**7. Cornea models**

The cornea is an avascular tissue consisting of two, thin, transparent layers in rodents. Thus it is possible to gently cut a tiny pocket between the two layers, and in this pocket insert a pellet containing factors which are to be investigated for their angiogenic or anti-angiogenic activities *in vivo*. All vessels detected in the cornea following a few days stimulation with the implanted factors can be considered newly formed vessels, and due to the transparent nature of the cornea and the strong red color of perfused blood vessels, the angiogenic response can be followed kinetically by simply taking photographs of the eye at different time points. This makes it possible to study the effects of angiogenic factors, either alone or in combination on different processes of angiogenesis such as initial angiogenic expansion, vascular remodeling, maturation and stability in the same animal over time.

Usually angiogenic or anti-angiogenic factors are prepared in slow release pellets consisting of hydron or alumni sucrose octo-sulfate, which are dried on a nylon mesh to ensure equal size and thereby amount of factor in each pellet which is then implanted in the corneal micropocket. Using this delivery system, factors are released constantly at low amounts for weeks, thus giving rise to a continuous angiogenic stimulation or inhibition depending on the factor in question (R Cao et al, 2003). Alternatively, small pieces of tumor tissue can be implanted to study the angiogenic capacities of different tumors from different organs. Even primary human tumor samples can be investigated in this model by using immune-deficient nude or SCID mice (Jensen et al, 2009).

While blood vessels can be macroscopically studied by simply taking photographs of the cornea, more in-depth examinations on blood vessel structure and function including pericyte coverage, tip cell formation and vascular permeability require that the cornea is excised post mortem and subjected to immunohistochemical analysis such as those described for the fat tissue (Fig. 13).

Fig. 13. Visualization of lymphangiogenesis and angiogenesis. A: Confocal analysis of mouse FGF2-induced corneal lymphatic vessels (red) stained with an anti-LYVE-1 antibody and blood vessels (green) stained with an anti-CD31 antibody. B: T241-VEGF-C tumor lymphatic vessels (red) stained with an anti-LYVE-1 antibody and blood vessels (green) stained with an anti-CD31 antibody.

Following staining with antibodies against markers specific for lymphatic endothelial cells, this assay is furthermore one of the strongest assays for investigating lymphangiogenesis. Furthermore, by treating the mice with drugs or neutralizing antibodies against particular receptors, this model can be used to parse out the specific contributions of one or a few receptors for a particular angiogenic factor in inducing angiogenesis and lymphangiogenesis.

weeks, thus giving rise to a continuous angiogenic stimulation or inhibition depending on the factor in question (R Cao et al, 2003). Alternatively, small pieces of tumor tissue can be implanted to study the angiogenic capacities of different tumors from different organs. Even primary human tumor samples can be investigated in this model by using immune-deficient

While blood vessels can be macroscopically studied by simply taking photographs of the cornea, more in-depth examinations on blood vessel structure and function including pericyte coverage, tip cell formation and vascular permeability require that the cornea is excised post mortem and subjected to immunohistochemical analysis such as those

Fig. 13. Visualization of lymphangiogenesis and angiogenesis. A: Confocal analysis of mouse FGF2-induced corneal lymphatic vessels (red) stained with an anti-LYVE-1 antibody and blood vessels (green) stained with an anti-CD31 antibody. B: T241-VEGF-C tumor lymphatic vessels (red) stained with an anti-LYVE-1 antibody and blood vessels (green)

Following staining with antibodies against markers specific for lymphatic endothelial cells, this assay is furthermore one of the strongest assays for investigating lymphangiogenesis. Furthermore, by treating the mice with drugs or neutralizing antibodies against particular receptors, this model can be used to parse out the specific contributions of one or a few receptors for a particular angiogenic factor in inducing angiogenesis and lymphangiogenesis.

nude or SCID mice (Jensen et al, 2009).

described for the fat tissue (Fig. 13).

stained with an anti-CD31 antibody.

This model is considered to be "clean" compared to the tumor models described above – i.e. the angiogenic or lymphangiogenic response can be induced by a defined factor or couple of factors. Considering all these benefits – this assay is probably the strongest *in vivo* assay to study the molecular biology of angiogenesis and lymphangiogenesis (Jensen et al, 2009).

However, the assay has been criticized for being poor at describing physiologic or pathological angiogenesis as the amount of factor used in the pellet usually give rise to supra-physiological concentrations in the cornea. However, the amount of factor used per pellet can be lowered accordingly, or instead of using purified recombinant growth factors, researchers may implant tumor tissue, or a small piece of suture, which lead to corneal inflammation, which then subsequently drives the angiogenic response although via a less defined and controllable pathway. The major limitation of the assay is the technical difficulty of implanting pellets into the mouse cornea. In the beginning, researchers circumvented this problem by using larger animals such as rats or rabbits, but in order to take advantage of the growing number of transgenic or knock-out mice available today it is a major benefit to use this animal if possible.
