We read with interest the recent article by Zare-Abdollahi et al where it was suggested that two recessive missense variants in MSFD8 (c.1235C>T; p.P412L and c.1361T>C; p. M454T), which each… Click to show full abstract
We read with interest the recent article by Zare-Abdollahi et al where it was suggested that two recessive missense variants in MSFD8 (c.1235C>T; p.P412L and c.1361T>C; p. M454T), which each have been associated previously with CLN7 when homozygous, resulted in nonsyndromic macular dystrophy in homozygous (c.1361T>C; p.M454T) and compound heterozygous modes (1). However some aspects of that study require further discussion. First, in order to demonstrate that the retinal dystrophy is macular as opposed to widespread, a normal electroretinogram should be demonstrated. Second, the reason for a predilection for a specific involvement of photoreceptors in CLN7 disease, may be that the protein product is localized in the outer plexiform layer as demonstrated in a recent study, and may form part of synaptic vesicles in the photoreceptor terminals, possibly leading to photoreceptor synaptic dysfunction in CLN7 disease, whereas a different mechanism such as lysosomal dysfunction, may be responsible for neurological disease (2). Furthermore, this could potentially also explain some of the observed variability regarding macular vs. widespread retinal disease, because of a higher content of synaptic ribbons for vesicle docking, in peripheral cones compared to central cones (2). CLN7 is a variant form of late-infantile neuronal ceroid lipofuscinosis (vLINCL) (3), a lysosomal storage disease, caused by mutations in CLN7/MFSD8 (major facilitator superfamily domain containing 8, https://ghr.nlm.nih.gov/ gene/MFSD8) gene which encodes a transmembrane lysosomal protein (4). Detailed retinal manifestations including multimodal retinal imaging in vLINCL have not been documented previously. We present here the retinal manifestations in a child with CLN7 disease, some of which, such as pronounced central retinal degeneration demonstrated by optical coherence tomography, were very similar to the findings by Zare-Abdollahi et al (1). A 9 years old girl whose parents are consanguineous presented 3 years ago with progressive visual loss, and night blindness. Her visual acuity VA was 20/300 in both eyes. Fundus examination disclosed discrete salt-and-pepper depigmented retinal pigment changes throughout the retinal periphery in both eyes. Fundus autofluorescence (FAF) showed an unremarkable distribution of autofluorescence except for a mild hyperautofluorescence in the paramacular area in both eyes. Optical Coherence Tomography (OCT) showed a severe attenuation of the outer retinal layers (Figure 1). Importantly, full field electroretinogram (ffERG) showed flat scotopic and photopic responses in both eyes. During the observation period, the girl developed multiple episodes of seizures for which she had a Magnetic Resonance Imaging (MRI) and was unremarkable. She was prescribed topiramate 100 mg daily and Levetiracetam 800 mg BID by her neurologist. Because of the rapid progression and neurological symptoms in childhood, a metabolic or neurodegenerative disease was suspected, especially an infantile or juvenile onset neuronal ceroid lipofuscinosis. Next Generation Sequencing (NGS) for metabolic and neurodegenerative diseases (https://doi.org/10.1186/s13059015-0693-2) revealed a homozygous missense NM_152778: exon12:c.1235C>T:p.P412L mutation in MFSD8 gene, which has been described previously as causing vLINCL, however no clinical data such as multimodal retinal imaging or assessment of retinal function was presented (5). Visual loss was rapidly progressive and 3 years after the presentation, visual acuity had dropped to hand motion in each eye. Furthermore, the p. P412L reported here is a missense mutation and would be considered to yield a milder phenotype than nonsense or frameshift mutations, however it seems to similarly abolish the function of MFSD8, leading to a severe neuroretinal degeneration in a homozygous mode. On the other hand, certain compound heterozygous variants in MFSD8 have been associated with nonsyndromic macular dystrophy (1,6). It was suggested that truncating mutations in one MFSD8 allele combined with a mild variant (p.Glu336Gln) on the other allele resulted in macular dystrophy, while combined severe homozygous variants result in CLN7 phenotype. Thus at present the mechanism underlying the phenotype-genotype correlation in MFSD8-related disease is incompletely understood. This is the first description of a widespread retinopathy in CLN7 documented by multimodal retinal imaging and electroretinography. The increased paramacular autofluorescence is compatible with an increase in autofluorescent fluorophores in various retinal layers, as described in early stages in
               
Click one of the above tabs to view related content.