Sanfilippo syndrome is an untreatable form of childhood-onset dementia. Whilst several therapeutic strategies are being evaluated in human clinical trials including i.v. delivery of AAV9-based gene therapy, an urgent unmet need is the availability of non-invasive, quantitative measures of neurodegeneration. We hypothesise that as part of the central nervous system, the retina may provide a window through which to ‘visualise’ degenerative lesions in brain and amelioration of them following treatment. This is reliant on the age of onset and the rate of disease progression being equivalent in retina and brain. For the first time we have assessed in parallel, the nature, age of onset and rate of retinal and brain degeneration in a mouse model of Sanfilippo syndrome. Significant accumulation of heparan sulphate and expansion of the endo/lysosomal system was observed in both retina and brain pre-symptomatically (by 3 weeks of age). Robust and early activation of micro- and macroglia was also observed in both tissues. There was substantial thinning of retina and loss of rod and cone photoreceptors by ~ 12 weeks of age, a time at which cognitive symptoms are noted. Intravenous delivery of a clinically relevant AAV9-human sulphamidase vector to neonatal mice prevented disease lesion appearance in retina and most areas of brain when assessed 6 weeks later. Collectively, the findings highlight the previously unrecognised early and significant involvement of retina in the Sanfilippo disease process, lesions that are preventable by neonatal treatment with AAV9-sulphamidase. Critically, our data demonstrate for the first time that the advancement of retinal disease parallels that occurring in brain in Sanfilippo syndrome, thus retina may provide an easily accessible neural tissue via which brain disease development and its amelioration with treatment can be monitored.

Lysosomal dysfunction may be an important factor in the pathogenesis of neurodegenerative disorders such as Parkinson's disease (PD). Heterozygous mutations in the gene encoding the lysosomal enzyme glucocerebrosidase (GBA1) have been found in PD patients, and some but not all mutations in other lysosomal enzyme genes, for example, NPC1 and MCOLN1 have been associated with PD. We have examined the behaviour and brain structure of mice carrying a D31N mutation in the sulphamidase (Sgsh) gene which encodes a lysosomal sulphatase. Female heterozygotes and wildtype mice aged 12‐, 15‐, 18‐ and 21‐months of age underwent motor phenotyping and the brain was comprehensively evaluated for disease‐associated lesions. Heterozygous mice exhibited impaired performance in the negative geotaxis test when compared with wildtype mice. Whilst the brain of Sgsh heterozygotes aged up to 21‐months did not exhibit any of the gross features of PD, Alzheimer's disease or the neurodegenerative lysosomal storage disorders, for example, loss of striatal dopamine, reduced GBA activity, α‐synuclein‐positive inclusions, perturbation of lipid synthesis, or cerebellar Purkinje cell drop‐out, we noted discrete structural aberrations in the dendritic tree of cortical pyramidal neurons in 21‐month old animals. The overt disease lesions and resultant phenotypic changes previously described in individuals with heterozygous mutations in lysosomal enzyme genes such as glucocerebrosidase may be enzyme dependent. By better understanding why deficiency in, or mutant forms of some but not all lysosomal proteins leads to heightened risk or earlier onset of classical neurodegenerative disorders, novel disease‐causing mechanisms may be identified.

No abstract