Insulin signaling to the glomerular podocyte is critical for normal kidney function

Abstract

Diabetic nephropathy (DN) is the leading cause of renal failure in the world. It is characterized by albuminuria and abnormal glomerular function and is considered a hyperglycemic "microvascular" complication of diabetes, implying a primary defect in the endothelium. However, we have previously shown that human podocytes have robust responses to insulin. To determine whether insulin signaling in podocytes affects glomerular function in vivo, we generated mice with specific deletion of the insulin receptor from their podocytes. These animals develop significant albuminuria together with histological features that recapitulate DN, but in a normoglycemic environment. Examination of "normal" insulin-responsive podocytes in vivo and in vitro demonstrates that insulin signals through the MAPK and PI3K pathways via the insulin receptor and directly remodels the actin cytoskeleton of this cell. Collectively, this work reveals the critical importance of podocyte insulin sensitivity for kidney function.

Figure 1. In vivo podocyte specific IR knockdown in transgenic mice

A. Glomeruli of Z/EG reporter mice immuno-stained with an anti GFP antibody. Specific excision demonstrated in podocytes (arrowed). B. Glomeruli were extracted and purified using dynabeads. This is light microscopic analysis of the extract showing a pure glomerular population and the beads. C. PCR of genomic DNA from different tissues in podIRKO compared with (cw) IR^fl/fl cre negative mice. Only podIRKO glomeruli have a smaller excised genomic DNA band present [arrowed] (G=glomeruli S=spleen E= ear). D. Podocyte isolation of CFP positive podocytes from 3 week old podIRKO and control mice demonstrate that the IR mRNA is knocked down by over 90% in podocytes of the podIRKO mice compared with other glomerular cells (OGC) which were CFP negative. In control mice the IR is increased by 60% in the podocyte fraction. E. There is increased production (100%) of IGF-1 receptor mRNA in podocyte fraction of podIRKO mice compared with a 50% reduction in the control mice.

Figure 1. In vivo podocyte specific IR knockdown in transgenic mice

A. Glomeruli of Z/EG reporter mice immuno-stained with an anti GFP antibody. Specific excision demonstrated in podocytes (arrowed). B. Glomeruli were extracted and purified using dynabeads. This is light microscopic analysis of the extract showing a pure glomerular population and the beads. C. PCR of genomic DNA from different tissues in podIRKO compared with (cw) IR^fl/fl cre negative mice. Only podIRKO glomeruli have a smaller excised genomic DNA band present [arrowed] (G=glomeruli S=spleen E= ear). D. Podocyte isolation of CFP positive podocytes from 3 week old podIRKO and control mice demonstrate that the IR mRNA is knocked down by over 90% in podocytes of the podIRKO mice compared with other glomerular cells (OGC) which were CFP negative. In control mice the IR is increased by 60% in the podocyte fraction. E. There is increased production (100%) of IGF-1 receptor mRNA in podocyte fraction of podIRKO mice compared with a 50% reduction in the control mice.

Figure 2. PodIRKO mice are normoglycaemic and normally grown at 8 weeks of age

A. Macroscopically mice were indistinguishable. B. No difference between podIRKO and littermate controls after formal weighing (Mean+/− SEM). C. They had comparable random blood glucose levels in both podocin and nephrin cre groups (Mean+/− SEM). D. No difference in serum insulin levels in a subset of mice at 8 weeks. E. Macroscopically no difference in kidney size between podIRKO and control mice at 8 weeks.

Figure 2. PodIRKO mice are normoglycaemic and normally grown at 8 weeks of age

A. Macroscopically mice were indistinguishable. B. No difference between podIRKO and littermate controls after formal weighing (Mean+/− SEM). C. They had comparable random blood glucose levels in both podocin and nephrin cre groups (Mean+/− SEM). D. No difference in serum insulin levels in a subset of mice at 8 weeks. E. Macroscopically no difference in kidney size between podIRKO and control mice at 8 weeks.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 3. PodIRKO mice develop albuminuria and histological changes recapitulating features of diabetic nephropathy

A. At 8 weeks podIRKO mice become albuminuric. SDS page gel shows specific band at 66KD corresponding to Albumin only in PodIRKO mice (2μl of urine loaded for each). B. Significant proteinuria in podIRKO mice generated on different cre promoters, raised in different environments. However there is variability both between strains and also within the groups (***p<0.001 n≥10 each group, mean +/− SEM shown). C. Light microscopy for podocin cre podIRKO mice and controls age 8 weeks. Top panels are low power Periodic Acid Schiff (PAS) staining middle high power PAS and lower H and E staining - high power. PodIRKO (middle column) demonstrate dilated tubules containing proteinaceous casts. The glomeruli show increased mesangial matrix production. Several glomeruli appear shrunken, associated with foci of segmental glomerulosclerosis. The affected glomerular segments show adhesion to Bowman’s capsule, segmental scarring, and capillary insudative (hyalinotic) lesions. Occasional podocytes show small, hyperchromatic, condensed nuclei suggestive of apoptotic bodies (arrowed bottom middle panel). Focal glomeruli show podocyte and parietal epithelial cell swelling with large protein absorption droplets. D. Objective histological glomerular scoring reveals significant levels of glomerulosclerosis and matrix accumulation in both the podocin and nephrin cre mice. Histological abnormality is more marked in the podocin animals. No difference in mesangial hypercellularity detected. (*p<0.05 **p<0.005 ***p<0.0005 n≥10 each group, mean +/− SEM shown). Plasma creatinine levels were measured in a subset of mice at 8 weeks. No difference was detected in the nephrin cre mice but in the podocin cre IRKO animals there was a mildly raised creatinine level in mutants when compared to controls. E. Type IV collagen progressively accumulates in glomeruli of podIRKO mice (brown staining). F. Podocyte apoptosis is detected using Tunel stain in 13 week podIRKO mice (nephrin promoter). Apoptotic nuclei arrowed. G. Light microscopy of 13 week old podIRKO (nephrin promoter) showed massive amounts of matrix formation as well as increased glomerulosclerosis, mesangial expansion and protein in the tubules in comparison to the 8 week old mice. right = high power (x63 magnification), left = low power(x10 magnification) PAS staining. H. Some 13 week old podIRKO mice had small sclerosed kidneys. I. Transmission electron microscopy of podIRKO and control mice. PodIRKO mice have normal GFB at 3 weeks of age. By age 5 weeks the podocyte actin cytoskeleton is beginning to become deranged with broadening of foot processes (arrowed). Foot processes are completely lost by 8 weeks (arrowed). No obvious changes were detected in the GEnC. J. Thickening of the GBM of podIRKO mice when measuring the lamina densa. K. Multiple sections from 3 mutant mice were compared with 3 control mice. Each section had at least 10 measurements of Lamina densa thickness performed perpendicular with the edge of the GBM. **p=0.007 Mean +/− SEM shown.

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 4. Insulin signals to the podocyte in the glomerulus

A. MAPK signaling. Mice given insulin phosphorylated p42/44 MAPK only in the glomerulus (×10 magnification). Green =p42/44MAPK; red = PECAM (endothelial marker); blue = nephrin (podocyte marker). B. Using high power confocal microscopy (×63 magnification) insulin stimulated p42/44 MAPK co-localizes with nephrin demonstrating specific podocyte signaling (middle row). Representative of 3 independent experiments. PodIRKO mice were unable to phosphorylate p42/44 MAPK in response to insulin stimulation (lower row). C. p42/44 MAPK is switched on at 15 minutes with both low (1nM) and high dose (100nM) insulin in wild type human podocytes. This does not occur in IR knockdown podocyte cell lines and GEnC in response to low dose insulin. D. Densitometry in wild type podocytes reveals a 120% increase in insulin-stimulated p42/44MAPK at 15 minutes (p=0.0093 n=6. Mean+/− SEM shown). E. pAKT signaling is predominantly in podocytes after 15 minutes stimulation with insulin in wild type mice (upper and middle panels)[representative images from 3 independent experiments]. Insulin stimulated AKT phosphorylation was abrogated in podIRKO mice (lower panels). F. AKT is phosphorylated in ciPods by 15 minutes at both low and high dose insulin. This does not occur in insulin receptor knockdown podocytes. In contrast to MAPK there is a small amount of pAKT detectable in human GEnC in response to low and high dose insulin. G. Densitometry shows a 200% increase in pAKT phosphorylation in response to insulin in human podocytes (**p= 0.007 n=6. Mean+/− SEM shown).

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.

Figure 5. Insulin directly remodels the actin cytoskeleton of podocytes

A. Immortalized human podocytes and GEnC were insulin stimulated (100nM) for 15 minutes and their F-actin cytoskeleton analyzed with phalloidin staining. Podocytes cortically reorganized (arrowed) their cytoskeleton and this did not occur in GEnC. B. Insulin receptor knock down (IRKD) immortalized podocytes do not remodel their F-actin cytoskelton in response to insulin (phalloidin staining). C. Objective computer assisted analysis with the in cell analyzer showed a dose response in insulin stimulated actin reorganization in Wild type podocytes which was statistically significant from 6.25nM. (ANOVA p<0.001 bonferonni *p<0.05 **p<0.001 compared with (cw) baseline n=8. Mean+/− SEM). This did not occur in the IRKD treated podocytes. D. Atomic force microscopy representative of 4 independent experiments. Insulin causes retraction of cell processes. Insulin applied for 15 minutes. E. Cell motility is increased in podocytes given insulin but not with IGF-1. Number of cells per unit area was significantly higher in insulin treated group 15 hours after insulin stimulation. This did not occur in IGF-1 treatment. (n=4 ** p< 0.01. Mean +/− SEM shown) F. Electrical resistance is lost across podocyte monolayers within 5 minutes of insulin stimulation but returns to baseline by 30 minutes using ECIS analysis. ANOVA post hoc Bonferroni *p<0.05 ** p<0.01. No effect observed in GEnC treated in the same manner (n=12). Mean+/− SEM shown. G. Insulin activates RhoA by 5 minutes (*p=0.018) and inhibits CDC42 by 2 minutes (*p=0.02) of insulin stimulation. These are both back to baseline by 30 and 15 minutes respectively (n=4 each. Mean+/− SEM shown). H. Differentiated human podocytes transfected with dominant negative constructs v Rho (c3 transferase) and CDC42 (N17CDC42). There is diminished insulin induced cortical actin reorganization in Rho dominant negatively treated cells. CDC42 dominant negative transfection resulted in a high proportion of cells being cortically organized in their basal state.