OCRL1 function in renal epithelial membrane traffic

Abstract

The X-linked disorder Lowe syndrome arises from mutations in OCRL1, a lipid phosphatase that hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP(2)). Most patients with Lowe syndrome develop proteinuria very early in life. PIP(2) dynamics are known to modulate numerous steps in membrane trafficking, and it has been proposed that OCRL1 activity regulates the biogenesis or trafficking of the multiligand receptor megalin. To examine this possibility, we investigated the effects of siRNA-mediated OCRL1 knockdown on biosynthetic and postendocytic membrane traffic in canine and human renal epithelial cells. Cells depleted of OCRL1 did not have significantly elevated levels of cellular PIP(2) but displayed an increase in actin comets, as previously observed in cultured cells derived from Lowe patients. Using assays to independently quantitate the endocytic trafficking of megalin and of megalin ligands, we could observe no defect in the trafficking or function of megalin upon OCRL1 knockdown. Moreover, apical delivery of a newly synthesized marker protein was unaffected. OCRL1 knockdown did result in a significant increase in secretion of the lysosomal hydrolase cathepsin D, consistent with a role for OCRL1 in membrane trafficking between the trans-Golgi network and endosomes. Together, our studies suggest that OCRL1 does not directly modulate endocytosis or postendocytic membrane traffic and that the renal manifestations observed in Lowe syndrome patients are downstream consequences of the loss of OCRL1 function.

Fig. 1.

siRNA-mediated knockdown of oculocerebrorenal syndrome of Lowe (OCRL1) in human and canine cells. A: Madin-Darby canine kidney (MDCK) cells were fixed, processed for indirect immunofluorescence to detect OCRL1 and the trans-Golgi network (TGN) marker furin, and examined using confocal microscopy. The images are maximum projections of 10 confocal slices. Scale bar: 10 μm. B: low-passage MDCK or HK-2 cells were nucleofected with control or OCRL1 siRNA as described in materials and methods. Cells were plated directly onto permeable supports (MDCK) or in 12-well dishes (HK-2) for 3 days. Samples were harvested and analyzed by Western blotting (WB) to detect OCRL1 and actin (as a loading control). The migration of molecular mass standards is indicated on the right. C: siRNA knockdown of OCRL1 in HK-2 cells was confirmed using a PCR-based assay as described in materials and methods. D: primers specific for human INPP5B were used to amplify mRNA isolated from HK-2 or HeLa cells. HeLa cells transfected with a cDNA encoding human INPP5B were included as a positive control to demonstrate the efficacy of the primers.

Fig. 2.

Phosphatidylinositol 4,5-bisphosphate (PIP₂) levels and actin comet frequency are elevated upon OCRL1 knockdown in MDCK cells. A: MDCK cells were treated with either control or OCRL1 siRNA and plated directly onto filters for 3 days. Phospholipids were labeled with [³²P]orthophosphate and analyzed by TLC to determine relative phospholipids levels as described in materials and methods. PIP₂ values in cells nucleofected with OCRL1 siRNA were normalized to control in 3 independent experiments and means ± SE are plotted. The difference in PIP₂ levels between the 2 experimental conditions is not statistically significant by Student's t-test. B: MDCK cells stably expressing GFP-actin were electroporated with control or OCRL1 siRNA and plated onto filters for 2 days before being transferred to Bioptech 0.17-mm ΔT dishes for an additional day before imaging. Images were taken every 2 s. MetaMorph software was used to overlay multiple frames and filter out low-level fluorescence to reveal the path of the actin comets in the cell (right; arrows). Bottom: path of a single comet. In each image, the arrow represents the starting position of the actin comet in the series. The percentage of cells treated with control vs. OCRL1 siRNA that had detectable actin comets during a 3-min imaging window is noted underneath; n represents the number of cells examined for each condition. Scale bar: 10 μm.

Fig. 3.

Knockdown of OCRL1 does not enhance apical biosynthetic delivery kinetics in MDCK, HK-2 cells. A: MDCK cells were electroporated in buffer containing either control siRNA or an siRNA oligonucleotide directed against N-WASP or OCRL1. The efficiency of N-WASP knockdown was between 15 and 40% based on WB of cell lysates (not shown). The cells were seeded onto Transwell filters for 3 days and then infected with adenovirus (AV) expressing the apical protein hemagglutinin (HA) and either control AV or AV-PI5KIβ. The following day, cells were starved, radiolabeled for 15 min, and chased for 2 h at 19°C. Cell surface delivery kinetics of HA were measured after warming to 37°C using a cell surface trypsinization assay. Similar results were obtained in 3 independent experiments; results from a single representative experiment are plotted. B: HK-2 cells treated with either control or OCRL1 siRNA were infected with AV-HA 2 days after nucleofection. The following day, cells were radiolabeled for 15 min and plasma membrane delivery kinetics of HA were quantitated. Similar results were obtained in 4 independent experiments; results from a single representative experiment are plotted.

Fig. 4.

OCRL1 knockdown does not affect megalin-mediated uptake and degradation of lactoferrin (Lf) in MDCK and HK-2 cells. A: comparable levels of MDCK and HK-2 cell lysates were blotted with antibodies against ARH and Dab-2. The migration of molecular mass standards is indicated on the right. B: filter-grown MDCK cells were infected with replication-defective recombinant AV encoding V5- and GFP-tagged mini-megalin. Cells were incubated on ice with anti-V5 antibody and secondary antibody to label the surface population of mini-megalin (red) and then fixed and processed for confocal microscopy. The total cellular population of mini-megalin was visualized using the GFP label (green). Scale bar: 10 μm. C: filter-grown MDCK cells infected with control AV or AV encoding mini-megalin were incubated with apically added [¹²⁵I]Lf on ice, then washed, solubilized, and cell-associated radioactivity was quantitated using a gamma counter. MDCK (D, E) or HK-2 (F, G) cells treated with control or OCRL1 siRNA were incubated with [¹²⁵I]Lf as described in materials and methods. MDCK cells were infected with AV-mini-megalin 1 day before the experiment. The kinetics of Lf recycling (D, F) and degradation (E, G) were quantitated. Similar results were obtained in 3 independent experiments for each cell type.

Fig. 5.

OCRL1 knockdown does not affect endocytosis of megalin in HK-2 cells. A: effect of PI5KIβ overexpression (left) or OCRL1 knockdown (right) on endocytosis of mini-megalin was quantitated using the biotinylation-based assay described in materials and methods. HK-2 cells treated with control or OCRL1 siRNA or infected with control or PI5KIβ expressing AVs as indicated were biotinylated on ice and then warmed to 37°C for 0 or 6 min. Samples were stripped to remove surface biotin and endocytosis was quantitated by WB after recovery of residual biotinylated mini-megalin. One set of 0-min samples was left unstripped so that results could be normalized to total mini-megalin at the surface before warming up. The mean endocytosis (±SE) from 3 experiments comparing PI5KIβ overexpression to control AV and 4 experiments comparing OCRL1 and control siRNA are plotted. B: endocytosis of [¹²⁵I]IgA by polymeric immunoglobulin receptor (pIgR)-expressing HK-2 cells nucleofected with either control or OCRL1 siRNA was performed as described in materials and methods. The means ± range of duplicate samples is shown. Similar results were obtained in 2 experiments.

Fig. 6.

Delivery of newly synthesized lysosomal hydrolases is impaired in HK-2 cells lacking OCRL1. HK-2 cells treated with control or OCRL1 siRNA were radiolabeled for 2 h and chased for 4 h. NH₄Cl (10 mM) was included in the indicated samples. Radioactive cathepsin D secreted into the media during the chase was quantitated after immunoprecipitation and SDS-PAGE and normalized relative to control. The results from 3 experiments are plotted. *P = 0.029 by Mann-Whitney Rank Sum test. Samples of cathepsin D immunoprecipitated from the medium in a representative experiment are shown above the graph.