Neuropilin-1 Regulation of Tumor Vascularization and Growth

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1512

Neuropilin-1 regulation of tumor vascularization and growth

ERIC MORIN

ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6206 ISBN 978-91-513-0495-3 UPPSALA urn:nbn:se:uu:diva-364415 2018 Dissertation presented at Uppsala University to be publicly examined in Rudbecksalen, Dag Hammarskjölds väg 20, Uppsala, Friday, 14 December 2018 at 09:15 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in English. Faculty examiner: Professor Hellmut Augustin (German Cancer Research Center, Heidelberg, Germany).

Abstract Morin, E. 2018. Neuropilin-1 regulation of tumor vascularization and growth. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1512. 69 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0495-3.

Angiogenesis, the formation of new blood vessels from existing ones, is dysregulated during tumor progression as a result of chronic hypoxia and inflammation. Such alterations lead to a lack of vessel hierarchy, and the formation of poorly perfused, leaky and blunt-ended vessels, contributing to disease progression. This thesis explores the impact of neuropilin-1 (NRP1) presentation of vascular endothelial growth factor-A (VEGF-A) to its cognate receptor, VEGFR2. NRP1 presentation of VEGF-A occurs in cis (when NRP1 and VEGFR2 are present on the same cell) or in trans (when molecules are present on adjacent cells). As shown in this thesis, the different modes of NRP1 presentation influence endothelial cell signaling and tumor angiogenesis. The overall aim with the studies has been to identify new biomarkers for cancer survival and potential therapeutic targets. In paper I, we explored if signaling downstream of VEGFR2 was affected by NRP1 presentation in cis compared to trans. Complex formation in trans was readily identified, however, the kinetics were delayed and prolonged, inhibiting VEGFR2 internalization and downstream signaling. Additionally, in vivo tumor studies in mice demonstrated that trans presentation of NRP1 led to early inhibition of angiogenesis and suppressed tumor initiation. In paper II, the presence and clinical impact of trans VEGFR2/NRP1 complexes in human cancer was investigated. We first identified gastric and pancreatic adenocarcinomas (PDAC) as candidates for further investigation. VEGFR2/NRP1 complexes were identified in both tumor types but were more prevalent in PDAC. Trans presentation of NRP1 in PDAC correlated with a reduction in several vessel parameters and tumor cell proliferation. Importantly, this study identified the presence of trans complexes as an independent marker of longer overall survival for PDAC patients. In paper III, we explored the impact of NRP1 presentation modes on renal cell carcinoma (RCC) patient survival. We performed in situ proximity ligation assay (PLA) and immunofluorescence staining on a RCC cohort. Tumor cell NRP1, either trans-complexed with endothelial cell-expressed VEGFR2 as detected by in situ PLA, or alternatively, detected by immunofluorescent staining, was identified as an independent predictor of increased overall survival. These data reinforce the importance of the cell type-specific expression of cancer biomarkers.

Keywords: Angiogenesis, Neuropilin-1, VEGFR2, tumor biology, pancreatic ductal adenocarcinoma, renal cell carcinoma

Eric Morin, Department of Immunology, Genetics and Pathology, Vascular Biology, Rudbecklaboratoriet, Uppsala University, SE-751 85 Uppsala, Sweden.

ISSN 1651-6206 ISBN 978-91-513-0495-3 urn:nbn:se:uu:diva-364415 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-364415)

Little by little, one travels far

- J.R.R. Tolkien

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Koch S.*, van Meeteren L.*, Morin E.*, Testini C., Weström S., Björkelund H., Le Jan S., Adler J., Berger P., Claesson- Welsh L. (2014) NRP1 presented in trans to the endothelium arrests VEGFR2 endocytosis, preventing angiogenic signaling and tumor initiation. Developmental Cell, 28(6):633-46. doi: 10.1016/ j.devcel.2014.02.010 *Equal contribution by the authors

II Morin E., Sjöberg E., Tjomsland V., Testini C., Lindskog C., Franklin O., Sund M., Öhlund D., Kiflemariam S., Sjöblom T., Claesson-Welsh L. (2018) VEGF receptor-2/neuropilin1 trans- complex formation between endothelial and tumor cells is an independent predictor of pancreatic cancer survival. Journal of Pathology, 246(3):311-322. doi: 10.1002/path.5141

III Morin E., Lindskog C., Egevad L., Sandström P., Harmen- berg U., Claesson-Welsh L., Sjöberg E. (2018) Perivascular Neuropilin-1 expression is an independent marker of improved survival in renal cell carcinoma. Submitted manuscript

Reprints were made with permission from the respective publishers.

Contents

Introduction ...... 11 The blood vascular system ...... 11 Angiogenesis ...... 13 Vascular endothelial growth factor signaling ...... 16 Neuropilin-1 ...... 20 Tumor angiogenesis ...... 22 Anti-angiogenic therapy ...... 25 NRP1 in cancer ...... 28 In situ proximity ligation assay ...... 30 Renal cell carcinoma ...... 32 Pancreatic ductal adenocarcinoma ...... 33 Gastric adenocarcinoma ...... 34 Present investigations ...... 37 Aims ...... 37 Summary of the papers ...... 38 Paper I ...... 38 Paper II ...... 41 Paper III ...... 43 Graphical summary of the papers ...... 44 Future perspectives ...... 45 Acknowledgements ...... 49 References ...... 52

Abbreviations

ANG Angiopoietin BMP Bone morphogenetic protein BRET Bioluminescent resonance energy transfer CAF Cancer-associated fibroblasts CCM Cerebral cavernous malformation CDKN2A Cyclin-dependent kinase inhibitor 2A CFP Cyan fluorescent protein CREB cAMP response element binding protein Dll4 Delta-like 4 E Embryonic day EGFR Epidermal growth factor receptor eNOS Endothelial nitric oxide synthase ERK Extracellular signal-regulated kinase FAK Focal adhesion kinase FGF Fibroblast growth factor FOXO1 Forkhead box protein O1 FRET Fluorescence resonance energy transfer GAC Gastric adenocarcinoma Gal1 Galectin-1 GLUT1 Glucose transporter 1 HER2 Human epidermal growth factor receptor 2 HGF Hepatocyte growth factor HIF Hypoxia inducible factor IL Interleukin JAM Junctional adherens molecule KLF Krüppel-like factor MMP Matrix metalloproteinase mTOR Mammalian target of rapamycin NEM Neuropilin-1-expressing monocytes NFκB Nuclear factor κ-light-chain-enhancer of activated B cells NRP Neuropilin PDAC Pancreatic ductal adenocarcinoma PDGF Platelet-derived growth factor PDGFR PDGF receptor PI3K Phosphatidylinositol 3-kinase

PKC Protein kinase C PLA Proximity ligation assay PLCγ Phospholipase C γ PlGF Placental growth factor PTP Protein tyrosine phosphatase RAB Ras-associated binding protein RCC Renal cell carcinoma Sema Semaphorin SH Src homology domain SHB SH2 domain containing adaptor protein B TAM Tumor-associated macrophages TCGA The cancer genome atlas TGFβ Transforming growth factor β TMA Tissue microarray TNFα Tumor necrosis factor α TP53 Tumor protein 53 TSAd T-cell-specific adapter protein VE-cadherin Vascular endothelial cadherin VEGF Vascular endothelial growth factor VEGFR VEGF receptor VE-PTP Vascular endothelial PTP YFP Yellow fluorescent protein ZO-1 Zonula occludens-1

Introduction

The blood vascular system Blood vessels are essential for organ homeostasis by distributing oxygen, nutrients, metabolites, cytokines and hormones, immune and inflammatory cells, and regulating coagulation, inflammation and fluid exchange. Vessels are organized into a tree-like hierarchy. Arteries transport oxygenated blood, branching into smaller arterioles, which slows down blood flow. Arterioles split further before connecting to the capillary bed where blood flow is further reduced to optimize time for gas and nutrient exchange with the surrounding tissue. Capillaries then merge into post-capillary venules and eventually veins, to return de-oxygenated blood to the heart. Although smallest in size, with a diameter of 5-10 µm, capillaries constitute a majority of the total vessel area (75%) (Honkura et al., 2018; Thurston et al., 1998). Due to the diffusion properties of oxygen, all cells in the body, with very few excep- tions, need to be within 100-150 µm distance from a capillary (Augustin and Koh, 2017).

There is considerable variation in vessel properties based on their position in the hierarchy, but also between vessels of the same class in different organs due to the local requirements and function of the particular organ. However, blood vessels share common structural features. As a general rule, the vessel lumen is lined by a single layer of endothelial cells, the endothelium, surrounded by a basement membrane and mural cells, enclosed by a sub- endothelial fibro-elastic connective tissue layer, together forming the tunica intima (Borysenko and Beringer, 1984). Surrounding the tunica intima is the tunica media, consisting of smooth muscle cells and elastin fibers. Depend- ing on the physical forces projected on the vessel, the thickness and level of organization of this layer varies greatly, being thickest in large arteries, and significantly thinner or absent in veins. The outermost layer is the tunica adventitia that consists predominantly of elastic fibers (Pugsley and Tabrizchi, 2000). In contrast to larger vessels, capillaries consist of a single endothelial cell layer, surrounded by a basement membrane and a specialized type of mural cell, the pericyte. Capillary pericytes are embedded in the basement membrane and in direct contact with endothelial cells, allowing a more direct regulation of endothelial function (Hall et al., 2014; Hawkins

11 and Davis, 2005; Tilton et al., 1979). Pericyte coverage varies highly between vascular beds, dependent on organ-specific functions (Sims, 1991).

A central role of the endothelium is to regulate trans-endothelial transport of molecules. The endothelium can be either continuous, fenestrated or sinusoidal. Continuous endothelium is the most common and is present in arteries and veins, and in capillaries of the skin, lung, brain and heart. In capillaries with continuous endothelium, fluid and small molecules pass through para- cellular junctions. Passage of larger molecules is restricted and requires active vesicle transportation or substrate specific transporters. Substrate specific transporters such as glucose transporter 1 (GLUT1) are essential for the transport of nutrients across the blood brain barrier, the least permissive vascular bed (Zlokovic, 2011). Fenestrated capillaries are equipped with fenestrae, transcellular pores of approximately 70 nm in diameter, with a diaphragm across the opening (Rostgaard and Qvortrup, 1997). The size of the pore and properties of the diaphragm allow the size-selective passage of molecules. Fenestrated capillaries are observed in locations with high trans- endothelial exchange and increased filtration such as glomeruli, gastric and intestinal mucosa, and endocrine and exocrine glands (Aird, 2007). Sinusoi- dal endothelium is found in the liver, spleen, bone marrow and some endocrine organs and share similarities with fenestrated endothelium, however, sinusoidal endothelium is discontinuous, the basement membrane is poorly formed and the abundant fenestrae lack diaphragm and have a larger diameter (Aird, 2007; Augustin and Koh, 2017). This adaptation of the endothelium allows free exchange of fluids, larger plasma proteins and chylomicrons (Fraser et al., 1995; Wisse, 1974).

Endothelial cells are elaborately connected to each other by two types of junctions, tight junctions and adherens junctions, which are central to vessel barrier function. The main component of endothelial adherens junctions is the transmembrane molecule vascular endothelial cadherin (VE-cadherin), which forms homophilic adhesions between adjacent endothelial cells (Carmeliet et al., 1999; Corada et al., 1999; Dejana et al., 2008; Montero- Balaguer et al., 2009; Taddei et al., 2008). Tight junctions are assembled by scaffold proteins such as zonula occludens-1 (ZO-1) and consist of a large number of transmembrane proteins spanning between endothelial cells, including occludin, junctional adhesion molecules (JAMs) and claudins (Cummins, 2012; Dorland and Huveneers, 2017; Nitta et al., 2003; Tornavaca et al., 2015). The permeability of the endothelium is highly dependent on the expression pattern of tight junction components. In the central nervous system vasculature and in peripheral arteries, tight junctions are well established and the permeability of these vascular beds is low (Aird, 2007). On the other hand, post capillary venules, which have less dense tight junctions than e.g. arterioles, are the predominant site of induced leakage

12 through agents such as vascular endothelial growth factor (VEGF) and inflammatory cytokines such as histamine and bradykinin (Honkura et al., 2018; Majno and Palade, 1961; Majno et al., 1961; McDonald et al., 1999).

Angiogenesis New blood vessels can form either de novo (vasculogenesis) or from preexisting vessels (angiogenesis). Blood-vessel formation is an essential physiological process in vertebrates and is required both for embryonic development and survival of the adult individual. Vasculogenesis denotes the formation of the primitive vascular plexus through in situ differentiation of hemangiogenic progenitors (Demir et al., 2007). Sprouting angiogenesis is the most studied and predominating form of angiogenesis, and occurs both during development, in adults and in disease. Intussusception meanwhile is another form of angiogenesis, where new vessels form through the splitting of the original vessel.

Sprouting angiogenesis is a complex process that requires endothelial cells to react to stimuli and modify their microenvironment to allow invasive growth of a hierarchal sprout with a probing and migrating tip cell and proliferating stalk cells. This sprout forms a lumen, anastomizes with neighbor- ing sprouts, remodels and stabilizes to eventually become a new functional quiescent vessel (Figure 1).

The most common trigger of angiogenesis is tissue hypoxia and additional triggers include inflammation and tumor cell signaling. Cells experiencing hypoxia release angiogenic signals such as VEGFs, placenta growth factor (PlGF), fibroblast growth factors (FGFs), angiopoietin-2 (ANG2) and other chemokines (Carmeliet and Jain, 2011a; Wang et al., 2017). In response to angiogenic signals, pericytes detach from the endothelium and basement membrane degradation and extracellular matrix (ECM) remodeling is initiated by matrix metalloproteases (MMPs), such as MT-MMP-1, expressed on the endothelial cell surface, which allows the sprout to invade the surrounding tissue (Gálvez et al., 2001; Hiraoka et al., 1998; Koike et al., 2002; Potente et al., 2011).

Activation of VEGF receptor 2 (VEGFR2) by VEGF results in endothelial cell activation, proliferation, migration and tip cell formation (Potente and Makinen, 2017). Tip cell invasion and migration are guided by dynamic filopodia protrusions and result in formation of an angiogenic sprout organized in a leading tip cell and trailing stalk cells. The tip cell migrates along a VEGF gradient towards the hypoxic area, while stalk cells elongate and proliferate to extend the sprout (Gerhardt et al., 2003). MMPs on the tip cell

13 also induce the proteolytic cleavage of proangiogenic factors deposited in the matrix, thus increasing their bioavailability (Lee et al., 2005).

Figure 1. Sprouting angiogenesis. A) Tissue hypoxia (blue) leads to upregulation and release of proangiogenic factors, such as VEGF, that stimulate nearby quiescent blood vessels. B) Endothelial cells become activated and express proteases to de- grade the basement membrane (red). C) Endothelial cells form a sprout with a leading tip cell (magenta) that extends filopodia protrusions to sense its environment. The tip cell migrates along the cytokine gradient towards the hypoxic area. Trailing cells take on a stalk cell phenotype (lilac) and proliferate to support the elongation of the sprout. D) Sprouts may meet and form a new vessel circuit through anastomosis initiated by the tip cells. E) Upon vessel fusion a new lumen is formed. To stabilize the newly formed vessel the endothelial cells release factors to attract mural cells (green) and induce deposition of extracellular matrix to form a new basement membrane. F) Improved perfusion restores normoxia in the tissue. Establishment of blood flow and vessel stabilization promotes vessel quiescence.

14 Tip and stalk cells play distinct roles during sprouting and their identity is governed by Notch signaling. Increased VEGFR2 signaling results in the upregulation of Delta-like protein 4 (Dll4), a Notch receptor ligand. Tip cell- expressed Dll4 induces increased Notch signaling in adjacent endothelial cells, promoting stalk cell behavior. Notch activation downregulates Dll4 and VEGFR2, and upregulates VEGFR1, a VEGF decoy receptor. Thereby the cell’s ability to respond to VEGF is suppressed, establishing the stalk cell phenotype (Hellström et al., 2007; Phng and Gerhardt, 2009). In addition to the Notch-Dll4 axis, Neuropilin 1 (NRP1), a VEGF co-receptor, is a key determinant of tip cell positioning and regulation of sprout formation (Aspalter et al., 2015). Due to the changing microenvironment and fluctua- tion in VEGFR2 activity, the tip cell phenotype is dynamic. Indeed, endothelial cells compete for the tip cell position and tip and stalk cells switch posi- tions constantly during sprouting (Jakobsson et al., 2010).

When to sprouts encounter each other, they may undergo anastomosis to form a new vessel circuit. This process is initiated by the contact of filopodia between separate tip cells, followed by junction establishment through VE- cadherin homophilic interactions. In parallel, lumen formation is initiated, allowing perfusion of the new vessel. Lumen formation can occur by separate mechanisms (Charpentier and Conlon, 2014; Gebala et al., 2016):

I Intracellular lumen formation; inverse blebbing or vacuole coales- cence creates a lumen through individual endothelial cells in a sprout. Upon anastomosis with another sprout, the lumen expands through the newly formed contact and the apical membranes fuse to form a continuous lumenized vessel.

II Extracellular lumen formation; endothelial cells have formed a cord like structure where the cells have established adherens junctions along their interface. The junctions become redistributed from the center of the cord to the circumference, thus allowing a lumen to form between the abutting cells

In order to stabilize and mature the newly connected vessel, endothelial cells need to recruit mural cells and modulate the surrounding extracellular matrix (Potente et al., 2011). Transforming growth factor β (TGF-β) released by endothelial cells promote vessel maturation through recruitment, differentiation and proliferation of mural cells, and also induces mural cell extracellular matrix production (Pardali et al., 2010). Platelet derived growth factor B (PDGFB) secretion from endothelial cells stimulates the PDGF receptor β (PDGFRβ) on nearby mural cells to induce their proliferation and migration (Gaengel et al., 2009). Mural cells express angiopoietin-1 (ANG1) which activates Tie2 receptors on endothelial cells. ANG1 stimulation reduces ves-

15 sel leakiness through junction stabilization and promotes pericyte adhesion, thereby promoting vessel stabilization (Augustin et al., 2009).

After the establishment of a new, functional vessel, endothelial cells convert back to a quiescent state, the default state for the majority of endothelial cells. Quiescent endothelial cells are non-migratory and non-proliferating, and their metabolic activity is reduced. However, they still require low levels of VEGF- and FGF-stimulation to maintain vascular integrity and homeostasis (Lee et al., 2007; Murakami et al., 2008). Quiescence is accomplished through the concerted effect of several mechanisms. The phosphatase PTEN, which is induced by Dll4-Notch signaling, arrests cell cycle progression in the stalk cells by negatively regulating phosphatidylinositol 3-kinase (PI3K)- Akt signaling (Serra et al., 2015). The transcription factor Forkhead box protein O1 (FOXO1), inactivated by the serine/threonine kinase Akt, is also essential for quiescence. FOXO1 suppresses MYC, a strong inducer of endothelial cell proliferation and metabolism, by downregulating its expression and inducing the expression of other MYC suppressors (Wilhelm et al., 2016). Tie2-ANG1 contributes to maintaining vessel quiescence by redis- tributing to cell-cell contacts and interacting in trans with Tie2 on neighbor- ing endothelial cells. This changes the signaling output of Tie2 from stimulating migration and sprouting to promoting survival, due to its close interaction with vascular endothelial protein tyrosine phosphatase (VE-PTP) and VE-cadherin when located at junctions (Fukuhara et al., 2008; Saharinen et al., 2008). Other quiescence factors include cerebral cavernous malformation proteins (CCMs) 1-3 and bone morphogenic protein 9 (BMP9). Additionally, blood flow-mediated shear stress is in itself an inducer of endothelial quiescence and stability. Vessels lacking sufficient perfusion and shear stress regress as a part of the development towards a well-functioning vascular bed (Franco et al., 2015). Shear stress signaling is conferred by the vasculopro- tective transcription factor Krüeppel-like factor 2 (KLF2), which upregulates endothelial nitric oxide synthase (eNOS) and negatively regulates nuclear factor κB (NFκB) and VEGFR2 among a large number of other proteins (Chistiakov et al., 2017; Dekker et al., 2006).

Vascular endothelial growth factor signaling The most potent pro-angiogenic regulatory factor is VEGF-A, a member of the VEGF family of structurally related homodimeric glycoproteins, which also includes VEGF-B, VEGF-C, VEGF-D and placenta growth factor (PlGF) (Cross et al., 2003; Ferrara et al., 2003) (Figure 2). By alternative exon splicing, several isoforms of VEGF-A are formed e.g. VEGF-A121, VEGF-A165, VEGF-A189, and VEGF-A206 (Kowanetz and Ferrara, 2006) of which VEGF-A165 is the predominant isoform (Ferrara et al., 2003). The spatial pattern of isoform expression influences VEGFR2 signaling and en-

16 dothelial cell responses. For example, VEGF-A121 lacks the motif necessary to anchor the ligand to extracellular matrix present in longer slice variants. Moreover, conflicting results have been put forward about VEGF-A121’s capacity to bind the VEGFR2 co-receptor NRP1, but the consensus is that the VEGF-A121 splice variant is not sufficient to induce the VEGFR2/NRP1 complex (Pan et al., 2007b; Soker et al., 1998). Through differential en- gagement of co-receptors e.g. heparan sulfate and neuropilins, VEGF-A splice variant expression can influence VEGFR2 signal transduction kinetics, receptor turnover and degradation (Chen et al., 2010; Fearnley et al., 2016). VEGF-A is essential for endothelial differentiation during embryo- genesis, and inactivation of just one Vegfa allele in mice is sufficient to cause lethality at embryonic day (E)11-12 due to impaired blood-island formation and reduced angiogenesis (Carmeliet et al., 1996; Ferrara et al., 1996).

Figure 2. The VEGF ligand family and their receptors. The VEGF family consists of VEGF-A, VEGF-B, VEGF-C, VEGF-D and PlGF. The ligands bind different members of the VEGFR tyrosine kinase family and their co-receptors NRP1 and NRP2 as indicated by black arrows. Only the processed form (dashed lines) of VEGF-C and VEGF-D can bind VEGFR2. Processed VEGF-C but not VEGF-D can bind NRP2 (adapted from Koch et al., 2011).

VEGF-A exerts its effect through binding to two receptor tyrosine kinases, VEGF-receptor 1 (VEGFR1) and VEGFR2 (Ferrara et al., 2003), which are both expressed on endothelial cells (Fong et al., 1995; Katoh et al., 1995). These receptors consist of seven extracellular immunoglobulin-like domains, a transmembrane domain and a tyrosine kinase domain split by a kinase in- sert sequence (Shibuya et al., 1990). Interestingly, the two receptors are structurally similar but play different roles in angiogenesis. VEGF-A binds to VEGFR1 with a 10-fold higher affinity than to VEGFR2, however, signal

17 transduction by VEGFR1 is only weakly induced by its ligands (Waltenberger et al., 1994). Importantly, alternative splicing of the receptor generates a soluble VEGFR1 protein lacking the transmembrane domain but with an intact VEGF-A binding domain. This endogenous soluble splice variant is able to bind VEGF-A with high affinity and inhibits VEGF induced signaling and is overexpressed during preeclampsia (Kendall and Thomas, 1993; Maynard et al., 2003). VEGFR1’s main function in angiogenesis thus appears to be as a decoy receptor for VEGF-A, thereby acting as a spatial regulator of VEGFR2 signaling and sprouting angiogenesis (Kappas et al., 2008). Interestingly, VEGFR1 may have important roles in non-endothelial cells. Mice lacking VEGFR1 kinase activity show normal vascular development, VEGF-A induced macrophage migration is severely reduced (Hiratsuka et al., 1998).

VEGFR2 is considered the main mediator of VEGF-A’s effect on endothelial cells. VEGFR2 plays an important role in promoting angiogenesis by me- diating endothelial cell proliferation, survival and permeability (Cross et al., 2003; Shibuya and Claesson-Welsh, 2006). Upon binding of VEGF-A, VEGFR2 dimerizes and several tyrosine residues become phosphorylated (Y951, Y1054, Y1059, Y1175 and Y1214) (Matsumoto et al., 2005).VEGFR2 mediates its effects in endothelial cells by activating several signaling pathways through tyrosine phosphorylation sites. Phosphorylation of Y951 mediates the binding of T-cell-specific adapter molecule (TSAd). TSAd in turn binds to the Src Homology 3 (SH3) domain of c-Src (Sun et al., 2012). Through c-Src activation, VE-cadherin is phosphorylated and vascular permeability is induced (Li et al., 2016b). The PI3K/Akt pathway, important for cell motility and inhibition of apoptotic activity, is also activated by VEGFR2 phosphorylation, via the receptor tyrosine kinase Axl, which is activated downstream of TSAd and c-Src (Dimmeler et al., 1999; Eriksson et al., 2003; Gerber et al., 1998; Koch et al., 2011; Ruan and Kazlauskas, 2012). Tyrosine residues Y1054 and Y1059 located on the kinase activation loop are essential for induction of receptor kinase activity (Kendall et al., 1999). Y1175 phosphorylation allows the binding of phospholipase C γ (PLCγ), which mediates the activation of protein kinase C (PKC) and consequently, induction of the extracellular regulated kinase (ERK) pathway, which is central for VEGF-A induced endothelial cell proliferation (Meadows et al., 2001). In addition, PKC phosphorylates heat shock protein 27 (HSP27) and cAMP-response-element-binding protein (CREB) (Evans et al., 2008). Y1175 also binds the SH2-domain-containing adaptor protein B (SHB), which in turn promotes activation of focal adhesion kinase (FAK), of consequence for endothelial cell migration and cell attachment (Holmqvist et al., 2004). Y1214 binds to GRB2 and PI3Kp85, which are adaptor molecules in the ERK and Akt pathways, and activators of MYC. pY1214 signaling is important for endothelial cell proliferation, ves-

18 sel morphogenesis and vessel stability (Testini et al., 2018) (unpublished data).

Binding of VEGF-A to VEGFR2 induces receptor dimerization, activation of the kinase domain and transphosphorylation, which in turn leads to rapid clathrin and ephrin-B2 dependent internalization of the ligand-receptor complex (Lampugnani et al., 2006; Sawamiphak et al., 2010). NRP1 also influences VEGFR2 internalization by guiding the complex to Ras-associated binding protein 5 (RAB5) and early endosome antigen 1 (EEA-1) -positive endosomes for internalization This trafficking is conferred by synectin and myosin VI bound to NRP1’s C-terminal SEA motif (Clegg and Mac Gabhann, 2015; Horowitz and Seerapu, 2012; Jopling et al., 2009). Internali- zation does not terminate signaling, instead, endocytosis directs VEGFR2 to different compartments within the cell, necessary for downstream signaling (Miaczynska et al., 2004).

Protein tyrosine phosphatase 1B (PTP1B), located on the endoplasmic retic- ulum dephosphorylates VEGFR2 expressed on the cell surface (Lanahan et al., 2013). Retaining VEGFR2 at junctions decreases ERK pathway activity (Lampugnani et al., 2006). Endosomal trafficking is therefore central for VEGFR2 signaling. In agreement, arrest of VEGFR2 internalization at the membrane by interaction with NRP1 in trans (protein complex bridging between cells) affects PLCγ and subsequent ERK1 activation (Koch et al., 2014). Dephosphorylated VEGFR2 in endosomes is sorted for recycling (RAB11), a NRP1 dependent process, or for lysosomal degradation via RAB7 endosomes (Ballmer-Hofer et al., 2011). Additionally, a constitutive, VEGF-A independent RAB4 endosome recycling of VEGFR2 has been observed in quiescent endothelial cells (Horowitz and Seerapu, 2012).

VEGFR1 binds VEGF-B and PlGF in addition to VEGF-A. Although tyrosine kinase activity in endothelial cells is weak in response to VEGF-A (Ito et al., 1998), VEGF-B and PlGF have been shown to induce VEGFR1 signaling in sensory nerves (Selvaraj et al., 2015). VEGF-B signaling via VEGFR1 is neuroprotective in the retina and brain of mice (Li et al., 2008).

VEGFR3, the third member of the VEGFR family, is expressed by both blood and lymphatic endothelial cells, and in neuronal progenitors. VEGFR3 binds the unprocessed forms of VEGF-C and VEGF-D, while VEGFR2 binds their processed counterparts. VEGFR3 has been shown to be upregu- lated in tip cells during sprouting angiogenesis (Benedito et al., 2012). VEGFR3 is also essential for regulation of lymphangiogenesis. Both lymphangiogenesis and differentiation of endothelial progenitor cells into lymphatic endothelial cells is dependent on VEGFR3 and VEGF-C (Karkkainen et al., 2004; Zhou et al., 2010).

19 Neuropilin-1 Neuropilin 1 (NRP1) and neuropilin 2 (NRP2) constitute the neuropilin family. NRP1 was originally described as a mediator of axonal guidance during neuronal development through its binding of Semaphorin 3A (Sema3A) (He and Tessier-Lavigne, 1997). NRP1 was later shown to also bind VEGF-A, PlGF and VEGF-B (Soker et al., 1998). NRP2 binds PlGF and processed VEGF-C as well as the VEGF-A145 isoform (Koch et al., 2011). During development, NRP1 is expressed preferentially in arteries while NRP2 is found in veins and lymphatic vessels (Herzog et al., 2001; Yuan et al., 2002). NRP1 is also expressed by non-endothelial cells such as epithelial cells, smooth muscle cells, neurons, neuronal progenitors and immune cells (Chaudhary et al., 2014; Ko et al., 2010; Pellet-Many et al., 2011; Zhao et al., 2011).

NRP1 is a transmembrane glycoprotein consisting of a large extracellular domain with subdomains (a1-a2- and b1-b2 and c) for binding Sema3A, VEGFs and heparan sulfate, followed by a transmembrane domain and a short cytoplasmic domain (Gu et al., 2002) (Figure 3A). The transmembrane domain together with the extracellular c-domain are suggested to be important for NRP1 dimerization (Koch, 2012). The cytoplasmic domain contains a SEA-motif, which enables the PDZ-domain containing protein synectin, an adaptor protein involved in myosin IV-mediated endocytosis, to bind to NRP1 (Cai and Reed, 1999; Naccache et al., 2006). NRP2 shares 44% amino acid homology with NRP1 and is similar in its overall molecular design. However, it binds other members of the semaphorin and VEGF families and has a distinctly different expression pattern. NRP2 is important for neuronal development and lymphangiogenesis (Chen et al., 1997; Kolodkin et al., 1997; Yuan et al., 2002).

Over-expression of NRP1 in transgenic mice is embryonically lethal due to malformations in the cardiovascular system, including an excess number of blood vessels, hemorrhage and heart malformations, as well as neurological defects (Kitsukawa et al., 1995). NRP1-deficient mice die in utero at E12.5- 13.5 as a result of both neurological and vascular defects including disor- ganized blood vessels, deficient capillary networks and cardiovascular defects (Kawasaki et al., 1999).

Figure 3. NRP1 structure and binding partners. A) NRP1 is a single-pass transmembrane glycoprotein. The extracellular domain consists of a1-a2, b1-b2 and c subdomains. The cytoplasmic domain contains a C-terminal SEA-motif allowing the binding of the PDZ-domain of synectin. Synectin in turn binds myosin VI, an adaptor protein for endocytosis and trafficking. B) NRP1 interacts with a number of receptors and their ligands to modulate downstream signaling. Galectin-1 (Gal1) can link NRP1 to PDGFR or TGFβR, leading to increased receptor activation. Text boxes indicate the resulting biologic effect of NRP1 in each interaction (adapted from Koch, 2012).

Although VEGF-A and Sema3A are the most studied ligands, NRP1 also binds other growth factors and proteins such as hepatocyte growth factor (HGF), PDGFB, FGF2, TGFβ and galectin 1 (Gal1) (Klagsbrun et al., 2002) (Figure 3B). NRP1 can influence the signaling of the putative receptors for several of these ligands, and is suggested to promote cell migration in those contexts (Cao et al., 2010; Glinka et al., 2011; Koch, 2012; Zachary et al., 2009). In contrast, binding of Sema3A to NRP1 leads to a repulsive behavior in axonal growth cones (Kolodkin et al., 1997). Sema3A stimulation in neurons induces NRP1 complex formation with the plexin co-receptor family, leading to neuron-specific signaling (Takahashi et al., 1999). Of note, Se- ma3A and 3C suppress neoangiogenesis in a mouse ischemic retina model in a NRP1-dependent fashion (Joyal et al., 2011; Yang et al., 2015).

An essential role for NRP1 in endothelial cells is to form a ternary complex with VEGF-A and VEGFR2, and thereby potentiate and modulate the bioactivity of VEGFR2 and stimulate endothelial cell migration (Soker et al., 2002; Soker et al., 1998; Wang et al., 2003; Whitaker et al., 2001). Further- more, NRP1 is co-internalized with VEGFR2 through clathrin-dependent endocytosis, for which the ability of NRP1 to bind synectin is vital (Salikhova et al., 2008). The cytoplasmic domain of NRP1 is required for the trafficking of VEGFR2 to specific endosomal vesicles, thereby influencing signaling and VEGFR2 turnover (Ballmer-Hofer et al., 2011). Interestingly, NRP1’s cytoplasmic domain is essential for arteriogenesis, the formation of

21 arteries, but dispensable for angiogenesis (Fantin et al., 2011; Lanahan et al., 2013). A NRP1 mutant lacking the 30 amino acid cytoplasmic tail suppresses migration and focal adhesion turnover in endothelial cells in response to VEGF-A (Seerapu et al., 2013).

In addition to migration, NRP1 has been proposed to mediate increased vascular permeability in response to VEGF-A and other molecules, independently of VEGFR2, through a mechanism involving the NRP1 C- terminal domain (Roth et al., 2016). On the other hand, using a similar model to assess permeability, Fantin et al. show that both VEGFR2 and full length NRP1 are required for VEGF-A-induced permeability, while the C- terminal binding protein synectin is not essential (Fantin et al., 2017). Fur- thermore, mice expressing a mutant NRP1, lacking a functional binding site for VEGF-A, undergo an essentially unperturbed development, suggesting that NRP1 has VEGF-A-independent functions during developmental angiogenesis (Fantin et al., 2014; Gelfand et al., 2014). NRP1 may induce angiogenesis independently of VEGF-A by interacting with integrins in endothelial cells. In an integrin-dependent manner, NRP1 promotes activation of Abl1, a non-receptor tyrosine kinase, leading to phosphorylation of integrin- associated downstream targets such as paxillin, resulting in actin cytoskele- ton remodeling, increased cell migration and angiogenesis (Raimondi et al., 2014). Thus, VEGF-A-independent NRP1 functions are essential in vascular development (Raimondi et al., 2014; Roth et al., 2016).

NRP1 expression in immune cells has been associated with an activated state of regulatory T cells (Tregs) and dendritic cells. NRP1-dependent activation of Tregs is induced by TGFβ, but Sema3A, Sema4A and VEGF may also be involved, leading to immune suppression (Chaudhary et al., 2014). NRP1- positive dendritic cells contribute to NRP1-induced Treg activation and proliferation through the prolonged interaction between the two cell types by trans NRP1 homophilic binding (Chaudhary et al., 2014). NRP1 has fur- thermore been implicated in vessel anastomosis through NRP1-expressing monocytes (NEMs), guiding the fusion of vessel sprouts in developing zebrafish and releasing vessel maturating factors such as thrombospondin-1, TGFβ and PDGFB, which recruits mural cells to the newly formed vessel (Carrer et al., 2012; Fantin et al., 2010; Zacchigna et al., 2008).

Tumor angiogenesis Similar to healthy tissues, tumors require nutrients and oxygen as well as ways of disposing waste products. It has been postulated that tumors cannot grow beyond 1-2 mm3 unless they acquire their own vascular supply (Folkman, 1971; Holmgren et al., 1995). This step in tumor progression is

22 called the angiogenic switch (Bergers and Benjamin, 2003; Hanahan and Folkman, 1996). The angiogenic switch can be triggered by hypoxia, which activates hypoxia inducible factor 1α (HIF1α) (Dor et al., 2001), promoting increased tumor expression of VEGF-A and other proangiogenic factors (Shweiki et al., 1992). Oncogenes such as K-Ras, v-Src, human epidermal growth factor receptor 2 (HER2), and epidermal growth factor receptor (EGFR) may also trigger the angiogenic switch through up-regulation of proangiogenic factors such as VEGFs, FGF2, PDGF and interleukin (IL)-8, and down regulation of anti-angiogenic factors such as thrombospondins and statins (Bergers and Benjamin, 2003; Folkman, 1971; Rak et al., 2000). Up- regulation of proangiogenic or downregulation of inhibitory chemokines triggers angiogenic activity of the endothelium, leading to sprouting angiogenesis.

Unlike normal blood vessels in healthy tissues, tumor vessels are chronically stimulated by proangiogenic factors persistently produced at high levels in the tumor, due to the rapid growth of the tumor and chronic hypoxia within the center of the tumor. In a physiological setting, neoangiogenesis ceases when the hypoxic region is sufficiently vascularized, followed by vessel remodeling, stabilization and eventually endothelial quiescence. The continuous angiogenic stimulation in tumors results in irregularly shaped, blind- ended, immature blood vessels, lacking clear hierarchy. Increased branching as well as hemorrhaging and elevated permeability are hallmarks of tumor vessels (Baluk et al., 2005; Bergers and Benjamin, 2003; Nagy et al., 2010). The basement membrane of tumor vessels is often irregular, ranging from abnormal thickness to complete absence. Pericytes, which serve to stabilize capillaries in healthy tissues are scarce and loosely attached or even absent from the tumor vasculature. The faulty organization of the tumor vasculature causes suboptimal blood flow (Bergers and Benjamin, 2003), resulting in further hypoxia and thus elevated VEGF expression, generating a vicious circle of overstimulation and vascular aberrations (Shweiki et al., 1992) (Figure 4).

Due to the poor organization of the endothelial cells and persistent stimulation, cellular junctions are also immature, resulting in increased permeability. The increased permeability results in higher interstitial fluid pressure, further reducing tumor perfusion by obliterating the vessel lumens of fragile newly formed vessels. High interstitial fluid pressure also induces proteins associated with lymphatic metastasis and is correlated with worse prognosis in several tumor types (Rofstad et al., 2014; Yeo et al., 2009; Yu et al., 2014). The heterogeneity of vessel density and perfusion results in uneven response to cancer therapeutics. The hostile hypoxic environment caused by aberrant tumor angiogenesis may select for more malignant tumor cells (Jung et al., 2015). In addition, radiotherapy and many chemotherapeutic

23 approaches rely on the formation of oxygen radicals to kill the tumor cells, however, the hypoxic tumor environment reduces the efficiency of such treatments (Carmeliet and Jain, 2011a).

Figure 4. Tumor angiogenesis. In healthy organs, vessels are organized in an or- dered pattern. Most normal blood vessels have a continuous basement membrane (red) and mural cell coverage (green) which contributes to vessel quiescence and tissue homeostasis. In tumors, chronic hypoxia, low pH and inflammation results in dysregulated angiogenesis. Vessels formed in this environment are immature, lack normal hierarchy and may be blunt-ended. Tumor vessels lack proper mural cell coverage and the basement membrane is irregular or absent, leading to increased permeability and enabeling tumor cell intravasation (adapted from Carmeliet and Jain, 2011).

Tumor cells are not unique in driving tumor angiogenesis. Cancer-associated fibroblasts (CAFs), a cell type originating from the mesenchyme of the affected organ, promote angiogenesis by releasing angiocrine factors such as VEGF, FGF and PlGF, and in addition, act to recruit endothelial progenitor cells (Crawford et al., 2009; Erez et al., 2010).

Tumor associated macrophages (TAMs) and mast cells in the tumor environment contribute to angiogenesis by delivering proteolytic factors, such as MMP9 within the tumor microenvironment, which release matrix-bound proangiogenic factors. TAMs also release PlGF to stimulate the endothelium

24 (Deryugina and Quigley, 2015; Grivennikov et al., 2010; Huang et al., 2011; Rolny et al., 2011). In turn, tumor endothelial cells promote tumor growth by producing tumor stimulating factors such as interleukin-6 (IL-6), IL-8, insu- lin growth factor, TGFβ and FGF2 (Butler et al., 2010).

The tumor microenvironment is typically a highly inflammatory milieu, but an adequate immune-response is seldomly observed. Escape from immune destruction is a hallmark of cancer (Hanahan and Weinberg, 2011). The tumor endothelium has an important role in immune evasion. Normal quiescent endothelial cells play a significant role in the immune surveillance system by regulating leukocyte activation and infiltration into the tissue by up- regulating leukocyte adhesion molecules in response to inflammatory triggers. In tumors, proangiogenic signals downregulate leukocyte adhesion molecules on the endothelial cell surface and render them unable to respond to inflammatory stimuli. In addition, VEGF-A and FGF inhibit the activation of the pro-inflammatory factor NFκB (Klein, 2018).

The inadequate structure of vessels formed by tumor angiogenesis not only promotes the generation of more malignant tumor cells and contributes to evasion of immune response and therapeutics, but also tumor cell dissemina- tion. The harsh tumor microenvironment drives tumor cells towards a mesenchymal phenotype, increasing migratory behavior (Jung et al., 2015). In addition, the lack of proper basement membrane and weakness of endothelial cell-cell junctions facilitates tumor cell intravasation, thereby increasing the risk of metastasis (Bielenberg and Zetter, 2015).

Anti-angiogenic therapy Targeting tumor angiogenesis has been proposed as a therapeutic strategy for several decades (Folkman, 1971). Early studies on fragments of endogenous proteins with known anti-angiogenic effects, such as endostatin, showed great promise in vitro and in vivo, with marked effects on endothelial cell proliferation and regression or even disappearance of tumors (Boehm et al., 1997). Since then it has become apparent that the relationship between angiogenesis and tumor growth is more complex than anticipated. The increased overall survival of the original anti-angiogenic therapies have turned out to be modest in cancer treatment. Today, anti-angiogenic drugs are used either in combination with chemotherapy or as single drugs (Jayson et al., 2016). The primary targets of successful anti-angiogenic therapy have so far been VEGF-A and its receptor VEGFR2. In 2004, Bevacizumab became the first Food and Drug Administration (FDA) approved anti-angiogenic agent (Kowanetz and Ferrara, 2006). Bevacizumab is a VEGF-A-neutralizing antibody, approved for first line treatment of metastatic colorectal cancer, non- small cell lung cancer, ovarian cancer and metastatic renal cell carcinoma

25 (RCC) (Jayson et al., 2016). Treatment with Bevacizumab improves progression-free survival but overall survival benefits are smaller or not observed (Young and Reed, 2012). Bevacizumab was retracted from the treatment of advanced breast cancer due to the small improvement in overall survival (Miles et al., 2013; Sledge, 2015). After showing improved overall and progression free survival, an antibody directed against VEGFR2, Ramu- cirumab, was approved by the FDA for the treatment of gastric cancer or gastro-esophageal junction cancer, non-small cell lung cancer and metastatic colorectal cancer (Fontanella et al., 2014).

Anti-angiogenic tyrosine kinases inhibitors targeting VEGFR2 in combination with other tyrosine kinases are also used in the clinic; Sorafenib, used for RCC, hepatocellular carcinoma and thyroid cancer, Sunitinib, which is used against RCC and pancreatic neuroendocrine tumors, Regorafenib for refractory metastatic colorectal cancer and Pazopanib against RCC and soft tissue sarcomas among others (Jayson et al., 2016). Aflibercept, a fusion protein containing the VEGF binding domains of VEGFR1 and 2 and the Fc domain of human immunoglobulin is approved for treatment of metastatic colorectal cancer in combination with standard chemotherapy (Van Cutsem et al., 2012b) (Figure 5).

Figure 5. Targets of anti-angiogenic therapy. Currently, most anti-angiogenic drugs target different steps of the VEGF-A/VEGFR2 pathway, and some may also target other receptors and pathways in combination. Bevacizumab and Aflibercept both target the ligands of VEGFR2 through different mechanisms. Ramucirumab inhibits the VEGF pathway by directly binding to the ligand-binding site in VEGFR2. Sunitinib and Pazopanib inhibit the tyrosine kinase activity of VEGFR2 and other receptor tyrosine kinases. Sorafenib and Regorafenib also target receptor tyrosine kinases but additionally inhibit downstream molecules such as Raf in the ERK signaling cascade.

26 Current efforts also involve targeting angiogenic pathways other than VEGF-A/VEGFR2. The Dll4-Notch signaling pathway is an attractive target because cancer stem cells as well as endothelial cells have been shown to be responsive to Dll4 inhibition. An anti-Dll4 antibody has shown promising results in phase I clinical trials (Smith et al., 2014). Regorafenib which is approved for refractory metastatic colorectal cancer inhibits Tie2 in combination with VEGFR2 and other tyrosine kinases. The results of Regorafenib treatment are encouraging for further exploration of therapies targeting Tie2 or its ligands ANG1 or 2. Currently, clinical trials are ongoing for Trebana- nib, an inhibitor of ANG1 and 2, and Rebastinib, a selective inhibitor of Tie2 (Cortes et al., 2017; Harney et al., 2017; Monk et al., 2014; NIH, 2018). Efforts are also ongoing to find useful combinations of anti-angiogenic therapy with other current treatment options such as radiotherapy, conventional chemotherapy, DNA repair inhibitors, immunotherapy, and other kinase inhibitors.

One of the suggested rationales of anti-angiogenic therapy is that it preferentially targets endothelial cells, which in general, do not have the plasticity and genetic instability of the cancer cells themselves, thereby making them a more stable and predictable target for therapy (Folkman et al., 2000). In addition, the tumor endothelium is often activated, while most healthy tissues have a relatively quiescent vasculature (Bergers and Benjamin, 2003; Hanahan and Folkman, 1996). By combining anti-angiogenic therapy with conventional treatment options, positive results have been generated. How- ever, a common challenge in anti-angiogenic treatment is therapy resistance, for which several mechanisms are proposed (Bergers and Hanahan, 2008). The plasticity of the tumor allows it to upregulate other pro-angiogenic factors, such as HGF/cMet, FGFs and ANG1, to stimulate VEGF-independent angiogenesis (Casanovas et al., 2005; You and McDonald, 2008). Another resistance strategy that tumors may utilize is local invasion along preexisting vessels into the surrounding tissue, denoted vessel co-option (Rubenstein et al., 2000). In addition, tumors may also become more prone to metastasize as a consequence of anti-angiogenic therapy (Ebos et al., 2009). The high heterogeneity of the tumor vasculature may also contribute to resistance. Nagy et al. identified six different vessel types in solid tumors, and only a subset of these were sensitive to VEGF-A targeted therapy (Nagy et al., 2010). Interestingly, by single-cell RNA-sequencing of mouse tumor endothelial cells, distinct intratumoral gene-expression profiles between cell subtypes was revealed, which were further abrogated by treatment with anti- angiogenic drugs (Patel et al., 2014; Zhao et al., 2018).

An interesting concept in anti-angiogenic therapy is the normalization of the tumor vasculature. As described above, the vasculature of growing tumors is poorly perfused which causes hypoxia. Hypoxia facilitates epithelial-to-

27 mesenchymal transition of tumor cells, preventing senescence and making the cells more invasive and stem-cell like (Lunt et al., 2009; Thiery et al., 2009). Further, the vessel walls are irregular and weakened due to a lack of basement membrane, which might allow tumor cells to enter the bloodstream and metastasize (Mazzone et al., 2009). Anti-angiogenic therapy has been shown to facilitate a transient normalization of the tumor vasculature, resulting in increased perfusion, reduced hypoxia and a more stable vasculature (Batchelor et al., 2007; Carmeliet and Jain, 2011b). In addition, in mouse models, the normalization of the vasculature improved the sensitivity to radiation and chemotherapy (Tong et al., 2004; Winkler et al., 2004). Howev- er, continuous anti-angiogenic therapy may destroy the vessels completely or force the tumor to develop a compensatory mechanism for angiogenesis (Carmeliet and Jain, 2011b). It is therefore proposed that anti-angiogenic therapy is most efficient when given at a dose that leads to vessel normalization rather than at a higher dose, leading to vessel eradication. To achieve vessel normalization is a clinical challenge requiring a finely balanced treatment regimen to achieve effect without treating the heterogeneous endothelium too harshly. Therefore, translation of this therapeutic approach to the clinic requires further titration of both dose and treatment cycles.

NRP1 in cancer Aberrant angiogenesis has been recognized as an indicator of progression and dedifferentiation of tumors (Carmeliet and Jain, 2011b). NRP1 is nor- mally expressed in endothelial cells, epithelial cells and neurons and other cell types and in many cancers forms the tumor cells may also express NRP1 (Jubb et al., 2012). The effect of NRP1 expression differs between tumor types and there is no clear relationship between tissue specific expression and clinical outcome that is consistent between tumor types.

In a study of cutaneous squamous cell carcinoma, NRP1 was found to be expressed by the suprabasal epithelial layers of healthy tissue as well as in tumor cells. NRP1 was more highly expressed in highly differentiated tumors, suggesting that it could function as a reservoir to sequester VEGF-A to the epithelial compartment and thereby limit its bioactivity (Shahrabi- Farahani et al., 2014). Similarly, in colon cancer, tumor expression of NRP1 was associated with better prognosis (Kamiya et al., 2006). On the other hand, in human glioma, breast and prostate cancer, NRP1 expression in the tumor correlated with higher tumor grade and worse prognosis (Latil et al., 2000; Naik et al., 2017; Osada et al., 2004), and in oral squamous carcinoma NRP1 expression correlated with poor prognosis and disease relapse (Chu et al., 2014). In the gastrointestinal tract, increased tumor cell NRP1 expression has been observed in tumors of the pancreas (endocrine and exocrine),

28 gallbladder, ampulla of Vater and colon, as well as in precursor lesions such as Barrett’s esophagus and colorectal adenomas (Fukahi et al., 2004; Hansel et al., 2004). Additionally, tumor NRP1 expression levels increased with histological progression (Hansel et al., 2004). NRP1 is less well studied in hematological malignancies, but a study of acute myeloid leukemia patients showed higher expression in the cancer cells compared to normal hematopoietic cells, and the degree of overexpression correlated with reduced overall survival (Kreuter et al., 2006). A proposed rationale to explain the correlation between higher expression levels of NRP1 and tumor progression is that NRP1 would serve to promote cell migration, as demonstrated in mouse models of colon cancer (Parikh et al., 2004).

Immune modulation is an important capability of tumors. NRP1 is expressed by several types of immune cells, including NEMs, involved in tumorigenesis (Roy et al., 2017). When injected directly into tumors, NEMs contribute to vessel stability and mural cell recruitment reduce tumor leakiness and hypoxia (Carrer et al., 2012). TAMs also express NRP1 and can broadly be divided into two phenotypes, the tumor suppressive M1 and the protumoral M2 type. M1 TAMs secret pro-inflammatory factors such as interferons and tumor necrosis factor α (TNFα), while M2 TAMs secrete immune inhibitory signals such as IL-10 and TGFβ. NRP1 expression is higher in M2 TAMs (Wallerius et al., 2016). NRP1 positive Treg cells also contribute to immune evasion by promoting an immune suppressive environment in the tumor and preventing the recruitment of T-effector cells, shown to be associated with poor prognosis (Chaudhary et al., 2014).

Anti-NRP1 therapy has been shown, in mice, to have an additive effect on tumor growth inhibition when combined with an anti-VEGF antibody. This function of anti-NRP1 therapy is independent of the effect of VEGF-A binding to NRP1, since antibodies directed against the semaphorin binding site of NRP1 had the same effect (Pan et al., 2007a). Due to these promising findings a humanized antibody against the VEGF binding domain of NRP1 in combination with Bevacizumab was tested in phase 1 clinical studies but was discontinued due to higher than expected rates of proteinuria (Patnaik et al., 2014; Weekes et al., 2014).

NRP1 remains an intriguing therapy target and other targeting strategies have been explored. siRNA or miRNA targeting NRP1 have anti-angiogenic effects, reducing tumor growth in vivo in several tumor models (Berge et al., 2010; Raskopf et al., 2010; Wu et al., 2014; Zhang et al., 2012). Other antibodies targeting NRP1 have shown in vivo suppression of tumor growth (Ding et al., 2018; Kim et al., 2017). Small peptides interfering with the NRP1-VEGFR2 interaction by targeting NRP1’s transmembrane domain led to reduced tumor volume and metastatic spread in an in vivo breast cancer

29 model (Arpel et al., 2016). In a similar fashion to Aflibercept, based on the VEGF binding domains of VEGFR1 and 2, soluble NRP1 can act as a trap of several angiocrine factors (Gagnon et al., 2000). The NRP1 ligand Sema3A has potent tumor suppressing effects through a direct effect on endothelial cells but also by recruiting NEMs and regulating the proliferation of TAMs (Carrer et al., 2012; Wallerius et al., 2016). However, Sema3A treatment in vivo resulted in severe diabetic nephropathy and neuronal toxicity (Meyer et al., 2016).

NRP1 is also a target for improved cancer drug delivery, taking advantage of its expression on both endothelial and tumor cells. Generally, in cancer treatment, the delivery of drugs to the tumor is inefficient due to poor penetration. However, small tumor-homing cell-penetrating peptides that bind NRP1 through the exposure of a C-end Rule motif, have been shown to acti- vate an endocytosis pathway in endothelial and tumor cells (Pang et al., 2014). Peptide-binding triggers NRP1-dependent cell trafficking of co- administered drugs, resulting in improved tumor drug penetration and facili- tating the delivery of therapeutic agents (Roth et al., 2016; Sugahara et al., 2010).

In situ proximity ligation assay To understand the function and activity of a specific protein in any biological system, it is necessary to know both the expression pattern and interactions with other proteins. Antibody-based methods are versatile, can be used in many contexts, and may provide a good understanding of spatial distribution and expression levels. However, the resolution in conventional method- ologies is too low to study individual proteins or their interaction with binding partners.

Protein interactions are important to identify and understand protein function and determine suitable targets for therapeutic efforts. Mass spectrometry and yeast two-hybrid assays are valuable in finding interaction partners to the protein of interest, but do not provide information about subcellular localiza- tion or dynamics of the interaction (Ho et al., 2002; Uetz et al., 2000).

Fluorescence resonance energy transfer (FRET) is a powerful method to study protein interactions in cellular systems. The proteins of interest are tagged with fluorescent proteins, for example cyan fluorescent protein (CFP), and yellow fluorescent protein (YFP). Light excitement of CFP can transfer the absorbed energy to YFP if the tagged proteins are in close proximity to each other (<10 nm), resulting in emitted light of a longer wave- length (Sekar and Periasamy, 2003). Bioluminescent resonance energy trans-

30 fer (BRET) is a similar method that uses luciferase as an energy donor and thus does not require an external light source (Angers et al., 2000). One of the strengths of these approaches is the ability to study the interactions in real time, giving valuable information about their dynamics. However, due to the requirement of tagging the proteins, the system is rather artificial and important protein properties might be altered. Moreover, the requirement to modify the proteins of interest makes these methods not applicable to human samples.

Figure 6. In situ proximity ligation assay. In situ PLA can be employed to visualize protein interactions in situ with single protein resolution. A) Detection probes (antibodies with covalently bound oligonucleotides) bind the molecules of interest (red and orange). B) If the two detection probes are in close proximity, the two oligonucleotides can be bridged by specific added oligonucleotides (red and grey). Subsequent enzymatic ligation results in the formation of a circular DNA fragment. C) The circularized DNA fragment is amplified thousand-fold by rolling circle amplification. D) The formed DNA bundle is visualized by addition of labeled oligonucleotides complementary to the generated DNA strand, thereby amplifying signaling strength manifold (adapted from Söderberg et al., 2008).

In situ proximity ligation assay (PLA) is an antibody-based assay that can visualize endogenous protein interactions both in tissue samples and in cellular assays with single protein resolution (Figure 6) (Söderberg et al., 2006). PLA detection probes are primary or secondary antibodies covalently linked to specific oligonucleotides. The oligonucleotides, together with two addi-

31 tional oligonucleotides circularize by enzymatic ligation. The circularized sequence is amplified thousand-fold by polymerase-driven rolling circle amplification. This DNA bundle is visualized by the addition of small labeled oligonucleotides complementary to the circularized sequence, thus amplifying the signal significantly. The bundle remains attached to detection probes during the whole process, allowing the detection of protein-protein interactions or modifications in specific subcellular localizations (Jarvius et al., 2007; Söderberg et al., 2006; Söderberg et al., 2008). In situ PLA can also be used to study the activity of a single protein using one antibody that specifically recognizes the phosphorylated protein together with an antibody that recognizes an phosphorylation independent epitope (Jarvius et al., 2007).

Renal cell carcinoma Renal cell carcinoma (RCC) represents 3% of all cancers (Ferlay et al., 2015). Discovery is often incidental in conjunction with radiography and 25% of patients have advanced disease at discovery. For localized disease, partial or radical nephrectomy is performed. Nephrectomy may also be performed in metastatic disease to improve survival (Rini et al., 2009). Local recurrence occurs in 30% of patients after tumor resection (Cohen and McGovern, 2005). Metastatic disease is not curable despite the introduction of novel therapies in recent years, and the five-year survival is approximately 10-15% (Ghatalia et al., 2017; Lindskog et al., 2017).

Chemotherapy in RCC has a low response rate of 5-6% and provides no benefit to overall survival (Motzer and Russo, 2000; Yagoda et al., 1995). Immunotherapy with IL-2 and interferon alfa was previously the golden standard for treatment of metastatic RCC, although benefits were modest (Rini et al., 2009). Targeted therapies against the VEGF-A/VEGFR2 and mammalian target of rapamycin (mTOR) pathways are currently the first line therapy choices for metastatic RCC. These include Pazopanib, Sunitinib, Bevacizumab (in combination with interferon) and the mTOR inhibitor Temsirolimus (for patients with poor prognosis) (Kumar and Kapoor, 2017). Recently, combinatorial therapy with immune checkpoint inhibitors have shown improved response rate and overall survival in metastatic RCC (Motzer et al., 2018).

Metastatic RCC is one of the most responsive cancers to anti-angiogenic therapy. Indeed, sequential treatment with a second anti-angiogenic kinase inhibitor has been shown to be effective even when the RCC develops resistance to the first choice of angiogenic inhibitor (Escudier et al., 2012). The beneficial effect of anti-angiogenic therapy in RCC can be derived from

32 the molecular pathogenesis of the disease. There are two major subtypes of RCC, papillary and clear cell RCC. Clear cell RCC constitutes approximately 70-80% of RCC cases (Rini et al., 2009). The VHL gene is a tumor suppressor gene which is mutated in families suffering from the von Hippel- Lindau syndrome, a familial cancer syndrome causing retinal angiomas, hemangioblastomas and RCCs with clear cell morphology (Cohen and McGovern, 2005). Loss of one allele of the VHL gene is also observed in over 90% of sporadic clear cell RCCs, and inactivation of the other allele is found in at least 50% of RCCs (Gnarra et al., 1994; Schraml et al., 2002). VHL encodes the von Hippel-Lindau protein (VHL), a central component in an E3 ligase that ubiquitinates HIF1-under normoxic conditions, leading to proteasome degradation. Under hypoxic conditions HIF1- is transferred to the nucleus which leads to the transcription of pro-angiogenic factors such as VEGF-A, PDGF, FGF2 and TGFα. VHL loss or dysfunction leads to absence of oxygen-dependent regulation of HIF1-, resulting in constant production of proangiogenic and tumorigenic factors (Iliopoulos and Eng, 2000).

NRP1 is expressed in renal tubular epithelial cells and podocytes, a kidney specific mural cell type, as well as in the glomerular endothelium and small arteries (Karihaloo et al., 2005; Wnuk et al., 2017). However, NRP1 has not been extensively studied in the RCC context. Cao et al showed that aggres- sive RCC cell lines overexpress NRP1. Knocking down NRP1 resulted in reduced migration and invasion. In vivo, NRP1 tumor cell expression was associated with maintenance of a dedifferentiated phenotype and increased tumor growth (Cao et al., 2008).

Pancreatic ductal adenocarcinoma Pancreatic cancer is a disease with very poor prognosis. The five-year survival is around 6% and pancreatic cancer is the fourth leading cause of death among cancer diseases (Gillen et al., 2010; Kamisawa et al., 2016). Contrib- uting to the poor survival is the delay of discovery due to the lack of symp- toms until the disease is at a late stage. Only about 20% of patients are eligi- ble for resection at discovery, the only potentially curable treatment. Even in this group, 5 year survival is only 25% (Siegel et al., 2014). Another factor contributing to the poor outcome is that 90% of patients have metastatic disease at diagnosis (Kamisawa et al., 1995).

Pancreatic cancer treatment is dependent on the stage of the disease. In resected pancreatic cancer, adjuvant chemotherapy is given, most commonly gemcitabine (Oettle et al., 2013). For metastatic pancreatic cancer, the first line chemotherapy is FOLFIRINOX regimen (fluorouracil plus leucovorin,

33 irinotecan, and oxaliplatin) or gemcitabine plus nanoparticle-bound paclitaxel for those who tolerate it. The benefit of radiotherapy in the treatment of locally advanced disease has not been shown (Kamisawa et al., 1995). Pan- creatic ductal adenocarcinoma (PDAC) is the most common pancreatic cancer tumor type, comprising about 90% of cases (Ferrone et al., 2012). Fre- quently observed driver genes are KRAS , CDKN2A (cyclin-dependent kinase Inhibitor 2A), TP53 (tumor protein 53) and SMAD4 (Wood and Hruban, 2012). Mutations in the proto-oncogene KRAS is one of the earliest events in pancreatic tumorigenesis (Kanda et al., 2012). Mutations in the tumor suppressor CDKN2A is also an early event while TP53 and SMAD4 mutations occur at later stages of cancer progression (Maitra et al., 2003).

In contrast to RCC, PDAC is considered a hypovascular disease, with a high degree of tumor stroma and relatively few tumor cells. However, high microvessel density and high VEGF-A expression levels are prognostic markers, which predict poor patient survival (Ikeda et al., 1999; Seo et al., 2000). A larger number of phase 2 and 3 clinical trials have been performed with anti-angiogenic drugs; unfortunately these trials have so far been unsuccess- ful (Craven et al., 2016).

NRP1 is highly expressed in pancreatic cancer and has been associated with angiogenesis and poor overall survival (Ben et al., 2014; Fukahi et al., 2004). NRP1 has been proposed to contribute to pancreatic tumorigenesis in several ways; studies in pancreatic cancer cell lines and in vivo models have shown that NRP1 expression promotes tumor cell proliferation, migration, invasion and metabolism as well as tumor angiogenesis, chemotherapy resistance and endothelial to mesenchymal transition (Matkar et al., 2018). In contrast, Gray et al. showed that expression of NRP1 led to a reduction in tumor cell migration and also inhibited tumor initiation and growth, while knocking down expression increased tumor growth in vivo (Gray et al., 2005). Currently, a small tumor penetrating peptide, CEND-1, targeting NRP1 through its C-end Rule motif in combination with nabpaclitaxel and gemcitabine is undergoing phase 1 clinical trials (NIH, 2018).

Gastric adenocarcinoma Gastric cancer is the fifth most diagnosed cancer and the third leading cause of cancer related death worldwide (Bray et al.). Similar to PDAC, the disease is often diagnosed at an advanced stage. H pylori infection is the most important risk factor for gastric cancer, driving inflammation-associated onco- genesis (Wang et al., 2014). Gastric adenocarcinoma (GAC) is the most common gastric tumor type and is divided into the diffuse and intestinal subtypes based on histological difference. Diffuse GAC is poorly differenti-

34 ated, lack gland formation and infiltrates the stroma as individual cells. In- testinal GAC is a well or moderately differentiated cancer, which forms glandular structures. Due to their growth pattern and low differentiation, diffuse GAC has a worse prognosis (Lauren, 1965).

Surgical resection is the only curative treatment for GAC. Patients with local advanced disease receive neoadjuvant or adjuvant chemotherapy to improve overall survival. However, the five year survival is 10-15%, and 50-90% of patients eventually die of disease relapse (Paoletti et al., 2010; Van Cutsem et al., 2016). Metastatic disease has a worse prognosis with a median survival of 12 months when receiving first line combination chemotherapy (Glimelius et al., 1997; Wagner et al., 2010). Around 20% of GAC tumors overexpress the oncogene HER2, encoding the human epidermal growth factor receptor 2 (HER2) protein (Van Cutsem et al., 2015). In patients with HER2-positive tumors, receiving Trastuzumab, a HER2-specific antibody, in combination with chemotherapy resulted in improved survival compared to patients receiving chemotherapy alone (Bang et al., 2010). Trastuzumab combined with chemotherapy has therefore been approved as first line treatment of advanced HER2-positive GAC.

Targeted therapy against the VEGF-A/VEGFR2 pathway has also shown some patient benefit. GAC tumors overexpress VEGF-A and other proangiogenic factors, and angiogenesis has been associated with poor prognosis (Becker et al., 2000; Chen et al., 2002). A clinical trial with Bevacizumab (AVAGAST) revealed an improvement in progression free survival, while overall survival was improved only in the non-Asian population (Ohtsu et al., 2011). This finding suggests that the therapeutic benefit of Bevacizumab requires patient stratification. Indeed, when further analyzing the cohort for biomarkers of therapy response, both circulating VEGFR1 and tumor- expressed NRP1 predicted therapy response (Van Cutsem et al., 2012a). A phase 3 clinical trial with Ramucirumab, an antibody targeting VEGFR2, showed prolonged overall survival in advanced GAC compared to best supportive care, and equivalent results to second line chemotherapy (Fuchs et al., 2014). Ramucirumab is therefore recommended as an option in second line treatment for refractory metastatic GAC (Van Cutsem et al., 2016).

NRP1 expression in GAC has been associated with low tumor cell differentiation in patients. Depleting NRP1 in vitro and in vivo showed reduced tumor cell migration, proliferation and tumor growth (Li et al., 2016a). Similar results were seen when using a NRP1-neutralizing antibody (Ding et al., 2018). In contrast, in the AVAGAST clinical trial, van Cutsem et al. observed that low NRP1 expression was associated with poor overall survival. Interestingly, these patients had the most benefit from receiving Bevaci- zumab (Van Cutsem et al., 2012a). Low tumor NRP1 expression may dis-

35 tribute VEGF-A towards the endothelium, making the tumor more dependent on VEGF-A as a driver of angiogenesis, thus more sensitive to anti- VEGF therapy.

36 Present investigations

Aims The work presented in this thesis explores the role of NRP1 spatial distribution in regulating VEGFR2 signaling and the impact of NRP1 presentation in trans for tumor angiogenesis and growth (Paper I). Additionally, VEGFR2/NRP1 complex formation is explored as a marker of patient survival in PDAC, GAC (Paper II) and RCC (Paper III). The specific aims of each paper are listed below:

I. Determine how VEGF-A-induced signaling downstream of VEGFR2 is regulated by NRP1 presentation in cis or trans. Explore the role of NRP1’s spatial distribution in regulation of tumor angiogenesis using in vivo tumor models.

II. Investigate the prevalence of VEGFR2/NRP1 complex formation in human cancers and explore the prognostic significance of NRP1 presentation in trans in GAC and PDAC.

III. Study the impact of NRP1 cis and trans presentation in RCC on clinical outcome. Examine the validity of vessel adjacent NRP1 and trans NRP1 as prognostic markers of survival.

37 Summary of the papers Paper I VEGF-A and its receptor VEGFR2 are essential for angiogenesis both in development and disease. NRP1 binds VEGFs with high affinity and is known to serve as a VEGF-A co-receptor on endothelial cells, forming a ternary complex with VEGF-A and VEGFR2. NRP1 is essential in embryonic development and shows broad expression in epithelial cells, smooth muscle cells, neurons, and immune cells in the adult. NRP1 is also expressed in tumors, however, conclusions on the impact of NRP1 expression varies between studies and tumor types. In addition to forming complexes when both VEGFR2 and NRP1 are expressed by the same cell, cis, NRP1 has also been shown to interact with VEGFR2 when expressed by an adjacent cell, trans, with VEGF-A bridging the complex. Here we explore in detail the consequence of NRP1 presence in cis versus trans on VEGFR2 biology and angiogenesis in both tumor and developmental contexts (Figure 7).

Figure 7. VEGFR2/NRP1 complex formation in cis and trans. VEGF can bridge complex formation between VEGFR2 and its co-receptor NRP1. Complex formation can occur when both molecules are expressed by the same cell, cis. Complexes can also form when NRP1 expressed by an adjacent cell, for example a tumor cell, is presenting VEGF to VEGFR2 on the endothelial cell, trans.

To explore the kinetics of complex formation we performed cell culture ex- periments with porcine aortic endothelial (PAE) cells expressing both VEGFR2 and NRP1 (cis) compared to mixed cultures of cells individually expressing VEGFR2 and NRP1. After VEGF-A stimulation, cis complexes formed within 3 minutes and declined after 30 min. In contrast, trans complex formation was delayed but prolonged, and complexes remained arrested at the plasma membrane. The arrest of complexes at the cell surface suggested that VEGFR2 internalization was modulated by NRP1 in trans. We therefore tracked the trafficking of fluorescently labeled VEGF-A and confirmed

38 that the labeled ligand internalized as expected into vesicle-like structures in cis while internalization in trans was significantly delayed.

Full bioactivity of VEGFR2 requires internalization of the active receptor into early endosomes. We therefore examined signal transduction downstream of VEGFR2. VEGF-A stimulation led to rapid activation of PLCγ and ERK1/2 in cis followed by rapid deactivation. In trans, PLCγ and ERK2 activation was prolonged while ERK1 was not activated at any timepoint, suggesting qualitative difference in VEGFR2 signaling output based on NRP1’s spatial distribution.

The C-terminal tail of NRP1 contains a SEA motif allowing it to bind the PDZ-domain of synectin, implicated to be of importance for VEGFR2 cis complex formation and internalization. We therefore investigated the impact of NRP1’s C-terminal tail on trans complex formation. PAE and T241 fibrosarcoma cells were transfected with a truncated NRP1 lacking the C- terminus (NRP1ΔC). Real time radio ligand binding assays with labeled VEGF-A revealed that NRP1 engages in both high-affinity and low-affinity interactions with VEGF-A. The NRP1ΔC-VEGF-A binding was markedly skewed towards the low-affinity interaction. We observed that NRP1ΔC, unlike full length NRP1, failed to arrest the VEGFR2 at the plasma membrane and did not affect signaling kinetics downstream of VEGFR2.

The effect of NRP1 in trans on endothelial cell signaling suggested a possible role in pathological angiogenesis. We explored this using two tumor models, T241 fibrosarcoma and B16F10 melanoma, subcutaneously injected in mice with tamoxifen-regulated endothelial-specific NRP1 expression. The tumor cell lines were modified to express NRP1 or not. We observed a reduction in tumor growth when tumor cells expressing NRP1 had been administered, despite observing no difference in cell proliferation in vitro. We examined whether the difference in tumor growth was due to abrogated angiogenesis, but observed no difference in CD31 area or mural cell coverage between tumors; however, we observed a reduction in tumor establishment in mice receiving NRP1 expressing tumors cells, suggesting that NRP1 may have a decisive role in tumor initiation, possibly through an early impact on tumor angiogenesis.

To study early events of tumor vascularization, T241 cells suspended in mat- rigel were injected subcutaneously and tumors were extracted already after one week. In this model, we observed a suppression of angiogenesis by tumor cells presenting NRP1 in trans. Similar to the in vitro observation on signaling downstream of VEGFR2, activation of ERK1 was reduced in trans tumors. These data suggest that NRP1 in trans inhibits tumor growth through an early suppression of vessel recruitment and invasion.

39 Endothelial NRP1 expression is essential for mouse development; however, the influence of the spatial distribution of NRP1 requires further attention. Based on the results from the tumor models, we speculated that a VEGFR2- NRP1 interaction between endothelial cells in trans could impact VEGFR2 signaling to maintain a non-sprouting phenotype also in a physiological setting such as developing the retina vasculature. To explore this, we generated mouse pups with mosaic endothelial cell expression of NRP1. Studying the retina stalk region where sprouting should be suppressed we observed that sprouting occurred in mosaic regions, specifically by NRP1 expressing cells whose neighbors lacked NRP1, thus where no lateral NRP1-mediated inhibition could occur.

With this study, we highlight the importance of NRP1 spatial distribution for VEGFR2 signaling and regulation, and consequentially angiogenesis. We also elucidate the role of NRP1 in cis and trans in tumor angiogenesis, showing the relevance of NRP1 presentation by non-endothelial cells such as tumor cells.

40 Paper II NRP1 expression in tumors has been correlated to both poor and improved prognosis depending on the tumor type and method of analysis. Our previous study indicated that tumor angiogenesis in mouse models was influenced by the expression pattern of NRP1, with trans VEGFR2/NRP1 complex formation being associated with reduced angiogenesis and tumor growth. We therefore wished to explore the effect of VEGFR2/NRP1 complexes in trans on human tumor vascular parameters and patient survival.

Having previously observed a reduction in vessel area at an early stage in T241 fibrosarcoma trans tumors, we analyzed these tumors for other vessel parameters and observed a reduction in vessel number and branching, correlating with reduced tumor cell proliferation.

To select candidate human cancer types for further study, we performed in situ hybridization, screening for NRP1 expression in both vessels and tumor cells on a tissue microarray (TMA) of 17 cancer types, with 12 patients for each type. We detected NRP1 expression in both vessels and tumor cells in GAC and PDAC, potentially allowing formation of VEGFR2/NRP1 complexes in cis or trans.

We explored VEGFR2 and NRP1 co-expression using RNA scope, a method to multiplex mRNA detection in situ. We observed a high heterogeneity in the expression of both VEGFR2 and NRP1, but in general, NRP1 expression was more abundant in PDAC samples. Similar results were obtained when performing immunohistochemistry for NRP1, where we observed NRP1 expression by tumor cells in close proximity to blood vessels, and expression levels were higher in PDAC than GAC.

To confirm that VEGFR2/NRP1 complexes form in human tumors, we performed in situ PLA. We detected complexes located at the interface between the endothelium and tumor cells, confirming that complexes existed in the trans configuration. Trans complexes were observed in approximately 40% of tumors for both GAC and PDAC but more abundantly in PDAC, which was in line with results from the immunohistochemistry and RNAscope analyses. In PDAC, tumors positive for VEGFR2/NRP1 trans complexes showed reduced vascular area, vessel numbers, vessel branching and also reduced tumor cell proliferation, in congruence with our previous observations in mouse tumor models. In GAC, trans presentation of NRP1 had no effect on vessel parameters or tumor cell proliferation, likely due to the low- er NRP1 expression level and therefore, a less robust trans inhibition effect.

41 To determine if VEGFR2/NRP1 complexes in trans affected patient prognosis we performed in situ PLA and scored for trans complexes in a larger cohort of 64 PDAC patients. We observed a similar effect of trans complexes on vessel parameters and tumor cell proliferation in this cohort as in the previous, smaller cohorts described above. Importantly, Kaplan-Meier analysis showed an improved overall survival in patients scored as trans. Further- more, multivariable analysis revealed VEGFR2/NRP1 complex formation in trans to be an independent marker of overall survival in PDAC patients.

In this study we show for the first time that VEGFR2/NRP1 complex formation in cis and trans occurs in human cancer. Additionally, in PDAC, NRP1 presentation by vessel-adjacent tumor cells correlates with reduced vessel parameters, reduced tumor cell proliferation and improved patient survival, suggesting that VEGFR2/NRP1 complex formation could be a valuable prognostic marker and potentially a predictive marker of anti- angiogenic therapy. These findings also encourage the study of trans NRP1 in other cancer types.

42 Paper III RCC tumors are generally highly vascularized due to loss or mutation of the VHL gene leading to sustained HIF1-mediated transcription of proangiogenic factors such as VEGF-A. Therapies targeting angiogenesis have been shown to prolong overall survival in metastatic RCC and several anti- angiogenic therapies are currently considered in first line treatment of metastatic disease. Despite the importance of angiogenesis in RCC, the role of NRP1 has not been well studied.

To confirm that NRP1 is expressed in RCC we acquired expression data from The Cancer Genome Atlas (TCGA) database on a number of tumor types and healthy controls. We noted a significant upregulation of NRP1 in RCC compared to healthy kidney; moreover, the expression levels were high in comparison with other malignancies, validating the relevance of studying NRP1 in the context of RCC.

To explore the spatial distribution of NRP1 we performed immunofluorescent staining for VEGFR2 and NRP1 as wells as in situ PLA on a cohort of 64 RCC patients with advanced disease. We observed that 74% of patients were positive for tumor cell NRP1, 72% for NRP1 expression specifically in perivascular regions and 35% had endothelial NRP1 expression. NRP1 perivascular expression was significantly associated with VEGFR2/NRP1 complexes in trans. We observed a reduced size of individual vessels in tumors positive for trans complexes. Interestingly, tumors with complexes only in cis showed increased total vessel area and individual vessel size compared to tumors positive only for trans complexes.

We examined whether NRP1 expression correlated with patient survival using Kaplan-Meier analysis. We observed that perivascular NRP1 and the presence of complexes in trans both associated with improved survival, while patients with NRP1 only in endothelial cells showed poor survival. Multivariable analysis also revealed that perivascular NRP1 and VEGFR2/NRP1 complexes in trans were independent markers of overall survival. Our findings emphasize the importance of the NRP1 expression pattern in RCC and the potential of perivascular NRP1 expression as a prognostic biomarker.

43 Graphical summary of the papers

Figure 8. Graphical summary of the papers included in this thesis. We have explored the impact of NRP1 presentation of VEGF-A on VEGFR2 biology and tumor angiogenesis. We show that NRP1 in trans (right column) prolongs VEGFR2/NRP1 complex formation resulting in delayed VEGFR2 internalization and attenuated downstream signaling (Paper I). Tumor initiation was reduced in trans tumors (Paper I), correlating with a reduction in vessel parameters (Paper I & II) and tumor cell proliferation (Paper II). Studying patient material we found that VEGFR2/NRP1 complex formation in trans was associated with better overall survival in both PDAC and RCC patients (Paper II & III) (adapted from Koch et al., 2014).

44 Future perspectives

Physiological angiogenesis is a strictly regulated process essential in response to changes in metabolic needs, in the maintenance of organ homeostasis and in wound healing. However, in a number of diseases angiogenesis is perpetual and the newly formed vessels dysfunctional, aggravating the disease. Targeting tumor angiogenesis has been proposed for many decades, the rationale being that if the tumor vasculature is suppressed, the tumor cells will starve and wither, leading to shrinkage and elimination of the tumor. Since all types of tumors require blood vessels to grow, targeting angiogenesis is conceptually intriguing due to its potential to be a broadly applicable cancer therapy. In the last two decades, anti-angiogenic therapies have been developed and approved for a number of cancers. However, the results have not been as overwhelmingly positive as initially predicted. Currently, anti-angiogenic therapy is typically used in late stages of tumor progression, when prolonged survival rather than cure is the sought outcome.

Current anti-angiogenic therapies target the VEGF-A/VEGFR2 pathway. In this thesis we have explored the role of the co-receptor NRP1 in regulation of VEGFR2 activity and angiogenesis. We show that NRP1 expressed by tumor cells in trans, ergo adjacent to the endothelium, can arrest VEGFR2 at the cell surface, resulting in delayed VEGFR2 activation kinetics and altered downstream signaling output. NRP1 is expressed by several cell types, in addition to endothelial and tumor cells, including inflammatory and immune cells that can interact with endothelial VEGFR2. Thus, there is potential worth exploring for trans presentation of VEGF-A by NRP1 expressed on a number of different cell types in the tumor microenvironment.

Moreover, NRP1 binds several other growth factors, such as FGF, TGFβ, PDGF and semaphorins, in addition to VEGF-A, and may form complexes with their putative receptors. Determining if trans interactions occur also in these contexts would shed light on the diverse reported effects of NRP1. Indeed, Muhl et al. recently showed that PDGFRβ on pericyte protrusions can interact with endothelial NRP1 in the presence of the ligand PDGFD, a potentially important function for pericyte-endothelial cell-cell communica- tion (Muhl et al., 2017).

45 In this work we show that the presence and distribution of in situ VEGFR2/NRP1 complexes in patient samples can provide valuable clinical information. Moreover, employing methods such as in situ PLA reveals information about the biological roles of the molecules of interest that would not have been detected by conventional staining methods for the same proteins. Knowledge of the balance of cis and trans complex ratios could potentially guide the choice of therapy. Based on our observations, interruption of cis complexes could be beneficial while interference with trans complexes might be unfavorable. We expect that studies of other protein interactions using in situ PLA on clinical samples could provide valuable clinical insight into tumor behavior and therapy response.

Biomarker studies have historically focused on the overall expression levels of the protein of interest. For cell surface proteins, this may not provide the full picture since the extracellular microenvironment could affect the biology in a manner that may differ very considerably between individuals. We show that studying NRP1 expression specifically in adjacency to blood vessels generated valuable clinical information that might potentially have been overlooked if focusing only on overall protein expression. Although in situ PLA is a more direct method of detecting interacting molecules, staining for vessel adjacent NRP1 by conventional methods may be more readily availa- ble in the clinical setting. It would be highly interesting to study the spatial distribution of other cancer associated membrane proteins with multiple proposed ligands/interaction partners. For example, Ephrin-B2 is a plasma membrane anchored binding partner of Eph receptor tyrosine kinases, an interaction that occurs in trans between cells. However, Ephrin-B2 has also been shown to regulate VEGFR2 internalization and signaling, thus regulating angiogenesis (Sawamiphak et al., 2010).

The rate and spatial distribution of VEGFR2/NRP1 complex formation is governed by the expression pattern of the two proteins. Additionally, the formation is influenced by the availability of VEGF-A. The expression of VEGF-A might be highly variable between cancer types and individual tumors. In the works of this thesis we did not study VEGF-A expression in detail but recognize that it would be of high importance in the continued work. In addition to determining VEGF-A expression levels, the balance in expression of its splicing isoforms could be important. VEGF-A165 but not VEGF-A121 can bridge between VEGFR2 and NRP1 to induce complex formation. Interestingly, in a mouse tumor model using tumor cells transfected with vectors expressing VEGF-A121, VEGF-A165 or an empty vector, tumors overexpressing the VEGF-A121 isoform showed increased tumor growth while VEGF-A165 overexpression had an inhibitory effect compared to empty vector tumors (Kazemi et al., 2016). The authors also observed that high VEGF-A121 expression in colorectal cancer patient samples correlated with

46 an immature vessel phenotype. Thus, VEGF-A expression and regulation of isoform expression may greatly influence tumor growth, potentially through the regulation of VEGFR2/NRP1 complex formation. Specifically blocking VEGF-A121 or skewing expression towards the VEGF-A165 isoform are thus attractive potential novel therapeutic approaches.

Basement membrane integrity may also affect the formation of trans complexes between endothelial cells and surrounding cells. This integrity is typically affected in the tumor vasculature. Increased disorganization of the basement membrane allows perivascular tumor cells to interact directly with the endothelium. It would be interesting to determine whether VEGFR2/ NRP1 complex formation in trans correlates with impaired basement membrane integrity.

Anti-angiogenic therapy has the potential to be beneficial for many cancer patients; however, the individual response varies greatly and some patients do not respond to this line of therapy. Finding markers able to predict clinical benefit is therefore essential. In this work we have shown that VEGFR2/NRP1 complex formation in trans is associated with prolonged overall survival in both PDAC and RCC. However, due to the retrospective design of our studies we could not evaluate the predictive value of this marker. Such studies would be highly interesting, especially in the light of the AVAGAST trial of Bevacizumab where no benefit to overall survival was observed on the cohort as a whole, but when comparing groups based on NRP1 expression, those with low NRP1 levels had worse prognosis but showed a significant increase in overall survival when receiving Bevaci- zumab (Van Cutsem et al., 2012a). Low NRP1 could potentially indicate a reduced inhibitory effect of trans-NRP1, explaining the worse prognosis for this subgroup and better response to Bevacizumab. Additionally, in patients with high NRP1 expression, Bevacizumab treatment may interrupt VEGFR2/NRP1 complexes in trans, explaining the lack of clinical benefit in this subgroup. It would therefore be very relevant to study if Bevacizumab therapy response could be predicted by absence of trans complexes.

In this work we show that the spatial distribution of NRP1 regulates tumor angiogenesis. Preclinical studies have shown that anti-NRP1 therapy has additive anti-angiogenic effect when combined with an anti-VEGF-A antibody. The observation of increased rates of proteinuria when combined with Bevacizumab in a phase Ib clinical trial prevented further study of a humanized anti-NRP1 antibody (Patnaik et al., 2014). Of note, an initial phase I study of the anti-NRP1 antibody alone showed an acceptable safety profile (Weekes et al., 2014). NRP1 is therefore a potential target for anti- angiogenic therapy worth exploring further in combination with drugs other than Bevacizumab. Our data suggest that the selection of patients for such

47 trials needs to be carefully evaluated based on the expression profile of NRP1.

In summary, we have shown that NRP1 can interact with VEGFR2 in trans, resulting in differential VEGFR2 signaling output and angiogenesis. We show that this interaction occurs in human cancer, where it is associated with prolonged overall survival in several cancer types. We look forward to learn- ing more about the potential presence of VEGFR2/NRP1 complexes in yet other cancer forms, and the impact and value of such complexes as a predictive marker for anti-angiogenic therapy.

48 Acknowledgements

This work was carried out at the Department of Immunology, Genetics and Pathology, Rudbeck Laboratory, Uppsala University, Sweden. This work was supported by grants to Lena Claesson-Welsh from the Swedish Cancer Foundation, the Swedish Research Council, and the Knut and Alice Wallen- berg Foundation. I want to express my gratitude towards Uppsala läns land- sting and AT-enheten at Akademiska sjukhuset for my employment as a Forskar-AT, allowing me to continue with research.

This PhD has been a long journey and I would like to express my gratitude to the people who supported me during all these years.

First I would like to sincerely thank my main supervisor Lena. I started in your group as a master student in 2010 and your encouragement and enthusiasm convinced me to continue as your PhD student in parallel with starting medical school. It has been challenging to be occupied with medical studies and later clinical employment and still continue to do research but thanks to you and your constant support it has been possible. You have taught me to manage not just my projects but also how to prioritize my time, for which I am very grateful. Thank you for all the discussions about science, but also career and longtime goals. Without you this journey would have been a lot bumpier.

I am also very grateful to Sina, my co-supervisor. You took me under your wing when I first started in Lena’s group and you have taught me so much in the lab. Thank you for your thoroughness and your enthusiasm about science and all the work we have done together, especially on the first paper of this thesis. Even though you have returned to Germany, I have always felt your support and your willingness to help.

I am lucky to have worked in such an inspiring and fun environment. I want to express my gratitude toward all members of the LCW group; Dom, Elin, Lakshmi, Mark, Miguel, Mimmi, Pernilla, Ross, Suvi, Takeshi, Tor and Yi for stimulating discussions, countless lab meetings, practical advice and support. Thank you also for all the good times not related to research, I’m talking masquerades, beer clubs, murder mysteries etc.! I am especially grateful to you Elin for all the work we have done together during these last

49 couple of years since you joined the group. Working together with you and sharing your knowledge has been invaluable for me and my PhD. Thank you Mark for all the time spent proofreading my thesis. I also want to specifically thank my fellow PhDs and office buddies in the group. Thank you Tor, for all the discussions we’ve had about NFL, sports, medical school and PhD life. Ross, we have shared a lot of ups and downs during all these years as PhDs together and I am grateful for your constantly positive attitude, your kindness and your enthusiasm. Also, thanks to you and Sam for all the game nights together.

I want to thank all the former LCW group members as well. During my PhD I have met a lot of fantastic people who have now left the group for new adventures. I particularly want to thank Anja, Chiara, Emma G, Frank and Jimmy for all the years together, poker nights, conferences (and conference road trips!) and all the other memorable moments we have shared.

Thanks to everyone in the Vascular Biology division for the ambitious yet open and accommodating atmosphere that we have created together. Thank you for all the interesting seminars, great retreats and parties!

I have received a lot of support from Christina and Helene at the IGP administration during my PhD for which I am truly thankful.

I want to extend my gratitude to the all collaborators and co-authors of the papers, thank you for all your valuable input and providing critical clinical samples and reagents for the projects of this thesis. I also want to thank BioVis, the tissue profiling facility and the PLA platform for providing their services.

To all the people I have met through my PhD at IGP, thank you for all the great times together and for getting to know you, whether it has been plan- ning masquerades together, attending courses, partying or just seeing each other in the lunch room. I especially want to thank Alberto, Annika, Colin, Elisabeth, Emma Y, Jelena, Khayrun, Leonor, Lucy, Malte, Matko, Marco, Nicky, Ram, Sara, Svea, Vasil, Viktor and Yang, for all the great times we’ve had both at work and elsewhere (especially elsewhere!). Thank you Luuk for trying to save the world with me.

Aleksi, Andreas, Fredrik, Patrik and Robert, together we make the Clas- sic 6! We have known each other for a really long time and experienced so much together. I’m so grateful to call you my friends. Thank you for your constant support, and thank you for all the Malung trips, CMLs, NHL tour- naments, bordshockey, basketball matches and all the adventures.

50 Verónica, you have been my greatest supporter and I am so grateful to have you in my life. I truly appreciate your unwavering belief in me and you have helped me stay motivated through this whole journey. You have been my source of inspiration. I can’t imagine doing this PhD without you!

I also want to thank Verónica’s family for accepting me into their life. Thank you for your warmth and kindness.

Last I want to thank my family. Thank you mormor Siv and farmor Inga for all the calls and words of encouragement! Thank you Mom, Dad and my brothers with partners for always being there for me and believing in me. I am so grateful for your support!

51 References

Aird, W.C. (2007). Phenotypic heterogeneity of the endothelium: I. Structure, function, and mechanisms. Circ Res 100, 158-173. Angers, S., Salahpour, A., Joly, E., Hilairet, S., Chelsky, D., Dennis, M., and Bouvier, M. (2000). Detection of β2-adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). Proceedings of the National Academy of Sciences 97, 3684-3689. Arpel, A., Gamper, C., Spenle, C., Fernandez, A., Jacob, L., Baumlin, N., Laquerriere, P., Orend, G., Cremel, G., and Bagnard, D. (2016). Inhibition of primary breast tumor growth and metastasis using a neuropilin-1 transmembrane domain interfering peptide. Oncotarget 7, 54723-54732. Aspalter, I.M., Gordon, E., Dubrac, A., Ragab, A., Narloch, J., Vizán, P., Geudens, I., Collins, R.T., Franco, C.A., Abrahams, C.L., et al. (2015). Alk1 and Alk5 inhibition by Nrp1 controls vascular sprouting downstream of Notch. Nature Communications 6, 7264. Augustin, H.G., and Koh, G.Y. (2017). Organotypic vasculature: From descriptive heterogeneity to functional pathophysiology. Science. Augustin, H.G., Koh, G.Y., Thurston, G., and Alitalo, K. (2009). Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system. Nat Rev Mol Cell Biol 10, 165-177. Ballmer-Hofer, K., Andersson, A.E., Ratcliffe, L.E., and Berger, P. (2011). Neuropilin-1 promotes VEGFR-2 trafficking through Rab11 vesicles thereby specifying signal output. Blood 118, 816-826. Baluk, P., Hashizume, H., and McDonald, D.M. (2005). Cellular abnormalities of blood vessels as targets in cancer. Curr Opin Genet Dev 15, 102-111. Bang, Y.-J., Van Cutsem, E., Feyereislova, A., Chung, H.C., Shen, L., Sawaki, A., Lordick, F., Ohtsu, A., Omuro, Y., Satoh, T., et al. (2010). Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. The Lancet 376, 687-697. Batchelor, T.T., Sorensen, A.G., di Tomaso, E., Zhang, W.T., Duda, D.G., Cohen, K.S., Kozak, K.R., Cahill, D.P., Chen, P.J., Zhu, M., et al. (2007). AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer cell 11, 83-95. Becker, K.-F., Keller, G., and Hoefler, H. (2000). The use of molecular biology in diagnosis and prognosis of gastric cancer. Surgical Oncology 9, 5-11. Ben, Q., Zheng, J., Fei, J., An, W., Li, P., Li, Z., and Yuan, Y. (2014). High Neuropilin 1 Expression Was Associated With Angiogenesis and Poor Overall Survival in Resected Pancreatic Ductal Adenocarcinoma. Pancreas 43, 744-749. Benedito, R., Rocha, S.F., Woeste, M., Zamykal, M., Radtke, F., Casanovas, O., Duarte, A., Pytowski, B., and Adams, R.H. (2012). Notch-dependent VEGFR3 upregulation allows angiogenesis without VEGF-VEGFR2 signalling. Nature 484, 110-114.

52 Berge, M., Bonnin, P., Sulpice, E., Vilar, J., Allanic, D., Silvestre, J.S., Levy, B.I., Tucker, G.C., Tobelem, G., and Merkulova-Rainon, T. (2010). Small interfering RNAs induce target-independent inhibition of tumor growth and vasculature remodeling in a mouse model of hepatocellular carcinoma. The American journal of pathology 177, 3192-3201. Bergers, G., and Benjamin, L.E. (2003). Tumorigenesis and the angiogenic switch. Nat Rev Cancer 3, 401-410. Bergers, G., and Hanahan, D. (2008). Modes of resistance to anti-angiogenic therapy. Nat Rev Cancer 8, 592-603. Bielenberg, D.R., and Zetter, B.R. (2015). The Contribution of Angiogenesis to the Process of Metastasis. Cancer journal (Sudbury, Mass) 21, 267-273. Boehm, T., Folkman, J., Browder, T., and O'Reilly, M.S. (1997). Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 390, 404-407. Borysenko, M., and Beringer, T. (1984). Functional Histology Textbook. Little Brown and Company Printed in the United State of America, Boston, Toronto. Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R.L., Torre, L.A., and Jemal, A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians 0. Butler, J.M., Kobayashi, H., and Rafii, S. (2010). Instructive role of the vascular niche in promoting tumour growth and tissue repair by angiocrine factors. Nat Rev Cancer 10, 138-146. Cai, H., and Reed, R.R. (1999). Cloning and characterization of neuropilin-1- interacting protein: a PSD-95/Dlg/ZO-1 domain-containing protein that interacts with the cytoplasmic domain of neuropilin-1. J Neurosci 19, 6519-6527. Cao, S., Yaqoob, U., Das, A., Shergill, U., Jagavelu, K., Huebert, R.C., Routray, C., Abdelmoneim, S., Vasdev, M., Leof, E., et al. (2010). Neuropilin-1 promotes cirrhosis of the rodent and human liver by enhancing PDGF/TGF-beta signaling in hepatic stellate cells. J Clin Invest 120, 2379-2394. Cao, Y., Wang, L., Nandy, D., Zhang, Y., Basu, A., Radisky, D., and Mukhopadhyay, D. (2008). Neuropilin-1 upholds dedifferentiation and propagation phenotypes of renal cell carcinoma cells by activating Akt and sonic hedgehog axes. Cancer research 68, 8667-8672. Carmeliet, P., Ferreira, V., Breier, G., Pollefeyt, S., Kieckens, L., Gertsenstein, M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C., et al. (1996). Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380, 435-439. Carmeliet, P., and Jain, R.K. (2011a). Molecular mechanisms and clinical applications of angiogenesis. Nature 473, 298-307. Carmeliet, P., and Jain, R.K. (2011b). Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nat Rev Drug Discov 10, 417-427. Carmeliet, P., Lampugnani, M.-G., Moons, L., Breviario, F., Compernolle, V., Bono, F., Balconi, G., Spagnuolo, R., Oosthuyse, B., and Dewerchin, M. (1999). Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 98, 147- 157. Carrer, A., Moimas, S., Zacchigna, S., Pattarini, L., Zentilin, L., Ruozi, G., Mano, M., Sinigaglia, M., Maione, F., Serini, G., et al. (2012). Neuropilin-1 identifies a subset of bone marrow Gr1- monocytes that can induce tumor vessel normalization and inhibit tumor growth. Cancer Res 72, 6371-6381.

53 Casanovas, O., Hicklin, D.J., Bergers, G., and Hanahan, D. (2005). Drug resistance by evasion of antiangiogenic targeting of VEGF signaling in late-stage pancreatic islet tumors. Cancer cell 8, 299-309. Charpentier, M.S., and Conlon, F.L. (2014). Cellular and molecular mechanisms underlying blood vessel lumen formation. BioEssays : news and reviews in molecular, cellular and developmental biology 36, 251-259. Chaudhary, B., Khaled, Y.S., Ammori, B.J., and Elkord, E. (2014). Neuropilin 1: function and therapeutic potential in cancer. Cancer immunology, immunotherapy : CII 63, 81-99. Chen, C.N., Cheng, Y.M., Lin, M.T., Hsieh, F.J., Lee, P.H., and Chang, K.J. (2002). Association of color Doppler vascularity index and microvessel density with survival in patients with gastric cancer. Annals of surgery 235, 512-518. Chen, H., Chedotal, A., He, Z., Goodman, C.S., and Tessier-Lavigne, M. (1997). Neuropilin-2, a novel member of the neuropilin family, is a high affinity receptor for the semaphorins Sema E and Sema IV but not Sema III. Neuron 19, 547-559. Chen, T.T., Luque, A., Lee, S., Anderson, S.M., Segura, T., and Iruela-Arispe, M.L. (2010). Anchorage of VEGF to the extracellular matrix conveys differential signaling responses to endothelial cells. J Cell Biol 188, 595-609. Chistiakov, D.A., Orekhov, A.N., and Bobryshev, Y.V. (2017). Effects of shear stress on endothelial cells: go with the flow. Acta Physiologica 219, 382-408. Chu, W., Song, X., Yang, X., Ma, L., Zhu, J., He, M., Wang, Z., and Wu, Y. (2014). Neuropilin-1 promotes epithelial-to-mesenchymal transition by stimulating nuclear factor-kappa B and is associated with poor prognosis in human oral squamous cell carcinoma. PLoS One 9, e101931. Clegg, L.W., and Mac Gabhann, F. (2015). Site-Specific Phosphorylation of VEGFR2 Is Mediated by Receptor Trafficking: Insights from a Computational Model. PLOS Computational Biology 11, e1004158. Cohen, H.T., and McGovern, F.J. (2005). Renal-cell carcinoma. N Engl J Med 353, 2477-2490. Corada, M., Mariotti, M., Thurston, G., Smith, K., Kunkel, R., Brockhaus, M., Lampugnani, M.G., Martin-Padura, I., Stoppacciaro, A., and Ruco, L. (1999). Vascular endothelial–cadherin is an important determinant of microvascular integrity in vivo. Proceedings of the National Academy of Sciences 96, 9815- 9820. Cortes, J., Talpaz, M., Smith, H.P., Snyder, D.S., Khoury, J., Bhalla, K.N., Pinilla- Ibarz, J., Larson, R., Mitchell, D., Wise, S.C., et al. (2017). Phase 1 dose- finding study of rebastinib (DCC-2036) in patients with relapsed chronic myeloid leukemia and acute myeloid leukemia. Haematologica 102, 519-528. Craven, K.E., Gore, J., and Korc, M. (2016). Overview of pre-clinical and clinical studies targeting angiogenesis in pancreatic ductal adenocarcinoma. Cancer letters 381, 201-210. Crawford, Y., Kasman, I., Yu, L., Zhong, C., Wu, X., Modrusan, Z., Kaminker, J., and Ferrara, N. (2009). PDGF-C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractory to anti-VEGF treatment. Cancer Cell 15, 21-34. Cross, M.J., Dixelius, J., Matsumoto, T., and Claesson-Welsh, L. (2003). VEGF- receptor signal transduction. Trends Biochem Sci 28, 488-494. Cummins, P.M. (2012). Occludin: one protein, many forms. Molecular and cellular biology 32, 242-250.

54 Dejana, E., Orsenigo, F., and Lampugnani, M.G. (2008). The role of adherens junctions and VE-cadherin in the control of vascular permeability. Journal of cell science 121, 2115-2122. Dekker, R.J., Boon, R.A., Rondaij, M.G., Kragt, A., Volger, O.L., Elderkamp, Y.W., Meijers, J.C., Voorberg, J., Pannekoek, H., and Horrevoets, A.J. (2006). KLF2 provokes a gene expression pattern that establishes functional quiescent differentiation of the endothelium. Blood 107, 4354-4363. Demir, R., Seval, Y., and Huppertz, B. (2007). Vasculogenesis and angiogenesis in the early human placenta. Acta Histochem 109, 257-265. Deryugina, E.I., and Quigley, J.P. (2015). Tumor angiogenesis: MMP-mediated induction of intravasation- and metastasis-sustaining neovasculature. Matrix biology : journal of the International Society for Matrix Biology 44-46, 94-112. Dimmeler, S., Fleming, I., Fisslthaler, B., Hermann, C., Busse, R., and Zeiher, A.M. (1999). Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature 399, 601-605. Ding, Y., Zhou, J., Wang, S., Li, Y., Mi, Y., Gao, S., Xu, Y., Chen, Y., and Yan, J. (2018). Anti-neuropilin-1 monoclonal antibody suppresses the migration and invasion of human gastric cancer cells via Akt dephosphorylation. Experimental and therapeutic medicine 16, 537-546. Dor, Y., Porat, R., and Keshet, E. (2001). Vascular endothelial growth factor and vascular adjustments to perturbations in oxygen homeostasis. Am J Physiol Cell Physiol 280, C1367-1374. Dorland, Y.L., and Huveneers, S. (2017). Cell–cell junctional mechanotransduction in endothelial remodeling. Cellular and Molecular Life Sciences 74, 279-292. Ebos, J.M., Lee, C.R., Cruz-Munoz, W., Bjarnason, G.A., Christensen, J.G., and Kerbel, R.S. (2009). Accelerated metastasis after short-term treatment with a potent inhibitor of tumor angiogenesis. Cancer cell 15, 232-239. Erez, N., Truitt, M., Olson, P., Arron, S.T., and Hanahan, D. (2010). Cancer- Associated Fibroblasts Are Activated in Incipient Neoplasia to Orchestrate Tumor-Promoting Inflammation in an NF-kappaB-Dependent Manner. Cancer Cell 17, 135-147. Eriksson, A., Cao, R., Roy, J., Tritsaris, K., Wahlestedt, C., Dissing, S., Thyberg, J., and Cao, Y. (2003). Small GTP-binding protein Rac is an essential mediator of vascular endothelial growth factor-induced endothelial fenestrations and vascular permeability. Circulation 107, 1532-1538. Escudier, B., Szczylik, C., Porta, C., and Gore, M. (2012). Treatment selection in metastatic renal cell carcinoma: expert consensus. Nature reviews Clinical oncology 9, 327. Evans, I.M., Britton, G., and Zachary, I.C. (2008). Vascular endothelial growth factor induces heat shock protein (HSP) 27 serine 82 phosphorylation and endothelial tubulogenesis via protein kinase D and independent of p38 kinase. Cell Signal 20, 1375-1384. Fantin, A., Herzog, B., Mahmoud, M., Yamaji, M., Plein, A., Denti, L., Ruhrberg, C., and Zachary, I. (2014). Neuropilin 1 (NRP1) hypomorphism combined with defective VEGF-A binding reveals novel roles for NRP1 in developmental and pathological angiogenesis. Development 141, 556-562. Fantin, A., Lampropoulou, A., Senatore, V., Brash, J.T., Prahst, C., Lange, C.A., Liyanage, S.E., Raimondi, C., Bainbridge, J.W., Augustin, H.G., et al. (2017). VEGF165-induced vascular permeability requires NRP1 for ABL-mediated SRC family kinase activation. The Journal of Experimental Medicine 214, 1049-1064.

55 Fantin, A., Schwarz, Q., Davidson, K., Normando, E.M., Denti, L., and Ruhrberg, C. (2011). The cytoplasmic domain of neuropilin 1 is dispensable for angiogenesis, but promotes the spatial separation of retinal arteries and veins. Development 138, 4185-4191. Fantin, A., Vieira, J.M., Gestri, G., Denti, L., Schwarz, Q., Prykhozhij, S., Peri, F., Wilson, S.W., and Ruhrberg, C. (2010). Tissue macrophages act as cellular chaperones for vascular anastomosis downstream of VEGF-mediated endothelial tip cell induction. Blood 116, 829-840. Fearnley, G.W., Smith, G.A., Abdul-Zani, I., Yuldasheva, N., Mughal, N.A., Homer-Vanniasinkam, S., Kearney, M.T., Zachary, I.C., Tomlinson, D.C., Harrison, M.A., et al. (2016). VEGF-A isoforms program differential VEGFR2 signal transduction, trafficking and proteolysis. Biology open 5, 571-583. Ferlay, J., Soerjomataram, I., Dikshit, R., Eser, S., Mathers, C., Rebelo, M., Parkin, D.M., Forman, D., and Bray, F. (2015). Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. International journal of cancer 136, E359-386. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O'Shea, K.S., Powell- Braxton, L., Hillan, K.J., and Moore, M.W. (1996). Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380, 439- 442. Ferrara, N., Gerber, H.P., and LeCouter, J. (2003). The biology of VEGF and its receptors. Nat Med 9, 669-676. Ferrone, C.R., Pieretti-Vanmarcke, R., Bloom, J.P., Zheng, H., Szymonifka, J., Wargo, J.A., Thayer, S.P., Lauwers, G.Y., Deshpande, V., Mino-Kenudson, M., et al. (2012). Pancreatic ductal adenocarcinoma: long-term survival does not equal cure. Surgery 152, S43-49. Folkman, J. (1971). Tumor angiogenesis: therapeutic implications. N Engl J Med 285, 1182-1186. Folkman, J., Hahnfeldt, P., and Hlatky, L. (2000). Cancer: looking outside the genome. Nature Reviews Molecular Cell Biology 1, 76. Fong, G.H., Rossant, J., Gertsenstein, M., and Breitman, M.L. (1995). Role of the Flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature 376, 66-70. Fontanella, C., Ongaro, E., Bolzonello, S., Guardascione, M., Fasola, G., and Aprile, G. (2014). Clinical advances in the development of novel VEGFR2 inhibitors. Ann Transl Med 2, 123. Franco, C.A., Jones, M.L., Bernabeu, M.O., Geudens, I., Mathivet, T., Rosa, A., Lopes, F.M., Lima, A.P., Ragab, A., Collins, R.T., et al. (2015). Dynamic endothelial cell rearrangements drive developmental vessel regression. PLoS biology 13, e1002125. Fraser, R., Dobbs, B.R., and Rogers, G.W. (1995). Lipoproteins and the liver sieve: the role of the fenestrated sinusoidal endothelium in lipoprotein metabolism, atherosclerosis, and cirrhosis. Hepatology 21, 863-874. Fuchs, C.S., Tomasek, J., Yong, C.J., Dumitru, F., Passalacqua, R., Goswami, C., Safran, H., Dos Santos, L.V., Aprile, G., Ferry, D.R., et al. (2014). Ramucirumab monotherapy for previously treated advanced gastric or gastro- oesophageal junction adenocarcinoma (REGARD): an international, randomised, multicentre, placebo-controlled, phase 3 trial. Lancet (London, England) 383, 31-39. Fukahi, K., Fukasawa, M., Neufeld, G., Itakura, J., and Korc, M. (2004). Aberrant expression of neuropilin-1 and -2 in human pancreatic cancer cells. Clin Cancer Res 10, 581-590.

56 Fukuhara, S., Sako, K., Minami, T., Noda, K., Kim, H.Z., Kodama, T., Shibuya, M., Takakura, N., Koh, G.Y., and Mochizuki, N. (2008). Differential function of Tie2 at cell–cell contacts and cell–substratum contacts regulated by angiopoietin-1. Nature Cell Biology 10, 513. Gaengel, K., Genove, G., Armulik, A., and Betsholtz, C. (2009). Endothelial-mural cell signaling in vascular development and angiogenesis. Arteriosclerosis, thrombosis, and vascular biology 29, 630-638. Gagnon, M.L., Bielenberg, D.R., Gechtman, Z., Miao, H.Q., Takashima, S., Soker, S., and Klagsbrun, M. (2000). Identification of a natural soluble neuropilin-1 that binds vascular endothelial growth factor: In vivo expression and antitumor activity. Proc Natl Acad Sci U S A 97, 2573-2578. Gálvez, B.G., Matı́as-Román, S., Albar, J.P., Sánchez-Madrid, F., and Arroyo, A.G. (2001). Membrane type 1-matrix metalloproteinase is activated during migration of human endothelial cells and modulates endothelial motility and matrix remodeling. Journal of Biological Chemistry 276, 37491-37500. Gebala, V., Collins, R., Geudens, I., Phng, L.K., and Gerhardt, H. (2016). Blood flow drives lumen formation by inverse membrane blebbing during angiogenesis in vivo. Nature cell biology 18, 443-450. Gelfand, M.V., Hagan, N., Tata, A., Oh, W.J., Lacoste, B., Kang, K.T., Kopycinska, J., Bischoff, J., Wang, J.H., and Gu, C. (2014). Neuropilin-1 functions as a VEGFR2 co-receptor to guide developmental angiogenesis independent of ligand binding. eLife 3, e03720. Gerber, H.P., McMurtrey, A., Kowalski, J., Yan, M., Keyt, B.A., Dixit, V., and Ferrara, N. (1998). Vascular endothelial growth factor regulates endothelial cell survival through the phosphatidylinositol 3'-kinase/Akt signal transduction pathway. Requirement for Flk-1/KDR activation. J Biol Chem 273, 30336- 30343. Gerhardt, H., Golding, M., Fruttiger, M., Ruhrberg, C., Lundkvist, A., Abramsson, A., Jeltsch, M., Mitchell, C., Alitalo, K., Shima, D., et al. (2003). VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. The Journal of cell biology 161, 1163-1177. Ghatalia, P., Zibelman, M., Geynisman, D.M., and Plimack, E.R. (2017). Checkpoint Inhibitors for the Treatment of Renal Cell Carcinoma. Current Treatment Options in Oncology 18, 7. Gillen, S., Schuster, T., Zum Büschenfelde, C.M., Friess, H., and Kleeff, J. (2010). Preoperative/neoadjuvant therapy in pancreatic cancer: a systematic review and meta-analysis of response and resection percentages. PLoS medicine 7, e1000267. Glimelius, B., Ekström, K., Hoffman, K., Graf, W., Sjödén, P.-O., Haglund, U., Svensson, C., Enander, L.-K., Linne, T., and Sellsröm, H. (1997). Randomized comparison between chemotherapy plus best supportive care with best supportive care in advanced gastric cancer. Annals of Oncology 8, 163-168. Glinka, Y., Stoilova, S., Mohammed, N., and Prud'homme, G.J. (2011). Neuropilin- 1 exerts co-receptor function for TGF-beta-1 on the membrane of cancer cells and enhances responses to both latent and active TGF-beta. Carcinogenesis 32, 613-621. Gnarra, J., Tory, K., Weng, Y., Schmidt, L., Wei, M., Li, H., Latif, F., Liu, S., Chen, F., and Duh, F.-M. (1994). Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet 7, 85.

57 Gray, M.J., Wey, J.S., Belcheva, A., McCarty, M.F., Trevino, J.G., Evans, D.B., Ellis, L.M., and Gallick, G.E. (2005). Neuropilin-1 suppresses tumorigenic properties in a human pancreatic adenocarcinoma cell line lacking neuropilin-1 coreceptors. Cancer Res 65, 3664-3670. Grivennikov, S.I., Greten, F.R., and Karin, M. (2010). Immunity, inflammation, and cancer. Cell 140, 883-899. Gu, C., Limberg, B.J., Whitaker, G.B., Perman, B., Leahy, D.J., Rosenbaum, J.S., Ginty, D.D., and Kolodkin, A.L. (2002). Characterization of neuropilin-1 structural features that confer binding to semaphorin 3A and vascular endothelial growth factor 165. J Biol Chem 277, 18069-18076. Hall, C.N., Reynell, C., Gesslein, B., Hamilton, N.B., Mishra, A., Sutherland, B.A., O’farrell, F.M., Buchan, A.M., Lauritzen, M., and Attwell, D. (2014). Capillary pericytes regulate cerebral blood flow in health and disease. Nature 508, 55. Hanahan, D., and Folkman, J. (1996). Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86, 353-364. Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674. Hansel, D.E., Wilentz, R.E., Yeo, C.J., Schulick, R.D., Montgomery, E., and Maitra, A. (2004). Expression of neuropilin-1 in high-grade dysplasia, invasive cancer, and metastases of the human gastrointestinal tract. Am J Surg Pathol 28, 347- 356. Harney, A.S., Karagiannis, G.S., Pignatelli, J., Smith, B.D., Kadioglu, E., Wise, S.C., Hood, M.M., Kaufman, M.D., Leary, C.B., Lu, W.P., et al. (2017). The Selective Tie2 Inhibitor Rebastinib Blocks Recruitment and Function of Tie2(Hi) Macrophages in Breast Cancer and Pancreatic Neuroendocrine Tumors. Molecular cancer therapeutics 16, 2486-2501. Hawkins, B.T., and Davis, T.P. (2005). The blood-brain barrier/neurovascular unit in health and disease. Pharmacological reviews 57, 173-185. He, Z., and Tessier-Lavigne, M. (1997). Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell 90, 739-751. Hellström, M., Phng, L.-K., Hofmann, J.J., Wallgard, E., Coultas, L., Lindblom, P., Alva, J., Nilsson, A.-K., Karlsson, L., and Gaiano, N. (2007). Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature 445, 776. Herzog, Y., Kalcheim, C., Kahane, N., Reshef, R., and Neufeld, G. (2001). Differential expression of neuropilin-1 and neuropilin-2 in arteries and veins. Mechanisms of development 109, 115-119. Hiraoka, N., Allen, E., Apel, I.J., Gyetko, M.R., and Weiss, S.J. (1998). Matrix metalloproteinases regulate neovascularization by acting as pericellular fibrinolysins. Cell 95, 365-377. Hiratsuka, S., Minowa, O., Kuno, J., Noda, T., and Shibuya, M. (1998). Flt-1 lacking the tyrosine kinase domain is sufficient for normal development and angiogenesis in mice. Proceedings of the National Academy of Sciences 95, 9349. Ho, Y., Gruhler, A., Heilbut, A., Bader, G.D., Moore, L., Adams, S.L., Millar, A., Taylor, P., Bennett, K., Boutilier, K., et al. (2002). Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415, 180-183. Holmgren, L., O'Reilly, M.S., and Folkman, J. (1995). Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression. Nat Med 1, 149-153.

58 Holmqvist, K., Cross, M.J., Rolny, C., Hagerkvist, R., Rahimi, N., Matsumoto, T., Claesson-Welsh, L., and Welsh, M. (2004). The adaptor protein shb binds to tyrosine 1175 in vascular endothelial growth factor (VEGF) receptor-2 and regulates VEGF-dependent cellular migration. J Biol Chem 279, 22267-22275. Honkura, N., Richards, M., Lavina, B., Sainz-Jaspeado, M., Betsholtz, C., and Claesson-Welsh, L. (2018). Intravital imaging-based analysis tools for vessel identification and assessment of concurrent dynamic vascular events. Nature communications 9, 2746. Horowitz, A., and Seerapu, H.R. (2012). Regulation of VEGF signaling by membrane traffic. Cellular signalling 24, 1810-1820. Huang, Y., Snuderl, M., and Jain, R.K. (2011). Polarization of tumor-associated macrophages: a novel strategy for vascular normalization and antitumor immunity. Cancer Cell 19, 1-2. Ikeda, N., Adachi, M., Taki, T., Huang, C., Hashida, H., Takabayashi, A., Sho, M., Nakajima, Y., Kanehiro, H., Hisanaga, M., et al. (1999). Prognostic significance of angiogenesis in human pancreatic cancer. British journal of cancer 79, 1553- 1563. Iliopoulos, O., and Eng, C. (2000). Genetic and clinical aspects of familial renal neoplasms. Paper presented at: Seminars in oncology. Ito, N., Wernstedt, C., Engstrom, U., and Claesson-Welsh, L. (1998). Identification of vascular endothelial growth factor receptor-1 tyrosine phosphorylation sites and binding of SH2 domain-containing molecules. J Biol Chem 273, 23410- 23418. Jakobsson, L., Franco, C.A., Bentley, K., Collins, R.T., Ponsioen, B., Aspalter, I.M., Rosewell, I., Busse, M., Thurston, G., and Medvinsky, A. (2010). Endothelial cells dynamically compete for the tip cell position during angiogenic sprouting. Nature Cell Biology 12, 943. Jarvius, M., Paulsson, J., Weibrecht, I., Leuchowius, K.-J., Andersson, A.-C., Wählby, C., Gullberg, M., Botling, J., Sjöblom, T., and Markova, B. (2007). In situ detection of phosphorylated platelet-derived growth factor receptor β using a generalized proximity ligation method. Molecular & cellular proteomics 6, 1500-1509. Jayson, G.C., Kerbel, R., Ellis, L.M., and Harris, A.L. (2016). Antiangiogenic therapy in oncology: current status and future directions. Lancet (London, England) 388, 518-529. Jopling, H.M., Odell, A.F., Hooper, N.M., Zachary, I.C., Walker, J.H., and Ponnambalam, S. (2009). Rab GTPase regulation of VEGFR2 trafficking and signaling in endothelial cells. Arteriosclerosis, thrombosis, and vascular biology 29, 1119-1124. Joyal, J.S., Sitaras, N., Binet, F., Rivera, J.C., Stahl, A., Zaniolo, K., Shao, Z., Polosa, A., Zhu, T., Hamel, D., et al. (2011). Ischemic neurons prevent vascular regeneration of neural tissue by secreting semaphorin 3A. Blood 117, 6024- 6035. Jubb, A.M., Strickland, L.A., Liu, S.D., Mak, J., Schmidt, M., and Koeppen, H. (2012). Neuropilin-1 expression in cancer and development. J Pathol 226, 50- 60. Jung, H.Y., Fattet, L., and Yang, J. (2015). Molecular pathways: linking tumor microenvironment to epithelial-mesenchymal transition in metastasis. Clin Cancer Res 21, 962-968. Kamisawa, T., Isawa, T., Koike, M., Tsuruta, K., and Okamoto, A. (1995). Hematogenous metastases of pancreatic ductal carcinoma. Pancreas 11, 345- 349.

59 Kamisawa, T., Wood, L.D., Itoi, T., and Takaori, K. (2016). Pancreatic cancer. The Lancet 388, 73-85. Kamiya, T., Kawakami, T., Abe, Y., Nishi, M., Onoda, N., Miyazaki, N., Oida, Y., Yamazaki, H., Ueyama, Y., and Nakamura, M. (2006). The preserved expression of neuropilin (NRP) 1 contributes to a better prognosis in colon cancer. Oncol Rep 15, 369-373. Kanda, M., Matthaei, H., Wu, J., Hong, S.M., Yu, J., Borges, M., Hruban, R.H., Maitra, A., Kinzler, K., and Vogelstein, B. (2012). Presence of somatic mutations in most early-stage pancreatic intraepithelial neoplasia. Gastroenterology 142, 730-733. e739. Kappas, N.C., Zeng, G., Chappell, J.C., Kearney, J.B., Hazarika, S., Kallianos, K.G., Patterson, C., Annex, B.H., and Bautch, V.L. (2008). The VEGF receptor Flt-1 spatially modulates Flk-1 signaling and blood vessel branching. J Cell Biol 181, 847-858. Karihaloo, A., Karumanchi, S.A., Cantley, W.L., Venkatesha, S., Cantley, L.G., and Kale, S. (2005). Vascular endothelial growth factor induces branching morphogenesis/tubulogenesis in renal epithelial cells in a neuropilin-dependent fashion. Molecular and cellular biology 25, 7441-7448. Karkkainen, M.J., Haiko, P., Sainio, K., Partanen, J., Taipale, J., Petrova, T.V., Jeltsch, M., Jackson, D.G., Talikka, M., Rauvala, H., et al. (2004). Vascular endothelial growth factor C is required for sprouting of the first lymphatic vessels from embryonic veins. Nature immunology 5, 74-80. Katoh, O., Tauchi, H., Kawaishi, K., Kimura, A., and Satow, Y. (1995). Expression of the vascular endothelial growth factor (VEGF) receptor gene, KDR, in hematopoietic cells and inhibitory effect of VEGF on apoptotic cell death caused by ionizing radiation. Cancer Res 55, 5687-5692. Kawasaki, T., Kitsukawa, T., Bekku, Y., Matsuda, Y., Sanbo, M., Yagi, T., and Fujisawa, H. (1999). A requirement for neuropilin-1 in embryonic vessel formation. Development 126, 4895-4902. Kazemi, M., Carrer, A., Moimas, S., Zandonà, L., Bussani, R., Casagranda, B., Palmisano, S., Prelazzi, P., Giacca, M., Zentilin, L., et al. (2016). VEGF121 and VEGF165 differentially promote vessel maturation and tumor growth in mice and humans. Cancer Gene Therapy 23, 125. Kendall, R.L., Rutledge, R.Z., Mao, X., Tebben, A.J., Hungate, R.W., and Thomas, K.A. (1999). Vascular endothelial growth factor receptor KDR tyrosine kinase activity is increased by autophosphorylation of two activation loop tyrosine residues. J Biol Chem 274, 6453-6460. Kendall, R.L., and Thomas, K.A. (1993). Inhibition of vascular endothelial cell growth factor activity by an endogenously encoded soluble receptor. Proc Natl Acad Sci U S A 90, 10705-10709. Kim, Y.J., Jung, K., Baek, D.S., Hong, S.S., and Kim, Y.S. (2017). Co-targeting of EGF receptor and neuropilin-1 overcomes cetuximab resistance in pancreatic ductal adenocarcinoma with integrin beta1-driven Src-Akt bypass signaling. Oncogene 36, 2543-2552. Kitsukawa, T., Shimono, A., Kawakami, A., Kondoh, H., and Fujisawa, H. (1995). Overexpression of a membrane protein, neuropilin, in chimeric mice causes anomalies in the cardiovascular system, nervous system and limbs. Development 121, 4309-4318. Klagsbrun, M., Takashima, S., and Mamluk, R. (2002). The role of neuropilin in vascular and tumor biology. Adv Exp Med Biol 515, 33-48. Klein, D. (2018). The Tumor Vascular Endothelium as Decision Maker in Cancer Therapy. Frontiers in Oncology 8.

60 Ko, J.-A., Mizuno, Y., Yanai, R., Chikama, T.-i., and Sonoda, K.-H. (2010). Expression of semaphorin 3A and its receptors during mouse corneal development. Biochemical and biophysical research communications 403, 305- 309. Koch, S. (2012). Neuropilin signalling in angiogenesis. Biochemical Society transactions 40, 20-25. Koch, S., Tugues, S., Li, X., Gualandi, L., and Claesson-Welsh, L. (2011). Signal transduction by vascular endothelial growth factor receptors. The Biochemical journal 437, 169-183. Koch, S., van Meeteren, L.A., Morin, E., Testini, C., Westrom, S., Bjorkelund, H., Le Jan, S., Adler, J., Berger, P., and Claesson-Welsh, L. (2014). NRP1 presented in trans to the endothelium arrests VEGFR2 endocytosis, preventing angiogenic signaling and tumor initiation. Dev Cell 28, 633-646. Koike, T., Vernon, R.B., Hamner, M.A., Sadoun, E., and Reed, M.J. (2002). MT1- MMP, but not secreted MMPs, influences the migration of human microvascular endothelial cells in 3-dimensional collagen gels. J Cell Biochem 86, 748-758. Kolodkin, A.L., Levengood, D.V., Rowe, E.G., Tai, Y.T., Giger, R.J., and Ginty, D.D. (1997). Neuropilin is a semaphorin III receptor. Cell 90, 753-762. Kowanetz, M., and Ferrara, N. (2006). Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res 12, 5018-5022. Kreuter, M., Woelke, K., Bieker, R., Schliemann, C., Steins, M., Buechner, T., Berdel, W.E., and Mesters, R.M. (2006). Correlation of neuropilin-1 overexpression to survival in acute myeloid leukemia. Leukemia 20, 1950-1954. Kumar, R., and Kapoor, A. (2017). Current management of metastatic renal cell carcinoma: evolving new therapies. Current opinion in supportive and palliative care 11, 231-237. Lampugnani, M.G., Orsenigo, F., Gagliani, M.C., Tacchetti, C., and Dejana, E. (2006). Vascular endothelial cadherin controls VEGFR-2 internalization and signaling from intracellular compartments. J Cell Biol 174, 593-604. Lanahan, A., Zhang, X., Fantin, A., Zhuang, Z., Rivera-Molina, F., Speichinger, K., Prahst, C., Zhang, J., Wang, Y., Davis, G., et al. (2013). The neuropilin 1 cytoplasmic domain is required for VEGF-A-dependent arteriogenesis. Dev Cell 25, 156-168. Latil, A., Bieche, I., Pesche, S., Valeri, A., Fournier, G., Cussenot, O., and Lidereau, R. (2000). VEGF overexpression in clinically localized prostate tumors and neuropilin-1 overexpression in metastatic forms. Int J Cancer 89, 167-171. Lauren, P. (1965). The two histological main types of gastric carcinoma: diffuse and so‐called intestinal‐type carcinoma: an attempt at a histo‐clinical classification. Acta Pathologica Microbiologica Scandinavica 64, 31-49. Lee, S., Chen, T.T., Barber, C.L., Jordan, M.C., Murdock, J., Desai, S., Ferrara, N., Nagy, A., Roos, K.P., and Iruela-Arispe, M.L. (2007). Autocrine VEGF signaling is required for vascular homeostasis. Cell 130, 691-703. Lee, S., Jilani, S.M., Nikolova, G.V., Carpizo, D., and Iruela-Arispe, M.L. (2005). Processing of VEGF-A by matrix metalloproteinases regulates bioavailability and vascular patterning in tumors. J Cell Biol 169, 681-691. Li, L., Jiang, X., Zhang, Q., Dong, X., Gao, Y., He, Y., Qiao, H., Xie, F., Xie, X., and Sun, X. (2016a). Neuropilin-1 is associated with clinicopathology of gastric cancer and contributes to cell proliferation and migration as multifunctional co- receptors. Journal of experimental & clinical cancer research : CR 35, 16.

61 Li, X., Padhan, N., Sjostrom, E.O., Roche, F.P., Testini, C., Honkura, N., Sainz- Jaspeado, M., Gordon, E., Bentley, K., Philippides, A., et al. (2016b). VEGFR2 pY949 signalling regulates adherens junction integrity and metastatic spread. Nature communications 7, 11017. Li, Y., Zhang, F., Nagai, N., Tang, Z., Zhang, S., Scotney, P., Lennartsson, J., Zhu, C., Qu, Y., Fang, C., et al. (2008). VEGF-B inhibits apoptosis via VEGFR-1- mediated suppression of the expression of BH3-only protein genes in mice and rats. J Clin Invest 118, 913-923. Lindskog, M., Wahlgren, T., Sandin, R., Kowalski, J., Jakobsson, M., Lundstam, S., Ljungberg, B., and Harmenberg, U. (2017). Overall survival in Swedish patients with renal cell carcinoma treated in the period 2002 to 2012: Update of the RENCOMP study with subgroup analysis of the synchronous metastatic and elderly populations. Urologic Oncology: Seminars and Original Investigations 35, 541.e515-541.e522. Lunt, S.J., Chaudary, N., and Hill, R.P. (2009). The tumor microenvironment and metastatic disease. Clin Exp Metastasis 26, 19-34. Maitra, A., Adsay, N.V., Argani, P., Iacobuzio-Donahue, C., De Marzo, A., Cameron, J.L., Yeo, C.J., and Hruban, R.H. (2003). Multicomponent analysis of the pancreatic adenocarcinoma progression model using a pancreatic intraepithelial neoplasia tissue microarray. Modern Pathology 16, 902. Majno, G., and Palade, G. (1961). Studies on inflammation: I. The effect of histamine and serotonin on vascular permeability: An electron microscopic study. The Journal of cell biology 11, 571-605. Majno, G., Palade, G., and Schoefl, G.I. (1961). Studies on inflammation: II. The site of action of histamine and serotonin along the vascular tree: A topographic study. The Journal of cell biology 11, 607-626. Matkar, P.N., Jong, E.D., Ariyagunarajah, R., Prud'homme, G.J., Singh, K.K., and Leong-Poi, H. (2018). Jack of many trades: Multifaceted role of neuropilins in pancreatic cancer. Cancer medicine. Matsumoto, T., Bohman, S., Dixelius, J., Berge, T., Dimberg, A., Magnusson, P., Wang, L., Wikner, C., Qi, J.H., Wernstedt, C., et al. (2005). VEGF receptor-2 Y951 signaling and a role for the adapter molecule TSAd in tumor angiogenesis. The EMBO journal 24, 2342-2353. Maynard, S.E., Min, J.-Y., Merchan, J., Lim, K.-H., Li, J., Mondal, S., Libermann, T.A., Morgan, J.P., Sellke, F.W., Stillman, I.E., et al. (2003). Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. Journal of Clinical Investigation 111, 649-658. Mazzone, M., Dettori, D., Leite de Oliveira, R., Loges, S., Schmidt, T., Jonckx, B., Tian, Y.M., Lanahan, A.A., Pollard, P., Ruiz de Almodovar, C., et al. (2009). Heterozygous deficiency of PHD2 restores tumor oxygenation and inhibits metastasis via endothelial normalization. Cell 136, 839-851. McDonald, D.M., Thurston, G., and Baluk, P. (1999). Endothelial gaps as sites for plasma leakage in inflammation. Microcirculation 6, 7-22. Meadows, K.N., Bryant, P., and Pumiglia, K. (2001). Vascular endothelial growth factor induction of the angiogenic phenotype requires Ras activation. J Biol Chem 276, 49289-49298. Meyer, L.A., Fritz, J., Pierdant-Mancera, M., and Bagnard, D. (2016). Current drug design to target the Semaphorin/Neuropilin/Plexin complexes. Cell adhesion & migration 10, 700-708. Miaczynska, M., Pelkmans, L., and Zerial, M. (2004). Not just a sink: endosomes in control of signal transduction. Curr Opin Cell Biol 16, 400-406.

62 Miles, D.W., Dieras, V., Cortes, J., Duenne, A.A., Yi, J., and O'Shaughnessy, J. (2013). First-line bevacizumab in combination with chemotherapy for HER2- negative metastatic breast cancer: pooled and subgroup analyses of data from 2447 patients. Ann Oncol 24, 2773-2780. Monk, B.J., Poveda, A., Vergote, I., Raspagliesi, F., Fujiwara, K., Bae, D.S., Oaknin, A., Ray-Coquard, I., Provencher, D.M., Karlan, B.Y., et al. (2014). Anti-angiopoietin therapy with trebananib for recurrent ovarian cancer (TRINOVA-1): a randomised, multicentre, double-blind, placebo-controlled phase 3 trial. The Lancet Oncology 15, 799-808. Montero-Balaguer, M., Swirsding, K., Orsenigo, F., Cotelli, F., Mione, M., and Dejana, E. (2009). Stable vascular connections and remodeling require full expression of VE-cadherin in zebrafish embryos. PLoS One 4, e5772. Motzer, R.J., and Russo, P. (2000). Systemic therapy for renal cell carcinoma. The Journal of Urology 163, 408-417. Motzer, R.J., Tannir, N.M., McDermott, D.F., Arén Frontera, O., Melichar, B., Choueiri, T.K., Plimack, E.R., Barthélémy, P., Porta, C., George, S., et al. (2018). Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. New England Journal of Medicine 378, 1277-1290. Muhl, L., Folestad, E.B., Gladh, H., Wang, Y., Moessinger, C., Jakobsson, L., and Eriksson, U. (2017). Neuropilin 1 binds PDGF-D and is a co-receptor in PDGF- D–PDGFRβ signaling. Journal of cell science 130, 1365-1378. Murakami, M., Nguyen, L.T., Zhuang, Z.W., Moodie, K.L., Carmeliet, P., Stan, R.V., and Simons, M. (2008). The FGF system has a key role in regulating vascular integrity. J Clin Invest 118, 3355-3366. Naccache, S.N., Hasson, T., and Horowitz, A. (2006). Binding of internalized receptors to the PDZ domain of GIPC/synectin recruits myosin VI to endocytic vesicles. Proc Natl Acad Sci U S A 103, 12735-12740. Nagy, J.A., Chang, S.H., Shih, S.C., Dvorak, A.M., and Dvorak, H.F. (2010). Heterogeneity of the tumor vasculature. Semin Thromb Hemost 36, 321-331. Naik, A., Al-Zeheimi, N., Bakheit, C.S., Al Riyami, M., Al Jarrah, A., Al Moundhri, M.S., Al Habsi, Z., Basheer, M., and Adham, S.A. (2017). Neuropilin-1 Associated Molecules in the Blood Distinguish Poor Prognosis Breast Cancer: A Cross-Sectional Study. Scientific reports 7, 3301. NIH (2018). www.clinicaltrial.gov (U.S. National Library of Medicine: National Institute of Health), pp. https://clinicaltrials.gov/ct2/home. Nitta, T., Hata, M., Gotoh, S., Seo, Y., Sasaki, H., Hashimoto, N., Furuse, M., and Tsukita, S. (2003). Size-selective loosening of the blood-brain barrier in claudin-5–deficient mice. The Journal of cell biology 161, 653-660. Oettle, H., Neuhaus, P., Hochhaus, A., Hartmann, J.T., Gellert, K., Ridwelski, K., Niedergethmann, M., Zülke, C., Fahlke, J., and Arning, M.B. (2013). Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer: the CONKO-001 randomized trial. JAMA 310, 1473-1481. Ohtsu, A., Shah, M.A., Van Cutsem, E., Rha, S.Y., Sawaki, A., Park, S.R., Lim, H.Y., Yamada, Y., Wu, J., and Langer, B. (2011). Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a randomized, double-blind, placebo-controlled phase III study. J Clin Oncol 29, 3968-3976. Osada, H., Tokunaga, T., Nishi, M., Hatanaka, H., Abe, Y., Tsugu, A., Kijima, H., Yamazaki, H., Ueyama, Y., and Nakamura, M. (2004). Overexpression of the neuropilin 1 (NRP1) gene correlated with poor prognosis in human glioma. Anticancer research 24, 547-552.

63 Pan, Q., Chanthery, Y., Liang, W.C., Stawicki, S., Mak, J., Rathore, N., Tong, R.K., Kowalski, J., Yee, S.F., Pacheco, G., et al. (2007a). Blocking neuropilin-1 function has an additive effect with anti-VEGF to inhibit tumor growth. Cancer Cell 11, 53-67. Pan, Q., Chathery, Y., Wu, Y., Rathore, N., Tong, R.K., Peale, F., Bagri, A., Tessier-Lavigne, M., Koch, A.W., and Watts, R.J. (2007b). Neuropilin-1 binds to VEGF121 and regulates endothelial cell migration and sprouting. J Biol Chem 282, 24049-24056. Pang, H.-B., Braun, G.B., Friman, T., Aza-Blanc, P., Ruidiaz, M.E., Sugahara, K.N., Teesalu, T., and Ruoslahti, E. (2014). An endocytosis pathway initiated through neuropilin-1 and regulated by nutrient availability. Nature communications 5, 4904. Paoletti, X., Group, G., Oba, K., Burzykowski, T., Michiels, S., Ohashi, Y., Pignon, J.P., Rougier, P., Sakamoto, J., Sargent, D., et al. (2010). Benefit of adjuvant chemotherapy for resectable gastric cancer: a meta-analysis. JAMA 303, 1729- 1737. Pardali, E., Goumans, M.J., and ten Dijke, P. (2010). Signaling by members of the TGF-beta family in vascular morphogenesis and disease. Trends in cell biology 20, 556-567. Parikh, A.A., Fan, F., Liu, W.B., Ahmad, S.A., Stoeltzing, O., Reinmuth, N., Bielenberg, D., Bucana, C.D., Klagsbrun, M., and Ellis, L.M. (2004). Neuropilin-1 in human colon cancer: expression, regulation, and role in induction of angiogenesis. The American journal of pathology 164, 2139-2151. Patel, A.P., Tirosh, I., Trombetta, J.J., Shalek, A.K., Gillespie, S.M., Wakimoto, H., Cahill, D.P., Nahed, B.V., Curry, W.T., and Martuza, R.L. (2014). Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 344, 1396-1401. Patnaik, A., LoRusso, P.M., Messersmith, W.A., Papadopoulos, K.P., Gore, L., Beeram, M., Ramakrishnan, V., Kim, A.H., Beyer, J.C., Mason Shih, L., et al. (2014). A Phase Ib study evaluating MNRP1685A, a fully human anti-NRP1 monoclonal antibody, in combination with bevacizumab and paclitaxel in patients with advanced solid tumors. Cancer chemotherapy and pharmacology 73, 951-960. Pellet-Many, C., Frankel, P., Evans, I.M., Herzog, B., Jünemann-Ramírez, M., and Zachary, I.C. (2011). Neuropilin-1 mediates PDGF stimulation of vascular smooth muscle cell migration and signalling via p130Cas. Biochemical Journal 435, 609-618. Phng, L.K., and Gerhardt, H. (2009). Angiogenesis: A Team Effort Coordinated by Notch. Developmental Cell 16, 196-208. Potente, M., Gerhardt, H., and Carmeliet, P. (2011). Basic and therapeutic aspects of angiogenesis. Cell 146, 873-887. Potente, M., and Makinen, T. (2017). Vascular heterogeneity and specialization in development and disease. Nat Rev Mol Cell Biol 18, 477-494. Pugsley, M.K., and Tabrizchi, R. (2000). The vascular system: An overview of structure and function. Journal of Pharmacological and Toxicological Methods 44, 333-340. Raimondi, C., Fantin, A., Lampropoulou, A., Denti, L., Chikh, A., and Ruhrberg, C. (2014). Imatinib inhibits VEGF-independent angiogenesis by targeting neuropilin 1-dependent ABL1 activation in endothelial cells. The Journal of experimental medicine 211, 1167-1183.

64 Rak, J., Yu, J.L., Klement, G., and Kerbel, R.S. (2000). Oncogenes and angiogenesis: signaling three-dimensional tumor growth. J Investig Dermatol Symp Proc 5, 24-33. Raskopf, E., Vogt, A., Standop, J., Sauerbruch, T., and Schmitz, V. (2010). Inhibition of neuropilin-1 by RNA-interference and its angiostatic potential in the treatment of hepatocellular carcinoma. Zeitschrift fur Gastroenterologie 48, 21-27. Rini, B.I., Campbell, S.C., and Escudier, B. (2009). Renal cell carcinoma. Lancet (London, England) 373, 1119-1132. Rofstad, E.K., Galappathi, K., and Mathiesen, B.S. (2014). Tumor interstitial fluid pressure-a link between tumor hypoxia, microvascular density, and lymph node metastasis. Neoplasia (New York, NY) 16, 586-594. Rolny, C., Mazzone, M., Tugues, S., Laoui, D., Johansson, I., Coulon, C., Squadrito, M.L., Segura, I., Li, X., Knevels, E., et al. (2011). HRG inhibits tumor growth and metastasis by inducing macrophage polarization and vessel normalization through downregulation of PlGF. Cancer Cell 19, 31-44. Rostgaard, J., and Qvortrup, K. (1997). Electron microscopic demonstrations of filamentous molecular sieve plugs in capillary fenestrae. Microvascular research 53, 1-13. Roth, L., Prahst, C., Ruckdeschel, T., Savant, S., Westrom, S., Fantin, A., Riedel, M., Heroult, M., Ruhrberg, C., and Augustin, H.G. (2016). Neuropilin-1 mediates vascular permeability independently of vascular endothelial growth factor receptor-2 activation. Science signaling 9, ra42. Roy, S., Bag, A.K., Singh, R.K., Talmadge, J.E., Batra, S.K., and Datta, K. (2017). Multifaceted Role of Neuropilins in the Immune System: Potential Targets for Immunotherapy. Frontiers in immunology 8, 1228. Ruan, G.X., and Kazlauskas, A. (2012). Axl is essential for VEGF-A-dependent activation of PI3K/Akt. The EMBO journal 31, 1692-1703. Rubenstein, J.L., Kim, J., Ozawa, T., Zhang, M., Westphal, M., Deen, D.F., and Shuman, M.A. (2000). Anti-VEGF antibody treatment of glioblastoma prolongs survival but results in increased vascular cooption. Neoplasia (New York, NY) 2, 306-314. Saharinen, P., Eklund, L., Miettinen, J., Wirkkala, R., Anisimov, A., Winderlich, M., Nottebaum, A., Vestweber, D., Deutsch, U., Koh, G.Y., et al. (2008). Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell– cell and cell–matrix contacts. Nature Cell Biology 10, 527. Salikhova, A., Wang, L., Lanahan, A.A., Liu, M., Simons, M., Leenders, W.P., Mukhopadhyay, D., and Horowitz, A. (2008). Vascular endothelial growth factor and semaphorin induce neuropilin-1 endocytosis via separate pathways. Circ Res 103, e71-79. Sawamiphak, S., Seidel, S., Essmann, C.L., Wilkinson, G.A., Pitulescu, M.E., Acker, T., and Acker-Palmer, A. (2010). Ephrin-B2 regulates VEGFR2 function in developmental and tumour angiogenesis. Nature 465, 487-491. Schraml, P., Struckmann, K., Hatz, F., Sonnet, S., Kully, C., Gasser, T., Sauter, G., Mihatsch, M.J., and Moch, H. (2002). VHL mutations and their correlation with tumour cell proliferation, microvessel density, and patient prognosis in clear cell renal cell carcinoma. J Pathol 196, 186-193. Seerapu, H.R., Borthakur, S., Kong, N., Agrawal, S., Drazba, J., Vasanji, A., Fantin, A., Ruhrberg, C., Buck, M., and Horowitz, A. (2013). The cytoplasmic domain of neuropilin-1 regulates focal adhesion turnover. FEBS letters 587, 3392-3399.

65 Sekar, R.B., and Periasamy, A. (2003). Fluorescence resonance energy transfer (FRET) microscopy imaging of live cell protein localizations. The Journal of cell biology 160, 629-633. Selvaraj, D., Gangadharan, V., Michalski, C.W., Kurejova, M., Stosser, S., Srivastava, K., Schweizerhof, M., Waltenberger, J., Ferrara, N., Heppenstall, P., et al. (2015). A Functional Role for VEGFR1 Expressed in Peripheral Sensory Neurons in Cancer Pain. Cancer Cell 27, 780-796. Seo, Y., Baba, H., Fukuda, T., Takashima, M., and Sugimachi, K. (2000). High expression of vascular endothelial growth factor is associated with liver metastasis and a poor prognosis for patients with ductal pancreatic adenocarcinoma. Cancer 88, 2239-2245. Serra, H., Chivite, I., Angulo-Urarte, A., Soler, A., Sutherland, J.D., Arruabarrena- Aristorena, A., Ragab, A., Lim, R., Malumbres, M., Fruttiger, M., et al. (2015). PTEN mediates Notch-dependent stalk cell arrest in angiogenesis. Nature communications 6, 7935. Shahrabi-Farahani, S., Wang, L., Zwaans, B.M., Santana, J.M., Shimizu, A., Takashima, S., Kreuter, M., Coultas, L., D'Amore, P.A., Arbeit, J.M., et al. (2014). Neuropilin 1 expression correlates with differentiation status of epidermal cells and cutaneous squamous cell carcinomas. Lab Invest 94, 752- 765. Shibuya, M., and Claesson-Welsh, L. (2006). Signal transduction by VEGF receptors in regulation of angiogenesis and lymphangiogenesis. Exp Cell Res 312, 549-560. Shibuya, M., Yamaguchi, S., Yamane, A., Ikeda, T., Tojo, A., Matsushime, H., and Sato, M. (1990). Nucleotide sequence and expression of a novel human receptor-type tyrosine kinase gene (flt) closely related to the fms family. Oncogene 5, 519-524. Shweiki, D., Itin, A., Soffer, D., and Keshet, E. (1992). Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 359, 843-845. Siegel, R., Ma, J., Zou, Z., and Jemal, A. (2014). Cancer statistics, 2014. CA: a cancer journal for clinicians 64, 9-29. Sims, D.E. (1991). Recent advances in pericyte biology--implications for health and disease. The Canadian journal of cardiology 7, 431-443. Sledge, G.W. (2015). Anti-vascular endothelial growth factor therapy in breast cancer: game over? J Clin Oncol 33, 133-135. Smith, D.C., Eisenberg, P.D., Manikhas, G., Chugh, R., Gubens, M.A., Stagg, R.J., Kapoun, A.M., Xu, L., Dupont, J., and Sikic, B. (2014). A phase I dose escalation and expansion study of the anticancer stem cell agent demcizumab (anti-DLL4) in patients with previously treated solid tumors. Clin Cancer Res 20, 6295-6303. Soker, S., Miao, H.Q., Nomi, M., Takashima, S., and Klagsbrun, M. (2002). VEGF165 mediates formation of complexes containing VEGFR-2 and neuropilin-1 that enhance VEGF165-receptor binding. Journal of cellular biochemistry 85, 357-368. Soker, S., Takashima, S., Miao, H.Q., Neufeld, G., and Klagsbrun, M. (1998). Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell 92, 735-745. Sugahara, K.N., Teesalu, T., Karmali, P.P., Kotamraju, V.R., Agemy, L., Greenwald, D.R., and Ruoslahti, E. (2010). Coadministration of a tumor- penetrating peptide enhances the efficacy of cancer drugs. Science 328, 1031- 1035.

66 Sun, Z., Li, X., Massena, S., Kutschera, S., Padhan, N., Gualandi, L., Sundvold- Gjerstad, V., Gustafsson, K., Choy, W.W., Zang, G., et al. (2012). VEGFR2 induces c-Src signaling and vascular permeability in vivo via the adaptor protein TSAd. J Exp Med 209, 1363-1377. Söderberg, O., Gullberg, M., Jarvius, M., Ridderstråle, K., Leuchowius, K.-J., Jarvius, J., Wester, K., Hydbring, P., Bahram, F., and Larsson, L.-G. (2006). Direct observation of individual endogenous protein complexes in situ by proximity ligation. Nature methods 3, 995. Söderberg, O., Leuchowius, K.J., Gullberg, M., Jarvius, M., Weibrecht, I., Larsson, L.G., and Landegren, U. (2008). Characterizing proteins and their interactions in cells and tissues using the in situ proximity ligation assay. Methods 45, 227-232. Taddei, A., Giampietro, C., Conti, A., Orsenigo, F., Breviario, F., Pirazzoli, V., Potente, M., Daly, C., Dimmeler, S., and Dejana, E. (2008). Endothelial adherens junctions control tight junctions by VE-cadherin-mediated upregulation of claudin-5. Nature Cell Biology 10, 923. Takahashi, T., Fournier, A., Nakamura, F., Wang, L.H., Murakami, Y., Kalb, R.G., Fujisawa, H., and Strittmatter, S.M. (1999). Plexin-neuropilin-1 complexes form functional semaphorin-3A receptors. Cell 99, 59-69. Testini, C., Smith, R., Jin, Y., Shibuya, M., Hellstrom, M., and Claesson-Welsh, L. (2018). VEGFR2 pY1212 signaling regulates Myc pathway signaling and endothelial cell proliferation in a strain-dependent manner. Unpublished data. Thiery, J.P., Acloque, H., Huang, R.Y., and Nieto, M.A. (2009). Epithelial- mesenchymal transitions in development and disease. Cell 139, 871-890. Thurston, G., Murphy, T.J., Baluk, P., Lindsey, J.R., and McDonald, D.M. (1998). Angiogenesis in mice with chronic airway inflammation: strain-dependent differences. The American journal of pathology 153, 1099-1112. Tilton, R.G., Kilo, C., and Williamson, J.R. (1979). Pericyte-endothelial relationships in cardiac and skeletal muscle capillaries. Microvascular research 18, 325-335. Tong, R.T., Boucher, Y., Kozin, S.V., Winkler, F., Hicklin, D.J., and Jain, R.K. (2004). Vascular normalization by vascular endothelial growth factor receptor 2 blockade induces a pressure gradient across the vasculature and improves drug penetration in tumors. Cancer Res 64, 3731-3736. Tornavaca, O., Chia, M., Dufton, N., Almagro, L.O., Conway, D.E., Randi, A.M., Schwartz, M.A., Matter, K., and Balda, M.S. (2015). ZO-1 controls endothelial adherens junctions, cell–cell tension, angiogenesis, and barrier formation. J Cell Biol, jcb. 201404140. Uetz, P., Giot, L., Cagney, G., Mansfield, T.A., Judson, R.S., Knight, J.R., Lockshon, D., Narayan, V., Srinivasan, M., Pochart, P., et al. (2000). A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature 403, 623-627. Wagner, A.D., Unverzagt, S., Grothe, W., Kleber, G., Grothey, A., Haerting, J., and Fleig, W.E. (2010). Chemotherapy for advanced gastric cancer. Cochrane database of systematic reviews. Wallerius, M., Wallmann, T., Bartish, M., Ostling, J., Mezheyeuski, A., Tobin, N.P., Nygren, E., Pangigadde, P., Pellegrini, P., Squadrito, M.L., et al. (2016). Guidance Molecule SEMA3A Restricts Tumor Growth by Differentially Regulating the Proliferation of Tumor-Associated Macrophages. Cancer Res 76, 3166-3178. Waltenberger, J., Claesson-Welsh, L., Siegbahn, A., Shibuya, M., and Heldin, C.H. (1994). Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem 269, 26988-26995.

67 Van Cutsem, E., Bang, Y.-J., Feng-yi, F., Xu, J.M., Lee, K.-W., Jiao, S.-C., Chong, J.L., López-Sanchez, R.I., Price, T., and Gladkov, O. (2015). HER2 screening data from ToGA: targeting HER2 in gastric and gastroesophageal junction cancer. Gastric Cancer 18, 476-484. Van Cutsem, E., de Haas, S., Kang, Y.K., Ohtsu, A., Tebbutt, N.C., Ming Xu, J., Peng Yong, W., Langer, B., Delmar, P., Scherer, S.J., et al. (2012a). Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the AVAGAST randomized phase III trial. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 30, 2119-2127. Van Cutsem, E., Sagaert, X., Topal, B., Haustermans, K., and Prenen, H. (2016). Gastric cancer. The Lancet 388, 2654-2664. Van Cutsem, E., Tabernero, J., Lakomy, R., Prenen, H., Prausova, J., Macarulla, T., Ruff, P., van Hazel, G.A., Moiseyenko, V., Ferry, D., et al. (2012b). Addition of aflibercept to fluorouracil, leucovorin, and irinotecan improves survival in a phase III randomized trial in patients with metastatic colorectal cancer previously treated with an oxaliplatin-based regimen. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 30, 3499-3506. Wang, F., Meng, W., Wang, B., and Qiao, L. (2014). Helicobacter pylori-induced gastric inflammation and gastric cancer. Cancer letters 345, 196-202. Wang, J.-C., Li, G.-Y., Li, P.-P., Sun, X., Li, W.-M., Li, Y., Lu, S.-Y., and Liu, P.-J. (2017). Suppression of hypoxia-induced excessive angiogenesis by metformin via elevating tumor blood perfusion. Oncotarget 8, 73892. Wang, L., Zeng, H., Wang, P., Soker, S., and Mukhopadhyay, D. (2003). Neuropilin-1-mediated vascular permeability factor/vascular endothelial growth factor-dependent endothelial cell migration. J Biol Chem 278, 48848-48860. Weekes, C.D., Beeram, M., Tolcher, A.W., Papadopoulos, K.P., Gore, L., Hegde, P., Xin, Y., Yu, R., Shih, L.M., Xiang, H., et al. (2014). A phase I study of the human monoclonal anti-NRP1 antibody MNRP1685A in patients with advanced solid tumors. Investigational new drugs 32, 653-660. Whitaker, G.B., Limberg, B.J., and Rosenbaum, J.S. (2001). Vascular endothelial growth factor receptor-2 and neuropilin-1 form a receptor complex that is responsible for the differential signaling potency of VEGF(165) and VEGF(121). J Biol Chem 276, 25520-25531. Wilhelm, K., Happel, K., Eelen, G., Schoors, S., Oellerich, M.F., Lim, R., Zimmermann, B., Aspalter, I.M., Franco, C.A., Boettger, T., et al. (2016). FOXO1 couples metabolic activity and growth state in the vascular endothelium. Nature 529, 216-220. Winkler, F., Kozin, S.V., Tong, R.T., Chae, S.S., Booth, M.F., Garkavtsev, I., Xu, L., Hicklin, D.J., Fukumura, D., di Tomaso, E., et al. (2004). Kinetics of vascular normalization by VEGFR2 blockade governs brain tumor response to radiation: role of oxygenation, angiopoietin-1, and matrix metalloproteinases. Cancer cell 6, 553-563. Wisse, E. (1974). Kupffer cell reactions in rat liver under various conditions as observed in the electron microscope. Journal of ultrastructure research 46, 499- 520. Wnuk, M., Anderegg, M.A., Graber, W.A., Buergy, R., Fuster, D.G., and Djonov, V. (2017). Neuropilin1 regulates glomerular function and basement membrane composition through pericytes in the mouse kidney. Kidney international 91, 868-879.

68 Wood, L.D., and Hruban, R.H. (2012). Pathology and molecular genetics of pancreatic neoplasms. Cancer journal (Sudbury, Mass) 18, 492-501. Wu, Y.Y., Chen, Y.L., Jao, Y.C., Hsieh, I.S., Chang, K.C., and Hong, T.M. (2014). miR-320 regulates tumor angiogenesis driven by vascular endothelial cells in oral cancer by silencing neuropilin 1. Angiogenesis 17, 247-260. Yagoda, A., Abi-Rached, B., and Petrylak, D. (1995). Chemotherapy for advanced renal-cell carcinoma: 1983-1993. Paper presented at: Seminars in oncology. Yang, W.J., Hu, J., Uemura, A., Tetzlaff, F., Augustin, H.G., and Fischer, A. (2015). Semaphorin‐3C signals through Neuropilin‐1 and PlexinD1 receptors to inhibit pathological angiogenesis. EMBO Molecular Medicine 7, 1267-1284. Yeo, S.G., Kim, J.S., Cho, M.J., Kim, K.H., and Kim, J.S. (2009). Interstitial fluid pressure as a prognostic factor in cervical cancer following radiation therapy. Clin Cancer Res 15, 6201-6207. You, W.K., and McDonald, D.M. (2008). The hepatocyte growth factor/c-Met signaling pathway as a therapeutic target to inhibit angiogenesis. BMB Rep 41, 833-839. Young, R.J., and Reed, M.W. (2012). Anti-angiogenic therapy: concept to clinic. Microcirculation 19, 115-125. Yu, T., Wang, Z., Liu, K., Wu, Y., Fan, J., Chen, J., Li, C., Zhu, G., and Li, L. (2014). High interstitial fluid pressure promotes tumor progression through inducing lymphatic metastasis-related protein expressions in oral squamous cell carcinoma. Clinical & translational oncology : official publication of the Federation of Spanish Oncology Societies and of the National Cancer Institute of Mexico 16, 539-547. Yuan, L., Moyon, D., Pardanaud, L., Breant, C., Karkkainen, M.J., Alitalo, K., and Eichmann, A. (2002). Abnormal lymphatic vessel development in neuropilin 2 mutant mice. Development 129, 4797-4806. Zacchigna, S., Pattarini, L., Zentilin, L., Moimas, S., Carrer, A., Sinigaglia, M., Arsic, N., Tafuro, S., Sinagra, G., and Giacca, M. (2008). Bone marrow cells recruited through the neuropilin-1 receptor promote arterial formation at the sites of adult neoangiogenesis in mice. J Clin Invest 118, 2062-2075. Zachary, I.C., Frankel, P., Evans, I.M., and Pellet-Many, C. (2009). The role of neuropilins in cell signalling. Biochemical Society transactions 37, 1171-1178. Zhang, Y., He, X., Liu, Y., Ye, Y., Zhang, H., He, P., Zhang, Q., Dong, L., Liu, Y., and Dong, J. (2012). microRNA-320a inhibits tumor invasion by targeting neuropilin 1 and is associated with liver metastasis in colorectal cancer. Oncol Rep 27, 685-694. Zhao, H., Maruyama, T., Hattori, Y., Sugo, N., Takamatsu, H., Kumanogoh, A., Shirasaki, R., and Yamamoto, N. (2011). A molecular mechanism that regulates medially oriented axonal growth of upper layer neurons in the developing neocortex. Journal of Comparative Neurology 519, 834-848. Zhao, Q., Eichten, A., Parveen, A., Adler, C., Huang, Y., Wang, W., Ding, Y., Adler, A., Nevins, T., Ni, M., et al. (2018). Single-Cell Transcriptome Analyses Reveal Endothelial Cell Heterogeneity in Tumors and Changes following Antiangiogenic Treatment. Cancer Research 78, 2370. Zhou, F., Chang, Z., Zhang, L., Hong, Y.K., Shen, B., Wang, B., Zhang, F., Lu, G., Tvorogov, D., Alitalo, K., et al. (2010). Akt/Protein kinase B is required for lymphatic network formation, remodeling, and valve development. The American journal of pathology 177, 2124-2133. Zlokovic, B.V. (2011). Neurovascular pathways to neurodegeneration in Alzheimer's disease and other disorders. Nature reviews Neuroscience 12, 723- 738.

69 Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1512 Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine”.)

ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-364415 2018