Targeted disruption of Müller cell metabolism induces photoreceptor dysmorphogenesis

Reliability assessment of a rod photoreceptor outer segment grading system

The purpose of this study was to assess the reliability of a rod photoreceptor outer segment (PR-OS) grading system based on the analysis of 1 microm thick retinal sections obtained from Xenopus laevis whole-eye organ cultures. Digitally captured images, representative of the entire spectrum of rod PR-OS organization levels, were selected and coded numerically. A total of 102 individual rod PR-OS profiles were graded according to a six-step classification scheme based on the percentage of rod PR-OS membrane organization. Unweighted (exact agreement) and weighted kappa (kappa) coefficients (for use with ordered categorical rating scales) were calculated. Differences between kappa coefficients were tested for by chi-square analysis. To investigate the intra- and inter-rater variability and the possible presence of an interaction of the measurements with time, a repeated-measures analysis of variance was performed. The overall unweighted and weighted intra-rater kappa coefficients were 0.78 and 0.92, respectively. The overall unweighted and weighted inter-rater kappa coefficients were 0.73 and 0.90, respectively. There was no significant difference between raters or between first and second reading, nor was interaction between raters and time of rating documented. Individual kappa coefficients were equivalent both between raters and between sessions. Intra- and inter-rater agreement was within one step in 100% of cases. The estimated values of the kappa coefficients are consistent with a good to excellent degree of reliability and reproducibility of this rod PR-OS grading system. This system will be useful in the assessment of rod PR-OS morphology in studies of photoreceptor physiology and pathology.

Investigating the mechanisms of retinal degenerations with antisense oligonucleotides

Utilizing antisense oligonucleotides coupled with an intact Xenopus eye rudiment model, we have effectively demonstrated that we are able to downregulate the expression of a photoreceptor-specific protein, rds/peripherin, and generate a loss-of-function model upon which to further study the function of the rds/peripherin gene. The ultrastructure and protein expression patterns very closely resemble those previously documented in both the rds mouse and in human autosomal dominant retinitis pigmentosa due to peripherin/RDS mutations. An identical strategy can be applied to any gene correlated with a degenerative retinal phenotype. As the entire array of genes is revealed through the various genome projects, including human and mouse, it is becoming increasingly critical to evaluate and determine the function of the corresponding gene products. Discovering which gene is responsible for a particular clinical phenotype is only the first of many steps in the development of a treatment or cure for that particular disease. Using our in vitro model, in which the retina is readily accessible to the antisense oligonucleotide yet the normal three-dimensional ultrastructure of the retina is maintained, we can evaluate the function of virtually any gene as the sequence becomes available. A thorough understanding of the function of individual genes will provide insights on the role of gene product in retinal health and pathophysiology. This experimental approach will also allow for specific therapeutic interventions to be evaluated so that targeted treatments can be designed for individuals with specific genetic mutations.