At the start of this 21st century, there would appear to be a strong increase in the number of ophthalmological pathologies of genetic origin in industrialised countries. But this increase is only relative. It is the result of two contrary events. The first is the spectacular reduction in acquired, particularly infectious, pathologies. The second is better knowledge of genetically determined disorders, made possible in large part by the creation, in the eighties, of the gene map. This marked the start of an explosion in genome investigation methods, where continued progress has resulting in the identification of a considerable number of genes whose mutations cause severe sight disorders in children.
The development, maintenance and functioning of ocular structures require the contribution of thousands of genes whose expression varies over time and is influenced, in part, by the environment. Mutations of these genes can be silent. But they are most often responsible for a change in visual performance that can go as far as total blindness (Table 1).
Tab. 1: Principal hereditary pathologies of vision, affecting young children 1.
1. Development abnormalities
The most severe ophthalmological disorders are frequently caused by mutations affecting the genes involved in the development of ocular structures.
Sometimes unilateral, often bilateral, eye development abnormalities can occur at different stages in embryonic life, leading to a spectrum of malformations that can affect all structures of the eye, one by one, or together.
Anophthalmia, the most serious of these disorders, is characterised by total absence of any kind of ocular structure, resulting from an absence of induction of the optical vesicle, three weeks into embryonic life. Sometimes in isolation (primary anophthalmia), it is often corre-lated with more extensive malformation of the anterior brain (secondary anophthalmia).
Microphthalmia, is often characterised by a wide spectrum of severity, from the most extreme forms, often confused with anophthalmia, through to a small size eye, sometimes formed and operational, depending on whether the abnormality occurs during invagination of the optical vesicle, closure of the embryonic fissure or, later, due to halting of the development of the ocular structures that have been formed normally.
These early and severe malformations, the incidence of which is around 1 to 2 per 100 000 births, can be screened by ultra-sound at around 18-20 weeks of amenorrhoea. Sudden discovery of this type of malformation in a foetus born of healthy parents can be a sign of the occurrence of a dominant novo mutation, often in a gene that programmes the eye, or of a rare autosomal recessive abnormality, particularly in blood-related couples. If abortion is chosen due to a bilateral blindness pathology, a foetal-pathological study should be requested, to search for associated malformations. In all cases, a genetic study will be offered to identify the mutation or mutations responsible, in order to offer an early molecular diagnosis for the next pregnancy, in case of genetic risk. Knowledge of the numerous genes responsible for these diseases is still only shaky, and genetic abnormalities cannot be discovered in every case.
Alongside these extreme forms, affecting the size and structure of the eye, there exists a variety of malformations of the anterior chamber, often the result of an abnormality in the migration of cells from the neural crest. These cases of irido-corneal dysgenesis, the prevalence of which is around 1 to 2 per 100 000 come in a wide anatomic spectrum, all with the same risk of occurrence of a sudden, massive glaucoma making them very severe and necessitating regular monitoring of ocular pressure in all babies with these malformations. An examination should be performed several times a year under general anaesthetic in very young children under three. These cases of dysgenesis of the anterior segment are almost all inherited with autosomal dominance. A large number of genes have been identified, which unfortunately does not cover all cases. Note that gene PAX6 is the major gene in ocular development, the mutations of which are responsible for a large number of malformations.
Primitive congenital glaucoma, the incidence of which is estimated at 5 for every 100 000 births, is the result of a cessation in the development of the iridocorneal angle at a stage corresponding to 6 months of intra-uterine life. It is them obstructed by thick trabecular meshwork that can lead, in utero, to ocular hypertension and buphthalmos that is visible during antenatal ultra-sound screening. Nevertheless, a large number of these children are born with a normal sized eye, even though at birth a megalocornea can already be observed. These children must receive an emergency operation (new-born, less than one week old) in order to release the iridocorneal angle. Primitive congenital glaucoma is a recessive autosomal malformation. The mutations of several genes may be responsible for it. Note, however, that the gene CYP1B1 alone represents 2/3 of cases.
Corneal dystrophies are extremely varied in terms of their anatomical presentation. Classification has been vastly improved by identification of a major gene, BIGH3, for which there exist genotypephenotype correlations. All methods of transmission can be shown, but these afflictions, which caused blindness in the past, have seen their prognosis considerably improved by the development of corneal grafts. Corneal dystrophies are rare afflictions, the frequency of which has not been established with any degree of precision.
Congenital cataracts also vary greatly in terms of their clinical presentation. After excluding acquired ante- and neo-natal causes, all methods of transmission can be envisaged, with a majority of dominant autosomal forms which should not be neglected in the face of a sporadic case. The incidence of congenital cataracts is around 1 to 6 for 10 000 births. Numerous genes have been identified, the mutations of which are responsible for these opacities of the crystalline lens. Nevertheless, success in terms of progress made with early surgical procedures (newborns less than a month old), involving a crystalline lens implant, has considerably reduced requests for genetic studies. Note, that most congenital cataracts are observed in utero enabling access to paediatric surgery in good conditions.
In terms of the posterior segment, two major categories of vision disorders can be distinguished: those resulting from a lack of normal retinal development and the large group of retinal degenerations.
With regard to the former, we will address only persistent primary vitreous, Stickler syndrome, juvenile retinoschisis linked to chromosome X and Norrie disease.
Persistent primary vitreous is in most cases unilateral and is not considered the result of any kind of transmissible genetic abnormality. Nevertheless, heredity is suspected but not elucidated in bilateral forms. The incidence of these affections is estimated at 1 in 30 000 births.
Stickler syndrome is an affection whose frequency is around 1 to 5 per 10 000. It combines alterations in the vitreous evolving towards retinal detachment in childhood, accompanied by facial dysmorphia, bone dysplasia and, in some cases, hearing problems. It is always a dominant autosomal affection of variable expressivity. Three collagen genes have now been identified, covering almost all cases. In children with such illnesses, monitoring of the retinal periphery is necessary with barrage laser treatment if required. The severity of this disease and the possibility of prior identification of the mutation that causes it mean that prenatal diagnosis is possible, if families ask for it.
Juvenile retinoschisis linked to X refers to the association of a maculopathy and a secondary vitreoretinopathy with abnormal splitting of the retina in its innermost layer. It comes in the form of a bilateral cystic macular lesion and, on the periphery, a lifting (schisis) accompanied by blurring and vitreous condensation again representing the risk of retinal detachment. Diagnosis is not easy, but once done, the discovery of mutation by the only gene responsible, RS1, confirms the diagnosis and enables the screening of female carriers in the family. The prevalence of this affection varies according to populations between 1 in 5000 and 1 in 25 000.
Norrie disease, also linked to chromosome X, is rare (< 1 in 100 000). It designates major dysplasia of the retina and presents in a newborn as a leukocoria, due to the appearance of the retina as a highly vascularized white-yellowish mass which, in the past, could have been mistaken for a retinal tumour. This is a fearsome disease since consistently blinding retinal dysplasia is accompanied in 2/3 of cases by mental disability and in 1/3 of cases by developing deafness. Underlying this disease in all cases are mutations of the gene NDP. In cases where these mutations are recognised they allow for screening of carriers in the family and prenatal diagnosis in women at risk.
2. Early onset and severe degenerative retinal disorders
These refer to a gradual death of the photoreceptor – pigmentary epithelium couple and are extremely heterogenic from a clinical and genetic point of view. They can begin at any time of life and many forms begin in the first two decades of life. Taking all forms into account, they affect one person in 3000.
These are inexorably evolutive diseases, which have a considerable impact on the visual function of the children involved.
Five major clinical groups of retinal degeneration can be distinguished, identifiable by the natural history of the diseases, their age and how they started and their development.
The most frequent are rod-cone dystrophies (66% of retinal dystrophies), recognisable by primary affliction of the rods resulting in problems with night vision and a concentric reduction in the peripheral field of vision. This impact on the rods will continue, 10 to 20 years later, with an impact on the function of the cones, recognisable by intolerance to light, dyschromatopsy and a reduction in visual acuity. These diseases are extraordinarily heterogeneous from a genetic point of view. Several dozen genes have already been identified, which cover 50% of cases.
Conversely, there are forms of cone-rod dystrophies (10-15% of retinal dystrophies), in which the first symptoms are primary disorders of the cones, which will be followed by a secondary affliction of the rods. The same genetic heterogeneity has been reported.
Isolated cone dystrophies are much rarer (2-3 % of retinal dystrophies), because when they are diagnosed they may correspond to the initial stage of cone-rod dystrophies.
Leber Congenital Amaurosis (LCA) was for many years described as a separate entity. It is recognisable in the first weeks or months of life in newborns with apparent blindness. It accounts for around 5-10% of all retinal dystrophies. This disease is autosomal recessive in almost all cases, based on the mutations of 17 different genes that cover only 70% of cases, certain of which have brought this type of blindness into the big group of ciliopathies (Fig. 1, 2). Genetic work carried out over the past 15 years has identified two major clinical types: a conerod type (type 1), which is dramatically severe and stationary and a rod-cone type (type 2) which is indeed severe but which evolves, representing the extremity of a spectrum of seriousness of retinal dystrophies. Of these 17 genes, at least four account for type 1 LCA, which is still the most frequently observed, whereas around 12 account for type 2. The mutations of the latter can also account for later onset rod-cone type retinal dystrophies (Fig. 3).
Fig. 1: Genetic heterogeneity of Leber Congenital Amaurosis(LCA) and relative frequencies of genetic subtypes. Genes currently identified account for 70% of LCA cases. All mutations are recessive with the exception of some rare changes in CRX and IMPDH1.
Fig. 2: Physiopathological heterogeneity of Leber Congenital Amaurosis(LCA). The disease is accounted for by alteration of variable retinal functions of the retinal pigmentary epithelium or photoreceptor cells. Most LCA genes are specifically or preferencially expressed in the retina but some of them are ubiquituously expressed. Considering the involvement of intraflagellar transport proteins, some of which are ubiquituous, some LCA cases are regarded as ciliopathies.
Fig. 3: Clinical variability of Leber Congenital Amaurosis (LCA) and genotype-phenotype correlations. VA: visual acuity; LP: light perception; CF: counting fingers; RP: retinitis pigmentosa.
Macular dystrophies are characterised by degeneration of the macular pigmentary epithelium, leading to the gradual death of central photoreceptors, the first of which are the cones. Within this group, Stargardt's disease is by far the most frequent of all juvenile macular degenerations since it accounts alone for almost 10% of retinal dystrophies. Paradoxically, this disease is homogenous from a genetic point of view, despite its great clinical variability. Contrary to all other macular degenerations, which are autosomal dominant, Stargardt's disease is consistently recessive. It is caused by mutations of the ABCA4 gene, the severity of which is closely correlated to that of the ensuing retinal disease.
The rapid discoveries that have been made in less than twenty years with regard to all these diseases have swept away the former dogma of “one gene, one disease” because today numerous diseases, even those that are extremely homogenous on a clinical level, can be caused by the mutations of very different genes (tissular expression and function) and, conversely, the mutations of one gene can be responsible for considerably different pathologies. One may have thought that the identification of the same gene in different pathologies would lead to the establishment of correlations between mutations and phenotype. Yet, with the exception of a few examples, particularly Stargardt's disease and, to a lesser extent, Leber congenital amaurosis, this is not the case. This ignorance of the causes of phenotype variability which is sometimes found even amongst siblings, calls genetic and environmental modifying factors into question. Definition of the complete genotype of individuals, technically possible today but too expensive to be realistic on a routine basis, should enable us in the years to come to find answers to the origins of this variability which is, without a doubt, one of the major future challenges that genetics will have to face.
1. The main information concerning the clinical effects of these diseases and the genetic and molecular data concerning the genes in question can be found on the following websites: Online Mendelian inheritance in Man http://www.ncbi.nlm.nih.gov/omim and Orphanet www.orpha.net/consor4.01/www/cgi-bin/?lng=FR