Lipofuscin Turnover

Arecent article in IOVS by Katz et al. ¹ reported the reversible accumulation of lipofuscin-like inclusions in the retinal pigment epithelium of rats given a single intravitreal injection of the protease inhibitor leupeptin. Although the experimental component of the study was comprehensive, we feel that the interpretation of the results was potentially misleading, particularly because the authors had ignored some fundamental aspects of lipofuscin biochemistry. First, RPE lipofuscin is likely to be heterogeneous, because it will be derived from both autophagy and the degradation of photoreceptor outer segments (POS). This is confirmed by a variety of studies that demonstrate that lipofuscin-like granules naturally accumulate in RPE cells in the absence of POS ^{2 3} and that this process can be accelerated by protease inhibitors and certain calcium antagonists. ^{2 3 4} Second, the lipofuscin-like granules generated by protease inhibitors and calcium antagonists in vitro have very different chemical constituents to those of “true” lipofuscin granules isolated from RPE cells. ⁴ These granules, while exhibiting the broad band fluorescence of lipofuscin (which is not retinoid dependent, as suggested by Katz et al. ¹ ), do not contain the classic lipid-soluble fluorophores associated with true lipofuscin. This discrepancy has been highlighted by both ourselves ⁴ and previously by Katz et al. ⁵ Third, even when protease inhibitors are used in conjunction with POS, the resultant granules generated in vitro fail to exhibit the classic fluorophores associated with true lipofuscin. ⁴ Fourth, the recently identified pyridinium bis-retinoid, ⁶ otherwise termed A2E, has a chemical structure that makes it resistant to lysosomal degradation. This also may be true for other fluorophores within true lipofuscin. Finally, we have loaded cultured RPE cells with true lipofuscin for periods up to 56 days and failed to observe any degradation or change in lipofuscin composition (Boulton M, unpublished results). Although it could be argued that the lysosomal system is less functional in cultured RPE, it is certainly sufficient to digest phagocytosed POS. We feel that the data presented by Katz et al. ¹ adds further support to the above raised points. Formation of high levels of large inclusions of lipofuscin-like material was observed in rats shortly after leupeptin administration, but this was not reflected by a corresponding increase in overall cellular fluorescence by 20 weeks. Thus, it is likely that the autofluorescent granules that are transiently, and rapidly, upregulated after leupeptin treatment in the Katz study simply reflect a buildup of phagosomal material (phagosomes are known to be weakly autofluorescent) which is rapidly destroyed once the effect of leupeptin has worn off. The loss of this leupeptin-induced material, together with the observed continuous and equal increase in cellular fluorescence in both the control and treated cell populations after prolonged incubation, is further evidence to the differences in composition between “true” lipofuscin and lipofuscin-like granules. 

Finally, a minor but important point: Katz et al. ¹ refer to the bimodal accumulation of lipofuscin reported by two studies that were limited by the number of donors in the young and middle-age groups. ^{7 8} In vivo fluorescence measurements as undertaken by Delori et al. ⁹ and Von Ruckman et al. ¹⁰ suggest that lipofuscin accumulation is linear throughout life. This is more logical but, sadly, is an observation ignored by many researchers when discussing lipofuscin accumulation.