Multifocal retinopathy in Dachshunds with CLN2 neuronal ceroid lipofuscinosis

Abstract

The CLN2 form of neuronal ceroid lipofuscinosis is an autosomal recessively inherited lysosomal storage disease that is characterized by progressive vision loss culminating in blindness, cognitive and motor decline, neurodegeneration, and premature death. CLN2 disease results from mutations in the gene that encodes the soluble lysosomal enzyme tripeptidyl peptidase-1. A null mutation in the TPP1 gene encoding this enzyme causes a CLN2-like disease in Dachshunds. Dachshunds that are homozygous for this mutation serve as a model for human CLN2 disease, exhibiting clinical signs and neuropathology similar to those of children with this disorder. Affected dogs reach end-stage terminal disease status at 10-11 months of age. In addition to retinal changes typical of CLN2 disease, a retinopathy consisting of multifocal, bullous retinal detachment lesions was identified in 65% of (TPP1-/-) dogs in an established research colony. These lesions did not occur in littermates that were heterozygous or homozygous for the normal TPP1 allele. Retinal changes and the functional effects of this multifocal retinopathy were examined objectively over time using ophthalmic examinations, fundus photography, electroretinography (ERG), quantitative pupillary light response (PLR) recording, fluorescein angiography, optical coherence tomography (OCT) and histopathology. The retinopathy consisted of progressive multifocal serous retinal detachments. The severity of the disease-related retinal thinning was no more serious in most detached areas than in adjacent areas of the retina that remained in close apposition to the retinal pigment epithelium. The retinopathy observed in these dogs was somewhat similar to canine multifocal retinopathy (CMR), a disease caused by a mutation of the bestrophin gene BEST1. ERG a-wave amplitudes were relatively preserved in the Dachshunds with CLN2 disease, whether or not they developed the multifocal retinopathy. The retinopathy also had minimal effects on the PLR. Histological evaluation indicated that the CLN2 disease-related retinal degeneration was not exacerbated in areas where the retina was detached except where the detached areas were very large. DNA sequence analysis ruled out a mutation in the BEST1 exons or splice junctions as a cause for the retinopathy. Perfect concordance between the TPP1 mutation and the retinopathy in the large number of dogs examined indicates that the retinopathy most likely occurs as a direct result of the TPP1 mutation. Therefore, inhibition of the development and progression of these lesions can be used as an indicator of the efficacy of therapeutic interventions currently under investigation for the treatment of CLN2 disease in the Dachshund model. In addition, these findings suggest that TPP1 mutations may underlie multifocal retinopathies of unknown cause in animals and humans.

Keywords: BEST1; Bullous retinopathy; CLN2; Canine; Dog; Neuronal ceroid lipofuscinosis; Retinal detachment; TPP1.

Figure 1

Representative combined scanning laser ophthalmoscopic (cSLO) fundus images of retinas exhibiting grades 1, 2 and 3 mutlifocal lesions. All images are from the left eye.

Figure 2

(A) Scanning laser ophthalmoscopic image of the fundus from a dog exhibiting multiple localized serous detachments. A scan line for the OCT image shown in (B) is shown in white. The edges of 3 of these lesions traversed by the scan line are indicated by colored arrows. (B) An OCT cross-section image of the retina from along the scan line shown in (A) with the edges of the same 3 lesions indicated by the same colored arrows. (C) A 3-dimensional reconstruction of the same area of the retina made by combining multiple OCT images showing the same 3 serous detachments.

Figure 3

(A) An SLO image of the fundus of a 12.5 month old dog at the tapetal/nontapetal boundary. Yellow line shows the OCT scan line used to generate the image in (B). (B) sdOCT scan of resorptive/flattened lesions demonstrating evidence of focal thinning and degeneration of the overlying retina (arrows). These areas of thinning are not apparent in the SLO image.

Figure 4

Retinal thickness in NCL-affected dogs. Lesions were defined as small [<0.5 disc diameter (DD)], large (>1 DD) and resorptive/flattened for retinal thickness measurements. Relative to areas where the retina remained attached to the RPE, retinal thickness was not affected except in large Grade 3 lesions. Retinal thickness in large, Grade 3 lesions was significantly thinner than retina without lesions (*p=0.012). Resorptive/flattened lesion retinal thickness was significantly reduced from all other categories (†, p<0.005).

Figure 5

(A) cSLO fundus image of a dog with Grade 2 retinopathy lesions. (B) cSLO fluorescein angiogram of the same dog detailing transition between choroidal and arterial phases with no obvious leakage of fluorescein dye around lesions. (C) cSLO fluorescein angiogram of the same dog detailing late venous phase with no obvious leakage of fluorescein dye around lesions.

Figure 6

ERG a-wave amplitude by age and retinal lesion status. Measures are from a dark-adapted dog, in response to a stimulus of 13.2 log photons/cm²/s (10cd.s/m²). No statistically significant differences exist amongst the groups. However, due to the limited number of dogs available the statistical power of the comparison was too low to rule out the possibility that the severity of the lesions did affect the ERG a-wave.

Figure 7

For 8 month old dogs with grade 3 lesions, PLR constriction amplitude is significantly reduced from that of other 8 month old affected dogs in response to a stimulus of 10 log photons/cm²/s (*p=0.011).

Figure 8

Light micrographs of an area of a retina near a grade 2 lesion. (A) The boundary between an area where the retina remains in close apposition to the RPE and the edge of the detached region (d). (B) An area from the same retina distant from the region of detachment. (C) An area of the retina within the region of detachment. Bar in (C) indicates the magnification in both panels (B) and (C).

Figure 9

Fluorescence micrographs of an unstained cryostat section of the retina from a region where the retina remained attached to the RPE (A) and a region within one of the localized retina detachments (B). In both areas the disease-related autofluorescent material was present primarily in the ganglion cells (gcl) and outer limiting membrane (olm). The retinal pigment epithelium (rpe) also contained autofluorescent material with similar spectral properties, but this likely consisted primarily of normal age pigment since similar RPE autofluorescence is present in genetically normal dogs. Scale bar in (B) indicates magnification for both micrographs.