Kaposi Sarcoma: A Model of Oncogenesis

2010

Editors

Liron Pantanowitz

Assistant Professor of Pathology, Tufts University School of Medicine, Departrnent of Pathology Baystate Medical Center, Springfield, Massachusetts, USA

Justin Stebbing

Consultant Medical Oncologistl Senior Lecturer, Department of Medical Oncology Imperial Collegel Imperial Healthcare NHS Trust, Charing Cross and

Hammersrnith Hospitals, London, UK

Bruce J. Dezube

Associate Professor of Medicine, Department of Medicine (Hematology-Oncology), Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts; USA

Research Signpost, T.C. 371661 (2), Fbrt P.O., Trivandrum-695 023 Kerala, India

Contents

Dedication

Contributors

Foreword Patrick S. Moore and Yuan Chang

Part I. Basic Science Section

Chapter l Overview of Kaposi sarcoma Liron Pantanowitz, Justin Stebbing and Bruce J. Dezube

Chapter 2 Inhibition of antigen presentation during Human Herpesvirus-8 infection Klaus Friih

Chapter 3 In vitro endothelial ce11 systems to study Kaposi sarcoma Ashlee V. Moses

Chapter 4 KSHV-encoded and KSHV-regulated microRNAs Pauline Chugh and Dirk P. Dittmer

Chapter 5 Chemokines and signal transduction targets in Kaposi sarcoma Ryan J. Sullivan, Liron Pantanowitz and Bruce J. Dezube

Chapter 6 The formation of new blood vessels in Kaposi sarcorna Giovanni Barillari, Elena Toschi, Cecilia Sgadari, Paolo Monini and Barbara Ensoli

Chapter 7 The origin of Kaposi sarcoma tumor cells Berenice Aguilar and Young-Kwon Hong

Chapter 8 Histological variants of Kaposi sarcoma Wayne Grayson and Liron Pantanowitz

Chapter 9 Dendntic cells in Kaposi sarcoma Giovanna M. Crisi and Liron Pantanowitz

Chapter 10 Environmental factors in the pathogenesis of Kaposi sarcoma Thierry Simonart

Part 11. Epidemiology-Clinica1 Section

Chapter 11 Transmission of Human Herpesvinis-8 Vickie Marshall, Nauarena Labo àhd Denise Whitby

Chapter 12 The epidemiology of Kaposi sarcoma Huong Q Nguyen and Corey Casper

Chapter 13 Kaposi sarcoma and gender Justin Stebbing, Tom Powles and Mark Bower

Cbapter 14 Kaposi sarcoma regression and exacerbation Liron Pantanowitz and Bmce J. D m b e

Cbapter l5 Contemporary perspective of Kaposi sarcoma in Africa Vivek Subbiah, Jachon Orem, Walter Mwanda. Margaret Borok and Scot C. Remick

Chapter l6 Transplant associated Kaposi sarcoma: What have we learnt h m studia in immune suppressed persons? Diego Serraino, Angela De Paoli, Pierluca Piselli and Lucia Fratino

Chapter 17 Kaposi sarcoma in the pediatric population Deborah Zeitlin, Justin Stebbing, Bmce J. Dezube and Liron Pantanowitz

Chapter l8 Ora1 Kaposi sarcoma L Feller and J Lemmer

Chapter 19 Kaposi sarcoma in unusual anatomica1 locations Liron Pantanowitz

Chapter 20 Imaging techniqua for Kaposi sarcorna Deirdre O 'Mahony, Amir H Gandjbakhche, Moinuddin Hassan, Abby Vogel and Robert Yarchoan

Chapter 21 Evaluation and management of patients with Kaposi sarcoma David M. Aboulafa

Chapter 22 Classic Kaposi sarcoma therapy: New look at an old disease Giuseppe Di Lorenzo and Rossella Di Trolio

Chapter 23 HIV related Kaposi sarcoma and antiretroviral therapy Simon D. Portsmouth

Chapter 24 Pathogenesis-related approaches to the treatment of Kaposi sarcoma SUSM E. Krown

Research Slgnpost 37/661 (2). Fort P.O. Tiivandrum-695 023 Kerala, India

Kaposi Sarcoma: A Model of Oncogenesis, 2010: 101-122 ISBN: 978-81-308-0380-7 Editors: Liron Pantanowitz, Justin Stebbing and Bruce J . Dezube

6. The formation of new blood vessels in Kaposi sarcoma -.

Giovanni Barillari, ElenaToschi, Cecilia Sgadari, Paolo Monini and Barbara Ensoli National AIDS Center. Istituto Superiore di Sanità. Rome. Iuly

Summary. Abnormal md intaise new bbod vessel formation is a generai feature of al1 clinicalepidemiological forms of Kaposi sarcoma (KS). This chapter sumrnarises key aspects of the scientific literature regarding the genesis of new vessels m KS lesbns, in an effort to draw a u n w g view of KS initiation and progression. The basic mechanisms leading to blood vessel generation, inchiding angiogenesis and vasculogenesis, are discussed focusing on the growth factors, mtegrin receptors, and extracellular matrix (ECM)-degrading proteinases which are altered andor de-regulated in KS. Features of vessels present m early- and late- stage KS lesions are illustrateci, paying a particular attention to the role that the KS spindle-shaped cells resembling activated endothelial cells or lymphatic endothelial macrophages play in KS-associated vasculogenesis or angiogenesis. We have also reviewed the pro-angiogenic pathways tiggered in KS by oncogenes and oncosuppremr gene products, cytokines and chemokines, Human Immunodeficiency Virus (HIV) Tat protein, and Human Herpesvirus 8 (HHV8). Finally, we have reported about anti-angiogenic molecules which had been or are currently being evaluated for KS treatment. In particular, we have emphasised H IV Protease Inhibitors (PI), a class of antiretroviral dmgs effectively inhibiting KS onset or promoting KS regression. h fact, due to their capability of blocking both new blood vessel formation and tumor cell invasion, fundamental steps of tumor progression, P1 should be considered as promising anti-cancer tools.

1. Introduction 1.1. Physiologic new blood vessel formation

In ihe human body, cells are generally located within 200 pm fiom blood vessels, which is the maximal distance aiiowing the d i a s i o n of oxygen and nutrients: when tissues . Correspondaice/Reprint request: Dr. Barbara Ensoli, National AiDS Center, Istituto Superiore di Saniti, Rome, Italy E-mail: barbaraensoli@iss.it

102 GiovanniBarillari e! al.

and organs grow, the 200 pm iunit is exceeded, and new blood vessels have to be formed [l]. During human embryonic development, vessels are generated rnainly by vasculogenesis, which is accomplished through the mobilization of enbthelial cell precursors (ECP) fiom yolk sack or bone rnarmw, theù homing to sites of new vessel formation, and their differentiation into mature endothelial c e k (EC) [l]. Later, new capiliaries sprout fiom existing vessels b u g h a multi-step process termed angiogenesis [l].

Angiogenesis is started when EC degtade and penetrate the blood vessel basement membrane (BM) and, subsequently, the peri-vascular extracellular matrix (ECM): this is foUowed by EC proliferation and migration toward chernotactic stimuli, leading to EC organization into solid cords [l]. Subsequently, vacuoles form inside endothelial cords, which then evolve into tubes with hoilow lumens penniaing blood flux [l]. As a consequence of EC shear stress, pericytes are recmited at angiogenic sites, and produce molecules hrming the new blood vessel BM [2]. Upon their adhesion onto the reconstituted BM, EC become quiescent, and newly forrned capillaries are stabilised [l].

Both vasculogenesis and aagiogenesis are regulated by chernical s i b l s [3]. Promoters of new blood vessel formation (the angiogenic factors) include molecules inducing one, few, or al1 vasculogenic or angiogenic steps [3]. Some molecules act directly on ECP or W=, while others stimulate local strornal or immune ceUs to release angiogenic factors [3]. Arnong human angiogenic factors, the vascular endothelial growth factors (VEGF) and the fibroblast growth factors (FGF) are particularly effective at promoting ECP and EC survival, growth, and motility [4,5]. Upon release by producing cells, angiogenic hctors remain diffisible or bind to ECM components [4,5]. In most cases, ECM sequestas the angiogenic molecule without compromising its biologica1 activity. For example, ECM-bound, immobilized FGF or VEGF drive migrating EC, guiding the formation of EC cords [4,5]. However, some ECM components, such as thrombospondin (TSP>l, can prevent FGF or VEGF binding to their receptors [4,5].

indeed, although it is mggered by soluble factors, new blood vessel formation is modulated by ECM components. In particular, the interactions occun-ing between EC and the ECM are mediateci by the integrins, a family of transmembrane adhesion receptors composed by ct and 0 s u b u ~ t s [6]. Among the integrins, crlpl, 02/31, ob/3l, aorp3 and avDS are intensely expressed at angiogenic sites, their levels aniilor a f h i t y being up- regulated by angiogenic factors [6]. When angiogenesis is started, EC are exposed to both i interstitial collagen and non-coilagen proteins: the latter are recmited fiom plasma i

i because of the increase in vascular permeability and form a provisional matrix [7]. EC i adhesion to provisional matrix tnggers the above rnentioned pro-angiogenic integrins, activating signalling pathways which, in tum, support al1 angiogenic steps [6]. The cross talk of pro-angiogenic integrins and the VEGF or FGF receptors (VEGFR and FGFR) increases the signalling of al1 these molecules associated with EC migration, survival and growth [6]. In addition to integrin-mediated adhesive interactions, angiogenesis requires ECM remodelling: i.e., degradation of the pre-existing BM and ECM and synthesis of novel BM and ECM [l ,2]. When EC invasion is impaired angiogenesis is blocked [8].

EC penetrate the blood vessel BM and perivascular ECM because of the activity of ECM-degrading enzymes. Among them aie heparanases, serine proteases (including plasmin and its activators), and cysteine proteases (such as the cathepsins) [8]. However, angiogenesis mostly depends on the matrix-metalloproteinases (MMP), a family of zinc- containing enzymes which digest ECM components including collagens, laminin, fibronectin, and proteoglycans [8]. The human MMP iàmily compnses 23 members which possess distinct but overlapping substrate specificities: al1 MMP share a basic structural organization comprising a signal peptide that targets thern for secretion, a

i I I C fomatioii ol'iicw klootl vcssels iii K;il>osi siircx)ma

, ,iopeptidc doiiiniii aiid an N-teriiiinal catalytic domain [8]. M M P activity is i~gu la ted at :aur points gciic cxpressioii. compartmentalization (i.e., pericellular accumulation o f iizyme), piu-ciizyiiic nctivatioii. and enzyme inactivation - and i t is furthcr conh.olled by ~ibstrate availability aiid affiiiity. Specifically, MMP are synthcsized as pro-enzyiies. :tid then sccicicul in n dillusible fotm, or anchored to the cell membrane (ineinbirine-typc .i/IMP, MI'-MMP) 181. Suhsty~icntly. soluble MMP pro-foims arc activatrd througli tlie Ictachnieiit ol'tlic liciiiopexiii doiiiain promoted by proteinases including otlier M M P . F o i xainple. MMP-3 piaccsscs proMMP-l into fully active M M P - I [q].

MMI' octivity is inliibite(l principally by the Tissue Inhibitors oSMMP (TIMP). Four I'IMP Iiiivc hccii isolotcd io diite: types-l, -3. and -4 arc sciet td. tliltiisible pioteiiis ~vhich ii i ipii ir tlic iictiviiy of i i iost soluble M M P by bincling tliriii i11 a 1 : 1 stoicliioiiictry; Iiy contiast. 'l'lM1'-3 is ;i niatrix-associateci inoleculc which i i ih ib i f~ soinc MT-MMP i 10.1 l ] . MMI' c;iii ;ilso hc iiiliibitcd by piacollaycii C'-tcriiiiiial protcinasc ciiliaiicci-. <.crversioii-iidiiciiig cys~ciii-rich protein witli Kazal iiiotifv. tissuc hc lor pathway iiihibitor-2, (L?-iiiiicii)globuliiis. aiid TSP-I or -2 1 1 0.12.1 3 1.

MMI' arc cxprcsscd by cluicsceiit EC at a very low levcl: wlicii angiogeiiesis is iriggercd. I :C' cx pi.cssioii 01' M M P is up-rcgulattui b y growtli Sactors. cyok ines. andlor iiiechaiiical sircss 18 1. Major rolcs in initiatitig aiigiogencsis havc b e n attributed to M T I - MMP. MM1'-2, ;iiid MM1'-O. Iiitercstiiigly. MMP-3 :ind/oi M ' f I -MMP arc pi-t?;eiit ol i thc 13.2 surliicc iii iissociiitioii widi dic trvfi3 inicgriii 1 14.1 5 1. Iii aciditioii. M ' f l -MMP caii co- localizc witli Il l-iiitcgriiis iii l'('-I.:(' cciiiiacis I 161, wliile MMP-9 interacts wi th the CD44 cell odlicsioii iiiolcculc 1 17 l . 'l'licsc. iiiulti-protciii coiiiplcxcs Iielp iii Socusing M M P proteolytic iictiviiy. I3csi(lcs tlicir nhility oF tlcgiadiiig tlie EC'M io ciuate space for inigi~i l i i ig I:('. :ic~ivc MMI': I ) gciicratc pro-iiiigratory IX 'M liagiiiciits; 2 ) piacess integriiis (siicli ;is twD3) iii ii iiiaiiiier tliat iiicreiiscs thrir al l inity br die ECM aiid, therchy. ilicii. pio-;iiigiogciiic ;ictivitics; 3 ) coiivcrt lateiit 1i)riiis oS iiiigiogciiic cytokiiies (e.g. itiicrlciikiii-X ;iiid ttanslì)riiiiiig growtli Ihctor) iiito ai1 active forni; aiid 4) rcleasc ECM- or ccll siirliicc-hoiiiitl iiiiiaiiiinatory nicdiators or angiogenic factors (iiicludiiig FGF ;iiid Vli(;l:) i i i io ;i soluhlc 1bi1ii 18.18.191. Moirover. M M P influente EC growtli. This is bccnusc MMI' ciiii break IX'-liC' iidliesioiis (by ciigesting intercellular adhesion rewpiors). ;iii(l dci;icli pcricytes (wliicli pro(liicc EC growth inhibitors) from the outer sidc ol'hloocl vcr;scl ( 20 1.

1.2. Al)crraiit i icw I~lood vcssel fornirtioii iii KS

111 tlic Iicaltliy a<liilt viisculogci~csis is cxtrciiicly nirc 121 1, wlii lc angiogenesis occurs alniost cxclusivcly iii associiition with tissue wpair or feiiialc rcproductive cycles [3]. This is (Iiic io tlic nctivity o f aiigiogenesis repres.wi.s which. b y countcracting angiogenic factors. prcbaiigiogciiic iiitegriiis. and EC'M-degradiiig proteaws. cause EC apoptosis aiidtoi iiiliihii 1iC' i i ioii l i ly. Ncgaiivc irgulatois o f ncw blood vcsscl hrmation comprise natiiial aiitiigoiiists of niigiognic kctois (iingiopoictin-2). soluble foims o f angiogenic factoi rcccpiors (VEGFRI ), inliibitors o f EC'M-degrading cnzynics ('f IMP). ECM coiiipoticiits ('N" I ). or E C M Ii-ligineiiis (angii)siiiiiii and endostiitiii) gciierated by M M P [3.8.321.

Iii cliroiiic iiillammation. inetabolic syndroiiies. or tumors, the balance betwcen ncgriiivc oiid positive i-egulators o f angiogenesis is altercd in favour o f tlie positive oiies. aiid abrrixnt iiew blood vessel fonnation occurs (31. Tliis is particularly proininciit in KS, a disciise c1iaractcrij.r.d by multiple skin or mucosal ang i~~ro l i fe ra t ive lesions, which in tiiiic evolve iiito iiodulai-, angiofibrosarcomatous lesions [iavicwed in rcf. 331. Indced. the

Giovanni ~arillari ri

Table 1. Features of blood vessels present in hte-stage KS lesions

- Vessel BM is discontinuous: vascular permeability is increased, activated leukocytes and plasma-deriv

ECM proteins are recruited in KS lesions

- Vessel wall is composed ofa mosaic of EC, E-KSC, and PB-KSC which align together in a

VE-cadherin-dependent fashion

- Vessels are very wtable, parily because of their high expression of angiopoietie2

- Vessels are tortuous, occasionally have a dilated diameter, or are compressed by proliferating KSC

Abbreviations are: BM, basement membrane; EC, endotheliai ceUs; ECM, ex~acellular matru; E-KS(', endothelial- KS cells; KSC, KS cells; PB-KSC. peripheral blood- KS cells; VE-cadherin, vascular endothelial- cadhenn

formation of new capillaries is véfy intense ai al1 stages of KS progression and in ali KS forms, including classical, post-transplant, Afiican or AIDS-associated KS [24-271. h the initial stage of KS progression, newly formed capillaries have a regular, continuous wall [24,26]. Soon afterwards, however, the vessel BM becomes discontinuous and locally disrupted: this leads to an increase in vascular permeability that, in turn, causes the recruitment of activated leukocytes and plasma-derived ECM proteins (including fibrin, fibronectin, and vitronectin) in KS lesions [25,27] (Table 1). Within a short time, proliferating celis with an elongated ("spindle") shape appear in the lesions, becoming the KS histological haiimark (KS cells, KSC) [23].

KSC are a quite heterogeneous cell population (Figure l). In fact, some of them express macrophage markers such as PAM-L CD6fLand D14 @naaqha& KsCI M-KS~>,while%hers display markers of activated EC (endothelial KSC, E-KSC) of either blood vessel or lymphatic origin [26,28]. EC markers include Fdctor VIiI-related antigen (NIII-RA), Ulex europaeus lectin-l (UEA), vascular endothelial (VE)-cadherin, platelet endothelial cell adhesion molecule (PECAM>l (or CD31), intercellular adhesion molecule (1CAM)- 1, D2-40, LYVE-1, VEGFR3 andlor the pro-angiogenic ephrin- l [26,28,29]. Finally, some KSC express both macrophage and endothelial markers, resembling the so-called endothelial macrophages of lymph node sinuses [26,28]. i

Irrespective of these phenotypic traits, KSC present in al1 forms of KS, in either . early- or late -stage lesions, are highly proliferating and invasive [30,31]. Atter the pre- existing ECM of the peri-vascular area has been destroyed, invading KSC secrete a provisional rnatrix which is sirnilar to that praent in early-stage wound healing, being composed by fibronectin, tenascin and collagen-VI [32,33]. Owing to the expression of severa1 pro-angiogenic integrins [26,28], KSC adhere on to the provisional matrix as a support for migrating and proliferating. At the same time, KSC internalize plasma proteins, including fibrinogen [34].

In late-stage KS lesions, KSC become the predominant cell type [23]. Bundles of collagen-I line the spaces between KSC, while BM cornponents including larninin and collagen-IV are distributed at the interface between KSC or EC and the collagen-I bundles [25,35]. At this stage of KS progression, KSC loose EC rnarkers such as FVIII- RA and UEA [36] and, unlike KSC 6om early stage lesions, they can display features of transformed cells [37-391. in addition, KSC establish stable VE-cadherin-mediated cell- to-ce11 contacts with EC or other KSC forming channels which allow blood influx. These chamels (the abnormal vessels of late-stage KS), are quite unstable due to the high expression of angiopoietin-2 [33,35,40]. Because of the above mentioned features, KS

firination of new blood vessels in Kaposi sarmrna

Figure 1. KS windle cells (KSC) expms both endothelial and macrophage markers. KSC present in a human KS lesion stain iòr the endothelial ceii marker CD3 1, the activated endothelial cell inarker ICAM-I, and the macrophage marka CD68 (irnmunohistochemistry, alkaline phosphatase iuiti-alkaline phosphatase, 100 X magnification). Modified From Ensoli et a l [23], and reproduoed in part by kind pennission &om Elsevier Company.

abnormal blood vessels may favor thrombosis, andlor compromise immune cells homing and binding to EC, extravasation and cytotoxicity. Moreover, these aberrant vessels are tortuous, and sometimes present with a dilated diameter or are compressd by proliferating, densely aggregated KSC [33,35,40]. This impairs blood supply, causes hypoxia and acidosis, and interferes with drug delivery, thus rendering KSC resistant to cytotoxic drugs [3].

Although the intense new blood vessel formation which characterizes KS is generally believed to depend upon neo-angiogenesis [24-271, a particular form of vasculogenesis occurs in KS patients [41,42]. Specifically, adherent elongated, KSC-like cells differentiate fiom cultured peripheral blood mononuclear ceUs of patients with KS or individuals at risk for developing KS [41-451. As fòr lesional E-KSC, the peripheral

e

Giovanni Banllari et al.

blood KSC (PB-KSC) express both ECP and EC markers (including CD34, UEA, FVIIi- RA, PECAM-l, ICAM-l, VEGFR2 and endothelial nitric oxide synthase), or endothelial macrophage markers (VE-cadherin and CD68 or the rnannose receptor), while they are negative for the leukocyte markers CD14 and CD45 [41-451. Most likely, PB-KSC are mobilized 60m the bone marrow into the blood in response to angiogenic growth factors released by either KSC or activated leukocytes infiltrating KS lesions [41,43,46-541. This may explain the presence of multiple KS lesions at different body sites. These cells are believed to be KSC progenitors which precede over KS development, and evidence indicates that they can initiate KS lesions when transrnitted by otherwise KS Eee organ donors to tmnsplant recipients [41-451. Furthermore, as for ECP [55], the spindle-shaped morphology and the expression of endothelial intercellular adhesion molecules [41,42] could allow PB-KSC to align and form, together with lesional EC andior E-KSC, cord- like structures which will ultimately differentiate into the new, abnormal vessels.

2. Mechanisms for KS-associated new blood vessel formation

The intense, aberrant formation of new capillaries which characterizes KS results from the concurrent action of seueral factors, rnany of which are also involved in angiogenesis assaciated with other tumors andlor microbial products (Table 2).

Table 2. Pathways leadmg to KS-associated abnormal new blood vessel formation.

- Ras onwgene adivation: upregulation of VEGF expression

- PTEN bctional loss: P13 WAkt activation, EC invasion and proliferation

- p53 inadivation: reduction of p53-induced, anti-angiogenic TSP- I; increase in p53-repressed,

pro-angiogenic Bcl-2, FGF-2 and MMP-I

- Hff-la stabilization: up-regulation ofVEGF and VEGFR expression

- Increase in inflammatory mediators levels: I) up-regulation of pro-angioge~c chemokines and angioge~c

factors promoting angioproiifaative lesion development; 2) increase in the number of cuculating PB-KSC;

3) acquirement of the KSC phenotype by EC; 4) KSC proliferation; 5) up-regulation and activation of

MMP and other ECM-degrading proteases

- EC andlor KSC infection by HHVI: I) NF-kB. AP-I, Ets-I and HIF-I adivation upregulating the expression

of angiogenic chemokines, angiogenic factors, and MMP; 2) onwgene deregulation and tumor suppressor

functional loss leading to ECI KSC invasion, survival, uncontrolied growth and, probably. transformation

- HIV-I Tat release by acutely infected lymphocytes andlor macrophages: stimulation ofECIKSC survival,

growth and locomotion through a molecular mimiay of ECM niolecules, VEGFM higgering and

extracellular-bound FGF release into a diffusible form

Abbreviations are: EC, endothelial cells; ECM, extracellular matrix; FGF, fibroblast growth factor; HIF, hypoxia inducible factor; HHVI, human herpesvirus 8; KSC, KS cells; M W , matrix metalloproteinase; NF- kB, nuclear factor kappa-B; PB-KSC, penpheral blood-KS cells; P13K. phosphatidylinositol-3 kinase; PTEN, phosphatase and tensin homolog; TSP, thrombospondin; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor recepor.

Lutnuiion of new blood vessels in Kaposi sarwma

I, Oncogenes and oncosuppressors

l umor angiogenesis can be preceded by proto-oncogene mutation into oncogenes, a / o r by the fùnctional impahent of tumor suppressor genes [3]. In KS lesions, the Ras mogene is activated, and this could contribute to the up-regulation of VEGF expression d c c t e d in al1 KS foms [56]. Moreover, the hnction of the phosphatase and tensin hci~iiolog (PTEN) tumor suppressor is often lost in KS [57]: as for other types of tumor,

i \ followed by the activation of the phosphatidylinositol-3 kinase (PUK)/Akt puthway, leading to EC invasion and proliferation [58]. Of note, the hypoxia inducible kccor (H1F)-la which also plays a key role in tumor angiogenesis by activating both V I (iF-A and VEGFR2 gene transcription, is expressed in KS lesions, and this directly tcliitei to KS clinica1 progression [59]. However, at variance with many human tumors (601, the p53 tumor suppressor is not mutated in KS [61]. Nevertheless, the p53 protein is

5 cxpressed at very low levels in early-stage KS lesions [62,63], andlor it is found in the ; cytoplasm of KSC, suggesting p53 functional inactivation [64]. Furthemore, although : p53 protein leveis progressively increase in late-stage KS lesions [65,66], this is

niwciated with the expression of Mdm-2, a potent inhibitor of p53. function [67]. oii.sistently, the p53-induced anti-angiogenic TSP-1 is reduced or absent in KS lesions

(20.681, while the p53-Uihibited pro-angiogenic Bcl-2 is highly expressed [63]. Loss of p53 function could also explain the increase in FGF and MMP levels accornpanying KS tlcvelopment and progression [69,70].

2.2. Inflammatory cytokines and angiogenic factors

Chical evidence indicates that al1 individuals with KS or at nsk for developing KS (AlDS-KS patients included) are characterised by a disturbante of the immune system iliat leads to CD8+ Tceli activation [48,52,71-731. Early lesions 6-om al1 KS forms are itifiltrated by a large number of activated leukocytes which release high levels of iiiflarnmatory mediators, including interleukin (IL)-1, IL6, tumor necrosis factor (TNF), iind interferon (1FN)-y [48,50,!52,53]. As for infiltrating leukocytes, EC activated by iiitlammatory cytokine produce severa1 chemokines (MCP-l, RANTES, IP-10, MIP-1% iind MIP-l P) which attract granulocytes, lymphocytes, monocytes and macrophages [46]. I'hese inflarnmatory cells, in tum, produce additional cytokines and chemokines, thus iimpiifjmg the inflarnrnatory reaction associateci with KS initiation. At this time, KSC are also usually detected in lesions. Importantly, inflarnmatory cytokines either favour the irans-differentiation of peripheral blood mononuclear cells into PB-KSC [41,43], or induce EC to acquùe features of the KSC phenotype, including the elongated shape, the expression of ICAM-l, and the disappearance of FVIII-RA [48,51,74,75]. As for cytokine-activated EC, KSC also produce chemokines acting in an autocrine fashion on KSC, which express both the CXCR and CCR chemokine receptors [76-801. Finally, inflammatory mediators stimulate EC, leukocytes, lesional KSC, and PB-KSC to synthesise angiogenic factors which, in tum, promote the development and progression of KS angioproliferative lesions.

In KS lesions, the most abundant angiogenic factors are FGF-2 and VEGF-A [47,51,54]. These two potent promoters of angiogenesis are produced and released by infiltrating leukocytes, KSC, and inflarnrnatory cytokine-activated EC, and they are present at high levels in sera eorn KS patients [78,81]. FGF-2 stimulates the rnotility and growth of both EC and KSC, while VEGF-A enhances FGF-2 effects, and is responsible for vascular permeability and edema formation [47,51,54]. Other angiogenic molecules

108 Giovanni ~ariiian e/ al.

expressed in KS lesions are scatter factorthepatocyte growth fkctor and platelet-denved growth factor, which both stimulate KSC locomotion and growth [49,53], and insulin-like growth factor, that stabilixs HIF-le hence promoting VEGF-A expression [59]. Platelets adhere to vessels damaged upon KSC and EC invasion, and release angiogenic factors including VEGF, platelet-derived endothelial cell growth fkctor and transforming growth factor /3 [82]. In such a milieu, recmited granulocytes release platelet activating factor, which, in turn, promotes KSC locomotion [83].

Inflammatory mediators and angiogenic factors synergize in activating the expression andlor function of pro-angiogenic enzymes with a role in KS progression. Among thern are cyclooxigenass2 and endothelial nitric oxide synthase [84,85], both expressed at high levels in lesions from al1 forms of KS [86,87]. Other enzymes up- regulated by inflammatory mediators andtor angiogenic factors are ECM-degrading proteinases. In particular, activated EC, leukocytes or KSC ( h m al1 forms of KS, at either early or late-stages) synthesize and secrete high levels of plasmifiogen activators (PA) andlor MMP including MMP-I, -2, -3, -7, -9, and -13 [3 1, 8 1,88-921 (Figure 2). With regard to MMP-2 and MMP-9, high levels of these enzymes are also detected in sera from KS patients [81,88,92]. Both PA and MMP are key, not only for invasion, but also for the proliferation of EC or KSC. In fact, MMP break intercellular adhesions and release ECM-bound FGF-2 andlor VEGF-A in a diffisible f o m [89-911.

A

C

Figure 2. The malrix meialloproleinase (MMP)-2 is highly expressed in KS lesions. Thin sections of a human KS lesion had been hybridised with a 3%-radiolabeled antisense MMP-2 RNA probe, as previously descnbed [92]. Bnght field and dark field show MMP-2 expression at the periphery (panels A and B) or in the center (panels C and D) of the nodular lesion. Modified from Toschi et al. [92], and reproduced in part by kind pemission 6om the American Society for Ce11 Biology ( ASCB).

&nni~on ofnew blood vesscls in Kaposs sarwma

Humnn herpesvirus 8 (HHV8)

tlllV8 is present in a latent fonn in EC and KSC in al1 forms of KS [93]. PB-KSC IHO infected and, because of theu capability of supporting produttive viral niion. they represent a reservoir of HHV8 [42-441. Evidence also indicate that

mtiiatory cytokines upregulated in KS induce HHV8 reactivation in latently infected 43 1. This leads to viremia and spreading of the vixus to al1 circulating cell types eg monocytes and T cells [43,52]. Reactivation of HHV8 in inflammatory cells iting KS lesions rnay also be key for the infection of resident EC and KSC, and ad to the increased vual load that is obsetved as lesions progress fiom early to late-

mluliir stages [43,94,95]. Inflammatory cells also promote the recruitment of HHV8 h f c c ~ d cells into tissues via the activation of the vascular endothelium and induction of &c~nokines [96,97]. Finally, viral proteins with potential paracrine effects on KS lesion -. Lsntiaiion are produced d u ~ g lytic gene expression [97].

IlHV8 deeply modifies EC physiology, favoring KS-associated abnormal i~~yiogenesis. In particular, as compared to contro1 EC, HHVS-infected EC, or EC aaposd to HHV8 products, synthesize low amounts of the anti-angiogenic TSP-1 [98], rtnl produce high levels of angiopoietin-2, pro-angiogenic chemokines, VEGF, cyclooxigenase2 and MMP [46,85,99-1061. Consistent with these fmdings, injection of IlilV8-infected EC in mice promotes the development of angioproliferative KS-like

The acquirernent of such a strong angiogenic phenotype by EC depends on HHV8 ciipability of activating the Nuclear Factor-kappa B (NF-kB), Ap-l, Ets-l and HIF-la iriinscription factors [46,99,103,104,106,108-11 l]. At the same time, HHVS deregulates

3 i ~ f t i ~ ~ ~ e ~ ~ ~ ~ i m p a i r s t h _ e PTEN, W L and p53 cuinor suppressors, leading to EC invasion, migration, survival and uncontrolied growth lh1.109,112-1161. Noticeably, upon infection by HHV8 or exposure to HHV8 proteins, I C no longer express typical endothelial surface makers, and acquire an elongated shape 199,102,110,111,117]. This phenotype, together with the increase in invasive and iingiogenic properties, makes HHV8-infected EC very sirnilar to KSC.

Indeed, ihe long-lasting functional deregulation of cellular transcription factors, proto-oncogenes andlor oncosuppresors promoted by HHV8 could favor EC or KSC iransformation, and KS progression toward a real tumor. In fact, although KSC are considered to represent KS tumor cells, they display features of transfonned cells only when they are obtained fiom latsstage KS lesions [37-391. This suggests that KS, @or to evolve into a tme sarcoma, initiates as a reactive process. Indeed, the fmding that in early KS lesions only a small fraction of EC or KSC is infected by HHV8, whereas in la% stage lesions most EC and KSC are infected [93,118-1201, strongly supports a pathogenetic role for HHV8 afier KS initiation. In this regard, it has to be highlighted that the activation of CD8+ T cells triggering KS development can be fbllowed by an impairment of the immune cytotoxic responses [l 2 1,1221, which may favor the escape of HHV8-infected EC and KSC fiom immune suweillance.

2.4. HN-1 Tat protein

The higher incidence and aggressive nature of KS in HIV-infected individuais suggests that HN-1 could have a role in KS development and progression [123]. However, the hding that EC infection by HIV occurs only in specific anatomica1 districts and that this is followed by EC death [reviewed in ref. 1241, indicateti that i f H N

Giovanni Barillari et al.

had any role in KS pathogenesis, this should be induect. Consistently, severa1 in vitro and in vivo studia had shown that the Tat protein of HTV-l, a trans-activator of the vira1 gene expression, effectively links HIV-l infection and the highly aggressive AIDS-KS [reviewed in ref 1251. in fact, upon its release by acutely infected cells, Tat promotes the motility and growth of activated EC and KSC [125]. These angiogenic, KS-promoting effects are mediated by two Tat regions which recognize receptors expressed on activated EC or KSC surface: the arginine-glycine-aspartic acid sequence, which binds pm- angiogenic integrins, and Tat highly basic residues that interact with VEGFR2. Tat binding to these receptors triggers intracellular signals leading to ce11 survival and PA/MMP activation [125]. In addition, Tat basic residues compete with ECM-bound FGF-2 for heparin-binding sites, retrieving the sequatered angiogenic factor into a soluble form which, in turn, promotes angiogenesis and KSC proliferation [125]. In fact, Tat synegises with FGF-2 in inducing angioproliferative, KS-like laions in mice [125].

3. Pathogenetic and anti-angiogenic therapies for KS

Treatment of KS must be detemined according to the patient status and to hisl her overall disease. However. at Dresent there is no uniform treatment for KS. nor is there a . . consensus regarding best treatnitnt. Local therapies, such as surgical removal, laser therapy, cryotherapy, radiation therapy, or topical chemotherapy are generally indicated for patients with early disease, characterised by limited extension and confined to the skin [126,127]. Although these therapies are easy to perform and relatively safe, KS recurrences are frequent. Patients with wide-spread or recurrent KS r'equire extensive treatment including radiotherapy and systemic cytotoxic chemotherapy. As a genera1 nile, the same treatment modalities apply to al1 KS forms, while response rates and their duration may vary [128-1301. However, the adverse si&-effects of cytotoxic chemotherapy are a major concem, particularly in elderly KS patients [128-1301.

in re&nt years, progress & understanding the mwhanism underlying KS , pathogenesis has fostered a great number of preclinical and clinical studies based on the use of molecules directed against KS pathogenetic targets. The strong association of HHV8 with KS, including individuals at nsk of developing KS [93], suggested that drugs directed against this virus could be effective in KS treatment andtor prevention. in one study, the risk of developing KS was decreased in individuals receiving ganciclovir, while foscarnet-treated patients had a delayed KS clinical progression [131]. Severa1 other studia, however, found no correlation between anti-HHV8 therapy and a decreased risk of KS development [l 32-1 341.

Since angiogenesis plays such a prominent role in KS development, other clinical trials were designed at targeting pro-angiogenic cytokines and MMP (Table 3). Other anti-KS therapeutic approaches employed anti-angiogenic molecules including thalidomide, IFNa, or IG12. SpecificaUy, phase 11 clinical studies of oral thalidomide (100-1000 mgiday for 8 weeks)showed a dosedependent response in 3540% of treated AIDS-KS patients [135,136]. Thalidornide inhihits FGF-2 and VEGF activity [4], and reduces TNFa production by monocytes, while it increases IL-12 expression [137]. Sirnilarly, AIDS-KS treatment with IFNa (at doses as high as 20 million UIm2 body surface &ea) led to tumor reduction in up to 38% of patients [138]. This is consistkt with results f?om in vitro studies indicating that IFNa decreases both FGF-2 and MMP-9 gene expression [137], inhibits H N replication and HHV8 reactivation [139-1411, and increases NK and monocyte-mediated cytotoxicity against KS-denved targets [138]. More promisingly, L 1 2 (100-625 ngkg, subcutanwusly, twice a week) gave a response

% Itirmation of new blood vessels in Kaposi sarcoma

Table 3. Anti-angiogenic molecules evahiated for KS treatment.

IFN-a: immunomodulator. antiviral, FGF-2 ami MMP-9 inhibitor

IL-12: immunomodulator, IP-IO and Mig inducer - Thalidomide: anti-infiammatory, FGF and VEGF inhibitor

. IM862: VEGF inhibitor; promoter of NKcell activity

. TNP-470: inhibitor of retinoblastoma protein

Imatinib mesylate: inhibitor of c-kit and platelet-derived growth factor receptor

Sorafenib and sunitinib: tyrosine kinase inhibitors - Bevacizumab: monoclonal antibody to VEGF - COL-3: inhibitor of MMP activity, EC and tumor ce11 invasion

9.

- Protease inbibitors: anti-infiammatory; inhibitors of MMP advity. EC and t u m cell invasion

Ahbreviations are: EC, endothelial celi+ FGF, fibroblast growth facta; IFN, intafmn; IL, interkukin; Il'. interferon-inducible potei& MMP, madx metalloproteinase, NK cell; natura1 killa cell; VEGF, vascular endothelial growth fador, VEGFR, vascular endothelial growth factor receptor.

n,, of 71% in AIDS-KS patients [142]. This muld depend on IL12 capabiiity of iricreasing cytotoxic T lymphocyte and NK cell activity [142], and inducing the ' pridudion of the anti-angiogenic hctors IP- l O and Mig [l 43,1441.

i Additional anti-angiogenic compounds which were evaluated in clinical trials in AlDS-KS include TNP-470, a synthetic analogue of fùmagillin tùnctionally impairing the rctinoblastoma protein [l451 and imatinib mesylate, which targets c-kit and platelet- tlcrived growth factor receptor signalling [146]. In particular, when admmistered at cicalating doses (10-70 mg/m2, iv weekly), TNP-470 promoted a partial response in 18% ol'early-stage AIDS-KS patients, and was well-tolerated even at the highest dose tested 11471. Imatinib mesylate oral treatment (600 mglday) resulteci in partial clinical regression in 50% of KS patients [146]. Interestingly, the hding that only leukemia cells displaying wild-type p53 respond to imatinib [148], suggested that imatinib-resistant KSC have inactive p53. A conha to ry phase ii study is in progress. Further anti- angiogenic molecules under evaluation in AIDS-KS patients include the tyrosine kinase inhibitors sorafenib and sunitinib [149], the VEGFR-2 inhibitor SU5416 [150], and bevacizumab, a monoclonal antibody to VEGF [151].

Recently, clinical and epidemiological studies reported a reduced KS incidence and risk of progression in HIV-infected individuals receiving PI-based therapeutic regimens 194,152,1531. By efficiently blocking the production of infectious HIV particles [154], 'in fact, P1 reduce the deregulation andfor the functional impairment of the immune system, which are key to AIDS-KS initiation and progression, respectively [155]. In addition, P1 capabiiity of suppressmg HIV replication [154], suggested that the P1 inhibitory effect on AIDS-KS possibly depends on a decrease in the levels of HIV-1 Tat protein, a potent promoter of angiogenesis and KS development or progression. However, other work indicated that P1 can also inhibit tumor progression in HIV-fiee experimental models,

* independently of the immune system [156-160; reviewed in ref. 1551. This raised the possibility that P1 muld target cellular pathways involved in tumor induction and/or progression. In this regard, the P1 indinavù (IDV) or saquinavir (SQV) had been shown to block EC and KSC invasion and MMP activity both in vitro and in vivo, at

112 Giovanni Barillari er al.

concentrations which are similar to the steady-state or to the peak concentrations present in plasma from treated patients [161]. Due to these actions, D V or SQV impair angiogenesis, thus inhibiting the development of KS-like lesions, or induce the regression of established KS in anima1 models [161]. These results are supported by P1 capabiliiy of affecting the activity andlor expression of severa1 human molecules [157,159,162-1731.

When investigating on the anti-angiogenic, anti-KS effects of IDV or SQV, i t should be considered that most of the molecules promoting angiogenesis and KS, namely inflammatory cytokines, angiogenic factors, HHV8 proteins or HIV-1 Tat, also activate the NF-kE3 transcription factor [46,99,108,110,111 ,l 59,174-1 831. The main mediator of NF-kB activation is the cellular proteasome, a protease complex that regulates intracellular proteins twnover [reviewed in ref 1841. For example, TNFa or I L l B stimulate the cellular proteasome to degrade the IkB proteins which, by anchoring NF-kB in the cytoplasm, block it trans-activating functions [179, 1841. As a consequence of IkB proteins degradation, NF-kB translocates into the nucleus, and this is followed by the increase in the expression of NF-kB-targeted genes, including those encoding for MMP [ l 75,176,179,185-1921, In agreement with these fuidings, inhibitors of the cellular proteasome reduce or block NF-kJ3 activation, MMP expression and EC or tumor cell invasion in vitro [179,188,190,191,193], and angiogenesis and tumor growth in vivo [194-1961. The &ti-angiogenic effect of cellular proteasome inhibitors is also consistent with the fact that, in addition to MMP, NF-kJ3 activates the transcription of other pro-angiogenic molecules, including PA, IL-8, VEGF and VEGFR [186,192,194,197-2001. Strikingly, SQV was shown to inhibit proteasome function, NF-kB activation, and MMP expression in a wide variety of cell types, EC included [162,165-1691. In contrast, experiments on IDV gave diverging results. Specifically, IDV was found to inhibit proteasome function and NF-kB pro-invasive '

effects in some in vitro or in vivo experimental settings, compnsing models of angiogenesis [l 56,160,166,170,20 1,2021, but to leave proteasome unaffected in other systems [157,162,165,167,172]. Future work will clarify whether the inhibitory effect D V and SQV exert on angiogenesis and KS progression occurs through the functional impairment of EC or KSC proteasome.

Based upon P1 capability of inhibiting KS-promoting events both in vitro and in animal models, we have conducted a phase I1 trial for the b-eatment of classical (HIV- negative) KS patients with DV. The interim analysis indicated that a favourable clinica1 course is associated with high dmg plasma levels, and with a decrease in angiogenesis and tumor invasion surrogate markers including FGF-2 and MMP-2 plasma levels, and circulating EC. As a result, a phase I1 study is underway for the use of P1 in combination with debulking chemotherapy in individuals with advanced, late stage, classical KS.

New blood vessel forrnation and tumor cell invasion are key to the progression of most human cancers [1,3]. Consistently, blocking cellular invasion and angiogenesis prevents tumor formation, and causes necrosis and regression of established tumors [3,5,8,10,13,84,149-15 l]. In agreement with these findings, our recent studies indicate that nude mice treatment with anti-invasive, anti-angiogenic concentrations of IDV or SQV potently inhibits the development and/or progression of a variety of prevalent human tumors including lung, breast, colon, and Iiver carcinoma [155,203]. In this regard, it has to be highlighted that, in addition to inhibiting angiogenesis and tumor cell invasion, P1 can also impaù tumor cell survival and growth [157,169]. These results, together with the large body of data on P1 pharmakokinetics and with the low toxicity of these dnigs, as compared to standard anti-cancer chemotherapeutics, strongly encourage the investigation and exploitation of P1 as nove1 anti-tumor tools.

AC knowledgements

! i'he authors u ish to thank Mrs. P. Sergiatnpielri for editorial assistance. The studies 5 by tlie authors were supported by grants h m the Italian Association for the Research on '

Caticer (AIRC) and 6 o m the National AIDS Research Program, Ministry o f Health, Italy.

Rc fe rences

I . Scmenza GL. Vasculogenesis, angiogenesis, and arteriogenesis: mechanisms of blood vessel ilmnation and remodeling. J Cell Biochem. 2007; 102:MO-847.

2 Betsholtz C, Lindblom P, Gerhardt H. Role of pericytes in vascular morphogenesis EXS. 2005; 115-1 25.

i Nishida N, Yano H, Nishida T, Karnuni T, Kojiro M. Angiogenesis in c a n p . Vasc Health Risk Manag. 2006; 2:213-219.

4 Ensoli B, Sgadan C, BariUan G. Monini P. The fihoblast growth factor. in the Cytokine Handbook (W Edition); Thomson A.W. & Lotze M.T. Editors. Elsevier Science Ltd., London W).

5 . Ferrara N. The role of VEGF in the regulation of physiological and pathological angiogenesis. EXS. 2005; 209-23 1.

h. Stupack DG. The biology of integrins. Oncology (Williston Park). 2007; 2 1 :6-12. 7. Senger DR. Molecular framework for angiogenesis: a complex web of hteractions between

extravasated plasma proteins and endothelial cell proteins induced by angiogenic cytokines. Am J Pathol. 1996; 149:l-7.

H. Rundhaug JE. Matre metalloprotemases and angiogenesis. J Cell Mol Med 2005; 9:267-285. . S w k i K, Enghikl JJ, Morodomi T, Sahresen G, Nagase H. Mechanisms of activation of tissue

procollagenase by maaix metaIloproteinase 3 (stramelysm). Biochemisby. 1990,29:1026 1 -10270. 10. Bakcr AH, Edwards DR, Murphy G. Metalloprotemase inhibitors: biologica1 adions and

therapeutic opportunities. J Celi Sci. 2002; 1 15371 9-3727. I I. Brew K, Dinakarpandian D, Nagase H. Tissue inhibitors of metalloprotemases: evolution,

shucture and function. Biochim Biophys Acta 2000; 1477:267-283. 12. Herman M', Suldiova GK, Kisiel W et al. Tissue factor pathway mhibitor-2 is a nove1

mhibitor of matrix metalloproteinases with implications for atherosclerosis. J Clin invest. 2001; 107:1117-1126.

13. Oh J, Takahashi R, Kondo S et al. The membrane-anchored MMP mhibitor RECK is a key neguiator of extracellular matrix integnty and angiogenesis. CelL 2001 ; 107:789-800.

14. B m k s PC, Silletti S. von Schalscha Ti, Fnedlander M, Cheresh DA. Disruption of angiogenesis by PEX, a noncatalytic metaiioproteinase fragment with integrm binding activity. Cell. 1998; 92:39 1400.

15. Siiletti S. Kessler T, Ooldbcrg J, Boger DL, Chenesh DA. Disuption of matrix metalloprotemase 2 binding to integrin alpha vbeta 3 by an organic molecule mhibits angiogenesis and tuma growth m vivo. Proc Natl Acad Sci U S A. 2001; 98:119-124.

16. Galvez BG, Matias-Rornan S, Yanez-Mo M, Sanchez-Madrid F, Amyo AG. ECM regulates MTI-MMP localization with betal or alphavbeta3 integrins at distinct cell capamnents modulating its mternalization and activity on human endothelial cells. J Cell Biol. 2002; 159:509-52 1.

17. Yu Q, Starnenkovic I. Localization of matrix metalloprotemase 9 to the ceU surface provides a mechanism for CD44-mediated tumor invasion. Genes Dev. 1999; 13:35-48.

18. Hendrix MJ, Setlor EA, Meltzer PS a al. Expression and functional significance of VE- cadherin in aggressive human melanoma ceiis: role in vasculogaiic mimicry. Proc Natl Acad Sci U S A. 2001; 98:8018-8023.

19. Sefior RE, Seftor EA, Kirschmann DA, Hendrix MJ. Targetmg the tumor microenvironment with chemically modified tetracyclines: inhibition of laminin 5 gamma2 chain promigratory fhgments and vasculogenic mimicry. Mol Cancer Ther. 2w2; I : 1 173-1 179.

114 Giovanni Barillari et al.

20. Raffetto JD, Khalil RA. Matrix metauoprotemases and their inhibitors m vascular remodelmg and vascular disease. Biochem Pharmacol. 2008; 75346-359.

21. Zammaretti P, Zisch AH. Aduk 'endothelial progenitor cells'. Renewmg vasculature. Int J Biochem CeU Biol. 2005; 37:493-503.

22. Nyberg P, Xie L, Kalluri R. Endogenous inhibitors of angiogenesis Cancer Res. 2005; 65:3967-3979.

23. Ensoli B, Sgadari C, Barillan G, Sirianni MC, Sturzl M, Monini P. Biology of Kaposi's sarcoma Eur J Cancer. 2001; 37:125 1-1269.

24. Becker J, Schuppan D, Rabanus JP, Gelderblom HR, Reichart P. Immunoelectron microscopy shows an atypical pattem and a quantitative shifi of coiiagens type 1, I11 and VI in oral Kaposi's sarcoma of AIDS. Virchows Arch A Pathol Anat Histopathol 1991 ; 419:237-244.

25. Dictor M, Bendsoe N, Runke S. Witte M. Major basement membrane components in Kaposi's sarcoma, angiosarcoma and benign vascular neogenesis J Cutan Pathol. 1995; 22:43541.

26. Kaaya EE, Castanos-Vela E, Amir H et al. Expression of adhesion molecules in endernic and epidemic Kaposi's sarcoma. Histopathology. 1996; 29:337-346.

27. Nerlich A, Zietz C, Schleicher E. Distribution of basernent mhbrane-Gsociated heparan sulfate proteoglycan in idiopath% and ADS-associated Kaposi's sarcoma Pathol Res Pract. 1991; 187:444450.

28. Uccmi S, Ruco LP, Monardo Fet al. Co-expression of endothelial cell and macrophage antigens in Kaposi's sarcoma cells. J Pathol. 1994; 17323-3 1.

29. Ogawa K, Pasqualmi R. Lindberg RA, Kain R, Freeman AL, Pasquale EB. The ephrin-Al ligand and its receptor, EphA2, are expressed d u ~ g tumor neovascularization. Oncogene. 2000; 19:60436052.

30. Lhcoln DT, Singl PK, Al-Banaw A. Growth hormone in vascular pathology: neovascularization and expression of receptors is associated with cellular proliferation. Anticancer Res. 2007; 27:420 1-42 18.

3 1. Simonart T, Degraef C, Mosselmans R et al. Early- and late-stage Kaposi's sarcomadaived ceUs but not activated endothelial ceUs can mvade deepidermized dermis. J Invest Dermatol. 2001; 1 16:679-685.

32. Becker J, Schuppan D, Reichart P. The extracellular matru m oral Kaposi sarcorna (AIDS): the immunohistochemical distribution of collagens type W , V, VI. of procollagens type I and iìl, of larninm andof undulm. Virchows Arch A Pathol Anat HistopathoL 1987; 412:161-168.

33. Bendelac A, Kanitakis J, Chouvet B, Thivolet J. Basement membrane in Kaposi's sarcoma: an immunohistochemical and ultrasimctural study. Pathol Res Pract. 1985; 180:626432.

34. Massarelli G, Scott CA, Mura A, Tanda F, Cossu A. Hyaline bodies in Kaposi's sarcoma: an immunocytochemical and uhrastnictural study. Appl Pathol. 1989; 7:26-33.

35. Kramer RH, Fuh GM, Hwang CB, Conant MA, Greenspan JS. Basement membrane and connective tissue proteins in early lesions of Kaposi's sarcoma associated with AIDS. J invest Dermatol. 1985; 845 16-520.

36. Green TL, Meyer JR, Daniels TE, Greenspan JS, de Souza Y, Kramer RH. Kaposi's sarcoma m ADS: basement membrane and endothelial ce1 markers in late-stage lesions. J Oral Pathol. 1988; 17:266-272.

37. Rabkin CS, Bedi G, Musaba E et al. AIDS-related Kaposi's sarcoma is a clonal neoplasm. Clin Cancer Res. 1995; 1 :257-260.

38. Rabkin CS, Janz S, Lash A et al. Monoclonal origin of multicentric Kaposi's sarcoma lesions. N Engl J Med. 1997; 336:988-993.

39. Saikevych IA, Mayer M, White RL, Ho RC. Cytogenetic study of Kaposi's sarcoma associated with acquired immunodeficiency syndrome. Arch Pathol Lab Med. 1988; 1 1 2:825-828.

40. AutieHatmainen H, Kamunen T, ApajaSarkkinen M, Dammert K, Risteli L. Laminin and type IV collagen in different histological stages of Kaposi's sarcoma and other vascular lesions of blood vessel or lymphatic vessel origm. Arn J Surg Pathol. 1988; 12:469476.

4 1. Browning PJ, Sechler JM, Kaplan M et al. Identification and cuhure of Kaposi's sarcoma-like spindle cells from the peripheral blood of human imrnunodeficiency virus-l-mfected mdividuals and norma1 conh.ols. Blood. 1994; 84:27 1 1-2720.

famation of new blood vessels in Kaposi sarcoma 115

, Della Bella S, Taddeo A, Calabro ML et al. Peripheral blood endothelial progenitors as potential reservoirs of Kaposi's sarcoma-associated herpesvims PLoS ONE. 2008; 3 :el520.

1, Monmi P. Colombini S. Sturzl M et al. Reactivation and persistente of human herpesvirus-8 mfection m B cells and monocytes by Th-l cytokines increased in Kaposi's sarcoma. Blood. 1999; 93:40444058.

. 44. Sirianni MC, Uccini S, Angeloni A, Faggimi A, Cottoni F, Ensoli B. Circulatmg spmdle cells: correlation with human herpemims-8 (HHV-8) mfection and Kaposi's sarwma Lancet 1997;

45. Uccini S, Sirianni MC, Vmcenzi L et a l Kaposi's sarcoma cells express the macrophage- associated qtigen mannose receptor and develop m peripheral blood cultures of Kaposi's sarcoma patients. Am J Pathol 1997; 150:929-938.

46. Caselli E, Fiorentmi S, Amici C, Di Luca D, Caniso A, Santoro MG. Human herpesvirus 8 acute infection of endothelial cells induces monocyte chemoattractant pmtem ldependent capillary-like shicture formation: role of the IKKNF-kappaB pathway. Blood. 2007; 109971 8-2726.

47. Cornali E, Zietz C, Benelli R et al. Vascular endothelial growth factor regulates angiogenesis and vascular pemeabiiity m Kaposi's sarcoma Am J Pathol. 1996; 149:185 1-1 869.

48. Fiorelli V, Gendeiman R. Sinanni MC et al. gammaInterferon produced by CD8+ T cells mfiltrating Kaposi's sarcoma mduces spmdle ceUs with angiogenio.phenotype and synergy with human immunodeficiency virus-l Tat protein: an immune response to human herpesvirus-8 infection? Blood. 1998; 91 :956-967.

49. Maier JA, Matiotti M, Albmi A et al. Over-expression of hepatocyte growth factor m human Kaposi's sarwma Int J Cancer. 19%; 65:168-172.

50. Miles SA, Rezai AR, SalazarGonzalez JF et al. AIDS Kaposi sarcoma-derived cells produce and respond to interleukin 6. Proc Natl Acad Sci U S A. 1990; 87:4068-4072.

5 1. Samaniego F, Markham PD, Gendeiman R et ai. Vascular endotheiial growth factor and basic r: fibroblast growth factor present in Kaposi's sarcoma (KS) are induced by inflammatory

cytokmes and synergize to promote vascular permeability and KS lesion development Am J --

Pathol 1998; 152:1433-1443. 51 SùìmiìttG Vhcmzkk f i dY+taLgmma- InWm production m peripheral blood

monmuclear cells and tumor mfiltratmg lymphocytes from Kaposi's sarwma pafiiitsi correlation with the presence qf human herpesvi~s-8 m peripheral blood mononuclear cells and lesional macrophages. Blood 1998; 9 1 :968-976.

53. Sturzl M, Brandstetter H, Zietz C et al. Identification of interleuh-l and plateletderived growth factor-B as major mitogens for the spmdle ceUs of Kaposi's sarcoma: a wmbmed m viiro and m vivo analysis Oncogene. 1995; 10:2007-2016.

54. Xerri L, Hassoun J. Planche J et al. Fibroblast growth tàctor gene expression m AIDS-Kaposi's sarcoma detected by in situ hybridization. Am J Pathol. 1 99 1 ; 138:9-15.

55. Drake CJ, Little CD. Exogenous vascular endotheiial growth factor induce5 malformed and hyperfùsed vessels during embryonic neovascularization. Proc Natl Acad Sci U S A. 1995; 92:7657-7661.

56. Spandidos DA, Kalotaakis A, Yiagnisis M, Varatsos A, Field JK. Ras, C-myc and CerbB-2 oncoprotem expression m non-AIDS Mediterranean Kaposi's sarwma Anticancer Res 1990; 10:1619-1625.

57. Sodhi A, Montaner S, Patel V et al. Akt plays a centra1 role in sarcomagenesis induced by Kaposi's sarcoma herpesvinisencoded G protein-coupled receptor. P m Natl AcadSci U S A. 2004; 1 O1 :482 1-4826.

58. Maeharna T. PTEN: its deregulation and tumorigenesis Biol Pharm Bull 2007; 30:1624-1627. 59. Catrma SB, Botusan IR, Rantanen A et a l Hypoxia-inducible factor-lalpha and hypoxia-

mducible factor-2alpha are expressed in kaposi sarcoma and modulated by msulm-like growth factor-l. CIm Cancer Res. 2006; 12:450645 14.

60. O m M. Regulation of the p53 tumor suppressor protem. J Biol Chern. 1999; 2743603 1-36034. 61. Katano H, Sato Y, Sata T. Expression of p53 and human herpesvinis-8 (HHV-8)-encoded

latency-associated nuclear antigen with mhibition of apoptosis in HHV-8-associated malignancies Cancer. 200 1 ; 92:3076-3084.

116

62. Bagmw R, Ramon M, Kiiim S, Lichtig C, FriedmaaBimbaum R An study of p53 protein expression in classieal Kapesi's sarcoma Am J 18367-370.

63. Dada MA, Chetty R, Biddolph SC, Schneida JW, Gatter KC. The immunoexpressim of bc and p53 m Kaposi's sarcoma Histopathology. 19%; 29:159-163.

64. Flarnini G, Magaiini S, Curigliano G et al. Immunohistocheanica1 analysis of p53 p& transplant recipients with Kapasi's samma J Canw Res Clin Oncol. 1997; 123:240-242.

65. Hodak E, Hannnei i, Femmesser M et al. Diff~aitial expression of p53 and Ki67 pmtehs & classic aad iatrogaiic Kaposi's sarcoma Am J Dematopathol. 1999; 21 :138-145.

66. PiUay P, Ch* R, Rcddy R. Bcl-2 and p53 immunoprofile m Kaposi's sarcama Pathd O n d Res. 1999; 5:17-20.

67. Li JJ, Huang YQ, Coclcerell CJ, Zhang WG, Nicolaides A, Friednm-Kim AE. Exprcasion m# mutaticm of the hima wippressor gene p53 m AiDS-associated Kaposi's sarcoma. Am 3' DgmatopathoL 1997; 19:373-378.

68. Taraboletti G, Beneili R, Borsoiti P et al. T h r o m m d i n - l inhibits Kaposi's sarcoma (KS) cell md HIV-l Tat-induced angiogaiesis and is poory expressed m KS lesions J Pathd 1999, 188~76-81.

69. Galy B, Cresncier L, Zanibellato C, Prats AC, Prats H. Tumour s u p p l e s ~ p53 mhibits hirmaa fibroblast growth factor 2 expression by a post-tmscripticmai mechanism. Oncogare. 2001; 20: 1669-1677. ..

70. Suo Y, Sun Y, Waigex L, Rutta Ji, Brincluerhoff CE, Cheing HS. p53 downregulates hinnan maham-l (Cohgmse-l) gere expshn. J Bio1 Chen. 1999,27411535-11540.

71. RiPardiii O, P h i S, R-te S ei al hnmmobgical activetion marlcas in bis sgum of Anican md Eumpean HIV ' ' rnd smmegaik indivihk AIDS. 19%; 10:1535-1542 9

72. Schlesmger M, C h z a d m c h i a n M et aL A distinctive forni of soluble CD8 is sraad by .i &imulatd CD8+ cells in HIV-l-mfeckd and higti-risk mdividualp. Clm immunol i bmopathol. 1994,73252-260. p$

73. Trniltnimi G. Hatzalris A, Potouridou I et aL The mie of i m m u n m ~ b and immune 3 activation m classic Kaposi's sarcoma Int J Canoer. 1999; 82:817-821. 1

74. BPiUui (3, Sgadari C, Palledino C et al. I n b a t o r y cytoldnea synagize with die H N - Tat 4 pr& to pnmiote angbgenesis and Kaposi's sarcoma via mduction ofbasic fibroblaat p w t h 4 Wor and the aipha v beia 3 mtcgrin. J hnmunol 1999; 163:1929-1935. 4

75. Zietz C, Hotz B, Sturzl M, Rauch E, Pmmg R, Lohrs U. Aortic aidothehm m HN-l m M m : chic mjury, activation, and mcreased leukocyte adhemce. Am J Pathot 1996, & 149:1887-1898. t

76. Bmkmeyer NH, Willers CP, Andes S, Mertms L, Rockstroh JK, Shixzi M. Cytokine profile 'i of HN-positive Kaposi's sarcana derived cells m vitro. Eur J Med Res. 1999; 495-100. :3

r! 77. Hemians P, Gori A, h o n e M, Franchioly P, Clumeck N. Possible role of granubcyta :

m8cr- colony stimulating factar (GMCSF) m the rapid progression of AIDS-reW e

Kaposi's samma lesims m viva Br J Haematol. 1994; 87:413414. 78. Sciacca FL, Stwzl M, Bussolino F et al. Expression of adhaion moleculeq platelet-activating v

facror, and chemokines by Kaposi's sarcorna d a J ImmunoL 1994; 153:4816-4825. 79. Uccmi S, Scarpmo S, Bailarini F ei aL In situ study of chemokine and chernokine-mx@w

expressiai m Kaposi sarcoma Am J Dermatopathol. 2003; 25377383. 80. Wang JF, LN ZY, Anand AR et al. Alpha-chemdcinqediated signal transchidion m hrmian

Kaposi's sarcoma spmdle cells. Biochim Biophys Acta 2004; 1691 :l 29-1 39. 81. Ciankocca M. Cooley l'P, Lee JY ct al. Matrix metaiioprotemase mhibitor COL3 m the

treatment of AIDS-mlated Kaposi's sarcoma: a phase I AiDS maliguancy consmhm study. J Clm Oncol. 2002; 20:153-159.

82. Li J, Zhang YP, K i m a RS. Angiogaiesis m wound repair: angiogeaiic growth factors and the extnicehlar matrix. Microsc Res Tech. 2003; 60:107-114.

83. Bocallino M, Camussi G, Giovane A et al. Platelet-activating factor regulates cadherin- catenin adhesim system expression and beta-catniin phosphorylatim during Kaposi's saroom cell motility. Am J Pathol. 2005; 166:15 15-1522

Florio T, Marini M, V i k V et al. Somabstath inhbits himm angiogaiesis and growth via somatostatin receptar-3-mediated regulatiai of endotheliai nitric oxide synthase wd mitogai- actiwted pmtcm kinase activities Endocrinobgy. 2003; 1441574-1584. Shanna-Walia N, Raghu H, Sadegopon S et al. Cplooxygaiase 2 inducui by Kaposi's sarcorna-associated herpesvinis early during m vitto infectim of targct cells plays a role m the maintamce of latait virai genc expreda. J ViroL 2006; 80:65344552. Rossieh L, Ruocco E, Signorielb G et aL Evfdence of COX-l and COX-2 expressiai m Kaposi's ssrcoma tissues. Ew J Canca. 2007; 43:1232-1241. Waiinw W, Raidl M, Panmia J et ai Nitric oxide synbases in Kaposi's sarcana m expressed predaninently by veosels aad tisue macrophagea Lab Invest. 1998; 78949955. D& BJ, Krown SE, Lee JY. Bmer KS, AbouMl DM. Randomized phase I1 trial of mahix metrilioproteiiiase inhbitor COL3 m AlDS-relateci Kaposi's sarcana: an AIDS Malignancy Cmsortium Study. J Clm OncoL 2006; 24:1389-1394.

89. M e a i P, D'Aleseio S, Serati S et al. The urokmae-type plasminogen activrdor, its receptor and u-PA inhbitor typel d k t m vitro growth aad mvasim of Kaposi's sarcoma and capillary mdothelial cells: mle of HIV-Tat pmtein. int J Oncol. 2005; 27:223-235.

90. Meade-ToIlin LC, Way D, W b MH. Exprcssh of multiple rnatrix metalloprateinases and urokinsee tyRs phmmogen &ivator m adtured K q d sanxnna cella Acta Histochem. 1999; ~~~~316: -

91. Pmtmowitz L, D& BJ, Hemandez-B~~~ntes S, Tahm SR, Dabbous MK. Malrix mdalloppaokiesa, P the pgressim and reg~ssial of Kaposi's samma J Cutan PathoL 20062 33 r193-798,

92. T e B, Banllari 0, Sguhi C et al. Activaticm of m&-metalJoproteinaeaZ and manbmne&ypl~-metakpmteinase m endathelid cclls and mductiai of vascular pem@iii& h vivo by human immunodefichcy virus-l Tat protem and basic fibroblast growdi,facta. MQI Bio1 Ce& 2001 ; 12:2934-2946.

93. Boshoff C, Schuk TF, Kameùy MM et al. Kaposfs mcorna-associated h a p e s h s m f m ai- ~JIUL@I& cella Nat Med 1995; 1 :l2741 278.

94. hipin N, Rubb DC, V, Gain I et aL The mfhraica of highly active antkimvital therapy on AlDS-asmiated b p i ' s sarcoma Br J DermatoL 1999; 140:875-881.

95. S tud q, Bbig,C, Schceier A ct al. Expressiai of HHV-8 1atmcy-assocat.d T0.7 RNA m spmdle cells aud aidothdial cells of AIDS-associatd, classical and Afncan Kaposi's sarcoma. ht J 1.997; 72.68-7 1.

i 96. Bh$16; Zkh C, Hasr B et al. Monocytes in Kaposi's sarcana lesions are prcductively m-by human heqeminis 8. J ViroL 1997; 71:7963-7968.

97. ShifzS M, Zie@ C, Maimi P, Ensoii B. Human hapesvinw-8 and Kaposi's sarcoma: mlatiònship wiih t h e m u ~ concept of tumorigaiesis. Adv Cmca Res. 2001; 81 :125-159.

98. Samc&.w, Skalsky RL, Maldonado Ah4 et aL Identificatiai of cellular gmes targeted by K!&V~lCUdacry~&roR~Aa PLoS Padiog, 2007; 3x65.

99. da tal,^, Surabhi m, Zbao J et al. induction of spindle ce11 morphobgy in humrni vasculir aido@elial ce& by humrri hapesvinis 8aicoded viral FLICE inhibitay protem K13. On+& 2b07; 263656-1660.

100.Pq ,q,., Lee Y,, Sqnanicgo F. Urinary protems with pro-apoptotic and antitumor activity. A'ppto$s. 2 W ; 5:rS -28.

l 01. Q& '&W, XS Ye F, Gao SJ. Kaposi's sarcoma-associated hapesvinis mfectiai pmmotes i u v d h , of primary human mbilical vein aidodielial cells by inducing matrix r n d ~ a s c s . J ViuoL 2007; 81 :7001-7010.

102.S!~iir,, BS), LaMm HL, McFerrin HE et ai, Kaposi's sarcoma associated heipsvirus G- prw c c i u p ~ ~ activation of cyclooxygaiase-2 in vascular aidodielial celis. Viro1 J. m; 4;87.

O 103. ~hg Yc,, JOQ cfi, MU, Lee HR, Jung N. Kaposi's sarcoma-associated hapesvi~s virai m regulatgr fac€or 3 stabiiizes hypoxia-inducible factor-l alpha to induce vascuiar etpJf$d@ goy@ expressiai. Caacer Res. 2008; 68:175 1-1 759.

104;s" 1 & e&&, $, Patel V et al. The Kaposi's sarcomaassociated herpes virus G -m- coipled mé$br upregulates vascuiar aidothelial growth factor expression and secretion

thmugh mitogen-dvated grotein base and p38 pathways actmg cm hypoxia-mducible fbd# laipha Canca Re. 2000,60:48734880.

105.Wang HW, Troaer MW, Lagos D at a l Kaposi sarcama hapesvinis-mduced celiub reprogramming contnbutes to the lymphatic endothelial gaie expressicm m Kaposi s a r d Nat Oenet 2004; 36:687-693.

106.Ye FC, Biackboum DJ, Maigel M et a l Kaposi's sarcoma-associated heapesvmis promotm angbgenesis by mducing angiopoietin-2 expression via AP-l and Etsl. J Virol. 20(M: 81 :3980-3991.

107.Fonsato V, Buttiglieri S, Deregibus MC et al PAXZ expression by HHV-8-inf& endothelial ceiis mduced a pmngbgaiic and promvasive phenoQp. B l d 2008; 1 Il! 2806-2815.

108.Brinkmann MM, G h M, Rambow L, Kieser A Hai-do C. Schuk W. Activaticm oB: mitogen-activated protein kase and G-kappa~ .pathways by a &si's sarcoma-associated herpesvirus K15 manbrane protein. J Virol. 2003; 779346-9358.

109. Cai Q, Lan K, Vama SC, Si H, Lm D, Robertscm ES. Kaposi's sarcoma-aso&ed herpesvmi8 latent pmtem LANA mterads wiih HP-l alpha to upreGlate RTA e x m m during hypoxi& Latencv ccmtrol under low oxyaai conditims J Virol. 2006; 80:7965-7975.

l l~.Gros& C, podgrabmskaS Skobe M, Ganem D. Activation of NF-kappaB by the lataat vFLIP geae of Kaposi's sarcoma-associated h a p e s h s is required for the spmdle shape of vkwmfected endothelial cells and contributn, to their promflammabry phenotype J ViPol. 2006; 80:7179-7185.

11 1.Pati S, Cavrois M, Guo HG et ai. Activation of NF-kappaB by the human herpesvinis 8 chemokine receptor ORF74: evidence f a a parscrine mode1 of Kaposils samina pahpesisesis J ViroL 2001; 75:8660-8673.

112.Gwadc Y. Hwang S, Byun H et al Kaposi's sarcama-associated herpesvirus o p read- h e 50 npresses p53-mduced transcriptional adivity and apoptosis. J Virol 2001; 752 62454248.

113.Nakamura H, Li M, Zarycki J, Jung Ri. inhibition of p53 tmor suppr- by virai mterferoa regulatory factor. J V h L 2001; 75:7572-7582.

114. Seo T, Park J, Lee D, Hwang SG, Choe J. Vira1 interferon regulatory factor 1 of Kaposi's soma-associated herpesvinis bmds to p53 cmd represses p53depmdent transcription and apoptosis. J Virol. 2001; 75:61936198.

115.Si H, Robatson ES. Kaposi's s a r c o m 8 ~ herpesvinisaicoded lataicy-associated nuclear antigen mduces chmmomnai mstability through inhbition of p53 function. J Virol. 2006,80697-709.

116.Tomlinson CC, DamaiPt B. The K1 p t e m of Kms sarcoma-associateti hepesvirua activates the Akt signaling pathway. J ViroL 2004,78: 191 8-1 927.

l 17.Foglieni C, Scabmi S, B h i D et al ProdudTve mfection of HUVEC by HHV-8 is associated wih chaiges campatible wati aigiogenic hansfmations. Eur J H h h a n . 2005; 49-273-284.

1 18.Dupin N, Fisha C, K e l h P et ai. Distriition of human heipesvini.+8 iatently mfected cells m Kaposits sarooma, muhiceniric Castlemants disease, and primary effision lymphoma Proc Natl Acad Sci U S A. 1999,96:45464551.

119.Stashis KA, Zhmg W, Gebhard K et al. Kaposi's samma-associated herpavin~~ gene expression m endothelial (spmdle) tumor cells. J Viro1 1997; 71 :7 15-71 9.

120. Stunl M, Wunklich A, Aschcrl G et al. Human herpesvinis-8 (HHV-8) gene ecrpressim m Kaposi's sarcorna (KS) lesions: an m situ hybridization study. Leukemia. 1999; 13 Suppl1:S110-S112.

12l.Guihot A, Dupin N, Marcelin AG et ai. Low T ce11 responses to human hapesvin~s 8 m patients with AIDS-reiated and classic Kaposi sarcoma J Infect Dis. 2006; 1941078-1088.

122.Thomas M, Bcmame IM, Field S et al. Down-regulation of NKG2D aid NKp80 ligands by Kaposi's sucorna-associated heqesvinis K5 ptects against NK celi cytotoxicity. Proc Natl AcadSciU S A. 2008; 105:1656-1661.

123. Safki B, Johnson KG, Myskowski PL a al. The natura1 history of Kaposi's sarcoma m tfic acquired inununodefic'mcy syndrome. Ann Jnten Med 1985; 103:744-750.

124. Bussolino F, Mitola S, Serini G, Banllan G, Pnsoli B. iataactions between endoiheiial cells andHN-1. iat J Biochem Cell BioL 2001; 33.371-390.

125. Barillari Ci, Ensoli B. Angiogmic effects of atnicelluh humrni immunodef1ciaicy v& type 1 T& pmtem and its role m thc pahgenesis of AIDS-associated Kaposi's mmm. Clm Microbi01 Rev. 2002; 153 10-326.

126. Bodsworth NJ, B W M, Bowm M, Donneil D, Yocum R. Phase iIi vehicle-controlied, multi- =temi snidy of topical alitein& gel0.1% m cutaneom AiDS-related K a p d s sarcoma. A m J Glh DamatoL 2001; 2:77-87.

127.Bmyier B, Rakowsky E, Katz A et aL Tailoring treritment for clwical Kaposi's sarcoma: aqa&msive clinica1 guidtlmca. Int J Oncol. 1999,14: 1097-1 102

128.Ammvibh SA Dematologkai management of Kaposi sarcoma Damatol Nuns 1995; 7: 191-195.

129.Bwnbitla L, Labienocr R, Baneschi V et al. Meditmean Kaposi's sarcom8.m the elderly. A ~ ~ ~ d o m i z c d study of oral etopaside vasus vmblastine. Cancer. 1994,742873-2878.

130. C h SC, im H, Tsao CJ, Tnatment of classiad type Kaposi's m a with paclitaxel. -Rea 2001; 21571-573.

13 1. R o b h & Lugo D, Gec L, Jacobson MA. E- of antivirai drugs used to tmt oytanssabvmis eedsgaa d k w cm subsequent course of pviously diagnosed Kaposi's ssffxma in pnthts with AJDS. J Acquir immune Defii Syndr Hum Retrovirol 1999, 20: *M.*

132.Catagliola D, Mary--so M. Can antivirai agents denasa the mcmma of Kaposi's -P Clinica1 Epidaniology Group frain Centres d'infonnation et de Soms de Phmnunodefieiaice H&. L a n e 1995; 346:578.

133.Glesbpc M& Hoovw DR, Weag S d al. Use of astihapas drugs and the risk of Kaposi's sa&amwlata f b n theMuki~~1ter AiDS Cohm Study. J infect Dia 19%; 173:1477-1480.

134.RisPza a, Andnaii M, Domicoi M et al. Human hapesvims 8 seropositivity aud risk of Kapon's samma and otha acquircd immtll lWency syndrome-related diseases. J Natl Capcet In& 199% 91 :1468-1474.

135.FS &.H- MR, W i e E, PhiUips RH, B w s r M. Activity of thalidomide in AiDS- &sui K@'s mmnn and comiation with HHV8 titre. int J STD AIDS. 1998; 9:75 1-75 5.

136.Lilile W, Wyvill KM, Pluda JM et d AOtivity of thalidomide m AiDS-related Kaposi's -*H3Iar ChcoL 2000; 189593-2602.

137. KìWn SE. Maiagemmt of Kaposi sanxrma: the role of mtaferan and thalidomide. C m Opin OncoL 20011- 13374-381.

138.Kmwn SE, Lee N, Lm L, Fischl MA, Ambinder R, Von Roain JH. Interfaon-alpha2b with pt#easm iohiiitm-bad antireiroviral thexapy in patients with AiDS9ssociated Kaposi sawma aa AIDS maiignamy consortium phase I trd. J Acquir immune Defic Syndr. 2006; 41-:149&53

139.Chmg(4 Rame R, Ditimer D, Gmem D. inflammatory cytokinea and the reactivation of Kaposi's samnna-aaociateci hedpesvirus lytic replication. Virology. 2000; 266:17-25.

140.Ho DD, Haftshqn #L, Rota TR a al. Rscombmant human mterferon alfa-A suppresses HTLV-III reglhth h v i h . Lancet 1985; 1 :602-604.

141.MmW P, Carhi F, sturzl M et al. Aipha mtaferm mhibits hurnan herpesvirus 8 (HHV-8) m c t k a t b m primary efhsion lymphama cells and reduces HHV-8 load m culbind paipbaal bbod mononuclear cells J Virol. 1999; 73:4029-4041.

142.LiitieM. AIDS, T cdls, chemothgapy: HAART-bfalking?Bbod. 2008; 11 13921-3922. 143.S- C, Angiolilio AL, Tosato G. Inhibihn of angiogenesis by mterleukh-l2 is mediated

by tha~mterferm-Mwible protein 10. B M 1996; 873877-3882. 144. S & A C, Farber JM, Angiolillo AL et al Mi& the monokine mduced by mterferon-gamma,

ppmtes tumor neaosis m vivo. Blood. 1997; 892635-2643. 145,Abei &-=m WiTaleiwa N et a l A fumagillin derivetive ceigiogenesis mhibitor, AGM-1470,

inIiiblts a c t i h of cyclmdependait kinases and phosphorylation of retinoblastoma gaie pdu&+but n& pateh tyrosyl phqhorylation or protqncogesie expresiai m ~ s c u l a r cdt#bWcell9. C m Res. 1994,54:3407-3412.

120 Giovanni E&

146.Koon HB, Bublcy GJ, Pmtanawitz L et al bthib-induccd regression of AIDS Kaposi's sarcama J Clin Oncol. 2005; 23982-989.

147.Deaibe BJ, Von Roenn JH, HoldcQ-Wikse J et aL Fumagillm analog m the tr Kapods s a r m a phase I AIDS Clinioai Trial Group study. AiDS Clinical Trial 21 5 Team. J Clin Oncol. 1998; 16:1444-1449.

148. Waidcl HO, de Stanchina E, Cepero E et d Loss of p53 impedes the antileukemic BCR-ABL inhbition. Proc Natl Acad Sci U S A. 2006; 1035'444-7449.

1 4 9 . k g H, Bowai JP. Mdoailar design and clniical &velopmait of VEGFR hase hhibitd.] Curr Top Med Chem. 2007; 7:1379-1393. t . .'

15O.Via LE, GomLangtoa RE, Phida JM. Chical triats rcf'enai resource. Cumnt clinical admmistering the antiangiogaiesis agent SU5416. Oncobgy (Williaon P&). 2000; 14:1312, 1315-3. 1

151. Shih T, Lmdley C. Bcvacizumab: an angiogenesis inhibitor for the treabnait of SOM malignanciea Clin Ther. 2006,28:1779-1802.

152. Lebbè C, Bhrm L, PeW C et a l Clinica1 and biologica1 impact of anhroviral t h q d protease inhibitars m HIV-related Kaposi's sarcoma AIDS. 1998; 12345-F49.

153.Rabkin CS, Testa MA, Huwg J, Von Raenn JH. Kaposi*s &coma -&d non-HodgWs lymphoma incidaicc a n d s in AiDS Ckical Triai Group sbudy participmts. J Acquk hnmune D& Syndr. 1999,21 Suppl 1 :S3 1433.

154.Hexner CW, Piscitelli SC. Caim~on-$irgeted therapy and the fuaire of H N management. AIDS. 2002; 16 Suppi l:SlS3.

155.Mmiii P, Sgadari C, Toschi E, Bariliai G, Ensoli B. Antihnnuur &ects,of mtP.etmvBiti d i q . Nat Rev Canoer. 2004; 4:861-875.

156. Esposito V, Paleacaidolo E, Spugnmi EP et al. Evaluatim of anthmoral propedes of the protesse inhibitor mdmavir m a muriae mode1 of hepatocarcinoma Clin Canccr Rea 1226342639.

157.Gaedicks S, FUatGeier E, ConsEantiniu O et d Anthnor &ect of the himum imm~11odeFwimcy virus protease inhibitor rbavir: induction of tumor+ell apoptoeis amciakd with padmtion of proieasoanal proteoiysis Canca Rea 2002; 62:690 16908.

158.i.aurait N, & Baiard S, Guillamo JS et al. Effects of the protearome mhibitor ritonanr on glioma growth m v h and in vivo. Mol C m m Tha. 2004; 3 :l294 36.

159.Pati S, P e k CB, Dunaine J, Bryant JL, Re& MS Jr., Weichold W. Antitumorigaiic effeds of HIV protmse inhbitor ritonavir: inhibition of Kaposi sarunna Blood. 2002; 993771-3779.

160. Spummi EP, Esposito V, Groeger AM et al. EBects of mdmavir m a prelmiinary p h w I shidy on dogs with stage iii spleaiic hemangiosamxna. h Vivo. 2006; 20:125-127.

161 C, B d a r i O, Tmhi E et a l H N protease inhibiters are potent anti-mgiogenic rnoiecuks and prmote regression of Kaposi sarcoma Nat M d 2002; 8225-232.

162.Ancire P, Groemup M, Klaiemian P et al. An inhibitor of HN-1 protease modulates proteasane activity, antigai pracntation, and T celi respaasea Roc Natl Acad Sci U S A. 1998; 95:13120-13124.

163.Autran B, Carcehin G, Li TS et al Positive &ec& of combined mtiretroviral therapy on CD4+ T ce11 homeostasis and functim m advand H N diseah. Sciace. 1997; 2779 12-1 16.

164.Caron M, Auclair M, Vigouroux C, Glorian M, Fored C, Capeair J. The HIV protease inhbitor mdmavir impairs sta01 regukdory elunent-binding protein-l intmnuclear bcalkation, inhibits preadipocyte difFaaitiation, and induca msulm wsistance. Dia- 2001; 50:1378-1388.

165.De Bamts S, Zakaroff-GUard A, Lafmtm M, Galitzky J, Bourtier V. hhibition of human preadipocyte proteasomal actMty by HIV paoteasc mhibitors or specifi mhibitor lactacystin leads to a d& m adipogenesiq whkh mvolves matrix metalioprotemase-9. J Pharmacol Exp nier, 2007; 32029 1-299.

166. Equils O, Shapiro A, Madak Z, LN C, Lu D. Hwnaa immunodenciency virus type 1 protease mhibitors bbck toil-like receptor 2 (TiR2)- and TLR4hiduced NF-kappaB activation. Antmiiuob A m t s Chwodia. 2004: 48:3905-3911.

167. F h d d i n ~ ~ : ~ a u r e n c e J. H N envelope gpl20-mediated ngulatim of oskaclastogenesis via receptor activator of nuclear tàctor kappa B ligand (RANKL) secretion and its rnodulation by

! Thc fonnation of n m blood vessels in Kaposi sarcoma

certam HiV protease mhibitors thmugh mterferon-gammaRANKL cross-talk. J Biol Chm. 2003; 278.1825 148258.

168. Liang JS, Distler O, Coope~ DAet a l HIV protease mhbitors proted apolipoprotem B fkom degradatiai by the proteasome: a potential mechanism for protease mhibitor-mduced hyperlipidernia Nat Med 2001 ; 7: 1327-1 33 1.

169. Pajonk F,'Hirnmelsbach J, Riess K, Somrner A, McBride WH. The human immunodeficiency virus (HIV)-l protease mhibitor saqumavir mhibits proteasome function and causes apoptosis and radiosensitization m non-HN-associated human cancer cells. Cancer Res 2002; 62: 5230-5235.

170.Piccinmi M, Rmaudo MT, Chiapello N et a l The humm 26s pmteasome is a target of annitiretrovhl agents. AIDS. 2002; 16693-700.

I71.Riddle TM, Schildmeyer NM, Phm C, Fichtaibaum CJ, Hui DY. The HiV protease mhibitor ritonavir mcreases lipoprotem production and has no &t on lipoprotem clearance m mice. J Lipid Res. 2002; 43 :!458-1463. -.

172.Schmidtke G, Emch S, Groettmp M, Holzhutter HG. Evidence for the existence of a non- catalytic modifier site of pepti& hydmlysis by the 20 S proteasome. J Biol Chem. 2000; 27522056-22063.

173.Tovo PA. Highly active antiretroviral therapy mhibits cytokine production m HIV-unmfected subjects AIDS. 2000; 14:743-744.

174.Grosjean J, Kmakidis S. Reilly K, Feldmann M, Paleolog E. Vascular endothelial growth factor signallmg m endothelial ce11 survival: a role for NFkappaB. Biochem Biophys Res Commun. 2006; 340:984-994.

175.Majka S, Maiuire PG, Das A. Regulation of matrix metalloprotemase expressicm by tumor necrosis factor m a murine mode1 of retina1 neovascularizaiion. Invest Ophthalmol Vis Sci. 2002; 43 960-266.

176.Meyer K, Ueberham.E, Gebhardt R. Infhience of organosulphur compounds 6om garlic on the secretion of matrix metalloprotemases and their mhibitor TIMP-l by cultured HUVEC cells. Ce11 Biol Toxicol. 2004,20:253-260.

177.Muddasani P, Norman JC, Elhnan M, van Wipen AJ, Im HJ. Basic fibroblast growth factor activates the MAPK and NFkappaB pathways that converge on Elk-l to contro1 production of matrix metalloprotemese-l3 by human adult articular chondrocytes. J Biol Chem. 2007; 28231409-31421.

178.Pati S. Foulke JS Jr., Barabitskaya O et al. Human herpesvirus 8-encoded GPCR activates nuclear factor of activated T cells and collaborates with human immunodeficiency virus Spe 1 Tat. J Virol. 2003; 775759-5773.

179.Sakai T, Kambe F, Mitsuyama H et al. Tumor necrosis factor alpha mduces expression of genes for matrix degradation m human chondrocytelike HCS-2/8 cells througti activation of NF-ka@: abrogation of the tumor n m s i s factor alpha effect by proteasome inhibitors J Bone Mmer Res. 2001; 16:1272-1280.

180.Song HY, Ryu J, Ju SM et al. Extraceilular HiV-l Tat enhances monocyte adhesion by u p regulation of ICAM-I and VCAM-l gene expression via ROSdependent NF-kappaB activation m astrocytes Exp Mol Med 2007; 39:27-37.

181. Sun Q, Mami H, Chaudhary PM. Kaposi's sarcoma associated herpes virusencoded vira1 FLICE inhibitory pmtem activates transcription fkom HIV-l Long Temiinal Repeat via the classica1 NF-kappaB pathway and tùnctionally cooperata with Tat Retrovirology. 2005; 2:9.

182.Uhmati C, Bugatti A, Giacca M, Schlaepfer D, Presta M, Rusiati M. alpha(v)beta3-mtegrin- dependent activation of foca1 adhesion kmase mediata NF-kappaB activation and motogenic activity by HiV-I Tat m endothelial cells. J Ce11 Sci 2005; 1 18:3949-3958.

183.Vanderrnoere F, El Yazidi-Belkoura I, Adriaenssens E, Lemome J, Hondemarck H. The antiapoptotic effect of fibroblast growth ktor-2 is mediated througti nuclear factor-kappaB activation mduced via mteraction between Akt and IkappaB kinase-beta m breast cancer cells. Oncogene. 2005; 245482-5491.

184.Volker C, Lupas AN. Molecular evolution of proteasomes. Curr Top Microbi01 Immunol. 2002; 2689 -22. I

122 Giovanni B a r M

185. Da. S, Bamji A, Frei E, Chaaejee A. Rapid exp~&on aid activation of MMP upm expome of hurnan breast cancer cells (MCF-7) to fibronectin m serum

186.

187.

humm skm. J Biol Chan. 2001; 276:2234 1-22350. 188.ikebe T, Takeuchi H, Jimi E, Bepp M, Shmohara M, Shirasuna K. Involvemcat"

proteascanes m migration a d matrix mcta~oproteinase-9 production of o d squamais carcinoma int J Cancer. 1998; 77578-585.

189.Ko HM, Park YM, Jung B et al. Invohement of matr8r metaiioproteinae9 m plare activatnig Factor-m&& angiogmesis. FEBS L . 2005; 5792369-2375.

190.L~ Y, Wahl LM. Production of m& metalloprotemase9 by adivated human mmocyt& invohres a phosphatidylmositol-3 kinase/AWMKaipha/NF-kappaB pathway. J Leukoc Biol. 2005; 78259-265.

191.h Y, Wahl LM. Oxidative stress augments the piduction of matrix metall~protemasek~ cyclooxygaiasb2, and prostaglandm E2 b u g h enhancanent of NF-kappa B activity m lipopolysawhide-&ated human primary monocytes. J ImmunoL.2005; 175 5423-5429. ,

192.Sung B, Pandey MK, Ahn KS &&l. A n d i c acid (6nonadecyl saiicylic acid), an mhbbir oE i hdone acetyltrmsfeme, suppresses expression of nuclear factor-kappaB-regulated gant r products mvolved m cell survival, proliferation, mvasim, and mfiammatim dinnigh mhibitfaa 1 of the mhibitory subunit of nuclear factor-kappaBalpha b a s e , leading to potentiation of apoposis. Blood. 2008; 1 1 1 :4880489 1.

193.Jadhav U, Chigumpati S, Lakka SS, Mohanam S. Inhbition of matrix metalloprotemase& , reduces m vitro mvasion and angiogaicsis m human micmv~scular endorhelial ce&. int J OnwL 2004; 25:1407-1414.

1 W. O h a T, Sasaki T, Nakarnura M et al The pmteasome is mvolved m angiogaiesis. Biochun - Biophys Res Cammun. 1998; 246:243-248.

195. Wang BW, Chang H, Lm S, Kuai P, Shyu KG. Induction of matrix me&~lloprotemases-l4 and -2 by cyclical meaimica1 stretch is mediated by tumor necrmis factor-alpha m cubnd h u m . umbiiical vein endothelial a lk . Cardiivasc Res 2003; 5946049.

l%.Williams S, Pettaway C, Song R. Papandnou C, Logothetis C. McConkey DJ. Differential effects of the proteasome inhibitor bort~omib on apoptosis and angiogaiesis m human prostate hunor xenografb. Mol Cancer Tha. 2003; 2:8354343.

197.Chai SU, Chau CH, Lee H, Ho CH, Lm €W, Yang YS. Lysophosphatidic acid upregulates *. expmsion of intaleub-8 and -6 m gnuiubsa-hitein cells through its receptors and n u c h h - k a p p a 8 depeident pathways: implications for angiogenesis of corpus hiteum ami ovarian hypastmiulath syndrome. J Clin Endocrinol M*. 2008; 93 935-943.

198.Meng F, Henson R, Patel T. Chanotherapeutic stress selectively activates NF-kappa B- dependait AKT md VEGF expression m liver canccrdmved endothelial cells. Am J Physiol Cdl PhysioL 2007; 293:C749C760.

199.Mquita J, Mezquita B, Pau M, Mezquita C. Down-regulation of Flt-l gaie expression by the protemome mhibitor MG262. J Cen Biochem. 2003; 89:1138-1147.

200.Tmg Q, Zheng L, Lm Let al. VEW is upregulated by hypoxia-mduced mitogenic factor via die PMWAkt-N E-hp& signalmg pathway. Respu Res. 2006; 7:37.

201. Liuzzi GM, Mastroianni CM, Latroniw T et al. Anti-HIV dmgs decrease the expressiai of matrix metallopiotemases m asimcytes and microglia Br&. 2004; 12739847.

202.Whelan KT, Lm CL, Cella M, McMichael Al, Austyn JM, Rowland-Jones S L The HIV proteasc mhbitor mdinavir reduce immature denclritic celi transendothelial migration. Eur J JmmunoL 2003; 33.5520-2530.

203.Monmi P, Toschi E, Sgadari C et ai. The use of HAART for biologicai tumour therapy. J HiV Ther. 2006; 1 1 33-56.

I C; - i

E .

5; !? 2. . e- : g . . . <; &, -. . .. e '%.

"' e-. %i

k:: &,'

185. Drr, S, Baneji A, Frei E, Chatkajee A Rapid expmsion a d actniation of MMP-2 an upon exposure of human breast cancer cells (MCF-7) t0 fibmnectm m serum free Li& Sci 2008; 82:467476.

186. Gmnd EM, Kagan D, Tran CA et d Tumor necrosis factor-alpha ngulates mesenchymai responses via mitogai-activated protein hase kinase, p38 kappaB m human endomeaiotic epithelial cek. Mol Phmacot 2008; 73:

187.Hm YP, Tusm TL, Hu&es M, Wu H, Garna WL. Transfamimg grow tumm n m i s br-alpha -mediateci mduction and V l y t k d v a t i m of MMP humm skin. J Bio1 Chm. 2001; 2769234 1-22350.

188.Ikebe T, Takeuchi H, Jimi E, Beppi M, Shinohm M, Shinwna K proteasomes m migration and matrix metalloptoteinase-9 production of o carcinoma Int J Cancer. 1998; 77578-585.

189.Ko HM, Park YM, Jung B et al. invokanait of matrEr metahproteinase-9 m pl&W-? activatmg fiidor-in&& angiogaiesis. FEBS Le& 2005; 579:2369-2375.

190.Lu Y, Wahl LM. Produdion of matm metalloprotemasa9 by activated human monocytib mvolves a ph~phatidylinositol-3 kindAWW(alpha/NF-kappaB pa4hway. J Leukoc BM. 2005; 78.259-265.

1 9 I . h Y, Wahl LM. Oxidative stress augments the production of mairix metailoproteinas-k, cyclooxygen-2, and prostaglandm k2 through-mhancanait of NF-kappa B activity m lipopolysaccharide-activated human primary mmocytes. J ImmunoL 2005; 17554236429. n

192. smg B, Pandey MK, Ahn KS &&l. hacardic acid (6iionadecyl salkylic acid), an inhbibor of histonc acetyhsferase, suppresses expwion of nuclear factor-kappaB-regulaied g d produe mvolved m ce11 survival, prolifération, mvasion, and mflawnatiai thmugti mhibitidn of the mhibitory subunit of nuclear factor-hppaBalpha b a s e , leading to potentiation af apoptosis. B M 2008; 1 1 1 :48804891.

193. Jadhav U, Chigumpati S, Laklsa SS, Mohanam S. inhbiticm of mabix metallopmtehase;9 d c e s m vitro mvasion and angiogenesis m humau microvaacular aidothelial cells. Int J Oncol2004; 251407-1414.

1 W. Oikawa T, Sasaki T, Nakamura M et ai. The proteasme is mvoked m angiogamis. Biocheri Biophys Res Commun. 1998; 246:243-248.

195. Wmg BW, Chang H, Lm S, Kum P, Shyu KG. Induction of matrix metalloproteinases-l4 an8 -2 by cyclical mechanical stretch is mediated by tumor nmosis factor-alpha in c u h d humm umbilical vein aicbihelial cells. Cardiovasc Res 2003; 59:46049.

l%.Williams S, Pettaway C, Song R, Papandreou C, Logothetis C, McCankey DJ. Diffemtisl effèds of the proteasome inhibitor bortaamib on apaptosis and augbgaiesis in human prostate tumor xenografts Mol Canca Tha. 2003; 2:835-843.

l97.Chai SU, Chou CH, Lee H, Ho CH, Lm €W, Yang YS. Lywphosphatidic acid upregulates upression of intaleukin-8 wd 4 m granubsa-htein cells through its rrce$ors and nuclear f8ctor-kapjxiB depeident pathways: implications for angiogenesis of corpus luteum wd ovarian hypastimulation syndrome. J Clin Endocrinol Metab. 2008; 93:935-943.

198.Meng F, Henson R, Paiel T. Chemodiaapeutic stress selectively activatea NF-kappa B- dtpendait AKT and VEGF expression in liver mcerdBived endothelial cells. Am J Physiol Ce11 Physid 2007; 293:C749-C760.

199. Mezquita J, Mezquita B, Pau M, Mezquita C. Down-regulation of Flt-l gene e x ~ s i o n by the proteasme inhibitor MG262. J Ceil Biochem. 2003; 89:1138-1147.

200. Tmg Q, Zhaig L, Lm i.& al. VEGF is upregulated by hypoxia-mdd mitogemic factor via the PMWAlb-NE-kappaB sig~aimg paihway. Respir Res. 2006; 7:37.

201. Liuzzi GM, Masttoianni CM, Latronico T et ai. hti-HIV dmgs decrease the expresiai of & metailoproteinases m astrocytes and microglia Br&. 2004; 12739847.

202.Whelan KT, Lm CL, Cella M, McMichael Al, Austyn JM, Rowland-Jones SL. The H N prdease mhbitor mdinavir reduces immature dendritic celi transendothelial migration. Eur J hnrnunoL2003; 3332520-2530.

203. Monini P, T d i E, Sgadari C et ai. The use of HAART for biologicai tumour thaapy. J H N Ther. 2006; 11 53-56.