DbdciOgtadgc(fc

A

Av

D

S

A

0©

S

Angiogenesis in Diabetic Nephropathy

Roy Zent, MD, PhD, and Ambra Pozzi, PhD

Summary: Angiogenesis, the formation of new blood vessels from pre-existing vasculature,plays a key role in both physiologic and pathologic events, including wound healing, cancer,and diabetes. Neovascularization has been implicated in the genesis of diverse diabeticcomplications such as retinopathy, impaired wound healing, neuropathy, and, most recently,diabetic nephropathy. Diabetic nephropathy is one of the major microvascular-associatedcomplications in diabetes and is the leading cause of end-stage renal disease worldwide. Inthis review we describe the major factors involved in the pathologic glomerular microvascularalterations in response to hyperglycemia and the possible use of anti-angiogenic therapies forthe treatment of diabetic nephropathy.Semin Nephrol 27:161-171 © 2007 Elsevier Inc. All rights reserved.Keywords: Angiogenesis, diabetes, kidney, therapy, endothelial cells

cpogsddetaetfacla

P

TgafVVdV(aa

iabetic nephropathy represents a majorcause of morbidity and mortality in type1 and type 2 diabetic subjects and has

ecome the leading cause of end-stage renalisease worldwide. Currently there is no spe-ific therapy for this condition, which almostnvariably progresses to end-stage renal failure.ne of the hallmarks of diabetic nephropathy islomerular microvascular injury, which poten-ially may be a therapeutic target for this dev-stating medical condition. In this review weescribe (1) the major steps involved in angio-enesis, (2) the pathologic glomerular vascularhanges observed in diabetic nephropathy, and3) the possible use of anti-angiogenic therapyor the treatment of diabetic-induced renal vas-ular damage.

NGIOGENESIS

ngiogenesis is the formation of new bloodessels from pre-existing vasculature. This pro-

epartment of Research Medicine, Veterans Affairs Hospital, Nashville, TN,and Department of Medicine, Division of Nephrology, Medical CenterNorth, Vanderbilt University, Nashville, TN.

upported in part by National Institutes of Health/National Institute of Diabetesand Digestive and Kidney Diseases (NIDDK) grants R01-DK074359 (A.P.);RO1-DK 69921 (R.Z.), National Institutes of Health/NIDDK O’Brien Centergrant P50-DK39261-16 (A.P., R.Z.), and a Merit award from the Departmentof Veterans Affairs (R.Z.).

ddress reprint requests to Ambra Pozzi, PhD, Department of Medicine,Division of Nephrology, Vanderbilt University, Medical Center North,B3109, Nashville, TN, 37232. E-mail: ambra.pozzi@vanderbilt.edu

270-9295/07/$ - see front matter

t2007 Elsevier Inc. All rights reserved. doi:10.1016/j.semnephrol.2007.01.007

eminars in Nephrology, Vol 27, No 2, March 2007, pp 161-1

ess plays a key role in both physiologic andathologic events, including embryonic devel-pment, menstruation, wound healing, tumorrowth, and diabetes. Angiogenesis is a multi-tep process that requires at least 4 indepen-ent events by endothelial cells, includingetachment from basement membranes, prolif-ration, migration, and maturation.1 Normallyhese events are regulated tightly by both pro-ngiogenic and anti-angiogenic factors, how-ver, in pathologic events such as diabeteshere is increased synthesis of pro-angiogenicactors with concomitant down-regulation ofnti-angiogenic molecules. This leads to in-reased proliferation and migration of endothe-ial cells, resulting in the formation of immaturend leaky vessels.

roangiogenic Factors

he soluble molecules vascular endothelialrowth factor (VEGF) and angiopoietins (Ang 1nd Ang 2), are the best-characterized growthactors that play a role in angiogenesis. TheEGF family consists of at least 4 members,EGF-A, -B, -C, and -D.2 VEGF-A, the most pre-ominant, consists of at least 8 isoforms, withEGF165 the major form expressed in humans

VEGF 164 in mouse). VEGF was first describeds a vascular permeability factor because of itsbility to induce leaky vessels. It exerts its ac-

ions by binding 3 different receptors selec-

71 161

tcVFpcVcraocritts

lteTcptldssoacgugatiiVgoad

A

Tot

mgmmtcSceXttplaldpmFpcsrfpitpavtecoahbic

at

DA

Tm

162 R. Zent and A. Pozzi

ively expressed on endothelial cells, VEGF re-eptor I (ie, VEGFR1, Flt-1), VEGF receptor 2 (ie,EGFR2, Flk-1), and VEGF receptor 3 (VEGFR3).3

lt-1 is required for the recruitment of hemato-oietic precursors and the migration of mono-ytes and macrophages, whereas Flk-1 andEGFR3 are essential for the functions of vas-ular endothelial and lymphendothelial cells,espectively.3 VEGF is probably the most potentngiogenic factors and its up-regulation often isbserved in pathologic conditions, includingancer, rheumatoid arthritis, and diabetes. Up-egulated VEGF synthesis is accompanied byncreased endothelial cell migration, prolifera-ion, and formation of immature vessels charac-erized by leakiness and decreased vascular re-istance.

The angiopoietins belong to a family of ateast 4 members, with Ang 1 and Ang 2 beinghe most predominant.4 Both Ang 1 and Ang 2xert their action by binding the same receptorie-2, selectively expressed on endothelialells. Interestingly, Ang 1 and Ang 2 exert op-osite effects on endothelial cell function. Al-hough Ang 1–mediated signaling via Tie-2eads to vessel maturation, quiescence, and re-uced leakage,5,6 Ang 2 blocks the Ang 1/Tie-2ignal resulting in increased angiogenesis, ves-el instability, and consequent leakage.7,8 More-ver, although Ang 1 promotes endothelial celldhesion, spreading, and formation of focalontacts, Ang 2 enhances endothelial cell mi-ration and tubulogenesis. Ang 1 and Ang 2,nlike VEGF, are not considered complete an-iogenic factors because they cannot triggerngiogenic responses by themselves, but ratherhey positively or negatively modulate VEGF-nduced endothelial cell function.9 Interest-ngly, Ang 2 expression can be up-regulated byEGF10 and it enhances VEGF-mediated angio-enesis.4 In pathologic events, such as cancerr diabetes, increased VEGF synthesis often isccompanied by increased Ang 2 levels withecreased and/or unchanged levels of Ang 1.

nti-angiogenic Factors

o ensure that there is not an overproductionf blood vessels there are endogenous inhibi-

ors of angiogenesis that can be classified into 2 i

ajor categories: proteolytic fragments andene products.11 Among the proteolytic frag-ents, extracellular matrix–derived and plas-inogen-derived fragments have been shown

o inhibit angiogenesis by inhibiting endothelialell migration, proliferation, and tubulogenesis.ome of these fragments include angiostatin (aleavage product of circulating plasminogen),ndostatin (a cleavage product of collagenVIII), the �lNC1 domain of collagen IV,12 and

he �3NC1 domain of collagen IV.11,13 In con-rast, most of the gene product inhibitors haveleiotropic effects that are not necessarily re-

ated to the regulation of angiogenesis. For ex-mple, thrombospondin-1 and pigment epithe-ium–derived factor (PEDF), which are well-escribed inhibitors of angiogenesis in bothhysiologic and pathologic conditions, can pro-ote endothelial cell apoptosis by inducing the

as ligand.14 PEDF originally was isolated as arotein secreted by cultured pigment epithelialells of fetal human retina,15 but later washown to possess plural effects, including neu-onal cell differentiation, protection of neuronsrom various neurotoxic agents, and, most im-ortantly, angiogenesis inhibition.16 Moreover,

n retinal endothelial cells PEDF down-regulateshe levels of VEGF, thus preventing vascularermeability and angiogenesis.17 Finally, therere 2 inhibitors, soluble VEGF receptor 1 andasohibin, which are expressed only in endo-helial cells, and have selective activities againstndothelial cells themselves.11 Soluble VEGF re-eptor 1 selectively blocks VEGF signaling andnly inhibits VEGF-mediated effects, includingngiogenesis and vascular permeability. Vaso-ibin is proposed to be the first negative feed-ack regulator of angiogenesis and it works by

nteracting with specific endothelial cell intra-ellular signaling pathways.

Fig. 1 summarizes the major pro-angiogenicnd anti-angiogenic factors that contribute tohe homeostasis of blood vessel formation.

IABETIC NEPHROPATHYND VASCULAR DAMAGE

he clinical entity of diabetic nephropathy, theost common cause of end-stage renal disease

n the developed world, is characterized ini-

tacpbaemapmatctcd

tuimossapanm

ant

Fgissf

Angiogenesis 163

ially by glomerular hyperfiltration, glomerularnd tubular epithelial hypertrophy, and mi-roalbuminuria. In established diabetic ne-hropathy, the glomeruli show basement mem-rane thickening, mesangial matrix expansion,rteriolar hyalinosis, and sclerosis, and there isvidence of interstitial fibrosis.18,19 A principalechanism whereby hyperglycemia induces di-

betic nephropathy is by stimulating excessiveroduction of reactive oxygen species (ROS) inultiple cell types, including mesangial cells20

nd podocytes.21 ROS in turn can up-regulatehe expression of profibrotic molecules such asonnective tissue growth factor (CTGF) andransforming growth factor-� (TGF-�), thus in-reasing the glomerular extracellular matrix

igure 1. Schematic representation of the major pro-enesis. In quiescent blood vessels (panel A) there is a tig

n panel B) and anti-angiogenic (see list in panel C) facpecific events in new blood vessel formation, includprouting (panel B). In contrast, maturation of blood vessactors (panel C).

eposition.22,23 Hyperglycemia also increases g

he production of advanced glycation end-prod-cts (AGEs) of extracellular matrix components

n the mesangium and glomerular basementembrane, resulting in changes in permeability

f the filtration barrier.24 Finally, mechanicaltress associated with intraglomerular hyperten-ion causes podocyte damage, which is associ-ted with down-regulation of nephrin, an im-ortant protein of the slit diaphragm with anti-poptotic signaling properties. The loss ofephrin correlates with foot process efface-ent of podocytes and increased proteinuria.25

Although the mesangial cells and podocytesre proposed as the major mediators of diabeticephropathy, diabetic-induced microvascula-ure injury also plays a key role in the patho-

ti-angiogenic factors involved in the control of angio-ance between the production of pro-angiogenic (see listcreased production of pro-angiogenic factors controls

etachment from matrix, proliferation, migration, andccompanied by increased production of anti-angiogenic

and anht baltors. Ining dels is a

enesis. Similar to diabetic retinopathy, an in-

crarbwdcnib

gnbgtgIaaldngr

Pi

ApuptVbfpamnsfgastiutVa

Ffd(KCa nalyz

164 R. Zent and A. Pozzi

reased density of glomerular capillaries,esulting from glomerular neovascularization,nd an increased number of efferent arte-ioles at the glomerular vascular pole haveeen seen in biopsy specimens of patientsith type 1 diabetes26 as well as in rats andb/db mice with diabetic nephropathy.27,28 In-reased glomerular vascular density accompa-ied by increased vessel leakage also was noted

n diabetic KK mice compared with nondia-etic controls (Fig. 2).

The mechanism whereby increased angio-enesis of abnormal vessels occurs in diabeticephropathy is understood poorly. One possi-ility is that the balance between pro-angio-enic and anti-angiogenic factors, critical forhe regulation of vascular permeability and an-iogenesis, is altered in the course of diabetes.n this context, increased expression of pro-ngiogenic factors and decreased expression ofnti-angiogenic molecules within the glomeru-us of diabetic patients or rodents has beenocumented.29,30 The major angiogenic andonangiogenic factors involved in the patho-enesis of diabetic nephropathy are summa-

igure 2. Increased glomerular vascularity and permeabrom control (KK) and 25-week-old diabetic mice (KK-dietermine the degree of vascularity. Ten minutes beforemw, 60 kd) (lower panel) to visualize the integrity ofK-diabetic mice compared with the KK mice, which iD31-positive structures per glomerulus was quantified us previously described.77 Thirty glomeruli/group were a

ized in Fig. 3. t

ro-angiogenic Factorsn Diabetic Nephropathy

mong the pro-angiogenic factors, VEGF isrobably the most potent permeability factorp-regulated in diabetes.31 This growth factor isroduced predominantly by the podocyte inhe glomerulus and high glucose increasesEGF synthesis in cultured podocytes32 and tu-ular cells.31 Moreover, a recent study per-ormed on renal biopsy specimens of diabeticatients showed increased expression of VEGFnd increased VEGF-receptor activation inildly injured glomeruli.33 This was accompa-

ied by increased endothelial cell proliferation,uggesting that VEGF activation in mildly af-ected diabetic kidney might lead to increasedlomerular angiogenesis.33 This study also par-llels the observation that VEGF gene expres-ion correlates with glomerular neovasculariza-ion in human diabetic nephropathy29 andncreased expression of VEGF 121 isoform isp-regulated in the kidneys of patients withype 2 diabetes.34 The mechanism wherebyEGF contributes to glomerular vascular dam-ge is not clear. On one hand, it is required for

KK diabetic mice. (A) Frozen sections of kidneys derived) were stained with anti-mouse CD31 (upper panel) tothe mice were injected intravenously with RITC-dextranlood vessels. Note the diffuse RITC-dextran pattern inndication of leaking vessels. (B) The area occupied bycion Image analysis (Scion Corporation, Frederick, MD),ed (Breyer and Pozzi, unpublished data).

ility inabeticdeaththe bs an ising S

he induction and maintenance of the fenestrae

itsgVlhAdc

bafiatva1itegrtalc

mtg

AiTltdgtpdp

satvaaVcbtcl

FRcu altere

Angiogenesis 165

n glomerular capillary endothelial cells.35 Onhe other hand, podocyte-specific overexpres-ion of the VEGF164 isoform leads to collapsinglomerulopathy.36 The cause of up-regulatedEGF synthesis in diabetes also remains specu-

ative. Of the known pathways up-regulated byyperglycemia, protein kinase C,37 ROS,38 andGEs39 have been shown to increase VEGF pro-uction in both glomerular and nonglomerularompartments.

As mentioned previously, Ang 1 and Ang 2elong to a family of both pro-angiogenic andnti-angiogenic factors that exert their cellularunctions by binding to the Tie2 receptor. Dur-ng kidney development, Ang 1, Ang 2, and Tie2re highly expressed and play pivotal roles inhe maturation of glomeruli and renal bloodessels,40 with Ang 2 promoting angiogenesisnd the establishment of leaky vessels, and Angenhancing endothelial cell stability and reduc-

ng vessel leakage. In the kidneys, Ang 1 coun-eracts the action of VEGF41 and increases trans-ndothelial electrical resistance in culturedlomerular endothelial cells.42 In contrast, up-egulation of Ang 2 has been documented in aype 1 mouse diabetic nephropathy model43,44

nd in mesangial cells exposed to high glucoseevels. In addition, high glucose levels signifi-

igure 3. Major factors involved in the pathogenesis oOS production, glycation of glomerular matrix, intraglooncomitant decreased synthesis of anti-angiogenic molar damage, neovascularization, matrix deposition, and

antly increased the levels of secreted Ang 2 by c

esangial cells,45 suggesting the involvement ofhis angiogenesis-associated factor in the pro-ression of diabetic nephropathy.

nti-angiogenic Factorsn Diabetic Nephropathyhe proteolytic fragments of both non-extracel-

ular and extracellular matrix have been showno improve diabetic nephropathy. These areiscussed later in the section on enhancing theeneration and/or action of anti-angiogenic fac-ors because they most likely only exert theirrotective effects when given in pharmacologicoses and play a minimal role in the normalathophysiology of this disease process.The anti-angiogenic factor PEDF has been

hown to be involved in both diabetes-medi-ted retinal and renal vasculature complica-ions. Interestingly, there appears to be an in-erse correlation between the levels of PEDFnd VEGF in both physiologic and pathologicngiogenesis. PEDF down-regulates the levels ofEGF in retinal endothelial cells17 and de-reased PEDF levels are associated with dia-etic retinopathy in human beings.46 Moreover,he levels of this anti-angiogenic factor are de-reased at both the messenger RNA and proteinevels in the kidney of diabetic rats.30 In this

etic nephropathy. Hyperglycemia can lead to increasedar pressure, and synthesis of pro-angiogenic factors with. These changes significantly contribute to the glomer-d filtration observed in diabetes.

f diabmerullecules

ontext, high glucose levels and AGEs decrease

PaPottkoca

Ri

TratseibigchsuaIgpAdgmcs

AT

BblmaAtps

fdfabti

BP

IamfclbaidsaancVvntmmopmttudabwttfspt

166 R. Zent and A. Pozzi

EDF production in human mesangial cells30

nd in microvascular endothelial cells.47 Finally,EDF prevents high glucose level–inducedverexpression of the profibrotic growth fac-or, TGF-�.48 Taken together, these data suggesthat decreased expression of PEDF in diabeticidneys contributes to extracellular matrixverproduction, increased angiogenesis, andonsequent development of diabetic nephrop-thy.

enin-Angiotensin Systemn Diabetic Nephropathy

he renin-angiotensin system (RAS) plays a keyole in the progression of diabetic nephropathynd inhibition of this system is one of the majorherapeutic strategies used to slow the progres-ion of this disease process. There is convincingvidence from multiple large trials that block-ng the RAS decreases the progression of dia-etic nephropathy and overall proteinuria. This

s proposed to be induced primarily by alteringlomerular hemodynamics, resulting in a de-rease in intraglomerular pressure. There is,owever, some new and interesting evidenceuggesting that the RAS might in part contrib-te to the diabetic-induced vascular damage byffecting glomerular endothelial cell functions.n this context, angiotensin II contributes tolomerular vascular growth and leakage byromoting VEGF synthesis by podocytes.49

nother interesting mechanism, currentlyescribed in tumor angiogenesis, is that an-iotensin II can promote endothelial celligration and tubulogenesis, thus directly

ontributing to angiogenesis in pathologicettings.50-53

NTI-ANGIOGENICHERAPY IN DIABETIC NEPHROPATHY

ecause the pathologic consequences of dia-etes, including diabetic nephropathy, are in

arge part caused by both microvascular andacrovascular complications, new strategies

re required to prevent these complications.s mentioned previously, increased genera-

ion of pro-angiogenic factors, decreased ex-ression of anti-angiogenic factors, hyperten-

ion, and high glucose level–mediated AGE i

ormation represent the major mediators ofiabetes-induced vascular damage. It is there-ore apparent that altering the activity of onend/or many of theses mediators might beeneficial in treating and ideally preventinghe glomerular vascular complications foundn diabetes types 1 and 2.

locking Traditionalro-angiogenic Factors

ncreased expression of pro-angiogenic factorsnd their receptors has been largely docu-ented in diabetes.54-56 Because most of these

actors might contribute to the glomerular vas-ular damage by increasing endothelial cell pro-iferation and permeability, many studies haveeen performed to block these pathways innimal models of diabetes. Anti-VEGF antibod-es administered to type 1 diabetic rats imme-iately after induction of diabetes have beenhown to decrease hyperfiltration, albuminuria,nd glomerular hypertrophy.57 VEGF blockadelso prevented the up-regulation of endothelialitric oxide synthase expression in glomerularapillary endothelial cells, suggesting that anti-EGF therapy might be beneficial in preventingascular damage in the early events of diabeticephropathy.57 Similarly, anti-VEGF therapy at-enuated increases in kidney weight and glo-erular volume in the diabetic db/db mouse, aodel of obese type 2 diabetes,58 and inhibition

f VEGF prevented early glomerular hypertro-hy in the Zucker diabetic fatty rat, anotherodel of obese type 2 diabetes.59 In contrast to

hese studies, anti-VEGF antibodies did not leado any beneficial effects to the kidney whensed in the Goto-Kakizaki rat, a lean type 2iabetes model.60 Although very promising innimal studies, anti-VEGF therapy has not yeteen used in human subjects with diabetes,hich makes it difficult to appreciate its poten-

ial therapeutic efficacy. However, when thisherapy was administered for colorectal cancer,requent adverse effects, including hyperten-ion, bleeding episodes, thrombotic events, androteinuria, were reported,61 suggesting thathese antibodies will be difficult to administer

n patients with diabetic nephropathy.

EGo

DtabtofiBadtTricPieCeo

csdgftabtltvtbaMcreushagb

slpcrpcpMccVtgerstk

plItbiddtln

obocmrip

B

AmAaclm

Angiogenesis 167

nhancing theeneration and/or Actionf Anti-angiogenic Factors

ecreased expression of anti-angiogenic fac-ors, including PEDF, non-extracellular matrix,nd extracellular-matrix cleavage products haseen documented in diabetic nephropa-hy,30,43,44,62 suggesting that the administrationf these anti-angiogenic factors might be bene-cial in the treatment of diabetic nephropathy.ecause the levels of the anti-angiogenic PEDFre decreased in kidneys of streptozotocin-in-uced diabetic rats30 and PEDF down-regulateshe production of high glucose level–mediatedGF-� synthesis by mesangial cells,30 diabeticats were treated with an adenovirus express-ng PEDF to evaluate its effects on diabeticomplications.48 Enhanced expression of renalEDF significantly alleviated microalbuminuria

n early stages of diabetes and prevented thexpression of the profibrotic agents TGF-� andTGF. These results suggested that exogenousxpression of PEDF is beneficial in a rat modelf diabetic nephropathy.

Among the non-extracellular matrix–derivedleavage products, angiostatin recently washown to ameliorate the glomerular vascularamage induced by diabetes.62 Angiostatin isenerated from circulating plasminogen by dif-erent enzymes, including matrix metallopro-einase (MMP)2, 3, 7, 9, and 12.13 Angiostatin ispotent inhibitor of endothelial cell functionsoth in vivo and in vitro and it has been showno ameliorate pathologic angiogenesis, particu-arly tumor-associated angiogenesis.63 Based onhe observation that angiostatin inhibits retinalascular permeability in diabetic retinopathy,64

he effect of angiostatin was analyzed in dia-etic nephropathy.62 In this study the levels ofngiostatin and one of its major generators,MP2, were found to be significantly de-

reased in the kidneys of streptozotocin-treatedats.62 When exogenous angiostatin was deliv-red by using an adenoviral delivery, albumin-ria and glomerular hypertrophy was alleviatedignificantly and there was a reduction in theigh glucose level–induced expression of VEGFnd TGF-�.62 Although this study strongly sug-ests that treatment with angiostatin might be

eneficial in diabetic nephropathy, a recent f

tudy suggested that use of this therapy mayead to potential complications. In this studyerformed on human subjects, mammary arteryapillary density and VEGF expression wereeduced significantly in diabetic patients com-ared with nondiabetic patients.65 This wasorrelated with an up-regulation in arterial ex-ression and the release of active MMP-2 andMP-9, as well as angiostatin.65 Therefore, in

ontrast to the glomerular vasculature, the in-reased angiostatin production and reducedEGF formation in the diabetic arterial vascula-

ure might lead to decreased and impaired an-iogenesis.65 This paradigm highlights the inter-sting concept that endothelial cells and theiresponses to specific stimuli differ within tis-ues and organs. Moreover, anti-angiogenicherapy might be beneficial in one organ (ie,idney), but deleterious in others.

Among the extracellular matrix–derivedroducts, endostatin (a cleavage product of col-

agen XVIII) and the �3-NC1 domain of collagenV (�3-NC1[IV]) have been shown to amelioratehe glomerular vascular complications inducedy diabetes.43,44 Treatment of streptozotocin-

nduced diabetic mice with either �3-NC1(IV)omain43 or endostatin44 reduced the vascularamage, glomerular hypertrophy, hyperfiltra-ion, albuminuria, and the number of glomeru-ar endothelial cells in the early stage of diabeticephropathy.

Interestingly, increased podocyte productionf collagen IV (particularly the �3 chain) driveny angiotensin II, VEGF, and TGF-� has beenbserved in diabetes.66 Although this chain mayontribute to the glomerular damage by pro-oting glomerulosclerosis, it also might play a

ole as a physiologic inhibitor of angiogenesis ifts cleavage product, �3-NC1(IV), is generatedroteolytically in this pathologic process.

locking AGEs and Their Receptors

GEs are proteins or lipids that are non-enzy-atically glycated after exposure to sugars.67

GEs are prevalent in the diabetic vasculaturend contribute to microvascular and macrovas-ular complications because they form cross-inks between molecules in the extracellular

atrix milieu. In addition, they alter cellular

unction by interacting with the RAGE recep-

tddbntadtaptsamkIttdmuddbpei2agmppgbI

C

Dpmprlggtic

ircolpdm

R

1

1

1

1

1

168 R. Zent and A. Pozzi

ors.67 In the endothelium, AGEs cause the pro-uction of ROS68 and in human glomerular en-othelial cells they regulate the expression ofoth VEGF and VEGFR2.69,70 Despite the recog-ition of their deleterious effects in diabetes,he pharmacologic treatment currently avail-ble for either type 1 or type 2 diabetic patientsoes not directly address the excess accumula-ion of AGEs. However, studies performed onnimal models of diabetes strongly suggest thatrevention of AGE formation and/or blockingheir cellular actions would be beneficial inlowing diabetic nephropathy. Treatment withnti-RAGE antibodies in obese type 2 diabeticice71 or mice with type 1 diabetes72 reduced

idney weight, albumin excretion, and collagenV production, thus supporting the hypothesishat RAGE is an important pathogenic factor inhe renal changes found in type 1 and type 2iabetes. Moreover, the suppression of AGE for-ation by aminoguanidine decreases albumin-

ria and the severity of glomerular lesions iniabetic rodents.73 Another compound, pyri-oxamine, which inhibits the glycoxidativereakdown of Amadori products to AGE,74 alsorevents the increase of plasma creatinine lev-ls, albuminuria, and glomerular hypertrophyn streptozotocin-diabetic rats75 and obese type

diabetic mice.76 Interestingly when pyridox-mine was administered together with the an-iotensin II receptor blocker valsartan, bothortality and the progression of diabetic ne-hropathy were reduced significantly com-ared with the single treatments,76 strongly sug-esting that combination therapy might beeneficial for the treatment of established typeI diabetes.

ONCLUSIONS

iabetic nephropathy is a significant medicalroblem because of its increasing incidence,orbidity, and mortality. One of the principalathogenic mechanisms of this disease is aber-ant angiogenesis. In this review we have high-ighted the major pro-angiogenic and anti-angio-enic factors that potentially could affect thelomerular vascular damage in the setting ofype 1 and type 2 diabetes. A better understand-ng of the role of angiogenesis in diabetes is

ritical for the development of successful inhib-

tors to this process. Because inhibition of theenin-angiotensin system is the only therapyurrently available to retard the progressionf this incurable disease, targeting the patho-

ogic angiogenesis associated with diabetic ne-hropathy might be viewed as a valid tool toecrease and ideally halt the morbidity andortality associated with this disease.

EFERENCES1. Pandya NM, Dhalla NS, Santani DD. Angiogenesis—a

new target for future therapy. Vascul Pharmacol.2006;44:265-74.

2. Roy H, Bhardwaj S, Yla-Herttuala S. Biology of vascu-lar endothelial growth factors. FEBS Lett. 2006;580:2879-87.

3. Olsson AK, Dimberg A, Kreuger J, et al. VEGF recep-tor signalling—in control of vascular function. NatRev Mol Cell Biol. 2006;7:359-71.

4. Eklund L, Olsen BR. Tie receptors and their angiopoi-etin ligands are context-dependent regulators of vas-cular remodeling. Exp Cell Res. 2006;312:630-41.

5. Thurston G, Rudge JS, Ioffe E, et al. Angiopoietin-1protects the adult vasculature against plasma leakage.Nat Med. 2000;6:460-3.

6. Thurston G, Suri C, Smith K, et al. Leakage-resistantblood vessels in mice transgenically overexpressingangiopoietin-1. Science. 1999;286:2511-4.

7. Scharpfenecker M, Fiedler U, Reiss Y, et al. The Tie-2ligand angiopoietin-2 destabilizes quiescent endothe-lium through an internal autocrine loop mechanism.J Cell Sci. 2005;118:771-80.

8. Roviezzo F, Tsigkos S, Kotanidou A, et al. Angiopoi-etin-2 causes inflammation in vivo by promotingvascular leakage. J Pharmacol Exp Ther. 2005;314:738-44.

9. Benest AV, Salmon AH, Wang W, et al. VEGF andangiopoietin-1 stimulate different angiogenic pheno-types that combine to enhance functional neovascu-larization in adult tissue. Microcirculation. 2006;13:423-37.

0. Zhang L, Yang N, Park JW, et al. Tumor-derived vas-cular endothelial growth factor up-regulates angiopoi-etin-2 in host endothelium and destabilizes host vas-culature, supporting angiogenesis in ovarian cancer.Cancer Res. 2003;63:3403-12.

1. Sato Y. Update on endogenous inhibitors of angiogen-esis. Endothelium. 2006;13:147-55.

2. Sudhakar A, Nyberg P, Keshamouni VG, et al. Humanalpha1 type IV collagen NC1 domain exhibits distinctantiangiogenic activity mediated by alpha1beta1 inte-grin. J Clin Invest. 2005;115:2801-10.

3. Zent R, Pozzi A. Extracellular matrix and the devel-opment of disease: the role of its component in can-cer progression. Adv Dev Biol. 2005;15:203-28.

4. Volpert OV, Zaichuk T, Zhou W, et al. Inducer-stim-ulated Fas targets activated endothelium for destruc-

tion by anti-angiogenic thrombospondin-1 and pig-

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

Angiogenesis 169

ment epithelium-derived factor. Nat Med. 2002;8:349-57.

5. Steele FR, Chader GJ, Johnson LV, et al. Pigmentepithelium-derived factor: neurotrophic activity andidentification as a member of the serine proteaseinhibitor gene family. Proc Natl Acad Sci U S A.1993;90:1526-30.

6. Dawson DW, Volpert OV, Gillis P, et al. Pigmentepithelium-derived factor: a potent inhibitor of angio-genesis. Science. 1999;285:245-8.

7. Zhang SX, Wang JJ, Gao G, et al. Pigment epithelium-derived factor downregulates vascular endothelialgrowth factor (VEGF) expression and inhibits VEGF-VEGF receptor 2 binding in diabetic retinopathy. JMol Endocrinol. 2006;37:1-12.

8. Schrijvers BF, De Vriese AS, Flyvbjerg A. From hyper-glycemia to diabetic kidney disease: the role of met-abolic, hemodynamic, intracellular factors andgrowth factors/cytokines. Endocr Rev. 2004;25:971-1010.

9. van Dijk C, Berl T. Pathogenesis of diabetic nephrop-athy. Rev Endocr Metab Disord. 2004;5:237-48.

0. Zent R, Yan X, Su Y, et al. Glomerular injury isexacerbated in diabetic integrin alpha1-null mice.Kidney Int. 2006;70:460-70.

1. Susztak K, Raff AC, Schiffer M, et al. Glucose-inducedreactive oxygen species cause apoptosis of podocytesand podocyte depletion at the onset of diabetic ne-phropathy. Diabetes. 2006;55:225-33.

2. Iglesias-De La Cruz MC, Ruiz-Torres P, Alcami J, et al.Hydrogen peroxide increases extracellular matrixmRNA through TGF-beta in human mesangial cells.Kidney Int. 2001;59:87-95.

3. Park SK, Kim J, Seomun Y, et al. Hydrogen peroxideis a novel inducer of connective tissue growth factor.Biochem Biophys Res Commun. 2001;284:966-71.

4. Forbes JM, Cooper ME, Oldfield MD, et al. Role ofadvanced glycation end products in diabetic ne-phropathy. J Am Soc Nephrol. 2003;14:S254-8.

5. Nagase M, Shibata S, Yoshida S, et al. Podocyte injuryunderlies the glomerulopathy of Dahl salt-hyperten-sive rats and is reversed by aldosterone blocker. Hy-pertension. 2006;47:1084-93.

6. Osterby R, Asplund J, Bangstad HJ, et al. Neovascu-larization at the vascular pole region in diabeticglomerulopathy. Nephrol Dial Transplant. 1999;14:348-52.

7. Nyengaard JR, Rasch R. The impact of experimentaldiabetes mellitus in rats on glomerular capillary num-ber and sizes. Diabetologia. 1993;36:189-94.

8. Guo M, Ricardo SD, Deane JA, et al. A stereologicalstudy of the renal glomerular vasculature in the db/dbmouse model of diabetic nephropathy. J Anat. 2005;207:813-21.

9. Kanesaki Y, Suzuki D, Uehara G, et al. Vascular endo-thelial growth factor gene expression is correlatedwith glomerular neovascularization in human dia-betic nephropathy. Am J Kidney Dis. 2005;45:288-94.

0. Wang JJ, Zhang SX, Lu K, et al. Decreased expression

of pigment epithelium-derived factor is involved inthe pathogenesis of diabetic nephropathy. Diabetes.2005;54:243-50.

1. Kim NH, Oh JH, Seo JA, et al. Vascular endothelialgrowth factor (VEGF) and soluble VEGF receptorFLT-1 in diabetic nephropathy. Kidney Int. 2005;67:167-77.

2. Iglesias-de la Cruz MC, Ziyadeh FN, Isono M, et al.Effects of high glucose and TGF-beta1 on the expres-sion of collagen IV and vascular endothelial growthfactor in mouse podocytes. Kidney Int. 2002;62:901-13.

3. Hohenstein B, Hausknecht B, Boehmer K, et al. LocalVEGF activity but not VEGF expression is tightly reg-ulated during diabetic nephropathy in man. KidneyInt. 2006;69:1654-61.

4. Bortoloso E, Del Prete D, Dalla Vestra M, et al. Quan-titative and qualitative changes in vascular endothelialgrowth factor gene expression in glomeruli of pa-tients with type 2 diabetes. Eur J Endocrinol. 2004;150:799-807.

5. Eremina V, Quaggin SE. The role of VEGF-A in glo-merular development and function. Curr Opin Neph-rol Hypertens. 2004;13:9-15.

6. Eremina V, Sood M, Haigh J, et al. Glomerular-specificalterations of VEGF-A expression lead to distinct con-genital and acquired renal diseases. J Clin Invest.2003;111:707-16.

7. Uchida K, Uchida S, Nitta K, et al. Glomerular endo-thelial cells in culture express and secrete vascularendothelial growth factor. Am J Physiol. 1994;266:F81-8.

8. Lee KS, Kim SR, Park SJ, et al. Hydrogen peroxideinduces vascular permeability via regulation of vascu-lar endothelial growth factor. Am J Respir Cell MolBiol. 2006;35:190-7.

9. Yamagishi S, Inagaki Y, Okamoto T, et al. Advancedglycation end product-induced apoptosis and overex-pression of vascular endothelial growth factor andmonocyte chemoattractant protein-1 in human-cultured mesangial cells. J Biol Chem. 2002;277:20309-15.

0. Woolf AS, Yuan HT. Angiopoietin growth factors andTie receptor tyrosine kinases in renal vascular devel-opment. Pediatr Nephrol. 2001;16:177-84.

1. Satchell SC, Harper SJ, Tooke JE, et al. Human podo-cytes express angiopoietin 1, a potential regulator ofglomerular vascular endothelial growth factor. J AmSoc Nephrol. 2002;13:544-50.

2. Satchell SC, Anderson KL, Mathieson PW. Angiopoi-etin 1 and vascular endothelial growth factor modu-late human glomerular endothelial cell barrier prop-erties. J Am Soc Nephrol. 2004;15:566-74.

3. Yamamoto Y, Maeshima Y, Kitayama H, et al. Tum-statin peptide, an inhibitor of angiogenesis, preventsglomerular hypertrophy in the early stage of diabeticnephropathy. Diabetes. 2004;53:1831-40.

4. Ichinose K, Maeshima Y, Yamamoto Y, et al. Antian-

giogenic endostatin peptide ameliorates renal alter-

4

4

4

4

4

5

5

5

5

5

5

5

5

5

5

6

6

6

6

6

6

6

6

6

6

7

7

170 R. Zent and A. Pozzi

ations in the early stage of a type 1 diabeticnephropathy model. Diabetes. 2005;54:2891-903.

5. Singh AK, Gudehithlu KP, Pegoraro AA, et al. Vascu-lar factors altered in glucose-treated mesangial cellsand diabetic glomeruli. Changes in vascular factorsimpair endothelial cell growth and matrix. Lab Invest.2004;84:597-606.

6. Spranger J, Osterhoff M, Reimann M, et al. Loss of theantiangiogenic pigment epithelium-derived factor inpatients with angiogenic eye disease. Diabetes. 2001;50:2641-5.

7. Yamagishi S, Matsui T, Inoue H. Inhibition by ad-vanced glycation end products (AGEs) of pigmentepithelium-derived factor (PEDF) gene expression inmicrovascular endothelial cells. Drugs Exp Clin Res.2005;31:227-32.

8. Wang JJ, Zhang SX, Mott R, et al. Salutary effect ofpigment epithelium-derived factor in diabetic ne-phropathy: evidence for antifibrogenic activities. Di-abetes. 2006;55:1678-85.

9. Kang YS, Park YG, Kim BK, et al. Angiotensin IIstimulates the synthesis of vascular endothelialgrowth factor through the p38 mitogen activatedprotein kinase pathway in cultured mouse podocytes.J Mol Endocrinol. 2006;36:377-88.

0. Uemura H, Nakaigawa N, Ishiguro H, et al. Antipro-liferative efficacy of angiotensin II receptor blockersin prostate cancer. Curr Cancer Drug Targets. 2005;5:307-23.

1. Yoshiji H, Noguchi R, Kuriyama S, et al. Suppressionof renin-angiotensin system attenuates hepatocarci-nogenesis via angiogenesis inhibition in rats. Antican-cer Res. 2005;25:3335-40.

2. Yoshiji H, Kuriyama S, Noguchi R, et al. Angiotensin-Iconverting enzyme inhibitors as potential anti-angio-genic agents for cancer therapy. Curr Cancer DrugTargets. 2004;4:555-67.

3. Yoshiji H, Kuriyama S, Kawata M, et al. The angioten-sin-I-converting enzyme inhibitor perindopril sup-presses tumor growth and angiogenesis: possible roleof the vascular endothelial growth factor. Clin CancerRes. 2001;7:1073-8.

4. Duh E, Aiello LP. Vascular endothelial growth factorand diabetes: the agonist versus antagonist paradox.Diabetes. 1999;48:1899-906.

5. Aiello LP, Wong JS. Role of vascular endothelialgrowth factor in diabetic vascular complications. Kid-ney Int Suppl. 2000;77:S113-9.

6. Satchell SC, Mathieson PW. Angiopoietins: microvas-cular modulators with potential roles in glomerularpathophysiology. J Nephrol. 2003;16:168-78.

7. de Vriese AS, Tilton RG, Elger M, et al. Antibodiesagainst vascular endothelial growth factor improveearly renal dysfunction in experimental diabetes.J Am Soc Nephrol. 2001;12:993-1000.

8. Flyvbjerg A, Dagnaes-Hansen F, De Vriese AS, et al.Amelioration of long-term renal changes in obese

type 2 diabetic mice by a neutralizing vascular 7

endothelial growth factor antibody. Diabetes.2002;51:3090-4.

9. Schrijvers BF, Flyvbjerg A, Tilton RG, et al. A neutral-izing VEGF antibody prevents glomerular hypertro-phy in a model of obese type 2 diabetes, the Zuckerdiabetic fatty rat. Nephrol Dial Transplant. 2006;21:324-9.

0. Schrijvers BF, De Vriese AS, Tilton RG, et al. Inhibi-tion of vascular endothelial growth factor (VEGF)does not affect early renal changes in a rat model oflean type 2 diabetes. Horm Metab Res. 2005;37:21-5.

1. Gupta K, Zhang J. Angiogenesis: a curse or cure?Postgrad Med J. 2005;81:236-42.

2. Zhang SX, Wang JJ, Lu K, et al. Therapeutic potentialof angiostatin in diabetic nephropathy. J Am SocNephrol. 2006;17:475-86.

3. Griscelli F, Li H, Bennaceur-Griscelli A, et al. Angiosta-tin gene transfer: inhibition of tumor growth in vivoby blockage of endothelial cell proliferation associ-ated with a mitosis arrest. Proc Natl Acad Sci U S A.1998;95:6367-72.

4. Sima J, Zhang SX, Shao C, et al. The effect of angiosta-tin on vascular leakage and VEGF expression in ratretina. FEBS Lett. 2004;564:19-23.

5. Chung AW, Hsiang YN, Matzke LA, et al. Reducedexpression of vascular endothelial growth factor par-alleled with the increased angiostatin expression re-sulting from the upregulated activities of matrix met-alloproteinase-2 and -9 in human type 2 diabeticarterial vasculature. Circ Res. 2006;99:140-8.

6. Chen S, Lee JS, Iglesias-de la Cruz MC, et al. Angio-tensin II stimulates alpha3(IV) collagen production inmouse podocytes via TGF-beta and VEGF signalling:implications for diabetic glomerulopathy. NephrolDial Transplant. 2005;20:1320-8.

7. Goldin A, Beckman JA, Schmidt AM, et al. Advancedglycation end products: sparking the development ofdiabetic vascular injury. Circulation. 2006;114:597-605.

8. Wautier MP, Chappey O, Corda S, et al. Activation ofNADPH oxidase by AGE links oxidant stress to alteredgene expression via RAGE. Am J Physiol. 2001;280:E685-94.

9. Pala L, Cresci B, Manuelli C, et al. Vascular endo-thelial growth factor receptor-2 and low affinityVEGF binding sites on human glomerular endothe-lial cells: biological effects and advanced glycosyl-ation end products modulation. Microvasc Res.2005;70:179-88.

0. Yamagishi S, Yonekura H, Yamamoto Y, et al. Ad-vanced glycation end products-driven angiogenesis invitro. Induction of the growth and tube formation ofhuman microvascular endothelial cells through auto-crine vascular endothelial growth factor. J Biol Chem.1997;272:8723-30.

1. Flyvbjerg A, Denner L, Schrijvers BF, et al. Long-termrenal effects of a neutralizing RAGE antibody in obesetype 2 diabetic mice. Diabetes. 2004;53:166-72.

2. Jensen LJ, Denner L, Schrijvers BF, et al. Renal

7

7

7

7

7

Angiogenesis 171

effects of a neutralising RAGE-antibody in long-termstreptozotocin-diabetic mice. J Endocrinol. 2006;188:493-501.

3. Kelly DJ, Gilbert RE, Cox AJ, et al. Aminoguanidineameliorates overexpression of prosclerotic growthfactors and collagen deposition in experimental dia-betic nephropathy. J Am Soc Nephrol. 2001;12:2098-107.

4. Booth AA, Khalifah RG, Todd P, et al. In vitrokinetic studies of formation of antigenic advancedglycation end products (AGEs). Novel inhibition ofpost-Amadori glycation pathways. J Biol Chem.

1997;272:5430-7.

5. Degenhardt TP, Alderson NL, Arrington DD, et al.Pyridoxamine inhibits early renal disease and dyslip-idemia in the streptozotocin-diabetic rat. Kidney Int.2002;61:939-50.

6. Zheng F, Zeng YJ, Plati AR, et al. Combined AGEinhibition and ACEi decreases the progression of es-tablished diabetic nephropathy in B6 db/db mice.Kidney Int. 2006;70:507-14.

7. Pozzi A, Moberg PE, Miles LA, et al. Elevated matrixmetalloprotease and angiostatin levels in integrinalpha 1 knockout mice cause reduced tumor vas-cularization. Proc Natl Acad Sci U S A. 2000;97:

2202-7.