Integrin Ligation Results in Nephrin Tyrosine Phosphorylation In Vitro

Abstract

Nephrin is expressed at the basolateral aspect of podocytes and is an important signaling protein at the glomerular slit diaphragm. In vitro studies have demonstrated that Nephrin phosphorylation-dependent signaling is able to assemble a protein complex that is able to polymerize actin. However, proximal signaling events that result in nephrin tyrosine phosphorylation are not well understood. Nephrin deletion in mice and human nephrin mutations result in developmental failure of the podocyte intercellular junction resutling in proteinuria. This has been presumed to be due to a failure to respond to an external polarized cue in the absence of nephrin or a failure to transduce an outside-in signal in patients with nephrin mutations. The nephrin extracellular domain binds to itself or neph1 across the foot process intercellular junction. Nephrin is tyrosine phosphorylation-silent in healthy glomeruli when presumably the nephrin extracellular domain is in an engaged state. These observations raise the possibility of an alternate proximal signaling mechanism that might be responsible for nephrin tyrosine phosphorylation. Here we present data showing that integrin engagement at the basal aspect of cultured podocytes results in nephrin tyrosine phosphorylation. This is abrogated by incubating podocytes with an antibody that prevents integrin β1 ligation and activation in response to binding to extracellular matrix. Furthermore, nephrin tyrosine phosphorylation was observed in podocytes expressing a membrane-targeted nephrin construct that lacks the extracellular domain. We propose, integrin-activation based signaling might be responsible for nephrin phosphorylation rather than engagment of the nephrin extracellular domain by a ligand.

Fig 1

(A) Schematic showing integrin chains and receptor specificity of integrin heterodimers. β1 integrin is common to many of the heterodimers that bind to laminin, fibronectin and collagen. (B) Integrin chain expression profile of cultured mouse podocytes cells and mouse whole glomerular lysate.

Fig 2. Nephrin is tyrosine phosphorylated when mouse podocytes are plated on laminin and fibronectin coated surface.

Mouse podocytes were plated on culture dishes coated with laminin (A) and fibronectin (B) for the indicated time points. At time 0 cells were maintained in suspension (Su) on a rocker prior to plating. Cells were lysed at various time points and lysates were resolved using SDS-PAGE. Membranes were probed with the indicated antibodies. FAK Y397 phosphorylation was used as a surrogate for integrin activation. (C and D) Bands densitometry using ImageJ software for 4 separate experiments showing increase in nephrin tyrosine phosphorylation as well as FAK Y397 phosphorylaton at different time points. Data are mean values ± SEM. *P<0.001.

Fig 3. Nephrin tyrosine phosphorylation on laminin-coated surface is abrogated by pre-incubation of podocytes with β1 blocking antibody.

(A). Mouse podocytes were incubated with β1 blocking mab and mouse IgG (control) prior to plating on laminin coated surface. Time 0 denotes cell in suspension (Su). Cells were lysed at various time points and lysates were resolved using SDS-PAGE. Membranes were probed with the indicated antibodies. FAK Y397 phosphorylation was used as a surrogate for integrin activation. (B) Bands densitometry using ImageJ software for 4 separate experiments showing increase in nephrin tyrosine phosphorylation as well as FAK Y397 phosphorylaton at different time points. Data are mean values ± SEM. *P<0.001, using two-tailed t test.

Fig 4. Nephrin tyrosine phosphorylation does not increase when mouse podocytes were plated on collagen-coated or uncoated surface.

(A) Lysates of mouse podocytes plated on fibronectin-coated, collagen I and IV-coated and uncoated surface were blotted for phospho-nephrin and p-FAK. The increase in nephrin phosphorylation observed on plating on fibronectin coated surface was not seen when cells were plated on collagen-coated or uncoated surface. Antibody that detects activated β1 integrin (HUTS5) was used to to assess integrin activation under non-reducing conditions. (B) Bands densitometry using ImageJ software for 4 separate experiments showing increase in nephrin tyrosine phosphorylation as well as activated β1 integrin at different time points. Data are mean values ± SEM. *P<0.001.

Fig 5. Nephrin extracellular domain is not required for nephrin tyrosine phosphorylation.

(A). HEK 293 cells expressing a nephrin construct where the extracellular domain is substituted by a myristoylated sequence is tyrosine phosphorylated following plating of cells on laminin coated surface. (B) Band densitometry using ImageJ software for 4 separate experiments showing increase in nephrin tyrosine phosphorylation as well as activated β1 integrin at different time points. Data are mean values ± SEM. *P<0.001, using two-tailed t test.

Fig 6. Src activation is abrogated in the presence of β1 blocking mab.

(A) Cultured mouse podocyte lysates show decrease in Src Y416 phosphorylation in the presence of β1 blocking mab suggesting decreased activation of Src kinases when integrin β1 activation is prevented. Mouse IgG was used as control. (B) Bands densitometry using ImageJ software for 4 separate experiments showing increase in nephrin and Src Y416 tyrosine phosphorylation, as well as activated β1 integrin at different time points. Data are mean values ± SEM. *P<0.001, using two-tailed t test.

Fig 7. Src Kinase Fyn is necessary for nephrin phosphorylation.

(A) SYF cells that lack Src, Fyn and Yes were transfected with full-length nephrin. Cells were plated on laminin-coated surface for indicated time points. Nephrin phosphorylation as well as β1 integrin activation was observed only after introduction of Fyn in SYF cells. (B) Band densitometry using ImageJ software for 4 separate experiments showing increase in nephrin tyrosine phosphorylation in the presence and absence of Fyn at different time points. Data are mean values ± SEM. *P<0.001, using two-tailed t test.

Fig 8

(A) Activated β1 integrin and phospho-nephrin colocalize in developing glomeruli. Frozen sections from newborn mouse kidneys were immunstained with activated β1 integrin and phospho-nephrin antibodes showing colocaliztion during glomerular devlopment. (B) Endogenous nephrin is seen in focal adhesions in cultured podocytes. Mouse podocytes expressing DsRED tagged CrkI (found in focal adhesions) were plated on laminin-coated glass cover slips. Immunostaining for nephrin shows presence of nephrin at the leading edge as well as in focal adhesions (Panel B). There is co-localization of CrkI and activated β1 integrin with phosphorylated nephrin (Panel C and D). Panel A shows the specificty of staining as the secondary antibody by itself does not give a signal. Scale bars, 20μm.

Fig 9. Schematic showing existing paradigm as well as proposed signaling that results in nephrin phosphorylation via integrin activation during development and following injury.