Fig 1.
Insulin induces albumin uptake in proximal tubule cells.
HKC-8 cells were treated with 100nm of insulin (I9278, Sigma) for 30 minutes followed by FITC-albumin (100μg/ml, A7016, Sigma) for an additional 30 minutes and transferred to 4°C, stripped off membrane bound albumin and lysed. Fluorescence was measured and normalized for the amount of protein. Insulin treatment resulted in an increase in albumin uptake (A). HKC-8 cells were pretreated with insulin (100nm) and incubated with increasing concentrations of albumin. Cell lysates were probed with rat albumin antibody (MP Cappel 55715). Albumin uptake was induced in all concentrations of albumin with insulin treatment in association with an increase in pser473-Akt (Cell Signaling) expression (B, C). Western blotting of cell lysates treated with insulin (100nm) revealed that Akt-pSer473 was upregulated significantly starting at 15 minute of incubation at 30 and 60 minutes (D). *p<0.05.
Fig 2.
Basolateral insulin treatment induces albumin endocytosis on polarized proximal tubule epithelial cells.
HKC-8, OK and primary mouse proximal tubule cells were grown on transwell permeable supports and treated with 100nm insulin basolaterally for 1 hour. FITC-alb (100μg/ml) was added to the apical site ½ hour into insulin treatment. Cells were lysed after PBS++ washes. Values were displayed as time-fold increase of albumin uptake in comparison to control cells. Proximal tubule epithelial cells of different origin exhibited an increase in albumin uptake by basolateral insulin treatment (A). Trans Epithelial Electric Resistance (TEER) was measured (ohm) by EVOM epithelial voltohmmeter. Resistance obtained from a well with only media was subtracted from the other wells. Cells reached a stable resistance at 4 days after seeding and expressed tight junction protein ZO1 (B). Localization of e-cadherin at the lateral aspect of the adherence junctions was captured by XY, XZ and YZ confocal section images (E). Primary mouse proximal tubule cells express megalin and form a confluent monolayer 6–8 days after seeding (C, D).
Fig 3.
Akt mediates insulin-induced albumin endocytosis.
Transfection of HKC-8 cells with a double dominant negative Akt construct resulted in a decrease in insulin induced endocytosis (p<0.05). The comparisons were performed between albumin treated and insulin+albumin treated cells. Insulin treatment resulted in a statistically significant increase in albumin uptake. In order to examine the role of Akt in insulin-induced albumin endocytosis, a comparison of albumin uptake was performed between control cells and cells transfected by DDN Akt plasmid after insulin treatment as depicted in the Fig. Western blot images of the samples revealed that cells transfected with DDN had a lower band with exposure to Akt-pser473 antibody (Cell Signaling) which is consistent with the lack of phosphorylation ability at the pser473-residue.
Fig 4.
Insulin treatment results in upregulation of Akt phosphorylation sites of AS160.
Akt substrate 160 kd (AS160) mediates GLUT4 trafficking in response to insulin in adipose tissue and muscle. The possible physiological effects of AS160 in proximal tubule are unexplored. We first showed that AS160 is expressed in human and mouse proximal tubule epithelial cells (Fig A-B). To explore the effect of insulin induced Akt activation on phosphorylation of AS160, we treated HKC-8 cells with insulin (100nm) for 15, 30 and 60 minutes and investigated expression of Akt phosphorylation sites of AS160. Densitometric measurements were normalized for β-actin and expression of AS160-Ser318, AS160-Ser 588, AS160-Thr 642 and total AS160 at 15, 30 and 60 min of insulin exposure were compared with baseline untreated samples (0 min). Insulin treatment induced phosphorylation of AS160 at Akt phosphorylation sites, S318, S588 and Thr 642 sites in proximal tubule epithelial cells (C). *: p<0.05 ** p<0.01.
Fig 5.
HKC-8 cells were transfected with HA-minimegalin and treated with only albumin (A) and insulin+albumin (A+I). Megalin was pulled down by immunoprecipitation with a HA antibody (Roche) and probed for AS160 (rabbit, Milipore). Mouse IgG is used as negative control. The protein expression in the pellet (P), supernatant (S) and input (I) were displayed. The pellet showed significant AS160 expression where as negative control did not display any protein expression indicating a robust interaction between AS160 and megalin. Insulin treatment caused a modest increase in this interaction. The same blot was stripped and reprobed by a HA antibody to show the efficiency of the pull-down experiments (A). Reciprocal immunoprecipitation was accomplished by transfecting the HKC-8 cells with Flag-AS160 and minimegalin plasmids. AS160 was pulled down by immunoprecipitation with a Flag antibody (Sigma) and the membrane was probed for megalin. It was confirmed that megalin has a strong interaction with AS160. Insulin treatment did not alter the degree of interaction between AS160 and megalin. Pull-down with mouse IgG was utilized as a negative control. The same membrane was stripped and reprobed by a Flag antibody (the image at the lower panel) to examine the efficiency of pull-down experiment (B). Both experiments revealed a strong interaction between megalin and AS160. Megalin and AS160 were successfully pulled down by tag antibodies.
Fig 6.
AS160 interacts with cytoplasmic tail of megalin.
In order to examine a possible direct interaction between cytoplasmic tail (CT) of megalin and AS160, in-vitro pull-down experiments were carried out. Cytoplasmic tail of megalin was cloned and GST-megalin fusion protein was generated as outlined in the methods section [21]. Approximately 100μg of GST, GST-megalin CT immobilized on GSH-Sepharose were incubated with HKC-8 cell lysate. After centrifugation, aliquots corresponding to 1/60 of each supernatant (S) and 1/5 of each washed pellet (P) were resolved by SDS-PAGE and probed for AS160. Pull-down experiments demonstrated a direct interaction between GST-megalin CT and AS160. Only GST was used as a negative control. Coomassie blue staining was utilized to ensure equal loading (B) (n = 3).
Fig 7.
Diabetic mice have normal kidney morphology.
Mice were sacrificed and kidneys were removed and stained to examine morphology five weeks after stz injection. Periodic acid-Shiff staining revealed normal glomerular and tubular structure in diabetic mice at 5 weeks after stz injection.
Fig 8.
Diabetic mice display a decrease in expression of endocytic proteins associated with alterations in their urinary shedding.
Diabetic and control mice kidneys were removed five weeks after stz injection. Western blotting to investigate the expression of albumin receptors megalin and cubilin and downstream Akt substrate was performed. Diabetic mice had downregulation of megalin, cubilin and AS160 expression in their kidney (A). Mice injected with stz to induce T1D and control animals underwent 24hr urine collection in metabolic cages 3 and 5 weeks after stz injection. Urinary proteins were precipitated and examined for megalin and cubilin expression. Urinary shedding of megalin was diminished whereas cubilin shedding was evident in diabetic mice at 3 weeks after stz injection at the time of minimal MA (A). At 3 weeks diabetic mice had minimal MA. A significant increase in MA was observed at 5 weeks after stz injection. Increase in urinary albumin excretion in diabetic mice was confirmed by Coomassie staining (B). Densitometric measurement of western blotting of megalin, cubilin and AS160 adjusted for β-actin revealed a statistically significant decrease in expression of these proteins in stz injected diabetic mice (C) *: p<0.05.
Fig 9.
Expression of active Akt is downregulated in diabetic mice kidneys.
Immunofluorescence staining of the control, stz injected diabetic mice for Akt-ser473 (Cell Signaling) was performed. Kidneys were double stained with anti-Rabbit FITC. Mouse kidney with knockout of Akt1 (Akt1KO) was utilized as a negative control. Immunofluorescence imaging revealed that phosphorylation of Akt at the 273 serine residue was diminished in diabetic mice kidneys conforming decrease activation of Akt in diabetic animal kidneys before occurrence of significant changes due to T1D.
Fig 10.
Megalin and cubilin expression is downregulated in diabetic mice kidney.
In order to confirm decreased expression of megalin and cubilin in kidneys of diabetic mice, immunofluorescence staining was performed. Kidneys of control and diabetic mice were stained with cubilin and megalin antibodies and double labeled with anti-sheep Alexa-546 and anti-rabbit FITC. Diabetic mice kidneys displayed a decrease in expression of megalin and cubilin in proximal tubule epithelial cells. Hoechst 33342 staining was utilized to delineate the nuclei of the cell.
Fig 11.
We propose that basolateral binding of insulin to its receptor (IR) is followed by receptor autophosphorylation and activation of tyrosine phosphorylation of insulin receptor substrates (IRS).
Binding of IRS to the regulatory subunit of phosphoinositide 3-kinase (PI3K) results in activation of PI3K, which phosphorylates phosphotidylinositol 4, 5 biphosphate (PIP2) on the 3’ position. This complex activates the 3-phosphoinositide dependent protein kinase, PDK-1, resulting in activation of Akt. Active Akt phosphorylates RabGAP AS160 which will increase GTP bound Rabs and recycling of megalin. Increase in megalin turnover will induce apical albumin endocytosis in proximal tubule epithelial cells. In DM this network of insulin induced protein-protein interactions will be perturbed resulting in altered trafficking of megalin-cubilin complex. Decreased membrane expression of megalin will lead to shedding of cubilin in the urinary space.