**11. Adult zebrafish models**

While embryonic zebrafish models have yielded many valuable insights into human vascular or vessel-related pathologies, most diseases strike adult patients, and accordingly, their clinical symptoms are expected to be more closely recapitulated by adult disease models (Dahl Ejby Jensen et al, 2009). To this end there are today also a growing number of adult zebrafish models available which are highly valuable in the study of angiogenesisrelated disorders including retinopathy, regeneration/wound healing and cancer.

### **11.1 Zebrafish retinal angiogenesis models**

Diabetic retinopathy and age-related macular degeneration are severely debilitating disorders in which angiogenesis is a major driving force of the pathology. During progression of these diseases, retinal hypoxia induces pathological growth of immature and fragile blood vessels in the retina, which is associated with the progression to severe states of the diseases. Mice are not convenient to use for studies on retinal hypoxia, as the traditional methods of generating tissue hypoxia in mice (vessel occlusion/ligation) are not applicable in the retina, as there are no easily reachable arteries which can be ligated without causing excessive damage to the animal. Furthermore, mice cannot be exposed to severely hypoxic environments, as their respiration system is not adapted to withstand low atmospheric oxygen levels.

Zebrafish are much more robust to environmental hypoxia, and can withstand even very low oxygen levels in the water for a long time. In contrast to mammals, by incubating zebrafish in an aquarium, where the oxygen levels in the water can be tightly controlled by regulating the perfusion of nitrogen gas, researchers are given full control on the precise degree of hypoxia, the amount of time the tissue experiences hypoxia and the possibilities of distinguishing hypoxia-effects from other effects of restricted circulation such as acidosis and accumulation of waste products.

The zebrafish retinal vasculature is furthermore particularly amenable to studies on angiogenesis due to its remarkably simple structure (Fig. 16). As in mice and humans, the

promoter (*zRag2*). Onset of tumors varied between 30 to 131 days post injection of the *Rag2 mMyc* expression plasmid. Fish developing T-ALL were characterized by inflated abdominal cavities and infiltration of malignant cells (transformed lymphoblasts) throughout the body, under the skin, into base of the pectoral fin, olfactory region, and the retrorbital soft tissue

These malignant lymphoblasts were transplantable into irradiated wild type adult zebrafish giving rise to small tumors appearing already one week post transplantation (Langenau et

Rhabdomyosarcoma (RMS) is a very aggressive soft tissue sarcoma with high incidence seen among children compared to other types of cancer. The zebrafish RMS model was generated by injecting a Rag2-kRASG12D construct, expressing a constitutively active isoform of RAS under the Rag2 promoter, into the zebrafish embryo at one cell stage. Visible highly invasive tumors in liver, intestine, kidney, and testes appeared already 10 days post injection, in accordance with the early onset of these tumors in human patients, distinguishing this model as one of the fastest spontaneous, vertebrate tumor models available (Langenau et al, 2007).

While embryonic zebrafish models have yielded many valuable insights into human vascular or vessel-related pathologies, most diseases strike adult patients, and accordingly, their clinical symptoms are expected to be more closely recapitulated by adult disease models (Dahl Ejby Jensen et al, 2009). To this end there are today also a growing number of adult zebrafish models available which are highly valuable in the study of angiogenesis-

Diabetic retinopathy and age-related macular degeneration are severely debilitating disorders in which angiogenesis is a major driving force of the pathology. During progression of these diseases, retinal hypoxia induces pathological growth of immature and fragile blood vessels in the retina, which is associated with the progression to severe states of the diseases. Mice are not convenient to use for studies on retinal hypoxia, as the traditional methods of generating tissue hypoxia in mice (vessel occlusion/ligation) are not applicable in the retina, as there are no easily reachable arteries which can be ligated without causing excessive damage to the animal. Furthermore, mice cannot be exposed to severely hypoxic environments, as their respiration system is not adapted to withstand low

Zebrafish are much more robust to environmental hypoxia, and can withstand even very low oxygen levels in the water for a long time. In contrast to mammals, by incubating zebrafish in an aquarium, where the oxygen levels in the water can be tightly controlled by regulating the perfusion of nitrogen gas, researchers are given full control on the precise degree of hypoxia, the amount of time the tissue experiences hypoxia and the possibilities of distinguishing hypoxia-effects from other effects of restricted circulation such as acidosis

The zebrafish retinal vasculature is furthermore particularly amenable to studies on angiogenesis due to its remarkably simple structure (Fig. 16). As in mice and humans, the

related disorders including retinopathy, regeneration/wound healing and cancer.

that led to splayed eyes.

**10.4 Rhabdomyocarcoma** 

**11. Adult zebrafish models** 

atmospheric oxygen levels.

and accumulation of waste products.

**11.1 Zebrafish retinal angiogenesis models** 

al, 2003)

retinal vasculature is supplied with blood from a major optic artery, which branches out at the center of the optic disc to form approximately 4-7 so-called grade I arteries which cover the inner surface of the retina. These arteries branch a few more times before anastomosing with protrusions from the circumferential vein. Thus, the zebrafish retina has a simple monolayer vasculature that is organized from the center to the periphery as arteriescapillaries-veins, which is in contrast to that in mice where the vasculature is multi-layered and arteries, capillaries and veins are co-localized throughout the retina (R Cao et al, 2008).

Fig. 16. Hypoxia-induced retinal angiogenesis model. Adult *Fli1:EGFP* transgenic zebrafish were incubated in 10% air-saturated water for 12 days, which induce marked angiogenic expansion of the capillary area of the retinal vasculature (middle figure) compared to fish in normoxic water (left figure). Hypoxia-induced retinal angiogenesis is dependent on VEGF signaling as the VEGF receptor inhibitor sunitnib is able to block the formation of new vessels under hypoxia in the zebrafish retina (right figure). Yellow arrowheads indicate angiogenic sprouts. Figure adapted from R Cao et al, 2008.

This retinal vasculature is highly sensitive to hypoxia. After only a few days of exposure to environmental hypoxia, a marked angiogenic expansion of the capillary area is clearly observed, which can be modulated either by the degree, or the amount of time, the fish is kept in hypoxia. An example of hypoxia-induced retinal angiogenesis in the adult zebrafish is given in Fig. 16.

In this model, adult zebrafish are put into a specifically designed aquarium where the water oxygen levels can be automatically controlled and monitored using an oxygen sensor coupled to a valve which opens for perfusion of nitrogen gas when the oxygen levels are above a set level defined by the researcher. Zebrafish can withstand very low oxygen levels better when being gradually adapted, so by slowly reducing the oxygen concentration in the hypoxia aquarium over the course of 1-2 days, the fish can be maintained in water with an oxygen concentration of only 8-10% of that in fully air-saturated water (Z Cao et al, 2010). After exposure to this level of hypoxia for a few days, retinal neovascularization is readily detected post mortem, under a fluorescent microscope. Also in this assay, it is convenient to add orally active drugs to the water during exposure to hypoxia, to study how such drugs may interfere with hypoxia-induced retinal neovascularization. Hypoxia-induced newly formed vessels can easily be distinguished due to the normally very simple and organized vasculature of the retina in adult zebrafish. Quantifying the number of sprouts, new branches and vascularization area are convenient ways of comparing the effects of the investigated drugs compared to non-treated controls (Z Cao et al, 2010). While capillary sprouting begins on day 3 after exposure to hypoxia, formation of a neovascular network becomes obvious on day 6 after exposure to hypoxia (R Cao et al, 2008).

Fig. 17. Regenerating tail fin assay. The distal 1/3 of adult *Fli1:EGFP* transgenic zebrafish tail fins was amputated under anaestisia, and the fish were placed in either clean water (left panels) or water containing the phosphatidyl-inositol-3-kinase (PI3K) inhibitor LY294002 (right panels). After 6 days the fish were sacrificed, the tail fins were fixed in 4% PFA and the regeneration could be determined by bright-field microscopy (A). The vasculature in the regenerating tail fin tissue was observed by fluorescence microscopy (B). Quantification of the regeneration length is an easy and convenient way to demonstrate the effects of the tested drugs (C). Bars indicate 0.5 mm. Figure reproduced from Alvarez et al, 2009.

Regeneration length/mm

Fig. 17. Regenerating tail fin assay. The distal 1/3 of adult *Fli1:EGFP* transgenic zebrafish tail fins was amputated under anaestisia, and the fish were placed in either clean water (left panels) or water containing the phosphatidyl-inositol-3-kinase (PI3K) inhibitor LY294002 (right panels). After 6 days the fish were sacrificed, the tail fins were fixed in 4% PFA and the regeneration could be determined by bright-field microscopy (A). The vasculature in the regenerating tail fin tissue was observed by fluorescence microscopy (B). Quantification of the regeneration length is an easy and convenient way to demonstrate the effects of the tested drugs (C). Bars indicate 0.5 mm. Figure reproduced from Alvarez et al, 2009.

This experiment allows researchers to study the mechanisms of hypoxia-induced angiogenesis *in vivo*, to study neovascularization, vascular remodeling, and leakiness under pathological settings, to correlate vascular changes with disease development, to screen potential orally active therapeutic agents, and to assess the beneficial effects of known antiangiogenic agents for the treatment of retinopathy. However, there are also limitations. Unlike the clinical situation, this model does not induce local tissue hypoxia but rather rely on systemic exposure to hypoxia. Also, owing to the limitation of available antibodies against zebrafish proteins, staining of particular cell subsets in the retina may be difficult, reducing the amount of information that can be obtained from histological examinations compared to mouse models.

## **11.2 Angiogenesis in the regenerating zebrafish fin**

Adult zebrafish have a remarkable regenerative capability. Many tissues which may not be regenerated in mammals are quickly regenerated in zebrafish. Among these are the heart, retina, maxillary barbell and fins (Poss et al, 2002; Vihtelic & Hyde, 2000; Alvarez et al, 2009). Importantly, as they regenerate, new blood and lymph vessels grow into the regenerating tissue – which enables studies on regenerative angiogenesis. One commonly used assay in the adult zebrafish, based on this principle is the regenerating tail fin. After amputation, the tail fin will re-grow and after approximately 1 month, the fin is back to its original size (Huang et al, 2003). This process of fin regeneration encompasses many of the same mechanisms as in human wound healing and regeneration, and is therefore a good model of regenerative angiogenesis. As the fin is largely transparent, and as morpholinos can be introduced by microinjection and electroporation, this assay is almost as versatile as the zebrafish developmental angiogenesis assays, the major difference being that it is performed in an adult animal. Also in the fins, the vasculature is remarkably simple (Fig. 17), and as the fin grows back various levels of vascular remodeling can be observed and therefore studied in detail (Huang et al, 2003). This regeneration model is probably the most commonly used, adult zebrafish angiogenesis model, and is considered to be complementary to the developmental angiogenesis models.
