Top of pageAbstract Late infantile neuronal ceroid lipofuscinosis (LINCL) is a fatal pediatric neurodegenerative disease resulting from mutation of the CLN2 gene encoding for a tripeptidyl peptidase (TPP-I). Deficiency of TPP-I results in aberrant degradation of membrane proteins in the lysosome with development of inclusion bodies, leading to diffuse neuronal cell death. We have compared CNS gene transfer of CLN2 with serotype 2, 5, 8 and rh.10 AAV vectors and observed that AAVrh.10 provided the highest levels of TPP-I protein in the CNS of rodents. In order to assess the functional consequences of AAVrh.10CLN2 CNS administration, a total of 1.6|[times]|1011 genome copies of AAVrh.10hCLN2, was administered into 4 locations per hemisphere of CLN2 |[minus]|/|[minus]| mice ages 7 to 8 wk and the mice were assessed weekly for performance in tests that combine motor and sensory skills. The treated CLN2 |[minus]|/|[minus]| mice were compared to untreated CLN2 |[minus]|/|[minus]| mice (negative controls) and wild type (CLN2 +/+) mice (positive controls). Analysis of TPP-I levels in the brain demonstrated the AAVrh.10CLN2-treated mice had levels higher than the wild type control (CLN2 +/+) in all regions of the brain. Analysis of neuronal lipofucsin-associated autofluorescence, a hallmark of the disease, demonstrated between 40 to 45 % decrease for the treated CLN2|[minus]|/|[minus]| mice in the injected areas in comparison to the untreated CLN2 |[minus]|/|[minus]| mice. In a test for gait analysis, for wild type mice, the footprints for the hindpaws are largely superimposed on those for the front paws and there was no dragging. In untreated CLN2 |[minus]|/|[minus]| mice, by the age of 21 wk, there was poor gait with dragging of feet and no clear coordination of location between the forepaw and the ipsilateral hindpaw. However, in the 21 wk CLN2 |[minus]|/|[minus]| mice treated with AAVrh.10hCLN2, there was better coordination of the paws and much less dragging. In a grip strength test, the treated mice had a better performance than untreated CLN2 |[minus]|/|[minus]| mice (p 0.05). In a balance beam test, the treated mice had a better performance than untreated CLN2 |[minus]|/|[minus]| mice (p<0.005, ANOVA with group as factor and time as covariate), although the performance of AAVrh.10hCLN2 treated CLN2 |[minus]|/|[minus]| mice was not as good as wild type mice (p<0.005). Kaplan-Meier analysis of survival curves also showed that there was a statistically significant survival advantage of AAVrh.10hCLN2 treated CLN2 |[minus]|/|[minus]| mice over the untreated CLN2 |[minus]|/|[minus]| mice. The median age of survival of the injected CLN2 |[minus]|/|[minus]| mice was 147 days compared to 118 days for the uninjected CLN2 |[minus]|/|[minus]| mice (p<0.01). Together, this data suggests the conclusion that AAVrh.10CLN2 is a suitable candidate for preclinical development for the treatment of LINCL.