Angiogenesis, the growth of blood vessels from an existing vasculature, can occur by sprouting from preexisting vessels or by vessel splitting (intussusception). Pathological angiogenesis drives choroidal neovascularization (CNV) in age related macular degeneration (AMD) which is commonly restricted under the retinal pigment epithelium (RPE), called occult CNV, but may also involve vessels penetrating through the RPE into the sub-retinal space. Pathological vessels are poorly developed, insufficiently perfused and highly leaky, phenotypes that are considered to drive disease progression and lead to poor prognosis. Currently, a number of anti-angiogenic drugs exists, the majority of which target vascular endothelial factor (VEGF), but although they often are highly beneficial for treating eye diseases in the short-term, they are generally of limited efficacy in other diseases such as cancer, and also have poorer efficacy when used for treatment of eye diseases in the long-term. A better understanding of the mechanisms underlying pathological angiogenesis can generate new targets for treatment leading to development of better drugs for cancer and retinopathies, but perhaps also other angiogenesis-dependent diseases, in the future. In this thesis mechanisms involved in developmental angiogenesis or pathological angiogenesis in the choroid, cornea or melanoma was identified. These findings highlight the need to further elaborate our knowledge related to angiogenesis in different tissues/conditions for a more targeted, and potentially effective treatment of diseases in the future.

In paper I, we for the first time identified the choriocapillaries (CCs) in adult zebrafish and found that occult CNV could be induced by exposing the fish to severe hypoxia. Interestingly, we found that occult CNV relied on intussusception, involving not only de novo generation of intussusceptive pillars but also a previously poorly understood mechanism called pillar splitting. This involved HIF-VEGF-VEGFR2 signaling and evidence that this also occurred in both rats and humans suffering from AMD suggested that the mechanism was conserved and clinically relevant.

In contrast, we found in paper II that the development of CCs in the zebrafish relies on sprouting angiogenesis, involve continuous remodeling, and delayed maturation of the vasculature in 2D. The initial development was found to occur by a unique process of tissuewide synchronized vasculogenesis. As expected, VEGFA via VEGFR2 was also critical for the development of these vessels in the zebrafish embryo, but surprisingly this was independent on hypoxia-inducible factor (HIF)-1.

Inflammatory nuclear factor-kB (NF-kB) signaling is involved in the progression of angiogenesis, but this signaling pathway has mainly been studied in the inflammatory cells and the role of NF-kB in the endothelial cells during angiogenesis is poorly understood. In paper III, we found that blocking NF-kB signaling using a specific IKK2 blocker IMD0354, specifically blocks pathological as well as developmental angiogenesis by targeting endothelial cell NF-kB signaling in the endothelial cells. Using a rat model for suture-induced corneal neovascularization, IMD0354 treatment lead to reduced production of inflammatory C-C motif chemokine ligand 2 (CCL2), C-X-C motif chemokine ligand 5 (CXCL5) and VEGF, and thereby reduced pathological corneal angiogenesis in this model.

Using the zebrafish tumor xenograft model in paper IV, we found an association between Microphthalmia associated transcription factor (MITF) and pigment epithelium derived factor (PEDF), which was involved in pathological tumor angiogenesis and metastasis. Similarly, in paper V we used zebrafish transplantation models to study and investigate the use of biocompatible polymers for the delivery of pro-angiogenic FGF-2 as a potential treatment strategy for ischemic diseases such as myocardial infarction (MI). Conclusively, this thesis provides new insights into diverse fields of angiogenic assays using zebrafish, and reveals new mechanisms of angiogenesis in health and disease. This work will hopefully provide a foundation for further studies into occult CNV related to AMD, a process that has not been possible to study previously in pre-clinical models. In addition, zebrafish xenograft or other transplantation models used in this work will likely be important to study cancer biology and to develop more attractive pharmaceutical preparations based on biocompatible hydrogels formulated as microspheres in the future.

Pathologic angiogenesis is involved in cancer and several blinding conditions such as wet age-related macular degeneration, proliferative retinopathies and corneal neovascularization.

In these dieseases, the angiogenic triggers are hypoxia and inflammation, and both involve the main angiogenic mediator, which is Vascular Endothelial Growth Factor (VEGF). Among available treatments, anti-VEGF often shows limited or temporary efficacy, while steroids are potentially responsible for many side-effects. This thesis presents a series of linked studies aimed at elucidating the early pathologic changes leading to inflammation and corneal neovascularization, and how various treatments affect this process. In this thesis, anti-inflammatory and anti-angiogenic treatments are applied in corneal neovascularization models, to identify VEGF-independent pathways and other novel factors as future therapy targets, as well as to investigate the endogenous modulation of angiogenesis.

A model of experimental neovascularization in the rat cornea was used as main model, where the neovascular response is triggered by a surgical suture placed into the cornea. Investigational treatments (anti-Vegf, dexamethasone, IMD0354, Gap27, or control substances) were then given topically, with the exception of IMD0354, which was given systemically. The effects in the cornea were studied in vivo with slit lamp photography to assess and quantify macroscopic vessel growth and using in vivo confocal microscopy (IVCM) to study cell infiltration and limbal vessel dilation and detect microscopic vessel sprouts; these examinations were performed longitudinally. Genomic analysis with RNA microarray, selected gene expression with q-RT-PCR, and selected protein expression in tissue (immunohistochemistry, immunofluorescence, Western blot) were performed at different time-points. Moreover, other experiments on cell cultures (HUVEC and HCEC), organ cultures (human corneas), ex vivo models (aortic rings) and in vivo studies (zebrafish vasculogenesis) were performed.

Dexamethasone suppressed limbal vasodilation and corneal neovascularization more than anti-Vegf, despite no difference in inflammatory cell infiltration into the cornea. Five-hundred eleven fewer genes were differentially expressed in dexamethasone-treated corneas relative to naïve corneas, compared to anti-Vegf. Among them, several major pro-angiogenic and pro-inflammatory factors and chemokines were suppressed only by dexamethasone and represent novel candidate factors to target in order to improve anti-VEGF treatment. On the other hand, selective inhibition of a single inflammatory pathway (NF-κB), despite showing similar early effects as dexamethasone in suppressing tissue inflammation, was not effective enough to suppress new vessel growth. The same factors suppressed by dexamethasone are also inhibited in endogenous modulation of angiogenesis. Surprisingly, dexamethasone activated several complement factors, which could possibly be beneficial in the anti-angiogenic response.

In a different therapeutic approach, promoting cell migration to accelerate epithelial wound closure similarly was not sufficient to avoid inflammation and angiogenesis in the cornea.

In conclusion, new and more effective treatments are needed for corneal inflammation and neovascularization with fewer side-effects. In this thesis, several novel factors and mechanisms related to inflammation are identified, factors that are not addressed by anti-Vegf therapy, and therefore represent interesting objects for further study, as they have the potential to be targets for adjuvant therapy. Specific anti-inflammatory treatment as well as therapeutic activation of endogenous regulatory pathways, and potentially complement modulation, might represent new strategies to improve anti-angiogenic therapy, but when used alone they do not seem to avoid corneal neovascularization.