized by the proliferation of synovial lining cells, 
infiltration by inflammatory cells and new blood vessel formation. VEGF is synthesized and 
released by a large number of the macrophages, fibroblasts and vascular smooth cells in the 
effected joints (Nagashima et al, 22 J. Rheumatol. 1624-30 (1995)). 

Tumor cells also express high levels of VEGF. Clinical trials are in progress to 
establish the efficacy of anti-angiogenic agents in the treatment of tumor cells. In 
cardiovascular disease, VEGF has been implicated in both pathologic and therapeutic effects. 



WO 00/67737 PCT7US00/12309 
Thus, VEGF appears to be up-regulated in artosclerotic arteries and has been implicated in the 
development of collateral circulation in ischemic myocardium. 

Based on the observations in recent clinical studies, VEGF protein and cDNA 
constructs expressing VEGF have been administered to patients and shown to inhibit intimal 
thickening following balloon angioplasty and improve blood flow in ischemic limbs. These 
effects were believed to be mediated through stimulation of endothelial cell growth and 
angiogenesis respectively (Abedi & Zachary, 272 J. Biol. Chem. 15442-51 (1997)). However, 
VEGF may also affect the neo-vascularization of atherosclerotic plaques (O'Brien et aL, 145 . 
Am. J. Pathol. 883-94 (1994)) and contribute to an increase in atherosclerosis. 

VEGF is a dimeric protein with a molecular mass of 45-46 kDa, composed of two 
23kDa subunits joined by sulfhydryl bridges. Five isoforms of VEGF, which arise as a result 
of alternate splicing, have been demonstrated. These isoforms differ in molecular weight and 
in their ability to bind to cell surface heparan-sulfate proteoglycans and VEGF receptors. 
VEGF increases vascular permeability, stimulate the expression of proteases required for the 
breakdown of the basement membranes of blood vessels in the early stages of angiogenesis 
and initiate cell proliferation and migration (Folkman & Klagsbrun, 235 Science 442-7 
(1987)). VEGF also affects the formation of focal adhesions required for cellular proliferation 
and migration. This effect is mediated via VEGF stimulation of focal adhesion kinase (FAK), 
a non-receptor kinase, which acts as a scaffold for the assembly of proteins required for the 
organization of the cytoskeleton and the formation of focal adhesions (Abedi & Zachary, 272 
J.Biol. Chem. 15442-51 (1997)). 

VEGF receptors are part of a family of tyrosine kinases distinguished by the presence 
of seven immunoglobulin-like loops in their extracellular domain and a split tyrosine-kinase 
domain in their intracellular portion (Folkman & Klagsbrun, 235 Science 442-7 (1987)). Two 
of these receptors, designated VEGF-R1 (Flt-1) and VEGF-R2 (Flk-l/KDR), are 
autophosphorylated in response to VEGF binding. The VEGF head to tail homodimer binds to 
two receptor molecules resulting in receptor dimerization. Ligand binding is followed by 
autophosphorylation of the receptor which is required for signaling. 

There are significant differences between the downstream response to VEGF 
stimulation of Flt-1 and Flk-l/KDR. Studies of porcine aortic endothelial cells over-expressing 
Flk-l/KDR demonstrated that the binding of VEGF to Flk-l/KDR results in the recruitment 
and phosphorylation of She, an SH2-phosphotyrosine-binding domain adapter. She recruits 



PCT/USOO/12309 

WO 00/67737 

Grb2, another adapter protein containing an SH3 domain which binds Sos, a guanine 
nucleotide exchange factor for Ras. The activation of Sos results in conversion of Ras to the 
activated GTP bound state. Similarly, Flk-l/KDR associates with Grb2 and Nek in a ligand 
dependent fashion (Kroll & Waltenberger, 272 J. Biol. Chem. 32521-7 (1997)). Hence, the 
activation of Flk-l/KDR stimulated the Ras dependent MAP kinase cascade with the resultant 
stimulation of cell proliferation. This conclusion is supported by the finding that PD98059, a 
specific MAP kinase inhibitor, inhibited the effect of VEGF on cell proliferation (Rousseau et 
al, 15 Oncogene 2169-77 (1997)). Both Flk-l/KDR and Flt-1 stimulate the phosphorylation 
and activation of P 38 kinase (stress activated protein kinase-2). VEGF activation of the p38 
kinase pathway stimulates the formation of stress fibers, the assembly of vinculin focal 
adhesions and cell migration and hence may have an important effect in angiogenesis 
(Rousseau et al, 15 Oncogene 2169-77 (1997)). 

In contrast to Flk-l/KDR, Flt-1 over-expressed in porcine aortic endothelial cells 
demonstrated only a minimal effect on the activation of MAP-kinase and a very weak 
phosphorylation of She. However, Flt-1 induced the phosphorylation of both phospholipase Cy 
and the p21 ras GAP P 62-pl90 complex, which stimulates the GTPase activity of p21 ras (Kroll & 
Waltenberger, 272 J. Biol. Chem. 32521-7 (1997), Seetharam et al. 10 Oncogene 135-47 
(1995)). Differences in the function of Flt-1 and Flk-l/KDR have been demonstrated in mice 
carrying the homozygous disruption in either receptor. Flk-l/KDR knockout mice, which die 
by embryonic day 8.5, lack endothelial cells and a developing hematopoietic system 
implicating Flk-l/KDR in the determination of hemato-angioblast progenitor cells and then 
endothelial cells. This is consistent with the coupling of Flk-l/KDR signaling to MAP-kinase 
stimulated cell division. In contrast, Flt-1 knockout mice, who also die at day 8.5, have 
abundant endothelial cells which migrate and proliferate, but do not assemble into tubes and 
functional vessels (Fong et al, 376 Nature 66-70 (1995)). 

Regulation of VEGF expression via hypoxia, growth factors and angiotensin II: VEGF 
expression is regulated by hypoxia, angiotensin II, thrombin, oncogenes, and cytokines 
including TGFp\ TNFa, IL-lp, and PDGF. 

Both hypoxia and oncogenes regulate VEGF expression at the level of transcription via 
the stimulation of hypoxia inducible factor (HIF-1). HIF is composed of a p subunit, which is 
stable under normoxic conditions, and an a subunit which has a half-life of <5 min. Hypoxia 
markedly inhibits the degradation of HIFa. Studies have shown that in PC12 cells hypoxia 



-10- 



WO 00/67737 PCT/US00/12309 
activated two stress activated protein kinases, p38cc and p38y while more prolonged hypoxia 
activated the Ras dependent p42/44 MAP kinase pathway (Conrad et aL, 214 J. Biol. Chem. 
23570-6 (1999)). Hypoxia has recently been shown to stimulate the p42/p44 MAP kinase 
dependent phosphorylation of HIF-la (Richard et aL, 214 J. Biol. Chem. 32631-7 (1999)). 
Flt-1 and Flk-l/KDR are regulated by hypoxia. While the Flt-1 promoter contains an HIF 
binding site, no such site has been found for the Flk-l/KDR receptor. VEGF up-regulates 
Flk-l/KDR gene expression via a feed back loop requiring VEGF binding to the Flk-l/KDR 
receptor. Since the response of Flt-1 to hypoxia is more immediate than that of Flk-l/KDR, the 
up-regulation of Flk-l/KDR is secondary to the hypoxic induction of VEGF (Shen et aL, 273 J 
Biol. Chem. 29979-85 (1998)). Inhibitor studies demonstrated that VEGF regulation of 
Flk-l/KDR expression was dependent on tyrosine phosphorylation, PKC, Src kinase and 
stimulation of the ERK pathway (Shen et aL, 273 J Biol. Chem. 29979-85 (1998)). 

Both thrombin and angiotensin II stimulate angiogenesis. Thrombin stimulates 
angiogenesis in the chick chorioallantoic membrane (CAM) assay. Incubation of HUVECs 
with thrombin increased the expression of VEGF and sensitized the cells to VEGF stimulation 
of [ 3 H] thymidine incorporation and cell growth. mRNAs coding for both Flt-1 and Flk-l/KDR 
were increased and Flk-l/KDR protein was increased by 200% (Tsopanoglou & 
Maragoudakis, 274 J. Biol. Chem. 23969-76 (1999)). Thrombin signals by the stimulation of 
the c-Jun N-terminal kinase/ stress activated protein kinase (JNK/SAPK) pathway, the p38 
kinase/stress activated protein kinase pathway and the extracellular signal-regulated kinase 
(ERK) pathway. Inhibitor studies have implicated the ERK pathway and protein kinase C in 
the regulation of Flt-1 and Flk-l/KDR by thrombin (Tsopanoglou & Maragoudakis, 274 J. 
Biol. Chem. 23969-76(1999)). 

Angiotensin II induces hypertension, and atherosclerosis in vivo (Li et aL, 143 
Atherosclerosis 315-26 (1999)). Angiotensin II also stimulates angiogenesis and markedly 
increase the expression of vascular endothelial growth factor (VEGF) in human vascular 
smooth muscle cells (Williams et aL, 25 Hypertension 913-7 (1995)) and angiogenesis, VEGF, 
Flt-1, and Flk-1 in cultured retinal microcapillary endothelial cells (Otani et aL, 82 Circ. Res. 
619-28 (1998)). Incubation of retinal microcapillary endothelial cells with angiotensin II 
increased the expression of Flk-l/KDR mRNA more than four fold and angiotensin II was 
shown to potentiate VEGF-stimulated tube formation on a three-dimensional collagen gel 
(Otani et aL, 82 Circ. Res. 619-28 (1998)). Like thrombin, angiotensin II stimulates the ERK, 



-11- 



WO 00/67737 PCT/US00/12309 
JNK, and p38 MAP kinase pathways. Angiotensin II-stimulation of KDR expression was 
shown to be dependent on tyrosine phosphorylation and activation of PKC by PMA. 

Effects of blood pressure and angiotensin II in animal models for atherosclerosis. The 
relationship of hypertension to atherosclerosis has been well established. Studies in a number 
of animal models have demonstrated that angiotensin II converting enzyme inhibitors or 
angiotensin II type 1 receptor blockers interfere with the progression of atherosclerosis in 
hypertensive animals. Recently ApoE-/- mice have been generated which demonstrate 
profound hypercholesterolemia and the propensity to develop atherosclerotic lesions with 
similarities to those found in humans. Treatment of ApoE-/- mice with a combination of an 
angiotensin II type 1 (AT,) receptor blocker losartan and an ot r adrenergic receptor blocker 
prazosin lowered blood pressure and decreased average plaque size by 43% (Makaritsis et ai, 
32 Hypertension 1044-8 (1998)). In a study in ApoE -/- mice intraperitoneal injection of 
angiotensin II (0.1 ml of 10* 7 M each day) for 30 days increased atherosclerotic lesions by 95% 
compared to placebo mice. Peritoneal macrophages from these animals demonstrated a 90% 
increase in cholesterol biosynthesis, as measured by incorporation of [ 3 H]-acetate into 
cholesterol. This effect was reversed by both the angiotensin converting enzyme (ACE) 
inhibitor fosinopril and losartan. Finally, in a macrophage cell line angiotensin II increased the 
expression of HMGCoA reductase in a dose dependent manner (Keidar et aL, 146 
Atherosclerosis.249-57 (1999)). Thus, these data suggest that angiotensin not only increases 
blood pressure, but also increases the production of cholesterol and the products of the 
cholesterol pathway. Hence, angiotensin may affect Ras-dependent and Rho-dependent gene 
expression. These data provide the basis for a new relationship between hypertension and 
agents (such as statins) that affect cholesterol metabolism which might affect the atherogenic 
effect of hypertension (as shown in EXAMPLE 6-9). 

Evidence for an effect ofVEGF and VEGF receptor expression in the 
neovascularization of atherosclerotic lesions. VEGF is involved in the neovascularization of 
atherosclerotic lesions. The expression of VEGF and VEGF receptors has been compared in 
normal and diseased coronary arteries. While the expression of both VEGF and VEGF 
receptors was undetectable in normal coronary arteries, a correlation was found between the 
severity of artherosclerotic involvement of vessels and the extent of expression of VEGF, 
Fit- 1 , and Flk-l/KDR. Hypercellular and atheromatous lesions showed positive staining for 
VEGF in endothelial cells, macrophages and smooth muscle cells. Large occlusive lesions 



-12- 



WO 00/67737 PCT/USOO/12309 

with extensive neovascularization demonstrated intense staining for VEGF, Flt-1 and 
Flk-l/KDR in macrophages, endothelial cells and microvessels (Inoue et al, 98 Circulation 
2108-16 (1998); Chen et al, 19 Arterioscler. Thromb. Vase. Biol. 131-9 (1999)). 

The effect of integrins in VEGF signaling. Angiogenesis involves the proliferation, 
migration, and differentiation of endothelial cells. Migration requires the formation of stress 
fibers and the assembly of focal adhesions. Signals from integrin receptors are integrated with 
those from VEGF signaling to organize the cytoskeleton, form focal adhesions, and stimulate 
migration (Kumar, 17 Oncogene 1365-73 (1998)). 

Integrin receptors are composed of noncovalently associated a and P chains which 
form heterodimeric receptor complexes. Both subunits contain a large extracellular domain 
and a cytoplasmic carboxy terminal of variable length. There are 17 a subunits and 8 P 
subunits which combine to form 22 different receptor complexes. The extracellular domains of 
the a and P chains form the ligand binding sites. Integrin receptors recognize the sequence 
RGD in their extracellular matrix ligands. However, integrins can recognize the differences 
between ligands with a degree of specificity: a v p 3 binds to vitronectin, binds to 
fibronectin, and a 2 p, binds to collagen and a v P 5 binds to laminin (Soldi et al, 18 EMBO J. 
882-92 (1999); Giancotti & Ruoslahti, 285 Science 1028-32 (1999)). Integrins not only bind to 
components of the extracellular matrix, but also bind to soluble ligands such as fibrinogen or 
to counter-receptors such as the intracellular adhesion molecule (I CAM) on nearby cells. 
Integrins can be cell type specific. Binding of integrins to the extracellular matrix results in the 
activation of members of the Rho family of small GTP-binding proteins leading to clustering 
of integrins, association with cytoskeletal proteins and the binding to molecules, which 
promote downstream signaling. These aggregates of extra cellular matrix proteins, integrins, 
and cytoskeletal proteins form focal adhesions where integrins link the outside matrix to the 
intracellular cytoskeletal complex. Signaling from these focal adhesions regulates cell 
adhesion, changes in cell shape and cell movement. The cytoplasmic tails of integrins are short 
and devoid of enzymatic activity. Hence integrins associate with adapter proteins which permit 
them to interact with the cytoskeleton, cytoplasmic kinases and transmembrane growth factor 
receptors. 

Focal adhesion kinase (FAK) is a non-receptor tyrosine kinase, which acts as a site for 
the assembly of other components of focal adhesions. FAK is recruited to the nascent focal 
adhesions by interacting directly with the tail of the integrin P subunit or indirectly through the 



-13- 



PCT/US00/12309 

^ho^adon of FAK V" which general a site for the binding of the SH2 domarn 
of Src and Fyn. These kinases phosphorate FAK associated proteins paxfflin, .etKtn, and 
p,3<r, which isadocking protein which recruits two adapter proteins, Ok. and Nek 
(Giancotti & Ruoslahti, 285 Science 1028-32 (1999)). 

ft e J NKp,hway.Exprcssio„ofpl30«( m a i orhindin g pro,einfor.heSH2do m aanofCrk) 
h as also been show to activate JNK. Rac (, member of the Rho fanri.y of GTPases) ,s 

Acad Sci. USA 15394-9 ,1998)). Sre also phosphoryla.es FAK Tyr« creattng a site of* 

, 0 te activation of Raa and theERK pathway (Schlaepfer e, a,. 372 Nature 786.91 OW 
Finally, FAK has been found to be associated with PI 3-kinase which activates Akt kinase 

is quired for optima, acrivation of VEGF receptors. Fur— , VEGF is a poor — 
VEGF. Significant cross-talk has been demonstrated between VEGF and integrin signaling. In 

^atedr.eptor^osine— ^,««t al ,^0», 
Moro - of.. .7 EMBO J. 6622-32 (,998), Schneiler « A 16 EMBO 3. 5600-7 0 97),. 

, phosphorylation ofFlk-l/KDR in the absence of an effect o„the expression of te rector. 

,35 43 (1999)). Vitronectin, ftbronectin, and thronrbospondin increase the expression of 
. ■ ^3,32-920,99,,^ 

io.esrin activation influences cell cycle passion, cefl «** and gene expression 
stimulated by VEGF aigna,htg in addition to their effects on cel. adhesion and ceU 



-14- 



WO 00/67737 PCT/USOO/12309 
morphology. Conversely, growth factors potentiate integrin signaling. Thus, VEGF stimulates 
the tyrosine phosphorylation of FAK and paxillin in HUVECs and the endothelial cell line 
ECV304 (Abedi & Zachary, 272 J. Biol. Chem. 15442-51 (1997)). VEGF increases the 
expression of a v and p 3 mRNA and the a v p 3 ligand osteopontin (OPN) in dermal 
microvascular endothelial cells (Senger el aL, 149 Am. J. Pathol. 293-305 (1996)). The 
importance of this cross-talk may be that VEGF stimulation alone or integrin stimulation alone 
might not be sufficient for the activation of certain pathways. Hence, cross-talk between 
integrin signaling and VEGF signaling might have an important effect in cellular proliferation 
and migration. 

Effect of Rho family members in angiogenesis and VEGF signaling. Although members 
of the Rho family of GTPases are known to have an important role in integrin signaling, their 
effect in angiogenesis and in VEGF signaling is not yet understood (Parsons, 8 Curr. Opin. 
Cell Biol. 146-52 (1996)). Three members of the Rho family of GTPases have been implicated 
in integrin signaling: RhoA, Rac, and Cdc42. The interrelationships between these three family 
members are not well understood. Microinjections of Swiss 3T3 cells have demonstrated that 
RhoA rapidly stimulated stress fiber and focal adhesion formation (Ridley & Hall, 70 Cell 
389-99 (1992)). Cdc42 stimulates actin polymerization to form filopodia, or microspikes. Like 
RhoA, Rac and Cdc42 stimulate the formation of focal complexes, which contain vinculin, 
paxillin and FAK, which differ from focal adhesions in both size and their lack of dependence 
on RhoA (Nobes & Hall, 81 Cell 53-62 (1995)). The activation of Cdc42 sequentially 
stimulates Rac and then RhoA, so that the formation of filopodia and lamellipodia is 
coordinately regulated in the control of cellular motility (Nobes & Hall, 81 Cell 53-62 (1995), 
Mackay & Hall, 273 J. Biol. Chem. 20685-8 (1998)). Stimulation by RhoA in scrape loaded 
Swiss 3T3 cells or stimulation by lysophosphatidic acid or bombesin in the presence of 
Cytochalasin D caused the phosphorylation of FAK, pl30 cas and paxillin in the absence of 
stress fiber formation demonstrating that the formation of focal adhesions and stress fibers 
were independent processes (Flinn & Ridley, 109 J. Cell Sci. 1 133-41 (1996)). Dominant 
negative mutants of Rho family members were used to demonstrate that adhesion of Rat- 1 
cells to fibronectin was independent of Rho family members. However, F-actin levels were 
decreased and cell spreading was decreased by 25-50%. Fibronectin stimulation of tyrosine 
phosphorylation of FAK was unaffected by Rac and Cdc42, but after an initial 10 minute lag 
period was decreased by a dominant negative RhoA mutant and C3 exotoxin. A dominant 



-15- 



WO 00/67737 

a****** However, in PDGF stored Rat-I cells Cdc42 had „ 0 effect on ERK2 
activation (Clark e/ a/., 142 J Cell Biol. 142-573-86 nogou 0' ■, , „ '^""'^ 
v 86 <1998 »- Similarly, PI 3-kinase and Akt 

' kinase were activated by Cdc42 onlv Rac and oul-> a v - 

IheJNK/SAPir j „ md Cdc42 have been mtplicated in the activation of 

tNK/SAPK ana p38 ^ ^ ^ .„ ^ ^ 

osmotic shock Ras activate p*,. c t j- "S«iana 
-Raa activates Rac. Studies ns, ng dominant negative Rae have demonstrated 
mat Rac is necessay for malignant transformation by Ras. 

Like Ras, Rho is activated by a large group (> 20 , of guamne nucleotide release factors 

nucleotide dissociation invito. (GDIs) w hlc b ac, as chaperons of GDP bound Rho fr„ m ,he 

llrr: T ftC,W Si! " a ' inS - ^ " -iated with 

Mo Cel. Bto,. 3,60.78 ( 1996)) , Stimu|ation ^ ^ by ^ ^ 

aid " ,h ; 'T^ of ° raf providins a — « 

Rho dependent effect on the cytoskeleton (Taylor „ m , Bio , Chm ^ 
■mmunoprecpitation studies have demonstrated tha, RhoA might directly effect grow, fact 

Angiogenesis and Disease. 

The invention provides a method for beating diseases and process, that are mediate, 
y angmgenes, The term "angiog^,. means the „ 

.-eororgan.Under„o™alphy s io, 0 gica,condi,io.,human S oranima, S u„derg„ 
angiogeneais only in very specific resrtcted ^ For 

served ,„ wound healing, feta, a„ d enrbryona, dw-w ^ ^ ^ 
luteum, endometrium and pl acent, Thetenrr ..endothelium- means athin layer of fla, 
eptthebal cells the, lines semus cavities, lymp h vessels, and blood vessels 

mamr ^ " a " 8i ° 8eneS,S " " « ^ar 

Wood vessels. Oogenesis begins with the erosion of the basemen, membrane by en^ 



-16- 



WO 00/67737 PCT7US00/12309 

blood vessels, then protrude through the basement membrane. Angiogenic stimulants induce 
the endothelial cells to migrate through the eroded basement membrane. The migrating cells 
form a "sprout" off the parent blood vessel, where the endothelial cells undergo mitosis and 
proliferate. The endothelial sprouts merge with each other to form capillary loops, creating the 
new blood vessel. 

A delicate balance between stimulatory and inhibitory factors regulates angiogenesis. 
Pro-angiogenic stimuli have a critical role in the pathogenesis of several disease states, 
including inflammatory diseases, and the growth and development of tumors {see, 
Iruela-Arispe, 78 Throm. Haemost. 672-7 (1977)). Vascular endothelial growth factor (VEGF) 
appears to be the most endothelial cell specific and unequivocal angiogenic factors {see, Leung 
et al., 246 Science 1306-9 (1989)). Basic fibroblast growth factor is another angiogenic 
cytokine. Thrombospondin I is one of a number of anti-angiogenic factors found in normal 
tissues which normally undergo physiologic remodeling and angiogenesis: including bone, 
endometrium, ovary and mammary gland. 

Persistent, unregulated angiogenesis occurs in a multiplicity of disease states, tumor 
metastasis and abnormal growth by endothelial cells. Persistent, unregulated angiogenesis also 
supports the pathological damage seen in these conditions. "Cancer" means 
angiogenesis-dependent cancers and tumors, i.e. tumors that require for their growth 
(expansion in volume and/or mass) an increase in the number and density of the blood vessels 
supplying then with blood. "Regression" refers to the reduction of tumor mass and size. 

Angiogenesis-related diseases include, but are not limited to, angiogenesis-dependent 
cancer, including, for example, solid tumors, blood born tumors such as leukemia, and tumor 
metastases; benign tumors, e.g, hemangiomas, acoustic neuromas, neurofibromas, trachomas, 
and pyogenic granulomas; rheumatoid arthritis; psoriasis; ocular angiogenic diseases, e.g., 
diabetic retinopathy, retinopathy of prematurity, macular degeneration, corneal graft rejection, 
neovascular glaucoma, retrolental fibroplasia, rubeosis; Osier- Webber Syndrome; plaque 
neovascularization; telangiectasia; hemophiliac joints; and angiofibroma. HMGCoA reductase 
inhibitors are also useful in the treatment of disease of excessive or abnormal stimulation of 
endothelial cells. These diseases include, but are not limited to, intestinal adhesions, 
atherosclerosis, scleroderma, and hypertrophic scars, z.e., keloids. HMGCoA reductase 
inhibitors are also useful in the treatment of diseases that have angiogenesis as a pathologic 



-17- 



PCT/USOO/12309 

WO 00/67737 

consequence such as cat scratch disease (Rochele minalia quintosa) and ulcers (Helobacter 
pylori). A further discussion of angiogenesis-related diseases follows: 

(a) Ischemia is associated with neovascularization and the release of VEGF. 

(b) Blindness is one of the most devastating complications of diabetes. In one form of 
diabetic retinopathy, new vessel formation and proliferation of glial cells has been 
demonstrated as part of the retinal lesion. The aqueous humor of the eyes of animals made 
hypoxic by photocoagulation contains increased levels of VEGF (Miller et al, 145 Am. J. 
Pathol. 574-84 (1994). Furthermore, in a preliminary study of diabetic patients treated with the 
HMGCoA reductase inhibitor pravastatin, significant improvement was found in the 
fundiscopic examination of all 6 patients studied compared to the untreated group (Gordon et 
al, 112 Am. J. Ophthalmol. 385-91 (1991)). 

(c) Rheumatoid arthritis is characterized by synovial membrane proliferation and 
outgrowth associated with erosion of articular cartilage and subchonral bone. The proliferating 
synovial membrane, the pannus, is vascularized by arterioles capillaries and venules. In 
collagen induced arthritis, an animal model for rheumatoid arthritis, the angiogenesis inhibitor 
AGM-1470 reversed pannus formation and neovasclarization as compared to control animals 
(Peacock et al. 175 J. Exp. Med. 1135-8 (1992)). An increase in VEGF has also been indicated 
in association with the angiogenesis of rheumatoid arthritis (Nagashima et al, 22 J. 
Rheumatol. 1624-30 (1995)). Furthermore, the pro-angiogenic cytokine TNFa has been 
implicated in the pathogenesis of rheumatoid arthritis. In clinical trials, treatment of patients 
with an antibody to TNFa deactivated the endothelial cells in the synovium, to reduce the 
expression of adhesion molecules and to decrease the levels of VEGF in association with a 
marked improvement of disease (Nagashima et al, 22 J. Rheumatol. 1624-30 (1995)).Thus, 
VEGF stimulated angiogenesis affects the pathogenesis of rheumatoid arthritis. 

(d) Psoriasis is a common inherited skin disease that is characterized by 
hyperproliferation of epidermal keratinocytes and excessive dermal angiogenesis. Medium 
conditioned by the growth of keratinocytes from patients with psoriasis induces a marked 
angiogenic response in the rabbit corneal pocket assay (see, EXAMPLE 2 below for a 
description of the assay). Furthermore, keratinocytes from patients with psoriasis expressed 
increased levels of the pro-angiogenic cytokine IL-8 and a decrease in the anti-angiogenic 
thrombospondin (Nickoloff * al, 144 Am. J. Pathol. 820-8 (1994)). 



-18- 



WO 00/67737 PCT/US00/12309 

(e) Angiogenesis has also been shown to affect atherogenesis. The ingrowth of blood 
vessels into atherosclerotic plaques may contribute to an increase in plaque size, an increase in 
infiltration by white blood cells, and the resultant destabilization and rupture of the plaque 
leading to acute myocardial infarction (Moulton et ai, 99 Circulation 1726-1732 (1999)). 

(f) Angiogenesis might affect the development of varicose veins. Several inhibitors of 
angiogenesis have been shown modulate the extent of venular dilation in an in vivo model. 

In summary, angiogenesis is important for the pathogenesis of a number of 
inflammatory and proliferative diseases. Agents which interference with angiogenesis might 
affect the treatment of these diseases. 

Birth Control Method 

HMGCoA reductase inhibitors can be used as a birth control agent, by preventing the 
uterine vascularization required for blastocyst implantation and for development of the 
placenta. Thus, the invention provides an effective birth control method when an amount of 
HMGCoA reductase inhibitor sufficient to prevent embryo implantation is administered to a 
female. In one aspect of the birth control method, HMGCoA reductase inhibitor sufficient to 
block embryo implantation is administered before or after intercourse and fertilization have 
occurred, thus providing an effective method of birth control, possible a "morning after" 
method. Inhibition of vascularization of the uterine endometrium interferes with implantation 
of the blastocyst. Similar inhibition of vascularization of the mucosa of the uterine tube 
interferes with implantation of the blastocyst, preventing occurrence of a tubal pregnancy. 
Administration methods of HMGCoA reductase inhibitors may include, but are not limited to, 
pills, injections (intravenous, subcutaneous, intramuscular), suppositories, vaginal sponges, 
vaginal tampons, and intrauterine devices. HMGCoA reductase inhibitor administration also 
interferes with normal enhanced vascularization of the placenta. 

Formulation and Dosage 

The HMGCoA reductase inhibitor of the invention can be provided in 
pharmaceutically acceptable formulations using formulation methods known to those of 
ordinary skill in the art. These formulations can be administered by standard routes. In general, 
the combinations may be administered by the topical, transdermal, intraperitoneal, intracranial, 
intracerebro ventricular, intracerebral, intravaginal, intrauterine, oral, rectal or parenteral (e.g., 
intravenous, intraspinal, subcutaneous or intramuscular) route. In addition, the HMGCoA 
reductase inhibitor may be incorporated into biodegradable polymers allowing for sustained 



-19- 



PCT/US00/12309 

WO 00/67737 

release of the compound, the polymers being implanted in the vicinity of where drug delivery 
is desired, for example, at the site of a tumor or implanted so that the HMGCoA reductase 
inhibitor is slowly released systemically. Osmotic minipumps may also be used to provide 
controlled delivery of high concentrations of HMGCoA reductase inhibitor through cannulae 
to the site of interest, such as directly into a metastatic growth or into the vascular supply to 
that tumor. The biodegradable polymers and their use are described, for example, by Brem et 

al, 74 J. Neurosurg. 441-446 (1991). 

HMGCoA reductase inhibitor formulations suitable for parenteral administration 
include aqueous and non-aqueous sterile injection solutions which may contain anti-oxidants, 
buffers, bacteriostats and solutes which render the formulation isotonic with the blood of the 
intended recipient; and aqueous and non-aqueous sterile suspensions which may include 
suspending agents and thickening agents. The formulations may be presented in unit-dose or 
multi-dose containers, for example, sealed ampules and vials, and may be stored in a 
freeze-dried (lyophilized) condition requiring only the addition of the sterile liquid carrier, for 
example, water for injections, immediately prior to use. Extemporaneous injection solutions 
and suspensions may be prepared from sterile powders, granules and tablets of the kind 
previously described. 

The HMGCoA reductase inhibitor formulations may conveniently be presented in unit 
dosage form and may be prepared by conventional pharmaceutical techniques. Such 
techniques include the step of bringing into association the active ingredient and the 
pharmaceutical carriers or excipients. In general, the formulations are prepared by uniformly 
and intimately bringing into association the active ingredient with liquid carriers or finely 
divided solid carriers or both, and then, if necessary, shaping the product. 

Therapeutically and prophylactically effective dosages of HMGCoA reductase 
inhibitor can be determined by those of skill in the art. The dosage of the HMGCoA reductase 
inhibitor depends on the disease state or condition being treated and other clinical factors such 
as weight and condition of the human or animal and the route of administration of the 
compound. For treating humans, between approximately 0.5 mg/kg to 500 mgrtcg of the 
HMGCoA reductase inhibitor can be administered. The preferred range for HMGCoA 
reductase inhibitor administration for reducing serum cholesterol is oral administration of from 
10-40 mg/day. Pravastatin is typically administered orally at a dose of 40 mg/day (West of 
Scotland Corcnaty Prevention Study Group, 97 Circulation 1440-5 (1998); Sacks et al, 97 



-20- 



WO 00/67737 PCT/US00/12309 

Circulation 1446-52 (1998)) for reducing hypercholesterolemia. The recommended starting 
dose is 10 or 20 mg once daily at bedtime. 

Therapeutic doses of simvastatin result in serum levels of 0.02-0.27 (iM (Desager & 
Horsmans, 31 Clin. Pharmacokinet 348 (1996)). In the EXAMPLES provided below, these 
concentrations had significant effects on cell division, cell migration and the formation of 
capillary-like structures by HUVECs. (see, EXAMPLE 1). Furthermore, the effects of 
HMGCoA reductase inhibitors are time dependent and hence lower doses given to patients 
over months and years are likely to have similar anti-angiogenic effects. In the corneal pocket 
and CAM assays, simvastatin suppressed bFGF and VEGF stimulated angiogenesis at 
somewhat higher concentrations than those seen in vitro. This may be because in these models, 
delivery of simvastatin is via diffusion from the pellet or mesh, which limits the effective 
concentration of the drug. However, this result could have clinical significance for humans, 
indicationg that the therapeutic or prophylactic dosage levels for the methods of this invention 
are higher than for the the levels for previous, cholesterol reducing, usages of statins. 

In general, a "standard therapeutic dosage" can be 5 to 40 mg/day of a statin, such as is 
described in the paragraphs and citations provided above. In general, a "higher than standard 
terapeutic dosagecan be a dose of as high as 120 mg/day or higher statin, such as is described 
in the paragraphs and citations provided above. In general, a "lower than standard terapeutic 
dosage" is a concentration as low as 0.5 M, as is shown in the EXAMPLES below. 

Guidance for therapeutically and prophylactically effective dosages of HMGCoA 
reductase inhibitors for anti-angiogenesis can differ from the dosage recommended for 
reducing hypercholesterolemia. Guidance for therapeutically and prophylactically effective 
dosages of HMGCoA reductase inhibitors can be determined by in vivo and in vitro assays. 
For example, HMGCoA reductase inhibitors may be quickly and easily tested in vitro for 
endothelial proliferation-inhibiting activity using a biological activity assay such as the bovine 
capillary endothelial cell proliferation assay (see, United States patents 5,885,795 and 
5,854,205, both to O'Reilly et al, both incorporated herein by reference). Other in vitro 
bioassays include the chick chorioallantoic membrane (CAM) assay and the mouse corneal 
assay. The chick chorioallantoic membrane assay is described by O'Reilly et al., 79(2) Cell 
315-328 (1994) and in EXAMPLE 2. The mouse corneal pocket assay is described in 
EXAMPLE 2. 



-21- 



WO 00/67737 



In vivo assays include the effect of *H m - • . • PCT/US00/12309 
Assays ean be performed ° f * faetors on imptated 

t*uees JLZ£T? 10 *' e,to ' - HMGC0A 
y rum el al > 362 Nature 841-844 n ocm a 
• '^^^HMGCoAreduetaseinhih-, " '° 

». . ^ y One ' <51Cancer Research6180-6184(199n 

1 ne /« v/ vo effect of HMnr n a j 1 

nectotHMGCoA reductase inhibitors can be tested in <, „ 
development not only of cancer the. „ ^ 3,50 P rese « opportunities for 

EXAMPLES we test how • 7** fc4 "*W * +m*m+. t. these 

■-hS.we.esIhows.gnatagbyVEGFandthepo.emia.ionofVFrF • ,■ 
tmegnns and angiotensin n m £ach " a "°" ° f VE °r ^almg by 

~- M of Rho Gipl te f I 08 ^ inh ' b,,i " g 
inhibitors interfere with the VEGF ^ 

a^eroseierosis and de r^Z 7 ^ " " * — - 



-22- 



WO 00/67737 PCT/US00/12309 
VEGF in the chick chorioallantoic membrane (CAM) and by bFGF in the mouse cornea is 
dependent on the posttranslational lipidation of a Rho GTPases; and (b) that the cellular 
response to VEGF, specifically VEGF stimulation of endothelial cell invasion, migration, and 
tube formation are dependent on the posttranslational lipidation of a Rho GTPase and inhibited 
by HMGCoA reductase inhibitors; 

(2) That VEGF signaling is dependent on a Rho GTPase and inhibited by HMGCoA 
reductase inhibitors at two levels: at the level of receptor activation and at the level of gene 
expression. Specifically, we provide a methodical plan for assessing therapeutic dosage by 
assaying for (a) that VEGF-stimulation of tyrosine phosphorylation of VEGF receptors, Flt-1, 
Flk-l/KDR, is regulated by a member of the Rho family of GTPases; and (b) that induction of 
VEGF, Flt-1 and Flk-l/KDR expression by angiotensin II, thrombin and hypoxia requires the 
Rho-dependent activation of a MAP kinase pathway. Hence, VEGF receptor activation and 
expression of VEGF and VEGF receptors are regulated by the posttranslational lipidation of 
Rho GTPases and inhibited by HMGCoA reductase inhibitors; 

(3) That activation of integrin signaling potentiates VEGF signaling via 
Rho-dependent pathways and HMGCoA reductase inhibitors disrupt the cross-talk between 
VEGF and integrin signaling. Specifically we provide a methodical plan for assessing 
therapeutic dosage by showing (a) that VEGF stimulation of FAK phosphorylation is 
dependent on a Rho GTPase; (b) that the effects of VEGF on endothelial cell invasion and 
migration are dependent in part on FAK; (c) that integrin-potentiation of VEGF stimulated 
phosphorylation of VEGF receptors is dependent on Rho and mediated through FAK; and (d) 
that integrin-stimulation of VEGF expression is dependent on the activation of a Rho 
dependent MAP kinase pathway. 

(4) That HMGCoA reductase inhibitors decrease the growth and size of atherosclerotic 
plaques by inhibiting the expression of VEGF and VEGF receptors and interfering with 
angiogenesis in an animal model of atherosclerosis. Using cholesterol-fed Apo-E-/- mice, we 
provide guidance for showing (a) that HMGCoA reductase inhibitors interfere with the 
expression of VEGF, Flt-1 and Flk-l/KDR in parallel with a decreased in neovascularization 
and plaque size; and (b) that angiotensin II treatment induces the expression of VEGF, Flt-1 
and Flk-l/KDR in parallel with increasing neovascularization and plaque size and these 
effects of angiotensin II are inhibited by HMGCoA reductase inhibitors. 

Other embodiments of the invention. 



-23- 



PCT/USOO/12309 

WO 00/67737 

The invention provides a method for identifying an inhibitor of angiogenesis. The 
practice of the method can be further detrmined using the guidance provided in the 
EXAMPLES below. The steps of the method include: (a) assaying the cellular response of 
endothelial cells to an angiogenic factor; <b) assaying the cellular response of endothelial cells 
to an angiogenic factor in the presence of an HMGCoA reductase inhibitor, such that the 
presence of the HMGCoA reductase inhibitor inhibits the cellular response of the endothelial 
cells; (c) assaying the cellular response of endothelial cells to an angiogenic factor in the 
presence of a test compound; and (d) comparing the cellular response of endothelial cells from 
step (a) with the cellular response of endothelial cells from step (b) and the cellular response of 
endothelial cells from step (c). An inhibition of the cellular response of endothelial cells from 
step (c) as compared with the cellular response of endothelial cells from step (a) identifies the 
test compound as an inhibitor of angiogenesis. 

The invention provides another method for identifying an inhibitor of angiogenesis. 
The practice of this method can also be further detrmined using the guidance provided in the 
EXAMPLES below. The steps of the method include: (a) assaying the activity of small GTP- 
binding protein activity from an endothelial cell; (b) assaying the activity of small GTP- 
binding protein activity from an endothelial cell that has been contacted with an HMGCoA 
reductase inhibitor, wherein the contact by the HMGCoA reductase inhibitor inhibits the 
activity of small GTP-binding protein activity in the endothelial cell; (c) assaying the activity 
of small GTP-binding protein activity from an endothelial cell that has been contacted with a 
test compound; and (d) comparing the activity of small GTP-binding protein activity from an 
endothelial cell from step (a) with the activity of small GTP-binding protein activity from an 
endothelial cell from step (b) and the activity of small GTP-binding protein activity from an 
endothelial cell from step (c). An inhibition of the activity of small GTP-binding protein 
activity from an endothelial cell from step (c) as compared with the activity of small GTP- 
binding protein activity from an endothelial cell from step (a) identifies the test compound as 

an inhibitor of angiogenesis. 

The invention provides yet another method for identifying an inhibitor of angiogenesis. 
The practice of this method can be further detrmined using the guidance provided in the 
EXAMPLES below. The steps of the method include: (a) assaying the formation of organized 
structures in vitro by endothelial cells; (b) assaying the formation of organized structures in 
vitro by endothelial cells in the presence of an HMGCoA reductase inhibitor, wherein the 



-24- 



WO 00/67737 PCT/USOO/12309 

presence of the HMGCoA reductase inhibitor inhibits the formation of organized structures in 
vitro by endothelial cells; (c) assaying the formation of organized structures in vitro by 
endothelial cells in the presence of a test compound; and (d) comparing the formation of 
organized structures in vitro by endothelial cells from step (a) with the formation of organized 
structures in vitro by endothelial cells from step (b) and the formation of organized structures 
in vitro by endothelial cells from step (c) An inhibition of the formation of organized 
structures in vitro by endothelial cells from step (c) as compared with the formation of 
organized structures in vitro by endothelial cells from step (a) identifies the test compound as 
an inhibitor of angiogenesis. 

The invention provides yet another method for identifying an inhibitor of angiogenesis. 
The practice of this method can be further detrmined using the guidance provided in the 
EXAMPLES below. The steps of the method include: (a) assaying the formation of blood 
vessels in vivo; (b) assaying the formation of blood vessels in vivo in the presence of an 
HMGCoA reductase inhibitor, wherein the presence of an HMGCoA reductase inhibitor 
inhibits the formation of blood vessels; (c) assaying the formation of blood vessels in vivo in 
the presence of a test compound; and (d) comparing the formation of blood vessels in step (a) 
with the formation of blood vessels in step (b) and the formation of blood vessels in step (c). 
An inhibition of the formation of blood vessels in step (c) as compared with the formation of 
blood vessels in step (a) identifies the test compound as an inhibitor of angiogenesis. 

The invention provides an article of manufacture (a kit), comprising packaging material 
and a primary reagent contained within said packaging material. The primary reagent is an 
HMGCoA reductase inhibitor, as described above. The packaging material includes a label 
which indicates that the primary reagent can be used for reducing angiogenesis in the tissue of 
a host (such as is also descibed above). 

The details of one or more embodiments of the invention are set forth in the 
accompanying description above. Although any methods and materials similar or equivalent to 
those described herein can be used in the practice or testing of the present invention, the 
preferred methods and materials have been described. Other features, objects, and advantages 
of the invention will be apparent from the description and from the claims. In the specification 
and the appended claims, the singular forms include plural referents unless the context clearly 
dictates otherwise. Unless defined otherwise, all technical and scientific terms used herein 



-25- 



PCT/USOO/12309 

WO 00/67737 , 

have A. same meaning as common!, unders.ood by one of ordinal skill in the art to whch 
this invention belongs. All patents and publications cited in this specification are incorporated 
by reference. 

The following EXAMPLES are presented to more fully illustrate the preferred 
embodiments of the invention. These EXAMPLES should in no way be construed as limiting 
the scope of the invention, as defined by the appended claims. 

EXAMPLE 1 

SHOwo A Ntr REGULATION OF ANGIOGENESIS 
HMGCoA reductase inhibitors inhibited angiogenesis in vitro. This EXAMPLE 
demonstrates that HMGCoA reductase inhibitors interfere with the proliferation and migratron 
of HUVECs in culture and their differentiation into blood vessel-like structures. Endothehal 
cells have a critical role in (he developmen, of new blood vessels. In response to vascular 
endothehal growth factor (VEGF) endothelial cells divide, migrate and difference mto 
Congated tubular structures which become blood vessels. To determine ore effect of regulatron 
„f,he cholesterol metabolic pathway on angiogenesis, we tested the effects ofHMGCoA 
reductase inhibitors, simvastatin and —tin on ihe formation of capil.ao.like structures 
by human umbilical vein endothelial cells (HUVECs) cultured on Matrigel. Within 16 hr of 
plating cells differentiated into a series of capillary-like structures (FIG. 1A). Matirgel® 
(Collaborative Research) is a basemen, membrane extract enriched win. laminin. Matngel® 
has the ability to promote the differentiation of endothelial cells into capillary like structures. 
When HUVECs are incubated for several hours on plates precoated with the extracellular 
matrix extract Matrigel®, they amnge themselves into polygonal shrrctures with walls 
composed of single HUVECs. In the presence of low concentrations of simvastaun (0.1 pM, 
,6 hr incubation) added a. the time of plating, the walls of these capillary-like structures 
became thickened and multicellular (FIG. IB). At higher concentrations, simvastatin drsnrpted 
, .he organization of the capillary-like stnrctu.es in a dose-dependent manner. (FIG. 1C, 1 uM; 
FIG. ID, 5 uM). Atorvastatin had a similar effect. Thus, the inhibition of the cholesterol 
metabolic pathway by HMGCoA reductase inhibitors interfered with angiogenesis ur 
HUVECs in vitro. 



-26- 



WO 00/67737 PCT/US00/12309 

HUVECs were isolated using the method of Gimbrone, 3 Prog Hemost. Thromb. 1 
(1976) and cultured in medium Ml 99 supplemented with 20% FBS, 2 mM L-glutamine, 50 
jig/ml endothelial cell growth factor, 100 |ig/ml heparin and 100 U/ml penicillin and 100 
(ig/ml streptomycin. Cells were used after the third passage. For tests of cell proliferation, 
HUVECs were plated at 1x1 0 5 cells per 60 mm dish with various concentrations of 
simvastatin. After incubation for three days at 37° C in 5% C0 2 , cells were trypsinized and 
viable cells determined by Trypan Blue exclusion. For growth of cells on Matrigel 6-well 
plates were coated with Matrigel (Collaborative Research, Inc., MA, USA), an extract of 
basement membrane secreted by the Englebreth-Holm-Swarm murine sarcoma containing a 
high concentration of laminin, and allowed to gel for one hour at 37° C. HUVECs, 5xl0 5 , were 
added to each well with various concentrations of simvastatin and incubated for 16-24 hr. The 
effect on the formation of capillary-like structures was determined by phase contrast 
microscopy. 

Effect of simvastatin on the proliferation and migration of endothelial cells. 
Angiogenesis involves proliferation, migration and differentiation of endothelial cells 
(Folkman & Klagsbrun, 235 Science 442-7 (1987)). To determine the effect of simvastatin on 
the proliferation of HUVECs, cells (1x10 s cells/60 mm dish) were plated and incubated at 
various concentrations of simvastatin. After 3 days, cells were trypsinized and counted. 
Simvastatin decreased cell number in a dose-dependent manner with a 33% decrease at 0.1 \iM 
and complete inhibition of cell growth at 2 jiM (FIG. 2A). 

The effect of simvastatin on migration of HUVECs cells was also tested using a 
cell-motility assay (FIG. 2B). For the migration tests, HUVECs cultured on 60 mm dishes 
were pretreated with various concentrations of simvastatin for 16 hours followed by a 1 hr 
incubation with 5 ^M Calcein-AM (Molecular Probes). Cells were washed, trypsinized, and 
resuspended in Ml 99 medium. The labeled cells were added to 3.0 ^m FluoroBlock inserts 
(FALCON) at a density of 50,000 cells/insert in the presence of the indicated concentrations of 
simvastatin. Medium Ml 19 supplemented with 10% FBS was used as a chemo-attractant in 
the lower wells, while medium Ml 19 alone was added to the control wells. Inserts were 
incubated for 2 hours at 37°C and fluorescence of cells which had migrated through the 
FluoroBlock inserts was measured on a CytoFluor 4000 plate reader at excitation/emission 
wavelengths of 485/530 nm. Medium Ml 99 alone is added to the upper chamber, while 1% 
serum with and without VEGF is added to the lower chamber. Migration is determined by 



-27- 



WO 00/67737 PCT/US00/123O9 
measuring the fluorescence of cells which migrate through the UV blocked membrane using a 
cytoFluor 4000 plate reader at excitation/emission wave lengths of 485/530 nm. 

Using a fluorescence assay, incubation of cells with simvastatin also demonstrated 
inhibition of cell migration in a dose-dependent manner, 54±3% (±SEM, N=3) at 5uM 
simvastatin, which was significant at 0.5 uM, pO.Ol. The more hydrophilic HMGCoA 
reductase inhibitor pravastatin, whose accessibility to non-liver cells is limited (Arai et ai, 40 
Sankyo Kenkyusho Nenpo 1-38), had no effect on the formation of capillary-like structures at 
concentrations as high as 20 pM. 

Inhibition of migration was significant at 0.5 pM, p<0.01. 

The finding that HMGCoA reductase inhibitors, simvastatin and atorvastatin, but not 
pravastatin inhibited angiogenesis in vitro might reflect that fact that although all three of these 
HMGCoA reductase inhibitors are quite similar in structure they exhibit markedly different 
hydrophobicities: simvastatin>atorvastatin>pravastatin. Although all three are transported into 
the liver, uptake into non-liver cells is dependent on relative hydrophobicity, which may 
account for differences in anti-angiogenic effect in vitro. The finding that pravastatin appeared 
to decrease the number of blood vessels in atherosclerotic lesions of cholesterol-fed monkeys 
in vivo, indicates that in vivo pravastatin or a metabolite of pravastatin may effect endothelial 
cell function. 



EXAMPLE 2 

EFFECT OF HMGCoA REDUCTASE INHIBITORS ON 
VEGF AND FGF-2-MEDIATED ANGIOGENESIS 

HMGCoA reductase inhibitors interfered with angiogenesis in vivo. To directly test the 
effects of simvastatin on angiogenesis, two models were used. The effects of simvastatin on 
VEGF stimulated angiogenesis were tested in a chorioallontic membrane (CAM) model of 
Nguyen et ai, 47 Microvasc. Res. 31-40 (1994) and FGF-2-stimulated angiogenesis in a 
corneal pocket model. This EXAMPLE shows that HMGCoA reductase inhibitors interfere 
with VEGF and FGF-2 stimulation of blood vessel formation in both models. In a chick 
chorioallantoic membrane (CAM) assay, the angiogenic response to VEGF was determined by 
computer-assisted imaging of the number of blood vessels that grew into a matrix polymer 
containing the angiogenic factor. Four independent determinations indicated that the 
HMGCoA reductase inhibitor simvastatin suppressed angiogenesis induced by VEGF in a 
dose-dependent manner. 

-28- 



WO 00/67737 PCT/USOO/12309 

Chorioallontic Membrane Model CAM assay was performed as described by Vazquez 
et ai, 274 J. Biol. Chem. 23349 (1999). Leghorn chicken embryos (Spafas) 12-14 days in ovo 
were used. Matrigel (750 (im/ml), VEGF, 250 ng/mesh alone or mixed with the indicated 
concentrations of simvastatin were loaded onto nylon mesh (pore size 250 (im; Tetko Inc.) 
incubated at 37°C for 30 min and 4°C for 2 hr to allow polymerization. For example, VEGF 
and other agents can be suspended at the desired concentrations in a mixture of aluminum 
sucrose octasulfate (sucralfate) which had been previously sterilized in boiling double-distilled 
water and Vitrogen (type I collagen) which had been diluted with water and neutralized with 
0.1 M NaOH. A 20 )il aliquot of this suspension is deposited onto a piece of mesh cut to the 
desired dimensions. The sample is allowed to gel on the top of the flat end of a 
1/8-inch-diameter Teflon rod cut into 1.2 cm length rods and mounted on a 100 mm petri dish. 
The dish is incubated at 37° C at 65-70% humidity for 20 min. 

Meshes were placed on the CAM and incubated for 24 hr. For example, the sample can 
be then transferred onto the CAM of a 8-day chick embryo. A smaller piece of mesh is placed 
on top of the collagen gel and incubation continued. 

Vessels were visualized by injecting 400 jil of fluorescein isothiocynate dextran into 
the embryo. Chicks were fixed with 3.75% formaldehyde and meshes dissected and mounted 
on slides. For example, the mesh is observed from day 3 to day 9 after implantation with a 
Zeiss stereoscope microscope. The stimulation of angiogenesis is expressed as a percentage of 
the squares in the top mesh which contains blood cells. The fluorescence intensity is analyzed 
with a computer-assisted image program (NEH Image 1.59, (Vazquez et al, 21 A J. Biol. Chem. 
23349-57(1999)). 

Quantitation of the capillary growth demonstrated a 2.7-fold increase in angiogenic 
response in the presence of 250 ng VEGF as compared to control. Treatment with VEGF plus 
increasing concentrations of simvastatin demonstrated a dose-dependent decrease in the 
angiogenic response. Effects were seen at concentrations of simvastatin as low at 0.5/lM {see, 
FIG. 3). The response at 10 \xM simvastatin was not statistically significantly different from 
control levels in chorioallantoic membranes treated with vehicle only (FIG. 3). 

Mouse Corneal Pocket Assay, The corneal pocket assay also demonstrated that 
simvastatin decreased angiogenesis in an animal model. Beads impregnated with FGF-2 
stimulated angiogenesis in the avascular mouse corneas. The corneas of mice were implanted 
with a polymer containing 10 ng of FGF-2 with and without either 5 jiM or 10 |iM 



-29- 



PCT/US00/12309 

JZZZ. ^oe of sbnvastatur, mis concenbation ofFGF-2 induced the — 

of numerous capillaries (FIG. 4A). 

For me cornea, pocke, assay, Swiss Webster mice (Char.es River Boston) were used a, 
8 to , 0 weeks of age for imitation of Hydron peUets. Comea pockets were perfomted as 
bribed by Kenyon e, a,., 37 Invest. Ophthalmol. Vis. Sci. 1625 (1996), Loughman e, a,, 24 
Aust N Z 1 Ophthalmol. 289-95 (1996). Pellets were generated by mtxtng 10 pg of 
recombinant FGF-2 plus 1 mg of sucralfate and ,0 p, of Hydron (200 mg/m, in ethanol: New 
Brunswick, New Jerseyland me indicated cone=n«rauonofsimvas«in(orC3=ao,ox,n or 

Ispens.onwassmearedontoasterilenylonmeshsouareCporesi^OOpm.Te, o,„c. and 

500 pm> that were slored at -MFC Five days after implantation corneal angiogenests was 
photographed and the presence of vessels deteimined. 

This angiogenesis was almost totally reversed in corneas in which pockels were treated 
with beads impregnated with both simvastatin and FGF-2 (FIG. 4B>. lite preseneeof 5 pM 
simvastatin completely reversed the effects of FGF-2 (FIG. 4B). The leper P in FIG. 
shown to indicate the position of the polymer. Of 29 corneas treated with FGF-2, 28 
dcuons.ra.eo an angiogenic response .o FGF-2. Of me 29 corneas seated wim FGF-2 plus 
uM simvastalin, .he angiogenic response was blocked in 26 corneas. 
' To better visualize the effects of simvastatin on comea. vascularity, corneas were 
treated as in F.G. 4A. The mouse, ,i, was also injected wim tomato lectin to visualize the 
Wood vesse,s. Sagittal sections of the mouse eye were ftxed and the vascu.ar bed vtsuahzed by 
photomicrography. FIG. 4C (top panel) demonstrates the vascular bod in a control cornea. 
m 4D demonsha.es the effects of 1 0 ng of FGF-2 following 48 hr after the iusertton of the 
polymer.Theeffectofthep* alone is shown .FIG. 40. Addition of lOOngFGF 2 uPo.he 
corneal pocket resulted in the marked prohferation of small capillaries. Thts effect of FGF-2 

sim vasta,„, The effect of .0 „g of FGF-2 plus simvas.atin (5 pM and 1 0 pM) on vasoutarUy ,s 
shown in FIG. 4E and FIG. 4F. Simvastatin decreased capillary growth back to conho. levels 
30 in a dose-dependent manner. 



-30- 



WO 00/67737 PCT/USOO/12309 

EXAMPLE 3 

FURTHER EFFECTS OF HMGCoA REDUCTASE INHIBITORS ON 
VEGF AND FGF-2-MEDIATED ANGIOGENESIS 

The HMGCoA reductase inhibitor simvastatin interfered with VEGF signaling. Assays 
were carried out to determine whether simvastatin interfered with VEGF signaling via an 
effect on the ligand-induced autophosphorylation of Fit- 1 and Flk-l/KDR. In HUVECs 
incubated in 1% serum without added growth factors, a 5 min incubation with 10 ng/ml VEGF 
resulted in a marked increase in tyrosine phosphorylation of Flk-l/KDR, measured by 
immunoprecipitation with antibody to the receptor followed by Western blot analysis with an 
anti-phosphotyrosine antibody. A 16 hr incubation of cells with increasing concentrations of 
simvastatin resulted in a marked dose dependent decrease in the tyrosine phosphorylation of 
Flk-l/KDR, while simvastatin had no effect on the expression of the receptor, as measured by 
Western blotting of aliquots of the same cell extracts with an antibody to Flk-l/KDR. These 
data are typical of 4 similar assays. In a similar assays, we demonstrated that simvastatin 
significantly decreased the VEGF-stimulated tyrosine phosphorylation of Fit- 1 while having 
no effect on total Flt-1 protein. These data are typical of three similar assays. 

These results indicated that the inhibition of the cholesterol metabolic pathway by the 
HMGCoA reductase inhibitor simvastatin interfered with the VEGF activation of Flt-1 and 
Flk-l/KDR, but not the expression of these receptors. To determine whether HMGCoA 
reductase inhibitors interfered with cross-talk between VEGF and integrin signaling, we 
determined the effect of simvastatin on VEGF-stimulated tyrosine phosphorylation of focal 
adhesion kinase (FAK). Cells were incubated overnight in 2% serum with and without 
simvastatin followed by a 5 min incubation with VEGF. Immunoprecipitation with anti-FAK 
antibody followed by Western blotting with anti-phosphotyrosine antibody demonstrated that 
VEGF-stimulated tyrosine phosphorylation of FAK and simvastatin decreased both basal and 
VEGF-stimulated phosphorylation of FAK, whereas simvastatin had no effect on the 
expression of FAK. 

Thus, HMGCoA reductase inhibitors might interfere with the cross-talk between 
VEGF and integrin signaling. In this EXAMPLE, we test how VEGF stimulation of FAK 
phosphorylation is dependent on a member of the Rho family of GTPase and that HMGCoA 
reductase inhibitors interfere with VEGF signaling by disrupting the cross-talk between VEGF 
and integrin signaling. 



-31- 



PCTAJS00/12309 



re pression of VEGF is b-« » be relied by g^th fbctors, cytokmes, 
ri"— of^.To^ew^i^o^ — 
hypoxia and th Werferes with VEGF stgnahng and 

m etabohcpau,waybyHMGCoAreduc.as ^ ^.ai. on the 

^ogenesisbyreguia.ing.heexpress.onofVEGF.w « sea 

We here demonstrate 4a. shnvastatm stgrnftcantly decreased th 

, fp^ftmilv members and is relatively specific for Kho a, d 
the ADP-ribosylation of Rho family membe ^ ^ 

prol , n :;rii. yrf - m — — — ■ 

Ws dependenceby inhibiting the geranylgeranylation of Rho. 

vascular - l SSSpvrophosphate 

This EXAMPLE shows that the effects of HMGCoA reductase inhibitors ° n 

reductase inhibitors exert then antt-angrogemc effects by tn 
30 smallGTP-bindingproteinssuehasRio. 

Since gerenyigeranytpyrophosphate and shouMbe 

effects of simvastatin on protein lipidation, we tested me 



-32- 



WO 00/67737 PCT/US00/12309 
in HUVECs grown on Matrigel®. Comparison of cells incubated with 5 simvastatin plus 
10 famesylpyrophosphate or 10 |iM geranylgeranylpyrophosphate (GGPP) with cells 
incubated with simvastatin alone demonstrated that famesylpyrophosphate had no effect on 
simvastatin interference with capillary structure formation, while 
geranylgeranylpyrophosphate completely reversed the effects of simvastatin on capillary 
structure- formation, cells treated with simvastatin plus geranylgeranylpyrophosphate and cells 
treated with simvastatin alone. 

The results of this EXAMPLE strongly support the conclusion that HMGCoA 
reductase inhibitors, at pharmacologically relevant concentrations exert a significant 
anti-angiogenic effect. This EXAMPLE provides additional guidance as to the level of 
HMGCoA reductase inhibitor in a therapeutic dose. 

The finding in this EXAMPLE that geranylgeranylpyrophosphate reversed the effects 
of simvastatin on the formation of capillary-like structures by HUVECs supports the 
conclusion that a geranylgeranylated protein has an important role in the angiogenic response 
of HUVECs platted on Matrigel®, Thus, inhibition of the geranylgeranylation reaction by 
HMGCoA reductase inhibitors is responsible for the interference of simvastatin with the 
formation of capillary-like structures. Taken together with the findings in EXAMPLE 2 that 
simvastatin interferes with angiogenesis as measured by the chorioallantoic membrane and 
corneal pocket assays in an in vivo model, these data show that simvastatin interferes with 
angiogenesis by a cholesterol-independent effect. 

The results of this EXAMPLE have important implications for the treatment of patients 
with diseases whose pathogenesis is dependent on neovascularization. The lack of significant 
side-effects of HMGCoA reductase inhibitors combined with their efficacy in the reduction of 
coronary events have made these agents important tools in the treatment and prevention of 
coronary artery disease. The addition of these newly described anti-angiogenic properties 
provide exciting new possibilities for their therapeutic use in the treatment and prevention not 
only of atherosclerosis, but also of cancer, arthritis and diabetic retinopathy. 



-33- 



WO 00/67737 

EXAMPLE 5 



PCTAJSOO/12309 



EXAMPLE 5 HP 
FORMATION OF ™S H OSPHATE 

. , «. m i™ HMGCoA reductase inhibitors limits the availably of 
metabohc pathway by HMGCoA r rf 

4 a/., 2 /u J. dioi cultured on Matngel. 

capillaiy-like stnretures. However, OGTI-288, a specific inhtbttor of 
effeotofsimvastatinonUtefomratiooofeapillarylikestructures. 
substt ate for transferase, and ^ 



-34- 



WO 00/67737 PCT/US00/12309 
geranylgeranyltransferase, respectively. Farnesylpyrophosphate had no effect on disruption of 
the formation of capillary-like structures by the HMGCoA reductase inhibitor simvastatin, 
whereas geranylgeranylpyrophosphate completely reversed the effects of simvastatin. Finally, 
we tested the effect of C botulinum C3 exotoxin, which specifically ADP ribosylates and 
inactivates the Rho family of small GTPases (Aktories, 5 Trends Microbiol. 282-8 (1997)). 
Treatment of HUVECs with C3 exotoxin at the time of plating on Matrigel mimicked the 
effect of simvastatin and disrupted the formation of capillary-like structures (see, FIG. 5). 
These data support the conclusion that simvastatin interfered with the formation of 
capillary-like structures by HUVECs grown on Matrigel by inhibiting the posttranslational 
geranylgeranylation of the Rho family of small GTP binding proteins. In this EXAMPLE, we 
test this using recombinant adenoviruses expressing dominant active and dominant negative 
mutants of the Rho family of GTPases, that the formation of capillary-like structures by 
HUVECs is dependent at least in part on a members of the Rho family of GTP binding 
proteins. 

The finding that the anti -angiogenic effects of simvastatin are reversed by GGPP, the 
substrate for geranylgeranyltransferase, and mimicked by GGTI-288, a specific inhibitor of 
geranylgeranyltransferase, show that HMGCoA reductase inhibitors interfere with 
angiogenesis via the inhibition of the geranylgeranylation reaction. 

The finding that C3 exotoxin which interferes with the function of Rho also inhibits the 
formation of capillary-like structures further shows the effect of a Rho GTPase in 
angiogenesis. These data are in agreement with a study in transformed endothelial cells from 
rat liver sinusoids, in which small GTP binding proteins were involved in the formation of 
tubular-like structures (Maru et al 9 176 J. Cell. Physiol. 223 (1998)). Although Rho has been 
implicated in processes such as cell division and cell migration which affect angiogenesis 
(Aepfelbacher et al, 17 Arterioscler. Thromb. Vase. Biol. 1623 (1997)), the direct 
involvement of Rho in angiogenesis has not previously been demonstrated. 

EXAMPLE 6 

VEGF-ST1MULATION, bFGF-STIMULATION, AND EXTRACELLULAR 
MATRIX-STIMULATION OF ANGIOGENESIS ARE DEPENDENT ON THE 
GERANYLGERANYLATION OF A Rho GTPase. 

In this EXAMPLE, we test how VEGF-stimulated angiogenesis in the CAM and bFGF 
stimulated angiogenesis in the mouse corneal pocket are dependent on specific members of the 



-35- 



PCTA3S00/12309 



m e„,betoftheRhofar„,lyofGTPases.l , e inhibit ors. 

HUGCoA reduce mhMors MM angW**" ™ « 
HMGCoA redu w e use GGT1, a specific inhibitor of 

, geranylgecaoylpyrophospta^ ftrase Md C 3 exotoxin wtach 

^pyrophosphate, the snbstrate foe fan- syl rf fc 

Anr-nbosyiatesRh.andin,^^ 

^cn^cpatbwayby— » -^^.n^OOn- 
bFCF-stirnutatedansiogeneaiabyinhibittngthagcranylgc y 

• effect of simvastatin on angiogcnesis and 
„ C3 exotoxin natmics ,he effect of snn ^ 

Se^y^pv^hospr^^fteefr^of, ^^.^ 

D^-lMr-**"""— ' bFG F inthemouse 

^^-a— ^ rrrget^^pbospbateor 

l-*^^^ ^geneaisisduoto^onofptoten. 
simvastatin suppression of WW angiogenic 

g Wl8 ^on, to g^ 
25 response. SHonid^benoreaponsetoge^W rf 

^esofHMGCoAreduCaa, tt— ^ , , te mW c.«. Monse 

corneal pocket assays are ^^^^j^^^ j simvastatin or pellets containing bFGF plna 
b FGF, pellets containing 10 ng bFGF plus 5 pM 



-36- 



WO 00/67737 PCT/USOO/12309 

10 jag of C3 exotoxin. If (as expected) a RhoA GTPase affects the anti-angiogenic effect of 
simvastatin, then C3 exotoxin reverses the angiogenic response to bFGF and mimic the effect 
of simvastatin. If the inclusion of C3 exotoxin in the pellet has no effect on bFGF stimulated 
angiogenesis at 5 fig/ml, higher concentrations are used. Alternatively, the toxin is given by 
injection into the tail vein. 

Dependence of VEGF -stimulated angiogenesis in the CAM assay on protein 
geranylgeranylation. Polymers containing 250 ng VEGF in combination with either 5 ^iM 
simvastatin or in the presence of 10 |iM GGTI or FTI are implanted. In a second set of assays, 
meshes containing VEGF plus 5 |iM simvastatin and 10 )iM geranylgeranylpyrophosphate or 
farnesylpyrophosphate are used. If direct application of these agents has no effect on 
angiogenesis, then they can be injected at the appropriate concentrations into the CAM vessels. 
As in the case of bFGF stimulated angiogenesis in the mouse corneal pocket assay. If (as 
expected) VEGF-stimulated angiogenesis is dependent on protein geranylgeranylation, then 
GGTI mimics the effect of simvastatin and geranylgeranylpyrophosphate reverses the effect of 
simvastatin on angiogenesis. To determine whether VEGF-stimulated angiogenesis is 
dependent on Rho, the effect of meshes containing VEGF and 5 jig /ml C3 exotoxin on 
angiogenesis are tested. 

Effect of expression of dominant negative Rho mutants on angiogenesis. To further test 
how members of the Rho family of GTPases are required for the stimulation of angiogenesis, 
dominant active and dominant negative mutants of RhoA, Rac-1 and Cdc42 is expressed either 
individually or in combination in CAMs and mouse corneal pockets and their effect on 
angiogenesis determined. Combinations of dominant activating and dominant negative 
mutants are not used. Retroviral vectors are used for the expression of genes in both HUVECs 
and chick cells and adenovirus vectors are used for the expression in HUVECs and in the 
corneal pocket assay. 

Recombinant retrovirus. We have generated constructs of pLNCX retroviruses 
containing myc-tagged dominant activating L63 RhoA, L61 Rac-1, and L61 Cdc42, and the 
dominant negative N19 RhoA, N17 Rac-1 and N17 Cdc42 each downstream from a 
tetracycline-controlled transactivator binding sequence. We have successfully cloned PT67 
cells which are high expressors of pLNCX virus encoding N19RhoA, L63 RhoA, 
P-galactosidase and a virus constitutively expressing the tetracycline-controlled transactivator. 



-37- 



WO 00/67737 



Initial tests with HUVECs have demonstrated a 7nv • r . PCT/USOO/12309 
ex Pr essi„ gaP . galactosfe 

Recombinant adenovirus We hav* «k,v a 

• ^ toteini0 , donof(te J"!" ^ ,0 - M ° I Bi ° I C * 

vin-s e* preKi „ g a " * Wta «* PLNCX 

pressmg a pgal and stained to detemiine%.i„fec,ed cell, r.n r , 
expressing a. mulants are ^ ^ «* * «* vituaas 

mutant Rho famih, mentbat M , ' nfeC "°" * a " d «P™*» *• 

determined by Western blot ana!v sis 0 f cell ext. , ■ iS 
negative RhoA and the daveloptnan, of apikv lube , ' " S d ° minM 

*• vimsaa toae Jil '° ^ "«"» «— •» 

HUVECs aae co-infected „,,„ pLNCX «^ °" ^ 

deva^en, of capilla7 slraclures ^ ~ *" * e 

individual virus or a cotnbination of viruses Tetracv , " 



-38- 



WO 00/67737 PCT/US00/12309 

animal model, mouse corneal pockets are treated with pellets containing 10 ng bFGF. The 
cornea is anesthetized and adenovirus expressing one of the dominant negative mutants of Rho 
mutant is dripped onto the corneas. Placing the virus directly on the cornea is necessary, since 
it is not possible to mix the virus into the pellets because of inactivation by the ethanol 
necessary for pellet preparation. To maximize viral infection the eye of the anesthetized mouse 
is maintained in a closed position for various times. The extent of infection and expression of 
the mutant Rho is determined by staining the corneas for c-myc. It is necessary to vary the 
concentration of virus and the time of incubation, to maximize the fraction of infected cells as 
measured by c-myc staining. The absence of an effect of expression of a dominant negative 
mutant does not constitute proof that Rho does not affect bFGF stimulated angiogenesis. For 
the same reason it is not possible to assess the effect of expressing more than one mutant at a 
time. 

Effect of dominant negative Rho mutants on angiogenesis in CAM assays. Since the 
efficiency of infection of corneas by the direct application of adenovirus may not yield levels 
of infection and expression of Rho mutants sufficient to effect angiogenesis, we use the CAM 
assay as an alternative animal model to test the effect of Rho in angiogenesis. The CAM has 
several advantages. (1) The pLNCX retrovirus readily infects chick cells. (2) Although the 
virus may not survive the preparation of the collagen mesh, it may be injected into the vessels 
of the chorioallantoic membrane which is more likely to permit their localization in the CAM 
vasculature. 

The CAM assays are designed as described in EXAMPLE 2 above for the corneal 
pocket assays. The chorioallantoic vessels of CAMs treated with patches containing 250 ng 
VEGF are injected with the pLNCX retrovirus expressing a dominant negative Rho mutant 
and the virus expressing the tetracycline-controlled transactivator and the effect on 
angiogenesis determined after 3 to nine days incubation. CAM assays are carried out according 
to the protocol. 

Effect of dominant activating mutants of Rho on angiogenesis (in vivo assays). This 
EXAMPLE provides in vivo assays to determine the mechanism by which Rho and HMGCoA 
reductase inhibitors regulate angiogenesis. Using dominant activating Rho mutants these tests 
address the question of whether the activation of Rho alone is sufficient to stimulate 
angiogenesis. 



-39- 



WO 00/67737 

r ntc ., tt u cr , PCT/US00/12309 
M .he effect of dominant activating mutants of ^ „„ angiogenesjs ^ ^ ^ 

.he mouse cornea, tee assays are earned ou« as described in EXAMPLE 2 above for ,he 

dominant negative mutants, except that arowth facto™ „,.„.,, i j ,, ■ u 

p « sruwm tactors are not included in the pellets and 

viruses expressing dominant activating mutants of Rh„ are used. 

Cu„„re ofendoiMal ceUs on a collagen „,„,„, „ f 

' mm, °" To « how dominant activating Rho mutants 

stimulate angiogenesis „, mr o, we do no, use the Matrige, mode,, since Matrige, stimulates 
the formation of capilla^-liKe structures in the absence of additional growth factor, However 
one in ,i, ro model for testing the , n d uct i„„ of angiogeneais is the three-dimensional collar ' 
matnx mode,. ,„ the collagen matrix model, VEOF induces the invasion of bovine aonic 
endothelial cells into the collagen matrix and the fixation of rube-like structures (Pepper e, 
189 B '° Chem - Bi ° PhyS - ReS C ™» 8 2«, (1992), Davis & Camanllo, 224 Exp Cell 
Res. 39-5, ( ,996),. The angiogenic effect of VEGF or the expression of Rho mutants can be 
quantity by photographing capi„ary,i ke structures using a phase contrast microscope 
focused a, a sing.e ,eve, beneath the surface mono,ayer. The tota, ,e„g,h of a,, cell cords which 
penetmte beneath me surface monolayer in each lie.d is determined. This is an excellent mode, 
for studying the physiologic consequences of VEGF signaling (Montesano, 22 Eur J Clin 
Invest. 504-15 ,1992); Montesano * Orci, 42 Cell 469-77 ,1985,; Peppers,.. ,„', Cell 
Biol. 743-55(1990)). 

So, in , his EXAMPLE, we use both BAECs and HUVECs. Bovine aortic endothelia, 
cells (BAECs) give a robust angiogenic response to VEGF stimulation in this assay Since the 
response of BAECs and HUVECs to VEGF is ,ui,c similar, BAECs provide a reliable mode, 
for these assays. ,„ addition, we determine that inhibition of the cholesterol metabolic pathway 
m BAECs tnh.bits aogiogenesis as demonstrated for HUVECs. BAECs a re cultured on 
Matngel and the effects of simvastatin, GGT1, and C3 exotoxin on the fonnation of 
capillary-like structures. 

EffeciofRHoin VEGF-stimulated endothelial cell invasion, m i gm ,ion. andiube 
/— we firs, detetntine w h e, h er Rho regula.es the angiogenic response in this mode, 
Since the collagen matnx mode, measures the ability of VEGF to s,imu,a,e endothelial cell 
mvaston and ,ube fonoation, we are also ,es,ing how HMGCoA reductase inhibitors inhibit 
angiogenesis by interfering with Rho dependent VEGF signaling. 



-40- 



WO 00/67737 PCT7US00/12309 
To test how VEGF signaling is dependent on Rho, BAECs are cultured on a 

three-dimensional collagen matrix until confluent and incubated for 24 hr in 5% serum cells 

are transferred to 2% serum and incubated for 24 hr with either sham, 5 |iM atorvastatin, 10 

(iM pravastatin, 5 |iM simvastatin, 10 \xM GGTI, or 5 jig/ml C3 exotoxin, VEGF is added, the 

incubation continued for three days and the formation of tubular structures determined. If (as 

expected) VEGF signaling is dependent on a Rho family member, then based on preliminary 

data, each of these treatments should interfere with invasion of the collagen matrix and tube 

formation. We further determine how the effect of simvastatin on invasion of the collagen and 

tube formation is reversed by incubation of monolayers with simvastatin plus 10 jiM 

geranylgeranylpyrophosphate. 

To determine the effect of dominant negative Rho mutants on VEGF stimulation of 

endothelial cell invasion and tube formation BAECs are infected with adenoviruses expressing 

dominant negative mutants of RhoA, Cdc42, or Rac-1 and the virus expressing the 

transactivator and incubated overnight in the presence of tetracycline. Cells are harvested and 

plated on a thick collagen gel in medium containing 5% serum at a titer sufficient to permit the 

rapid development of a confluent monolayer. Cells are transferred to 2% serum and incubated 

either in the presence or absence of tetracycline and the expression of Rho mutants determined 

by staining for c-myc. VEGF is added and incubation continued for 3 days in the presence and 

absence of tetracycline and the relative level of tube formation determined. Control plates of 

uninfected cells incubated with VEGF and tetracycline or with VEGF alone are included. Viral 

titer are varied to assure adequate levels of expression of the mutant Rho which are monitored 

by c-myc staining. 

The effect of Rho in VEGF-stimulated endothelial cell migration. To further test how 
HMGCoA reductase inhibitors interfere with angiogenesis by inhibiting VEGF signaling, we 
determine the effect of Rho in VEGF-stimulated migration in HUVECs. HMGCoA reductase 
inhibitors interfere with the migration of vascular smooth muscle cells via a process dependent 
on protein lipidation and that Rho is required for the migration of HUVECs in an in vitro 
wound repair assay (Aepfelbacher et al., 17 Arterioscler. Thromb. Vase. Biol. 1623-9 (1997), 
Corsini et al., 33 Pharmacol Res. 55-61 (1996)). To determine whether Rho affects 
VEGF-stimulated endothelial cell migration, HUVECs are incubated for 16-24 hr with either 
simvastatin, GGTI, FTI or 5 jig/ml C3 exotoxin, harvested and plated on FluroBlock inserts. 
Cells are incubated with simvastatin and/or other agents added to both the upper and lower 



-41- 



PCT/US00/12309 



„ r "TvEGF is added only .0 .he ..«er chanbe, In a second s=, of assays, e.Us are 
iri— .,a«— opb ^^^^ 

WO ,„ ' e*e„ — - - "* " " 

Cdc„ or Rac. and ,he v iro s expressing ,he ,ransac,iva,or and inenbaed over, ghr , he 
leo f ,e,racye 1 ,neXe, 1 sare tare es,edand P ,a,edona lhi e k eo, 1 a S e„ 8 e,,nnred, m 

„ 5V sel a, a .i.er sufficien, ,o penni. .be rapid deve.opnren. ofa confluent 
r a „rt the exnression of Rho mutants determined by staining 

VEGF are mediated at least in part by a Rho dependent pathway. 



-42- 



WO 00/67737 PCT/US00/12309 
Finally, we determine how dominant activating mutants of Rho stimulate migration of 
HUVECs. HUVEC monolayers are infected with recombinant adenovirus expressing dominant 
activating Rho mutants and the transactivator cells are incubated in the presence and absence 
of tetracycline taking care that the mutated Rho has been expressed and that excessive cell 
death has not occurred. Cells are labeled and migration determined as described in EXAMPLE 
1. 

EXAMPLE 7 

Rho REGULATES ANGIOGENESIS VIA THE CONTROL OF VEGF SIGNALING 

Rho regulates angiogenesis via the control of VEGF signaling at two levels: (1) 
activation of VEGF receptors, and (2) expression of genes coding for the VEGF ligand and 
VEGF receptors. In this EXAMPLE, we test how the VEGF-stimulated autophosphorylation 
of Fit- 1 and Flk-l/KDR is dependent on a member of the Rho family of GTPases and that 
HMGCoA reductase inhibitors interfere with VEGF signaling in part by inhibiting the 
VEGF-stimulated phosphorylation of Fit- 1 and Flk-l/KDR via an effect on the 
geranylgeranylation of Rho. Thus, this EXAMPLE provides guidance for testing how to 
determine therapeutic or prophylactic dosages of HMGCoA reductase inhibitors. 

Assays provided in this EXAMPLE determine how Rho regulates VEGF signaling. 
Specifically, we test how Rho regulates VEGF signaling by controlling the VEGF-stimulated 
auto-phosphorylation of the VEGF receptors Flk-l/KDR and Flt-1, which is required for 
downstream signaling. Assays provided in this EXAMPLE further test how VEGF signaling is 
also regulated by Rho at the level of gene expression. Specifically we test how pro-angiogenic 
stimuli such as thrombin, angiotensin II and hypoxia regulate the expression of VEGF and the 
VEGF receptors by a Rho dependent pathway and that inhibition of the geranylgeranylation of 
Rho family members by HMGCoA reductase inhibitors interferes with the induction of VEGF 
and VEGF receptors. 

VEGF stimulation of the tyrosine phosphorylation of Flt-1 and Flk-l/KDR is 
dependent on the geranylgeranylation of Rho. We first assay to determine whether 
VEGF-stimulated tyrosine phosphorylation of Flk-l/KDR and Flt-1 are Rho dependent and 
whether HMGCoA reductase inhibitors interfere with tyrosine phosphorylation of Flk-l/KDR 
and Flt-1 by inhibiting the geranylgeranylation of Rho. Monolayer HUVEC cultures are 
incubated for 16 hr in medium supplemented with 1% FCS in the absence of growth factors 



-43- 



PCT/US00/12309 



wo 00/67737 . . : ncu b a tion with 10 ng/ml 

^MGGTforEnorSpgOexotoxin --J-^ 

VEGF, cells ate homogenteed and me — — fe , % ^ 
antibody followed by PAGE and immunoblotung using ^ ^ 20% and basal 
cell death during the overnight premcubatton w,th smtv 

have 

, , „^ that VEGF stimulation of phosphorylat.on ts maxttnal 
alrc ady tndteated « V EG ^ ^ 

3 ™ Zand F1M is tevetaed by geranylgeranylpyrophosphatc 
P " ,a, '° n0F ™ Ithstnt.aatattnintheptesenceofeithet.OpM 
HUVECs are incubated for 16 hrw.tn VEGF . st imu1ated receptor 

phoS photy,a,ion detenn.ned. f (as expec ^ ^ q ^ 

Ration of the phosphorylatton of FH-1 and hosphale shou.d 

spttlfc a„„body,oF,M^a„dF„.lateusedtode,e m ,ne»he,het,he,e»e 

Bpr essionofF,M™KandFHlis*ered. ,„ o/n , ;/ , DR 0Dli 

" "* ' VEGF signaling, the effect of 

phosphorylation ^ Flk-l^s^o dependent, then dominant negative mutants interfere 

phosphorylation of Flt-1 and Flk 1 ta Kb mutants might mimic the effect of VEGF. 

Confluent monolayers ofHUVECs are ^ 
aaenovtros expressing dominant negative mutaots of ^ 

^activator and grown to confluence ,n teUaoyc.tne. Te»c* ^ 

3, aontioned ^^''""^^^..i--^ 
phosphotyiationofFlt-landFlWdetermtned. Ceils areata, 



expression of Rho. 

-44- 



WO 00/67737 PCT/USOO/12309 

The effect of HMGCoA reductase inhibitors on the binding of [ ns IJ VEGF and 
localization ofFlk-l/KDR and Fit- J. We determine the number of [ ,25 I]VEGF binding sites on 
the surface of intact cells from control HUVEC cultures or HUVECs incubated 16 hr with 5 
[xM simvastatin or 10 |iM GGTI. After incubation, cells are washed and incubated for 90 min 
at room temperature in M199 plus 20 mM HEPES, pH 7.4, 0.1% BSA and 100 soybean 
trypsine inihibitor and increasing concentrations of [ 125 1]VEGF in the presence and absence of 
unlabelled VEGF. The cells are washed and solubilized with 2% SDS in PBS and radioactivity 
measured. Non-specific binding is subtracted and the specific binding plotted by the method of 
Schatchard. (Soldi et al, 18 EMBO J. 882-92 (1999)). 

An alternative approach is to use FACS analysis to determine whether the receptors are 
on the cell surface and accessible to the ligand. Cells are rinsed and gently scrapped from the 
plate, gently resuspended and counted. Cells are incubated with antibodies to either Flt-1 or 
Flk-l/KDR followed by incubation with a secondary 1GG antibody conjugated to FITC. 
Resuspended cells are subject to FACS analysis. 

Rho regulates VEGF signaling by controlling the expression of VEGF, Flt-1 and 
Flk-l/KDR. In this EXAMPLE, we test how the hypoxia-stimulated expression of VEGF, 
Flt-1 and Flk-l/KDR is regulated by a member of the Rho family of GTPases and HMGCoA 
reductase inhibitors are capable of inhibiting the hypoxia-stimulated expression of VEGF and 
VEGF receptors. 

We then test how induction of VEGF, Flt-1 and Flk-l/KDR by angiotensin II, 
thrombin and hypoxia is dependent on Rho and how HMGCoA reductase inhibit the 
angiotensin II, thrombin and hypoxia induction of VEGF, Flt-1 and Flk-l/KDR by inhibiting 
the geranylgeranylation of Rho. 

Effect of Rho in the regulation of VEGF, Flt-1 and Flk-l/KDR expression by 
angiotensin II and thrombin. To test the effect of Rho in the expression of VEGF, Flt-1, and 
Flk-l/KDR in response to thrombin and angiotensin II, HUVECs are incubated for 16 hr with 
either GGTI, FTI or C3 exotoxin followed by the addition of thrombin or angiotensin II for 6 
hr and the effect on the level of expression of VEGF, Flt-1, and Flk-l/KDR determined. The 
effect of geranylgeranylpyrophosphate on simvastatin inhibition of thrombin and angiotensin 
II-stimulation of VEGF, Flt-1, and Flk-l/KDR are also determined. Since C3 exotoxin inhibits 
angiotensin II stimulated expression of VEGF, this stimulation is inhibited by GGTI, and 



-45- 



WO 00/67737 
VOEF expression. 

To tea, ,he effect of Mo i„ (ne expressio „ of VEpp Flt-1 mi Flk , /KDR ,„ 

set™ soppiemented wi,h ,% seron, aod then traiKfcrred t0 , modu|ator incubator 
(Btllups-Rothberg, and perfused for 30 mi„ witb a mixture of 5% CO, and 95% N, Under 
these conditions, the leve. of 0, in lh e chamber is undetectable. Tbe chamber, which is 
h-ifled by water in its base , is , he „ sea , ed ^ fc ^ ^ ^ ^ ^ 

and ,be effec, of hypoxia on the expression VEGF, Fit- 1, and F,k-,/KDR defined b, 

for vanous Umes to define ,he incnbarion ,, m e a, which eel, survival and ,he expression of 
VEGF, F„,, and Flk-l/KDR are „ P ,i mal . A , lema „ veIy , „ ^ ^ ^ " 
oxy g e„ conditions, 5% O, to nrore c ,ose,y reproduce hypoxic conditions which migh, exist in 
-o. To estahhsh the effec, ofRho in the response of VEGF, F.,-,, and Flk-l/KDR to 
hypoxia, cells are incubated in ,s serom f„ r 24 hr wjlh eilher Qm ^ Qr 

exotoxin, transfer ,„ tbe hypoxia chamber for 6 hr and the effect of hypoxia on tbe ieve, of 
express of VEGF, F„-l, and Flk-l/KDR defined. We a,so define whether 
pretreatmen, of cells with 8 erany, g erany,py r „ph osp h al£ and simvastatin reve.es the effec, of 
s,mvas,a,i„ on hypoxia-induced expression of VEGF, Flt-1, and Flk-l/KDR 

ne effec, of domimm „ egal , K Rho ,„„,„„„ m ;/ ^ 

^o xl a. MmeJevressiM m FM mdFlk _, /KDR ^ ms £xAMpLE ■ 

an g ,o,ensi„ „- s d m0 ,ated VEGF expression is re g „,a,ed by tbe activation of a Rho dependent 
down^ signabng pathway such as ERK, p 38 kinase, or tbe JNK pa,hway and ,ba, 
HMGCoA red uclase inhibitors interfere ^ 

par, by tnh,bmn g this VEGF expression. 

HUVECs are infected wi,h the vinases expressing the dontinan, negative ntutants and 

he vtrvas expressing the tetracycline transactor in the presence of tetracycline. CeHs are 
tatnd ,0 fresb media ,% in sena m witb and without tetracycline and incubated for 24 br 
The .eve, of expression of tbe mutam Rhos are defined as described above. Ce„s wil, then 
be -bated fo r 6hrwi,bei,bera„ g io,e»sin„oc t hro m bi„ or forohrintne hypoxia cbamber 



-46- 



WO 00/67737 PCT/USOO/12309 

and the level of expression of VEGF, Flt-1 and Flk-l/KDR determined. Dominant negative 
mutants of Rho A, Rac-1 and Cdc42 differentially inhibit ERK-2, JNK, and p38 kinase. 

Determination of Rho dependent pathways involved in angiotensin II, thrombin and 
hypoxia stimulated induction of VEGF, Flt-1 and Flk-l/KDR expression. In this EXAMPLE, 
we test how both thrombin and angiotensin II regulate the expression of VEGF, Flt-1, and 
Flk-l/KDR via a Rho dependent MAP kinase pathway and that HMGCoA reductase inhibitors 
interfere with thrombin and angiotensin II-stimulated expression of VEGF, Flt-1 and 
Flk-l/KDR and angiogenesis via the inhibition of the geranylgeranylation of Rho. 

Next, we test which of the Rho dependent pathways is involved in the angiotensin II 
and thrombin induction of VEGF, Flk-l/KDR and Flt-1 expression. Cells are incubated with 
either thrombin or angiotensin II and the time course and dose dependence of activation of 
ERK, JNK, and p38 pathways determined using Western blot analysis with commercially 
available antibodies to the phosphorylated forms of ERK-2, JNK and p38 kinase. The effect of 
angiotensin II and thrombin on kinase activity is also tested. ERK activity is tested by 
immunoprecipitating ERK and incubating the precipitated protein with [ 32 P]yATP and myelin 
binding protein followed by PAGE and autoradiography. JNK activity is tested by 
immunoprecipitating JNK and incubating the precipitated protein with commercially available 
c-jun followed by PAGE and Western blot analysis with anti-phos-jun antibody. p38 MAP 
kinase is assayed by immunoprecipitating p38 MAP kinase and incubating the precipitated 
protein with ATF 2 and [ 32 P]yATP followed by PAGE and autoradiography. 

We use a combination of dominant negative mutants and specific inhibitors of each 
pathway, to determine which pathway is involved in the induction of VEGF, Flt-1 and 
Flk-l/KDR. The ERK pathway is inhibited by PD 98059, p38 kinase pathway by SB203580 
and a dominant negative p38kinase and JNK/SAPK by a dominant negative JNK. The cDNAs 
coding for dominant negative mutants of JNK and p38 kinase is from Chen et al, 271 J. Biol. 
Chem. 31929-36 (1996). (We have already generated myc-tagged cDNAs under the control of 
the tetracycline repressor for the dominant active and dominant negative mutants of RhoA, 
Rac-1 and Cdc42.) We then generate similar constructs from the dominant negative mutants of 
p38 kinase and JNK. These constructs are used to generate recombinant adenoviruses 
expressing these genes. 

Cells are cultured for 24 hr in 1 % serum and incubated with either angiotensin II or 
thrombin for 7-15 min and with increasing concentrations of the ERK kinase inhibitor PD 



-47- 



WO 00/67737 PCT/US00/12309 
98059 or p38 kinase inhibitor SB203580 and the phosphorylation of ERK and p38 kinase 
determined as described above. Cells incubated for 24 hr in 1% serum are incubated for 6 hr 
with either thrombin or angiotensin II either under control conditions or with 30 uM PD 98059 
or 10 uM SB203580 and the expression of VEGF, Flt-1, and Flk-l/KDR determined by 
Western blot analysis. To test dominant negative mutants of Rho, cells are infected with 
viruses expressing either the dominant negative INK or the dominant negative p38 kinase and 
the virus expressing the transactivator and incubated for 24 hr in 1% serum plus tetracycline. 
Tetracycline is removed and incubation continued until expression of the mutant INK or p38 
kinase are demonstrated by staining with c-myc antibody, 9E10 or Western blot analysis of 
cell extracts with monoclonal antibody 9E10. Cells are further incubated for 6 hr with or 
without tetracycline and either thrombin or angiotensin II and the expression of VEGF, Flt-1, 
Flk-l/KDR determined. 

EXAMPLE 8 

THE POTENTIATION OF VEGF SIGNALING BY 1NTEGR1NS 
IS DEPENDENT ON A Rho GTPase. 

HMGCoA reductase inhibitors interfere with VEGF signaling by disrupting the 
cross-talk between VEGF and integrin signaling. Assays provided in this EXAMPLE test how 
the interaction between VEGF and integrin signaling is dependent on a member of the Rho 
family of GTPases and that HMGCoA reductase inhibitors interfere with VEGF signaling and 
angiogenesis by disrupting the interaction between VEGF and integrin signaling. 

VEGF-stimulated phosphorylation of FAK is dependent on Rho. The assays in this 
EXAMPLE are based on data presented in EXAMPLE 2, which demonstrate that simvastatin 
interferes with VEGF-stimulated tyrosine phosphorylation of FAK, but has no effect on the 
expression of FAK. That assay was carried out at a single concentration of simvastatin. To 
expand upon that data point and to provide guidance for determining a range of appropriate 
therapeutic or prophylactic dosages, we first determine the concentration dependence of 
simvastatin inhibition of VEGF-stimulated FAK phosphorylation. HUVECs are incubated for 
24 hr in 1% serum with increasing concentrations of simvastatin followed by a 5 min 
incubation with VEGF. Cell extracts are immunoprecipitated with anti-FAK antibody followed 
by PAGE and Western blot analysis with anti-phosphotyrosine antibody. An aliquot of each 
cell extract is subjected to Western blot analysis with anti-FAK antibody to determine the 
effect of simvastatin on FAK expression. 

-48- 



WO 00/67737 PCT/US00/12309 
HUVECs are cultured in the presence of GGTI, FTI or C3 exotoxin, followed by a 5 
min incubation with VEGF and tyrosine phosphorylation of FAK determined. To determine 
whether geranylgeranylpyrophosphate, the substrate for geranylgeranyltransferase, reverses 
simvastatin inhibition of VEGF stimulated FAK phosphorylation, cells are cultured with 
simvastatin with or without either geranylgeranylpyrophosphate or farnesylpyrophosphate 
followed by a 5 min incubation with VEGF and the phosphorylation of FAK determined. 
Then, tests are carried out to determine which members of the Rho family of GTPases is 
involved. 

Cells are infected with adenovirus expressing the dominant negative mutants of RhoA, 
Rac-1, or Cdc42, either individually or in combination and the virus expressing the 
transactivator in the presence of tetracycline. Once cells are confluent and infection is 
complete, fresh medium is added with or without tetracycline and incubation continued until 
expression of the myc-tagged Rho mutant can be detected by immunostaining. Cells are 
incubated for 5 min with VEGF and the phosphorylation of FAK determined. Since the 
phosphorylation of FAK is transient, a dominant activating Rho mutant should not have an 
effect on the steady state level of FAK phosphorylation, but may potentiate VEGF-stimulated 
FAK phosphorylation. Cells are infected as described in EXAMPLE 6 (above), with the 
dominant activating mutants of Rho family members and the phosphorylation of FAK in 
response to incubation with increasing concentrations of VEGF-determined in control cells 
(tetracycline) and cells expressing the Rho mutant the level of FAK phosphorylation 
determined. 

VEGF-stimulated invasion and tube formation by BAECs in a three-dimensional 
collagen matrix is mediated in part by FAK. We then test how VEGF signaling is dependent 
on FAK using the collagen matrix assay as a measure of VEGF signaling. If (as expected) 
VEGF stimulation of BAEC invasion and tube formation in the collagen matrix model for 
angiogenesis is regulated by a member of the Rho family of GTPases, we then use dominant 
activating and dominant negative mutants of FAK to determine whether Rho dependent VEGF 
signaling is dependent on the activation of FAK. We determine whether a dominant negative 
mutant of FAK interferes with VEGF-stimulated cell migration and tube formation and 
whether a dominant active FAK mutant reverses GGTI, simvastatin or C3 exotoxin inhibition 
of VEGF signaling. BAECs are infected with the recombinant adenovirus expressing a 
dominant negative mutant of FAK in the presence of tetracycline. Cells are harvested and 

-49- 



WO 00/67737 confluence in 5% serum plus 

.erracycltne. Medium is removed and replaced ^ 

expression of .he mutant FAK. VE ° F de(enm „ ed . lf (a s expected) a 

d „minan, negative FAK mutant inhibits VEGF ^ q ^ ^ 

slimu ,a,io»ofFAKphosphory.a.ion, which was inhibited hy 

an^onante^tinVEWin, ^^^n--.- 
,„ converse assays, BAECs m ^ „ atvesld 

the vims express the —or a d incnh M qqti i 

simvasl a,,n or C3 exotoxin are riM a ^ ^ of exp „ ss , on ott he 

treated with GGTI, simvastatin or C3 exotoxin, ^^^^^ ^ tream f rom Rho in rhis 
20 dEp e„d=nrVEG F si g na 1 in g a»dan S ,o S en K ,sandtha, F A 1 C, 

cross-talk signaling pathway. a „JFlk-l/KDR. To 

determine the effect of FAK ,n Vbta recomb i„ant adenovirus 

FurrvnGs cultured in 1 % serum are infected wu 
m o„o.ayersofHUVECscu„u d , he transactiv a.or in the presence of 

2S pressing rhe dominant negative FAK - ^ ^ of Klracydine , 

VEGF added for 5 mm. and the level of p P ^ ^ of 

wc Pennine whether rhe dominan, activated FAK 

rrr\ nr C3 exotoxin on the V tur bi» 
HMGCoA reductase inltibitorv GGT, ^ ^ ^ ^ 

30 phosphorylation of Flfland above „ incu ba,ed for 

d ominan. activating FAK mutant and . e transa , ^ ^ ? ^ 

24 hr with no additions, with simvastatin, GGTI, or C3 



-50- 



WO 00/67737 PCT/USOO/12309 
incubation with VEGF and the tyrosine phosphorylation of Fit-land Flk-l/KDR determined. 
This assay shows whether an activated FAK is capable of potentiating VEGF stimulated 
phosphorylation of Fit-land Flk-l/KDR and whether the phosphorylation of FAK is necessary 
for the phosphorylation of Fit- land Flk-l/KDR. 

The dominant inhibiting FAK mutant is the carboxy-terminal of ppl25 FAK designated 
pp41/43 FRNK , which interferes with the binding of ppl25 FAK (Richardson & Parsons, 380 
Nature 538-40 (1996)). The dominant activating mutant FAK is a transmembrane anchored 
chimeric receptor kinase consisting of the T cell CD2 receptor Iigated to ppl25 FAK with 
constitutively activated kinase activity (Chan ei ai, 269 J. Biol. Chem. 20567-74 (1994), 
Frisch etaL 134 J. Cell Biol. 793-9 (1996)). 

Effect ofRho in integrin potentiation of VEGF signaling. To determine the effect of 
cell adhesion on VEGF-stimulated phosphorylation of VEGF receptors, cells are cultured for 
24 hr in 1% serum on plastic dishes coated with gelatin under control conditions or in the 
presence of either GGTI, simvastatin, or C3 exotoxin. Cells grown under all 4 conditions will 
either be left adherent to the plate or detached by treatment with cold PBS plus 2 mM EGTA 
followed by suspension in fresh warm medium. Cells in suspension and adherent cells are 
incubated for 10 min with 10 ng/ml VEGF and the level of tyrosine phosphorylation of Fit- 1 
and Flk-l/KDR compared. The effect of cell attachment and inhibition of the 
geranylgeranylation of Rho on the relative level of expression of Flt-1 and Flk-l/KDR is 
compared to that in cells in suspension by subjecting aliquots of cell extracts to PAGE 
followed by Western blot analysis with antibodies to Flt-1 and Flk-l/KDR. These tests 
determine how the potentiation of tyrosine phosphorylation of Flt-1 and Flk-l/KDR in 
adherent cells is dependent on the geranylgeranylation of a member of the Rho family of 
GTPases and how this Rho-dependent stimulation affects the level of expression of VEGF 
receptors. 

Ligand specificity for integrin potentiation of VEGF-stimulated phosphorylation of 
Flt-1 and Flk-l/KDR, To determine how a specific matrix protein affects the Rho dependent 
potentiation of VEGF-stimulated tyrosine phosphorylation of Flt-1 and Flk-l/KDR, HUVECs 
cultured on either vitronectin (a v (3 3 specific ligand), fibronectin, (a 5 Pj specific ligand) or 
collagen (a 2 Pj specific ligand) is incubated for 24 hr in 1% serum in the presence of either 
GGTI, simvastatin or C3 exotoxin followed by a 10 min incubation in 10 ng/ml VEGF. Cells 
are harvested and the level of phosphorylation and expression of Fit-land Flk-l/KDR 



-51- 



PCT/USOO/12309 

WO 00/67737 

determined as described above. Prior studies have demonstrated that plating of cells on 
vitronectin results in the largest potentiation of VEGF-stimulated phosphorylation of 
Flk-l/KDR (Soldi et al, 18 EMBO J. 882-92 (1999)). Since poly-L-lysine does not 
significantly stimulate integrin signaling and hence does not potentiate VEGF-stimulated 
phosphorylation of Fit-land Flk-l/KDR, the use of poly-L-lysine provides a useful baseline 
for VEGF-stimulated phosphorylation of the receptors in the absence of integrin signaling. For 
assays using ploy-L-lysine, cells are incubated for 2 hr with 1 uM cycloheximide and 1 hr with 
1 uM monensin to block the synthesis of extracellular matrix prior to incubation with VEGF. 
Cells are detached in cold PBS containing 2 mM EGTA then plated on poly-L-lysine or 
fibronectin for one hour and the effect of VEGF on the phosphorylation of Fit-land 
Flk-l/KDR compared with and without pretreatment with GGTI, simvastatin or C3 exotoxin. 
This adhesion assay is used to determine the specificity of integrins for the potentiation of 
VEGF-stimulated phosphorylation of Flt-1 and Flk-l/KDR. Cells in suspension are incubated 
with increasing concentrations of antibodies against <x v , p 3 , P„ cc : , and a 5 at 4°C for 20 min and 
then plated on vitronectin for 1 hr, treated with VEGF and the effect on the tyrosine 
phosphorylation of Flt-1 and Flk-l/KDR determined. Alternatively, adherent cells cultured on 
vitronectin are preincubated with antibodies to integrin subunits, washed, and then incubated 
with VEGF and the level of tyrosine phosphorylation of Flt-1 and Flk-l/KDR determined. 

To determine the effect of members of the Rho family of GTPases in the integrin 
mediated potentiation of VEGF receptor phosphorylation, we compare the level of 
VEGF-stimulated phosphorylation of Flt-1 and Flk-l/KDR in cells cultured on matrix proteins 
and infected with adenovirus expressing dominant negative mutants of RhoA, Rac-1 and 
Cdc42 either singly or in combination. Cells expressing the dominant negative mutants plated 
on vitronectin or poly-L-lysine treated dishes are incubated with VEGF and the effect of the 
Rho mutant on tyrosine phosphorylation and expression of Flt-1 and Flk-l/KDR determined. 
Since the expression of a v p 3 is regulable, we test by Western blot analysis to determine 
whether the decreased response of Flt-1 and Flk-l/KDR phosphorylation to stimulation of the 
extracellular matrix is due to an effect of simvastatin, GGTI, or C3 exotoxin on the expression 
of a v [3 3 . The relative adhesion of cells under all of the above growth conditions is compared by 
growing cells in a 96-well plate under each condition, fixing the cells followed by staining 
with crystal violet and reading the absorbance at 540 nm in a microliter plate reader. 



-52- 



WO 00/67737 PCT/US00/12309 

The effect ofRho on the localization of VEGF receptors. It has previously been 
reported that integrin potentiation of VEGF-stimulated phosphorylation of Flk-l/KDR was not 
associated with an increase in the expression of Flk-l/KDR (Soldi et al t 18 EMBO J. 882-92 
(1999)). However, the absence of a change in total expression of Fit- 1 and Flk-l/KDR does 
not rule out the possibility that inhibition of the geranylgernylation of Rho effects integrin 
potentiation of VEGF stimulated phosphorylation of Fit- 1 and Flk-l/KDR by decreasing the 
availability of the receptors to VEGF at the cell surface. This possibility is ruled out by 
determining the effect of treatment of HUVECs cultured on extracellular matrix proteins with 
simvastatin, GGTI, or C3 exotoxin on the binding of [ ,25 I]VEGF to intact cells and the binding 
of anti-Fit- 1 and Flk-l/KDR antibody to cells measured by FACS analysis. 

Effect ofFAK in integrin potentiation of VEGF-stimulated receptor phosphorylation. 
Although a v p 3 in the presence of VEGF is associated with the Flt-1/KDR complex, no data 
have previously been presented which address the mechanism by which integrin signaling 
regulates VEGF-stimulated phosphorylation of VEGF receptors. To test how integrins 
communicate with the VEGF receptor through FAK, we determine the effect of expression of 
FAK mutants on vitronectin potentiation of VEGF-stimulated phosphorylation of Fit- 1 and 
Flk-l/KDR in HUVECs. Cells infected with recombinant adenoviruses expressing dominant 
negative FAK and the adenovirus expressing the transactivator cultured in the absence or 
absence of tetracycline to permit the expression of the mutant FAK are cultured for 24 hr on 
vitronectin or poly-L-lysine, and incubation continued for 10 min in the presence or absence of 
VEGF and the level of phosphorylation of Fit- 1 and Flk-l/KDR determined. An effect of a 
dominant negative FAK mutant on vitronectin potentiation of VEGF signaling would support 
the conclusion that FAK affects mediating this potentiation. If (as expected) the dominant 
negative mutant of FAK does not interfere with vitronectin potentiation of VEGF-stimulated 
phosphorylation of the VEGF receptors, integrins might communicate with VEGF signaling 
via a FAK independent pathway. If the dominant negative FAK mutant completely inhibits 
VEGF stimulated phosphorylation of FAK, then the dominant negative FAK interferes both 
with the integrin independent VEGF stimulation of Fit- 1 and Flk-l/KDR phosphorylation and 
with the integrin dependent potentiation of VEGF stimulated phosphorylation of Fit- 1 and 
Flk-l/KDR. 

If (as expected) assays outlined in this EXAMPLE (above) demonstrate that HMGCoA 
reductase inhibitors interfere with integrin potentiation of VEGF signaling, we then test the 



-53- 



PCT/US00/12309 

WO 00/67737 . . 

effect of FAK in mediating the cross-talk between in.egrins and VEGF by determmutg 
whether a dominant activating FAK mutan, is capable of reverstng simvasUUn, GGT1 or C3 
exotoxin inhibition ofintegrin potentiation of VEGF signaling. HUVECs expressing a 
dominant activating FAK mntant cultured on ftbronectin are treated for 24 hr under con.ro. 
conditions, or with simvastatin, GGTI or C3 exotoxm followed by a 10 min incubation wtth or 
wtthou, VEGF and the phosphorylation of Flt-1 and Flk-l/KDR detemtined. If (as expect) 
the dominant activattng FAK mutan, reverses the effect of Rho tnactivarion, men .he cross-talk 
between inregnns and VEGF is dependent on a Rho family member and that it is medtated 
through FAK. 

If (as expected) a dominant activating FAK mutant reverses the effect of GGTI, 
sim vastatin or C3 exotoxm on VEGF s lg na,in S in BAECs cultured on a collagen matnx, then 
FAK may potentiate VEGF stimulated phosphorylat.on of Flt-1 and Flk-l/KDR. Hence, a 
dominant negative FAK mutant interferes with VEGF-stimulated phosphorylation of Flt-1 and 
Flk-l/KDR and a dominant activating mutant FAK potentiates VEGF-strmulated 
phosphorylation of Fit-land Flk-l/KDR. 

Megrins potentiate VEGF signaling by inducing the expression of VEGF. In assays 
provided in this EXAMPLE, we test how integnns induce the expression of VEGF in 
HUVECs via a Rho-dependent MAP kinase pathway. HUVECs cultured on either vitronectm, 
f.bronectin or gelatin in the presence or absence of antibodies to either a, fc. P„ «, or a 5 and 
the expression of VEGF determined by Western blot analysis. If (as expected) an effect of 
interns on the expression of VEGF is determined, we then use simvastatm, GGTI, C3 
exotoxm and dominant negative mutants to determine how the mcreased expression of VEGF 
is Rho dependent and the expression of FAK mutants to determine whether it is dependent on 
FAK Finally, we use inhibitors of the ERK and P 38 kinase pathways and dominant negatrve 
mut ants of INK and p38 kinase, as described above, to determine which of these pathways 
affects the integrin stimulation of VEGF expression. 

EXAMPLE 9 

THE EFFECT OF SIMVASTATIN ON THE EXPRESSION OF VEGF- OB £EGF 
R F CFPTORS AND NEW BLOOD VESSEL FORMATION IN THE ATHEROSCLEROTIC 
' ^L™O^ESTEROL-FED OR ANGIOTENSIN II TREATED ApoE-/- MICE 

In this EXAMPLE, we assay in v/vo for the regulation of VEGF signaling by Rho and 
: inhibitors, to show that the inhibitors have a clinical effect in the 



HMGCoA reductase : 



-54- 



WO 00/67737 PCTYUSOO/12309 

therapeutic effect of HMGCoA reductase inhibitors in atherosclerosis. We test how HMGCoA 
reductase inhibitors interfere with the growth of atherosclerotic lesions by inhibiting VEGF 
signaling and angiogenesis in the ApoE-/- mouse. 

HMGCoA reductase inhibitors inhibit the expression of VEGF t Flt-1 andFlk-l/KDR 
and interfere with plaque formation and growth by inhibiting angiogenesis in cholestero-fed 
ApoE-/- mice, ApoE-/- mice are cholesterol- fed for 12 weeks prior to initiation of simvastatin 
treatment. The correlation between plaque development, plaque size, and the expression of 
VEGF, Flt-1 , and Flk-l/KDR are determined initially. The effect of simvastatin on expression 
VEGF, Flt-1, and Flk-l/KDR is tested. The new blood vessel formation in atherosclerotic 
plaques is then correlated with effects on plaque size and growth. 

Male ApoE-/- mice 6 to 8 weeks of age are fed a 0.1 5% cholesterol diet. At 20 weeks 
of age 10 animals are sacrificed to evaluate baseline extent of atherosclerosis. The remaining 
animals are divided into 2 groups and treated for 16 weeks. Group 1 continues with the same 
diet, but the feed for the animals in group 2 contains simvastatin, for a total daily dose of 30 
mg/kg. In previous studies, the anti-angiogenic agents TPN-470 and endostatin exerted the 
most significant effects when administered between weeks 20 and 36 (Moulton et al r 99 
Circulation 1726-32 (1999); Shepherd et al t 333 N. Engl. J. Med. 1301-7 (1995)). At 36 
weeks, the animals are euthanized and a sample of blood taken for determination of serum 
cholesterol. The heart and aorta are perfused with 2% paraformaldehyde for fixation and the 
heart and portions of the descending aorta embedded in parafin, sectioned, digested with 
protease XXIV, and incubated with either a rabbit polyclonal anti-von Willebrand Factor 
antibody for staining of blood vessels or rat monoclonal anti-mouse CD31 for staining of 
endothelial cells. For VEGF, a rabbit polyclonal antibody raised against the 20 amino-terminal 
residues of human VEGF are used (Santa Cruz). This antibody neutralizes VEGF activity and 
reacts specifically with native and denatured VEGF by Western blot. For the identification of 
Flt-1 and Flk-l/KDR, rabbit polyclonal antibodies are used (Santa Cruz). Primary antibodies 
are detected with a secondary antibody conjugated to horseradish peroxidase. Intimal vessels 
are detected under high power magnification and counted when both an endothelial nucleus 
and lumen can be seen and when the vessel can be seen in an adjacent section. 

To determine the extent of atherosclerosis, aortic sections are stained with hematoxylin 
and eosin. Plaque images are captured with a Hatachi HV-C203 CCD digital camera and 
measured with the Leica Q500 MC image-analysis program. Total surface area containing 



-55- 



WO 00/67737 PCT/US00/12309 
VEGF cells is quantified by using computer-aided planimetry and expressed as a percentage 
of total surface area of intima. In addition, the total surface occupied by VEGF + endothelial 
cells, the VEGF + EC area, is quantified in a similar manner and expressed as a percentage of 
the total surface area occupied by endothelial cells, as shown by von Willebrand factor 
staining. The luminal surface area occupied by von Willebrand factor staining is also estimated 
as a percent of the whole luminal surface area. The signal from Flt-1 and Flk-l/KDR may be 
more difficult to quantitate since it has been reported to be less intense than that for VEGF. 
Peritoneal macrophages from mice in each group are harvested from peritoneal fluid and of 
ApoE-/- mice and the level of VEGF determined by Western blot analysis. 

Angiotensin Il-treatment induces the expression of VEGF, Fit- 1, and Flk-l/KDR in 
parallel with increasing neovascularization and plaque size and these effects of angiotresin II 
are inhibited by HMGCoA reductase inhibitors in angiotensin Il-treated the ApoE-/- mice. 
This EXAMPLE provides guidance for testing how the anti-angiogenic effect affects limiting 
the growth and size of atherosclerotic plaques in the ApoE-/- mouse. In assays provided in this 
EXAMPLE, we test how angiotensin II stimulates the expression of VEGF, Flt-1, and 
Flk-l/KDR and the development of new blood vessels in atherosclerotic lesions induced by 
chronic administration of angiotensin II to ApoE-/- mice. Male ApoE-/- mice 6-8 weeks old 
are fed a normal diet and divided into 4 groups. Two are given a daily intraperitoneal injection 
either placebo or some ml of 10" 7 M angiotensin II daily. A third group is given the same dose 
of angiotensin II plus losartan, an angiotensin II type 1 receptor blocker. A fourth group 
receives daily intraperitoneal injections of angiotensin II plus simvastatin. Mice from the 
control group are sacrificed at the initiation of treatment to establish a baseline. 

To determine the time course of development of increased expression of VEGF, Flt-1 
and Flk-1 and new blood vessel formation animals are sacrificed at six week intervals 
following the initiation of angiotensin Il-treatment until week 30. Aortas analyzed as described 
in this EXAMPLE (above) for plaque density, microvessel formation, VEGF, Flt-1, and 
Flk-l/KDR staining. Since ApoE-/- deficient mice develop atherosclerotic lesions 
spontaneously on a normal diet, the inclusion of time points at 6 week intervals makes it likely 
that we are including the time period during which the potentiation of lesion formation by 
angiotensin II is significantly different than that of control animals. Peritoneal macrophages 
from mice in each group are harvested from peritoneal fluid and of angiotensin Il-treated and 



-56- 



WO 00/67737 PCT/USOO/12309 
control ApoE-/- mice. The level of VEGF is determined by Western blot analysis. Data are 
analyzed by ANOVA and Student's t test. 

In the in vitro and in vivo models for angiogenesis described in this application, 
HMGCoA reductase inhibitors interfere with angiogenesis in response to extracellular matrix, 
VEGF and bFGF. This effect can reasonably considered to be due to the interference of 
HMGCoA reductase inhibitors with VEGF signaling at the level of VEGF receptor activation 
and expression of VEGF and VEGF receptors. The assays outlined in this EXAMPLE provide 
guidance for testing how angiogenesis is dependent on the geranylgeranylation of proteins of 
the Rho family of GTPases. Hence, the assays of this EXAMPLE establish a new relationship 
between cholesterol metabolism and angiogenesis. 

The assays of this EXAMPLE provide guidance for the clinical relevance of HMGCoA 
reductase inhibitors, showing how HMGCoA reductase inhibitors inhibit the development of 
atherosclerotic plaques and the accompanying formation of new blood vessels. Thus, the 
assays of this EXAMPLE provide new insights into pathogenesis and treatment. 

The foregoing description has been presented only for the purposes of illustration and 
is not intended to limit the invention to the precise form disclosed, but only by the claims 
appended hereto. 



-57- 



WO 00/67737 



CLAIMS 



PCT/US00/12309 



WE CLAIM: 
1. 



2. 



6. 



A method for reducing angiogenesis in the tissue of a host, comprising: 

administering a therapeutically effective amount of an HMGCoA reductase 
inhibitor to the tissue of a host, wherein the administration reduces 
angiogenesis in the tissue. 

The method of claim 1, wherein the host has a disease selected from the group 
consisting of rheumatoid arthritis, diabetic retinopathy, psoriasis, a primary tumor, a 
metastatic tumor, or atherosclerosis. 

The method of claim 1, wherein the HMGCoA reductase inhibitor interferes with the 
vascularization of atherosclerotic plaques. 

The method of claim 1 , wherein the HMGCoA reductase inhibitor is selected from the 
group consisting of simvastatin, pravastatin, lovastatin, atorvastatin, fluvastatin, and 
cerevastatin. 

The method of claim 1, wherein the dosage is a standard therapeutic dosage. 

The method of claim 1, wherein the dosage is a higher than standard therapeutic 
dosage. 



7. The method of claim 1, wherein the dosage is a lower than standard therapeutic dosage. 

8. A method for preventing angiogenesis angiogenesis in the tissue of a host, comprising: 

administering a prophylactically effective amount of an HMGCoA reductase 
inhibitor to the tissue of a host, wherein the administration prevents 
angiogenesis in the tissue. 



-58- 



WO 00/67737 PCT/US00/12309 
The method for birth control, comprising: 

administering an effective amount of an HMGCoA inhibitor to the tissue of a 
female host, wherein the administration prevents uterine vascularization. 

A method for identifying an inhibitor of angiogenesis, comprising the steps of: 

(a) assaying the cellular response of endothelial cells to an angiogenic factor; 

(b) assaying the cellular response of endothelial cells to an angiogenic factor in the 
presence of an HMGCoA reductase inhibitor, wherein the presence of the 
HMGCoA reductase inhibitor inhibits the cellular response of the endothelial 
cells; 

(c) assaying the cellular response of endothelial cells to an angiogenic factor in the 
presence of a test compound; and 

(d) comparing the cellular response of endothelial cells from step (a) with the 
cellular response of endothelial cells from step (b) and the cellular response of 
endothelial cells from step (c), wherein an inhibition of the cellular response of 
endothelial cells from step (c) as compared with the cellular response of 
endothelial cells from step (a) identifies the test compound as an inhibitor of 
angiogenesis. 

A method for identifying an inhibitor of angiogenesis, comprising the steps of: 

(a) assaying the activity of small GTP-binding protein activity from an endothelial 
cell; 

(b) assaying the activity of small GTP-binding protein activity from an endothelial 
cell that has been contacted with an HMGCoA reductase inhibitor, wherein the 
contact by the HMGCoA reductase inhibitor inhibits the activity of small GTP- 
binding protein activity in the endothelial cell; 

(c) assaying the activity of small GTP-binding protein activity from an endothelial 
cell that has been contacted with a test compound; and 

(d) comparing the activity of small GTP-binding protein activity from an 
endothelial cell from step (a) with the activity of small GTP-binding protein 
activity from an endothelial cell from step (b) and the activity of small GTP- 
binding protein activity from an endothelial cell from step (c), wherein an 



-59- 



(c) 
(d) 



PCT/US00/12309 

,0 00/67737 binding protein activity from an 

UM*. of ft. activity of — GTP-bmdingp ^ 

compound as an inhibitorofangiogenesrs. 

, • hihitor of angiogenesis, comprising the steps of 
» — * f0m,a, '°" „ ,„ *» W enaothetia, cans ,n 

^t^-^-■--■'■-• , "-- ,,,,,,,, ■ 
comparing the fotmauon of orgaraoos 

^..(a.^thefo-on -^"^1,, 

^flestha.es.compoundaaaninhibitorofangioganea.s. 

-ingthe — of—,- £senctofanHMGCoA 

^ayiog the forn.at.on of blood 
encase inhibitor, wherein tbepresence of an HMOCOA 

M» *e formation of Hood vesseU; fatest 
^vangthe formation ofbtoodvesseis.v^mthepreaenc 

compound; and formation of 



(a) 
(b) 



(c) 
(d) 



-60- 



WO 00/67737 PCT/US00/12309 
formation of blood vessels in step (a) identifies the test compound as an 
inhibitor of angiogenesis. 



An article of manufacture, comprising packaging material and a primary reagent 
contained within said packaging material, wherein: 

(a) the primary reagent is an HMGCoA reductase inhibitor; and 

(b) the packaging material comprises a label which indicates that the primary 
reagent can be used for reducing angiogenesis in the tissue of a host. 



-61- 



WO 00/67737 



1 / 6 



PCT/US00/12309 




WO 00/67737 



2 / 6 



PCT/USOO/12309 




W0 00/67737 PCT/US00/12309 

3 / 6 




(IOJJUOQ JO o/ 0 ) 

esuodsay omeBoiBuv 



bi) 



e 8 

S » 

co Li. 

5 o 

•E > 
CO 



WO 00/67737 



5 / 6 



PCT/US00/12309 




WO 00/67737 

6 / 6 



PCT/USOO/12309 




WIS PAGE BLANK (uspto.