Ocular angiogenesis and retinopathy are the underlying causes of multiple sight-reducing ocular diseases such as macular degeneration, retinopathy of prematurity and diabetic retinopathy. Novel treatments for these diseases are desperately required as current treatments such as anti-vascular endothelial growth factor (VEGF) drugs, like bevacizumab, are delivered via intravitreal injections and can cause acute side effects.

Natural products chemistry encompasses the extraction of potentially pharmacologically active phytochemicals from plant sources. Homoisoflavonoids are a class of chemically active compounds that are isolated from the Hyancinthaceae, which contains four sub-families, the Ornithogaloideae, the Urgineoideae, the Oziroëoideae and the Hyacinthoideae. Species from the Hyacinthaceae family have been used traditionally to treat numerous aliments due to their anti-oxidant, anti-inflammatory and anti-bacterial effects. One homoisoflavonoid that has been isolated from the orchid Cremastra appendiculata, has been found to have an anti-proliferative effect on human retinal endothelial cells (HRECs).

In this study, six species from the Hyacinthaceae family were investigated in order to isolate homoisoflavonoid compounds to produce a library of compounds for screening against HRECs to determine their anti-angiogenic activity. The six species were Scilla peruviana, Scilla bifolia, Eucomis bicolor, Eucomis pallidiflora, Eucomis comosa and Muscari alaska and these species yielded eighteen homoisoflavonoids and one nortriterpenoid. In addition, seven homoisoflavonoid compounds were synthesised to prepare additional compounds that were not restricted to the functional groups present and position as with natural homoisoflavonoids.

All homoisoflavonoids both isolated and synthetised were screened against HRECs to determine their antiangiogenic activity. The natural compounds that were isolated showed no activity. However, the synthetic compounds yielded promising results, with each showing GI₅₀ values of less than 10 µM. Notably, removing the carbonyl group at C-4 significantly enhanced the activity of compound HJ5h, suggesting it should be incorporated into future synthetic strategies. Further research is needed to identify the optimal substitution patterns on both the A- and B-rings, which can be achieved by synthesising a wide range of compounds with slight variations in their substitutions. The selectivity of compounds HJ3 and HJ4 is particularly encouraging, as they were inactive against all other tested cell lines except Y-79. However, additional testing is required to evaluate the toxicity of these compounds.

In addition, Scilla bifolia was investigated for its potential as being a farmable source of active homoisoflavonoids, although this was proved to not be a viable option. Homoisoflavonoids were also subject to computational modelling in order to see how they might interact with enzymes within the eye. However, it was difficult to draw conclusions from these studies and further work is definitely required.