Development of a Hypoxia-Induced Zebrafish Choroidal Neovascularization Model - Zebrafish: Methods for Assessing Drug Safety and Toxicity - page 203

16.3.1 Optimization of CoCl 2 Treatment for
Generating a Hypoxia-Induced Zebrafish CNV Model

We treated 1dpf zebrafish with CoCl 2 concentrations ranging from 0.01 to 25mg/mL.
Using whole mount immunostaining with Phy-V antibody, which is specific for
activated endothelial cells (ECs) (Seng et al., 2004), we observed that 63.6

11.1%
treated zebrafish exhibited specific CNV phenotypes. Phy-V antibody staining was
observed throughout the vasculature including vessels, eyes, CVP, trunk/tail region,
and intersegmental vessels (ISVs) (Serbedzija et al., 2006). Although the ISV pattern
was normal (data not shown), a network of new vessels exhibiting strong staining in
the CVP indicated abnormal angiogenesis. In some zebrafish, edema was observed in
the region surrounding abnormal vessel growth, similar to the phenotype present in
AMD. Overall morphology of zebrafish treated with 0.1 mg/mL CoCl 2 was similar to
untreated zebrafish (data not shown), indicating that the pro-angiogenic effect in the
CVP was specific and there was no obvious toxicity. Although a higher CoCl 2
concentration (1mg/mL) did not increase CNV severity or frequency, it did increase
overall toxicity (data not shown). Therefore, treatment of 1dpf zebrafish with
0.1mg/mL CoCl 2 continuously for 4 days was selected as the optimal condition for
generating a hypoxia-induced zebrafish CNV model.

16.3.2 Removal of Intact Eyes from Whole Zebrafish

For optimal assessment of abnormal angiogenesis in CVP, which is located in the back
of the eyes, this region must be assessable. Initially, we used forceps to extract eyes
fromwhole animals; however, this procedurewas tedious and often resulted in damage.
We then tried treating fixed, whole mount Phy-V-stained zebrafish with collagenase
(100 units/mL, at 37 C for 40min) in an Eppendorf tube (McGrath and Seng, 2009).
This procedure loosened eyes from their sockets without damaging vessels. Eppendorf
tubes containing treated zebrafish were gently shaken a few times and loosened, intact
eyes were then dissociated from whole zebrafish (Fig. 16.3); the CVP, located in the
back of the eyes, was then accessible (Fig. 16.1). Compared to untreated zebrafish eyes
(Fig. 16.1a), ECs were highly activated (overall bright fluorescence and red arrows) in
vascularized CVP (yellow arrows) CoCl 2 -treated zebrafish eyes (Fig. 16.1b and c).
Abnormal activated EC aggregates (green box in Fig. 16.1c) were visible in CVP of
some CoCl 2 -treated zebrafish (Fig. 16.1d). These observations confirm that CoCl 2
induced abnormal angiogenesis in the CVP.

16.3.3 Immunohistochemistry of CoCl 2 -Treated
Zebrafish Eyes

To further characterize the zebrafish CNV model, we performed immunohistochem-
istry. Untreated 5dpf zebrafish eyes exhibited the characteristic five-layer retinal
structure (Fig. 16.2a, left panel, ND5) comprised of densely packed cells in the
ganglion cell layer (GCL), wide inner plexiform layer (IPL, long black bar), and