Statistics from Altmetric.com
The past decade has seen a resurgence of interest in the molecular defects underlying macular dystrophies. Firstly, this is because these diseases are important causes of incurable blindness and, secondly, the molecular defects highlighted by these studies may be relevant to the much commoner disease, age related macular degeneration (AMD). In this issue, Assink et al (p 682) have undertaken a molecular genetic study of one of these macular dystrophies: Sorsby's fundus dystrophy (SFD), an important addition to the literature, which also raises new questions.
The scientific literature describes SFD as a fully penetrant, autosomal dominant, retinal disease first described by Sorsby in 1949.1 Clinically, early, mid-peripheral, drusen2 and colour vision deficits are found.3 Some patients complain of night blindness.4 Most commonly, the presenting symptom is sudden acuity loss, manifest in the third to fourth decades of life, due to untreatable submacular neovascularisation.5Histologically, there is accumulation of a confluent lipid containing material 30 μm thick at the level of Bruch's membrane.
Parametric linkage analysis originally localised the SFD gene to chromosome 22q13-qter6 between marker lociD22S273 andD22S281. Subsequently, five different missense mutations7 and a splice site mutation7 have been identified inTIMP3 (tissue inhibitor of metalloproteinases 3). In the British Isles, to date, all SFD families carry the same Ser181Cys TIMP3 mutation and it has been suggested that this all relates to one single founder.8 TIMP3 encodes a retinal pigment epithelium expressed member of a group of zinc binding endopeptidases involved in retinal extracellular matrix remodelling, particularly in Bruch's membrane.9
Assink et al describe a large family with autosomal dominant maculopathy, the earliest features being drusen and “pisciform” lesions in the fourth decade of life. These are later complicated by disciform lesions that progress to chorioretinal atrophy. A key question is, does this truly represent the SFD phenotype? Pisciform lesions are not a feature normally associated with Sorsby's fundus dystrophy. Also, delayed choroidal filling on fluorescein angiography is thought to be a cardinal SFD finding but is not seen in this study family.2 Other maculopathies are known to lead to drusen and subretinal neovascularisation in this age group. For instance, the study family may represent an example of dominant drusen? Or, alternatively, does the study describe a new phenotype? Is it important that a family diagnosed as expressing SFD has exactly the symptoms and signs itemised in the literature or is it sufficient that they express the cardinal features as suggested in this study? Should SFD be confined to haemorrhagic maculopathies with proved TIMP3 mutation?
Have Assink et al truly excluded an association between TIMP3 and the phenotype expressed in their study? To completely excludeTIMP3 not only must coding sequence be screened but also the promoter and enhancer sequences that controlTIMP3 expression. The authors have almost but not entirely done this. Most pertinent here are the enhancer sequences which can be found quite distant from the relevant gene. These sequences could conceivably lie within theD2S275 and D2S274linked region and be composed of an as yet unknown sequence; mutation therefore becoming impossible to exclude.
Sorsby's fundus dystrophy, both clinically and histopathologically, shares some similarities with AMD. It is appropriate, therefore, thatTIMP3 coding sequence be screened in cohorts expressing this much more frequent cause of blindness in the developed world. No TIMP3 mutation, however, has yet been associated with AMD. Genes associated with other macular dystrophies have also undergone mutation screen in AMD (Table 1). No positive association has yet been identified except initially in the case of ABCA4.10 Recent work, however, has cast doubt on a link betweenABCA4 mutation and AMD. The initial, positive, association has not been reproduced in other studies andABCA4 has proved to be such a polymorphic gene that large numbers of base changes in unaffected control groups have cast doubt on the significance of ABCA4base changes in AMD patients. Despite this, screening genes causing rare macular dystrophies for mutation in AMD is still a valid undertaking. Mutation of the glucagon gene in rare families expressing early onset, autosomal dominant, diabetes mellitus leads to a significant association between this gene and diabetes mellitus as more commonly seen as a polygenic complex trait.11
Regardless of whether the study family truly has SFD or whether the work will prove helpful in identifying AMD genes, the work presented by Assink et al is an important contribution to the ocular genetics literature. Retinal dystrophies are proving to be among the most genetically heterogeneous conditions known, with dozens of genes now associated with inherited disease of the retina and with more to come. This is beginning to stimulate interest in unravelling of the molecular steps involved in these diseases. It is possible that as we more fully understand the complexity of the molecular consequences of mutation, new ideas will emerge of how to modify progression of disease in a way that may prove more practical than “conventional” gene therapy protocols that treat the primary gene deficit.