Critical Role for Phosphatidylinositol-3 Kinase Vps34/PIK3C3 in ON-Bipolar Cells

Abstract

Purpose Phosphatidylinositol-3-phosphate (PI(3)P), and Vps34, the type III phosphatidylinositol 3-kinase primarily responsible for its production, are important for function and survival of sensory neurons, where they have key roles in membrane processing events, such as autophagy, endosome processing, and fusion of membranes bearing ubiquitinated cargos with lysosomes. We examined their roles in the most abundant class of secondary neurons in the vertebrate retina, the ON-bipolar cells (ON-BCs). Methods A conditional Vps34 knockout mouse line was generated by crossing Vps34 floxed mice with transgenic mice expressing Cre recombinase in ON-BCs. Structural changes in the retina were determined by immunofluorescence and electron microscopy, and bipolar cell function was determined by electroretinography. Results Vps34 deletion led to selective death of ON-BCs, a thinning of the inner nuclear layer, and a progressive decline of electroretinogram b-wave amplitudes. There was no evidence for loss of other retinal neurons, or disruption of rod-horizontal cell contacts in the outer plexiform layer. Loss of Vps34 led to aberrant accumulation of membranes positive for autophagy markers LC3, p62, and ubiquitin, accumulation of endosomal membranes positive for Rab7, and accumulation of lysosomes. Similar effects were observed in Purkinje cells of the cerebellum, leading to severe and progressive ataxia. Conclusions These results support an essential role for PI(3)P in fusion of autophagosomes with lysosomes and in late endosome maturation. The cell death resulting from Vps34 knockout suggests that these processes are essential for the health of ON-BCs.

Multiple phosphatidylinositol(3)phosphate roles in retinal pigment epithelium membrane recycling

“…Autophagy was arrested prior to lysosome fusion, evidenced by accumulation of immature autophagic structures containing p62, ATG9, and LC3, but not lysosomal marker LAMP1 ( Fig 5). These observations are consistent with the autophagy defects reported in various Vps34 knockout cells and tissues (Zhou et al, 2010;Jaber et al, 2012;Bechtel et al, 2013;Reifler et al, 2014;He et al, 2016He et al, , 2019 There is no evidence that PI(3)P is needed for autophagosome formation or LC3 recruitment and lipidation in RPE, but we cannot rule out the possibility that it might be essential for them under some conditions. WIPI2, which was identified as a PI(3)P effector protein recruited to autophagosomes in a PI(3)P dependent manner (Dooley et al, 2014;Polson et al, 2010), was also dispensable for LC3 lipidation: treatment of RPE cells with a selective Vps34 inhibitor abolished CQ-induced WIPI2 puncta, but not LC3 puncta ( Fig 6); and transfection of RPE cells with WIPI2 shRNA greatly reduced WIPI2 protein levels, but did not affect LC3 puncta (Fig 10).…”

“…KO) cells the GFP-2xHrs probe was diffusely localized, indicating loss of the Vps34 product, PI(3)P ( Fig 1C). Similarly, punctate immunostaining of the endogenous PI(3)P binding protein EEA1 in endosomes was greatly reduced in KO RPE ( Fig 1D), and KO cells accumulated many puncta positive for the autophagy protein p62 as observed previously in Vps34 KO neurons (Zhou et al, 2010;He et al, 2016He et al, , 2019. These results confirm both the specificity of the PI(3)P probe and the success of the knockout in Cre-expressing cells.…”

“…Vps34 function is required for completion of autophagy, and knockout of Vps34 has been shown to result in arrest prior to autophagolysosome formation in every cell type tested (Zhou et al, 2010;Jaber et al, 2012;Bechtel et al, 2013;He et al, 2016He et al, , 2019Grieco et al, 2018). Similarly, in the KO RPE, puncta containing LC3 and autophagy proteins Atg9 and p62 accumulated ( Fig 4A,E,F,5A,E) and colocalized ( Fig 5C,D), along with ubiquitinated proteins ( Fig 5G).…”

Phosphoinositides in Retinal Function and Disease

“…One of the reasons for the dearth of knowledge has been a lack of tools for studying these very low-abundance lipids within the multiple cell types of the retina and the adjacent retinal pigment epithelium. Recently, tools developed by the broader PI field have begun to be applied to the unique challenges and opportunities posed by the retina [22,49]. Doing so will have enormous impact on our understanding of the cell biology of the retina and its disruption in disease.…”

“…When the Type III PI-3 kinase, Vps34, was knocked out in mouse rod cells and phosphoinositide levels measured, the results suggested complete ablation of PI(3)P despite the presence of the Type I enzyme [22]. Several of these have been reported to be expressed in retina or RPE at the level of protein, mRNA or enzyme activity and most, if not all, are likely present at some level; however, as far as their functions, only the Type I and Type III PI-3 kinase have been studied using gene knockouts in the retina and RPE [22,49,57,128]. Global knockouts have been produced for the α, β, and γ isoforms of Type I PIP-kinases, which are the major source of PI(4,5)P 2 in most cells [16].…”