PEER REVIEWED

Angiogenesis and Angiomaintenance in Pediatric Retinal Diseases

Michael T. Trese, MD • Kimberly A. Drenser, MD, PhD • Antonio Capone, Jr., MD

Many pediatric retinal diseases have been shown to have a blood vessel development, or angiogenic, defect. They include Norrie disease, familial exudative vitreoretinopathy, persistent fetal vasculature syndrome, Coats' disease and retinopathy of prematurity. Other diseases are also known to show an angiomaintenance defect, such as familial exudative vitreoretinopathy and juvenile diabetic retinopathy. What do we know about these diseases?

PATHOGENESIS OF PEDIATRIC RETINAL DISORDERS

Perhaps the best-described pathogenesis is for retinopathy of prematurity and the relationship between avascular retina, ischemia, VEGF drive and the effect on exudation, neovascularization and subsequent blood vessel development.¹

The genetics of these diseases have been looked at for years, and certainly at this stage our picture is still incomplete, but we do know that Norrie disease defects can lead to poor blood vessel and neural tissue development in the eye, ear, and central nervous system. We also know that mutations in the Norrie disease gene can lead to phenotypically different presentations than can appear typical for Norrie disease or familial exudative vitreoretinopathy (FEVR).² FEVR has had several mutations described, but perhaps the most common are the FZD4, LRP 5, and TSPAN12 mutations.

We surveyed our practice, which has a large number of FEVR patients, and found that there seems to be variability in expression, with FZDr mutations and LRP 5 mutations having more severe clinical appearances than TSPAN12 mutations. We think that this difference represents variability in the mutations, suggesting variability in effect on Wnt signaling. But what do we know about Wnt signaling and retinal development?

There are three described Wnt pathways, one of which, the canonical ß-catenin pathway, plays a role in gene regulation and cell proliferation of both vascular and neural tissue development.³ The other two pathways, the planar cell polarity and the calcium Ca2+ plush pathways, participate in cell migration and cytoskeletal structure and cell adhesion, as well as protein kinase C activity (Figures 1 and 2).

Figure 1. The frizzled surface cell marker and the pathway of ß-catenin/Wnt signaling.

Figure 2. The three Wnt pathways with known cellular impact.

GENE SIGNALING IN FEVR

Wnt signaling has been very well described in Norrie disease and to a large extent in FEVR. It has also been shown in an animal model that supplementation with norrin (the disease gene product of Norrie disease) can avoid the phenotypic presentation of Norrie disease–type changes in these animal models by supplementing norrin to drive the Wnt signaling pathway.⁴

FEVR has been a difficult clinical entity to manage because of its unpredictable reactivation. For many years, we have told patients that “FEVR is a lifelong, potentially active retinal vascular disease that can reactivate without warning.” We had no real idea why that might occur until recently, when we started doing widefield fluorescein angiography on children with FEVR. What we found was that, many years after the diagnosis of FEVR, there may be areas of capillary dropout posterior to the juncture of the vascularized and non-vascularized areas of retina, and these areas of capillary dropout precede reactivation and exudation and retinal detachment formation.

Over the last two years, we have treated these areas of capillary dropout and performed regular fluorescein angiography on these patients. It is our clinical impression at this time that laser treatment seems to reduce reactivation, although retinal tissue posterior to the juncture is destroyed by this ablative approach (Figure 3).

Figure 3. A child with FEVR who has developed capillary dropout posterior to the area of peripheral avascular retina. This area was treated with laser and no further exudation seen with 18 months of follow-up.

Why do areas of capillaries drop out in diseases such as FEVR? We think there is a function of Wnt signaling that controls a balance of growth and nurturing factors that contribute to the maintenance of capillaries. We also feel that norrin may be able to be used to maintain capillaries or re-establish vessels.

What evidence do we have to show that? It is very difficult to find animal models that show capillary dropout, but an effect on capillaries and capillary development can be demonstrated in the oxygen-induced retinopathy model commonly used in preclinical drug testing.⁵ In that model we do see the capillary loss, which is usually found reduced or eliminated by additional norrin (Figures 4–6). Similar capillary changes are seen in mouse pups treated with DKK, which blocks norrin Wnt signaling⁶ (Figure 7).

Figure 4. Capillary dropout of the oxygen-induced retinopathy model in the mouse.

Figure 5. A montage of OIR models in the mouse with large areas of capillary dropout.

Figure 6. Capillary sparing in the OIR model in a mouse treated with intravitreal injection of Norrin.

Figure 7. The pattern of capillary dropout seen in the mouse when Norrin/Wnt activity blocked.

We are currently involved in preclinical testing of norrin as a therapeutic agent for capillary dropout, as seen in FEVR. This mode of thinking also helps us with other pediatric retinal diseases, such as juvenile diabetes. It may be that patients who have very bad diabetic retinopathy with extensive capillary dropout and is chemically well controlled may have a mutation in their Wnt signaling, which, if minor, might not alter capillaries for many years. However, a more major mutation may lead to rapid capillary loss and severe diabetic retinopathy.

IMPLICATIONS FOR ROP

This two-hit hypothesis is not unheard of in ophthalmology and is perhaps most familiar when associated with retinoblastoma, but we have also described such an event in aggressive posterior ROP (APROP) associated with Wnt signaling mutations. We described two patients who were both twins, born prematurely. One twin in each pair developed very severe APROP, while the other twin did not. One of these APROP twins and the mother have a mutation in Wnt signaling, and the unaffected twin did not have this mutation or show these severe changes of APROP.³

This outcome suggests that the accident of premature birth and the development of ROP may be worsened by the mutation in Wnt signaling. This is perhaps similar to what could happen in diabetic retinopathy. We are currently performing a trial to determine the incidence of Wnt signaling mutations in patients with controlled hyperglycemia and yet severe diabetic retinopathy.

In summary, we feel that norrin may be able to be used to drive Wnt signaling to maintain or to repair retinal vessels injured or reduced by absent Wnt signaling, either alone or accompanying other issues, such as premature birth or diabetes. RP

REFERENCES

1. Chen J. Stahl A, Hellstrom A, Smith LE. Current update on ROP screening and treatment. Curr Opin Pediatrics. 2010 Dec 8 (E pub ahead of print).
2. MacDonald ML, Goldberg YP, Macfarlane J, Samuels ME, Trese MT, Shastry BS. Genetic variants of frizzled-4 gene in familial exudative vitreoretinopathy and advanced retinopathy of prematurity. Clin Genet. 2005;67:363-366.
3. Drenser KA, Dailey W, Vinekar A, Dalal K, Capone A, Trese MT. Clinical presentation and genetic correlation of patients with mutations affecting the FZD4 Gene. Arch Ophthalmol. 2009;127:1649-1654.
4. Ohlmann A, Scholz M, Goldwich A, et al. Ectopic Norrin induces growth of ocular capillaries and restores normal retinal angiogenesis in Norrie disease mutant mice. J Neurosci. 2005;25:1701-1710.
5. Penn JS, Tolman BL, Lowery LA. Variable oxygen exposure causes preretinal neovascularization in the newborn rat. Invest Ophthalmol Vis Sci. 1993;34:576-585.
6. Fedi P, Bafico A, Nieto Soria A, et al. Isolation and biochemical characterization of the human Dkk-1 homologue, a novel inhibitor of mammalian Wnt signaling. J Biol Chem. 1999;274:19465-19472.