Nephrin Tyrosine Phosphorylation Is Required to Stabilize and Restore Podocyte Foot Process Architecture

Abstract

Podocytes are specialized epithelial cells of the kidney blood filtration barrier that contribute to permselectivity via a series of interdigitating actin–rich foot processes. Positioned between adjacent projections is a unique cell junction known as the slit diaphragm, which is physically connected to the actin cytoskeleton via the transmembrane protein nephrin. Evidence indicates that tyrosine phosphorylation of the intracellular tail of nephrin initiates signaling events, including recruitment of cytoplasmic adaptor proteins Nck1 and Nck2 that regulate actin cytoskeletal dynamics. Nephrin tyrosine phosphorylation is altered in human and experimental renal diseases characterized by pathologic foot process remodeling, prompting the hypothesis that phosphonephrin signaling directly influences podocyte morphology. To explore this possibility, we generated and analyzed knockin mice with mutations that disrupt nephrin tyrosine phosphorylation and Nck1/2 binding (nephrin^Y3F/Y3F mice). Homozygous nephrin^Y3F/Y3F mice developed progressive proteinuria accompanied by structural changes in the filtration barrier, including podocyte foot process effacement, irregular thickening of the glomerular basement membrane, and dilated capillary loops, with a similar but later onset phenotype in heterozygous animals. Furthermore, compared with wild-type mice, nephrin^Y3F/Y3F mice displayed delayed recovery in podocyte injury models. Profiling of nephrin tyrosine phosphorylation dynamics in wild-type mice subjected to podocyte injury indicated site-specific differences in phosphorylation at baseline, injury, and recovery, which correlated with loss of nephrin-Nck1/2 association during foot process effacement. Our results define an essential requirement for nephrin tyrosine phosphorylation in stabilizing podocyte morphology and suggest a model in which dynamic changes in phosphotyrosine-based signaling confer plasticity to the podocyte actin cytoskeleton.

Blood filtration in the kidney is critically dependent on the size–selective barrier wall of the glomerulus, which is composed of an inner layer of fenestrated endothelium, a glomerular basement membrane (GBM), and an outer layer of podocytes. Differentiated podocytes adopt a unique and complex three–dimensional architecture that is fundamental to their function.^1,2 Their cell bodies extend numerous microtubule–based major processes, which branch into an interdigitating network of smaller actin–rich foot processes bridged by slit diaphragms that serve as the final filtration element to limit loss of urinary protein. Podocyte injury has emerged as a significant contributor to many forms of renal disease and is characterized by remodeling of the actin cytoskeleton, loss of slit diaphragms, and foot process effacement, leading to proteinuria.³

The slit diaphragm is a specialized cell-cell junction that is anchored to the actin cytoskeleton via a series of transmembrane proteins, including the central scaffolding protein nephrin.⁴ Mutations in nephrin in humans and rodents lead to foot process effacement and proteinuria.⁵ The Fin_minor mutation, which results in near-complete truncation of the short cytoplasmic tail of nephrin, causes a similar disease phenotype as complete loss of nephrin,⁶ suggesting that this segment is essential for nephrin function. Within this segment, there exists a number of highly conserved tyrosine (Y) residues (Table 1) that, on phosphorylation by Fyn kinase,^7–9 serve as docking sites for intracellular signaling proteins (Figure 1A).¹⁰ Through recruitment of actin adaptors, such as p85/PI3K,^11,12 the Cas/Crk complex,¹³ and Nck1/2,^8,9,14 nephrin phosphorylation is postulated to facilitate direct and dynamic connection to the podocyte cytoskeleton. Moreover, Nck enhances nephrin phosphorylation via activation of Fyn,¹⁵ and loss of Nck within podocytes leads to reduced nephrin tyrosine phosphorylation and widespread foot process effacement,^15,16 inferring a reciprocal relationship between Nck and nephrin in the maintenance of podocyte structure.

Table 1.

Conserved tyrosine residues in human, mouse, and rat nephrin

Human	Mouse	Rat
1114-YEESa	1128-YEESa	1127-YEESa
1138-YYRS	1153-YYSM	1152-YYSM
1158-YSRG	1172-YRQA	1171-YHQG
1176 – YDEVb	1191-YDEVb
1183-YPPS	1198-YGPP	1194-YGPP
1193-YDEVa,b	1208-YDEVa,b	1204-YDEVa,b
	1216-YDLR	1212-YDLR
1210-YQDP	1225-YEDP
1217-YDQVa,b	1232-YDQVa,b	1228-YDQVa,b

The human Fin_minor mutation (R1109X) results in deletion of all indicated tyrosine residues.

^aConserved between all three species.

^bYDxV motifs.

Figure 1.

Generation of nephrin-Y3F mice. (A) Diagram of human nephrin protein indicating the positions of phosphorylated (P) tyrosine (Y) residues within the cytoplasmic tail of nephrin with several binding partners as indicated. The three mouse nephrin tyrosines corresponding to human Y1176, Y1193, and Y1217 (yellow) were mutated to phenylalanine (F) to create nephrin-Y3F. (B) Schematic of the Y3F targeting vector used to introduce the modified exons (28 and 29; yellow) into the WT murine Nphs1 locus by homologous recombination. A loxP-flanked PGK-Neo selection cassette was later removed via Hprt-Cre. (C) PCR analysis of the Nphs1^WT and Nphs1^Y3F alleles using FW and RV primers (as in B) on genomic DNA samples from representative nephrin^WT/WT, nephrin^WT/Y3F (HET), and nephrin^Y3F/Y3F animals. All genotypes on both CD-1 and C57Bl/6 genetic backgrounds are obtained with normal Mendelian frequency. *, **Indicate presence of Y to F mutations.

Nephrin is tyrosine phosphorylated during glomerulogenesis and throughout life,^8,16,17 and decreases in nephrin phosphorylation on Y1217 are seen coincident with foot process effacement in human kidney diseases, including minimal change disease (MCD)¹⁸ and membranous nephropathy.¹⁹ Similarly, reduced phosphorylation of the downstream survival factor Akt on serine (S) 473 can be observed in MCD,^20,21 and inactivation of the actin binding protein cofilin is detected in patients with MCD, membranous nephropathy, and FSGS.²² These observations raise the intriguing possibility that uncoupling of phosphonephrin signaling from actin leads to pathologic disruptions in foot process morphology. However, additional investigation is required to determine whether perturbations in nephrin tyrosine phosphorylation directly contribute to the development of disease as a result of altered downstream signaling or rather, if decreases in nephrin tyrosine phosphorylation occur during the progression of disease simply as a consequence of changes in other signaling pathways.

To resolve the significant question of whether nephrin tyrosine phosphorylation directly controls podocyte function, we generated mice deficient in phosphonephrin signaling by altering the three tyrosine phosphorylation sites that mediate Nck binding (denoted nephrin^Y3F). Nephrin^Y3F/Y3 mice develop widespread foot process effacement, basement membrane alterations, and defects in filtration barrier integrity, and they show delayed recovery within experimental models of podocyte injury. Our findings provide compelling evidence for an ongoing requirement for nephrin phosphotyrosine–based signaling in stabilization and restoration of podocyte cytoarchitecture and suggest that there may be value in the pursuit of therapeutic strategies that can selectively enhance nephrin tyrosine phosphorylation.

Results

Generation of Nephrin^Y3F/Y3F Knockin Mice

To directly investigate the role of nephrin tyrosine phosphorylation in podocyte biology, we generated knockin mice in which the Y residues within the three Nck-binding YDxV motifs of nephrin were converted to phenylalanine (F), an amino acid that mimics the structure of tyrosine but cannot undergo phosphorylation (Figure 1A). These motifs correspond to tyrosine residues 1191, 1208, and 1232 in mouse and are herein referred to using the human amino acid numbering system with analogous positions at 1176, 1193, and 1217 (Table 1). A targeting vector containing appropriate nucleotide point mutations was constructed (Figure 1B), and the three mutations (herein denoted Y3F) were incorporated into the endogenous Nphs1 locus by homologous recombination. Successful targeting led to the generation of nephrin^{wild type (WT)/Y3F} mice on a mixed C57BL/6:129 background, which we subsequently bred onto the outbred CD-1 background for more than six generations and independently bred onto the C57BL/6N background for more than five generations. When we intercrossed heterozygous nephrin^WT/Y3F mice, we observed the birth of nephrin^WT/WT (WT), nephrin^WT/Y3F, and nephrin^Y3F/Y3F (Y3F) animals (Figure 1C) in the expected Mendelian ratios on both backgrounds. The presence of the three mutations was verified by DNA sequencing (data not shown).

The expression and biochemical function of the mutant nephrin allele were examined in both CD-1 and C57BL/6 mice, with comparable results seen in both genetic backgrounds (Figure 2) (data not shown). We first isolated mRNA from the kidney cortex of 1- and 6-month-old nephrin^WT/WT and nephrin^Y3F/Y3F animals and used quantitative PCR to compare the amount of mRNA expression of the two Nphs1 alleles as well as the podocyte markers Nphs2, Actn4, and Wt1 (Figure 2A). We observed similar amounts of Nphs1 mRNA in both young and aged nephrin^WT/WT and nephrin^Y3F/Y3F animals, suggesting that insertion of the knockin mutation did not affect nephrin promoter activity or mRNA stability. There were no significant differences in expression of the other podocyte markers. We next isolated glomeruli from 1-month-old nephrin^WT/WT, nephrin^WT/Y3F, and nephrin^Y3F/Y3F animals and used Western blot analysis to monitor nephrin expression and tyrosine phosphorylation (Figure 2B). Comparable levels of total nephrin relative to GAPDH were seen in all genotypes. Moreover, using our phosphospecific nephrin antibodies developed to recognize Y1176, Y1193, or Y1217,¹⁶ we detected strong phosphorylation of nephrin in WT animals, which was severely decreased in nephrin^WT/Y3F animals and completely absent from Y3F animals (Figure 2B).

Figure 2.

Validation of expression and function of Nphs1^Y3F mutant allele. (A) Real–time PCR analysis of mRNA levels in glomeruli isolated from 1- and 6-month-old nephrin^WT/WT and nephrin^Y3F/Y3F animals for Nphs1 and podocyte markers Nphs2, Actn4, and Wt1. For each gene, expression in WT animals was adjusted to 1.0. (B) Nephrin protein expression and phosphorylation was evaluated using immunoblotting (IB) for the indicated antibodies on glomeruli isolated from nephrin^WT/WT, nephrin^WT/Y3F (HET), and nephrin^Y3F/Y3F mice. (C) Dual-immunofluorescence staining for total nephrin (red) and phosphonephrin (Y1217; green) or total nephrin (red) and podocin (green) on kidney sections of 1-month-old WT and Y3F animals. (D) Transmission EM of 1-month-old WT and Y3F animals showing slit diaphragms (arrows). Scale bar, 100 nm. (E) IB for nephrin and Nck in anti-Nck immunoprecipitates (IPs) from WT and Y3F glomerular lysates showing disruption of nephrin-Nck interaction in Y3F mice. (F) IB of glomerular lysates (10 μg) from WT and Y3F mice showing altered nephrin phosphosignaling to Akt and Src. Similar results were obtained in CD-1 and C57BL/6 mice. All results shown are from CD-1 mice, except in E, which is from C57BL/6 mice.

Next, we investigated nephrin localization within the glomerulus of these animals via immunofluorescence staining, which revealed that both nephrin-WT and nephrin-Y3F proteins were expressed in a pattern characteristic of podocyte proteins (Figure 2C). Nephrin was phosphorylated on Y1217 in nephrin^WT/WT animals, whereas we could not detect any signal above background in nephrin^Y3F/Y3F animals (Figure 2C). Given that nephrin phosphorylation can affect nephrin-podocin interactions,^17,23 we performed dual-immunofluorescence staining for nephrin and podocin (Figure 2C) and found that both proteins were similarly expressed in a continuous pattern in the glomeruli of nephrin^WT/WT and nephrin^Y3F/Y3F animals. In accordance with this staining pattern, transmission electron microscopy (EM) revealed the presence of slit diaphragm structures in nephrin^Y3F/Y3F animals (Figure 2D).

Finally, we examined the effect of the mutant nephrin-Y3F protein on nephrin signaling by Western blot analysis of glomerular lysates isolated from nephrin^WT/WT and nephrin^Y3F/Y3F animals. Phosphorylation of these tyrosines is required for nephrin-Nck binding in vitro^8,9,14; thus, we validated that this interaction was disrupted in vivo using immunoprecipitation (Figure 2E). Consistent with the role of Nck recruitment in promoting nephrin tyrosine phosphorylation,¹⁵ we observed a significant reduction in total nephrin tyrosine phosphorylation in nephrin^Y3F/Y3F animals using a pan-phosphotyrosine antibody (Figure 2F). We also found decreased activation of nephrin downstream signaling targets, with reduced phospho-Akt S473 and phospho-Src Y416 levels in nephrin^Y3F/Y3F animals compared with nephrin^WT/WT animals (Figure 2F). Together, these analyses show that expression levels and localization of the mutant nephrin-Y3F protein are comparable with WT nephrin and that absence of the three YDxV tyrosine phosphorylation sites compromises a significant degree of nephrin phosphotyrosine signaling in vivo.

CD-1 Nephrin^Y3F/Y3F Mice Develop Severe Proteinuria, GBM Alterations, and Foot Process Effacement

Monitoring of nephrin^Y3F/Y3F mice revealed that, by 1 month of age on the CD-1 background, most male animals showed significant albuminuria, which progressed markedly with age, as seen by both Coomassie blue staining after SDS-PAGE (Figure 3A) and measurement of the albumin-to-creatinine ratio (ACR) (Figure 3B). Interestingly, a subset of heterozygous nephrin^WT/Y3F mice also showed signs of proteinuria by 4 months of age (Supplemental Figure 1A), suggesting that even partial reduction of nephrin tyrosine phosphorylation can induce glomerular damage over time.

Figure 3.

CD-1 nephrin^Y3F/Y3F mice develop proteinuria, GBM alterations, and foot process effacement. (A) Coomassie-stained gels of consecutive urine samples from nephrin^WT/WT and nephrin^Y3F/Y3F animals. (B) Quantification using ACR (n=4 per genotype). *P<0.05 by ANOVA. (C) Light microscopy analysis of periodic acid–Schiff-stained sections of WT and Y3F animals. Y3F animals show capillary loop dilation (arrows in 1, 2 and 6 months) and GBM spikes (inset in 4 months). (D) Scanning EM analysis of podocyte morphology in WT and Y3F animals. Y3F animals begin to show branched and disorganized foot processes (inset in 2 months), which progress to severe foot process disorganization and effacement (asterisk at 6 months). Scale bars, 20 μm in C; 2 μm in D.

Histologic examination of glomeruli from nephrin^WT/WT and nephrin^Y3F/Y3F mice by light microscopy was performed at four time points: at the earliest sign of proteinuria (1 month), at the onset of moderate proteinuria (2 months), after the progression to severe proteinuria (4 months), and finally, after a period of chronic proteinuria (6 months). At 1 month of age, we observed that glomeruli within nephrin^Y3F/Y3F mice contained enlarged capillary loops, which became much more extensive by 2 months of age (Figure 3C). At 4 months, there also appeared to be an irregular thickening of the GBM, and we observed pronounced GBM bulges or spikes in nephrin^Y3F/Y3F animals. By 6 months, glomeruli from nephrin^Y3F/Y3F were hypocellular, with numerous dilated capillary loops. Importantly, the mean arterial pressure was not different between nephrin^WT/WT and proteinuric nephrin^Y3F/Y3F mice at 4 months of age (Table 2), indicating that the vascular phenotype is not simply a result of increased BP. Furthermore, serum creatinine and BUN were not different between nephrin^WT/WT and nephrin^Y3F/Y3F mice ≤6 months of age, suggesting that the Y3F mutation did not affect on the GFR.

Table 2.

BPs of proteinuric male nephrin^Y3F/Y3F mice on different genetic backgrounds

	C57BL/6 10 mo		CD-1 4 mo
	WT, n=3	Y3F, n=3	WT/HET, n=4	Y3F, n=4
Weight, g	48.9±0.8	46.6±2.4	49.2±3.2	47.1±4.7
Systolic, mmHg	109±8	101±5	106±4	105±9
Diastolic, mmHg	74±7	70±2	80±3	78±7
Mean arterial pressure, mmHg	86±8	80±3	88±3	87±8
Heart rate, beats per min	498±24	510±20	571±43	543±56

Ultrastructural examination using scanning EM revealed differences in podocyte foot process architecture between nephrin^WT/WT and nephrin^Y3F/Y3F mice. At 1 and 2 months of age, we found that foot processes within nephrin^Y3F/Y3F mice adopted haphazard orientations, and they appeared short, thick, and wavy (Figure 3D). At 4 months of age, foot processes in nephrin^Y3F/Y3F mice showed signs of retraction, and by 6 months, nephrin^Y3F/Y3F mice displayed widespread podocyte foot process effacement.

Additional analysis using transmission EM and subsequent quantification of foot process widths revealed varying degrees of damage within glomeruli of individual nephrin^Y3F/Y3F animals ranging from podocytes that appeared almost normal to those where the majority of foot processes had undergone significant spreading, particularly as the animals aged (Figure 4, A and B). The GBM bulges were associated with regions of clearing, and they appeared in some cases to be detached from the main aspect of the GBM (Figure 4A). By 6 months, there was a diffuse thickening of most of the GBM, which showed an irregular scalloped contour. The nature of the GBM thickening in nephrin^Y3F/Y3F mice was further confirmed by silver staining of glomeruli at 6 months of age (Figure 4C). Notably, these GBM changes seemed to parallel the onset and progression of foot process alterations in nephrin^Y3F/Y3F mice as seen in Figure 3. Altogether, the morphologic alterations seen in the glomeruli of nephrin^Y3F/Y3F mice support the notion that nephrin tyrosine phosphorylation is required to maintain filtration barrier function.

Figure 4.

Progressive GBM abnormalities in CD-1 nephrin^Y3F/Y3F mice. (A) Transmission EM depicting changes in basement membrane morphology in nephrin^Y3F/Y3F animals compared with nephrin^WT/WT animals. At 1 and 2 months of age, there are focal regions of thickened GBM and evidence of splitting (arrows) in Y3F animals, and at 6 months, numerous subepithelial humps associated with extensive regions of effacement can be seen. (B) Box and whisker (5%–95%) plots of individual foot process widths measured via transmission EM (n=2–3 mice analyzed per group with a minimum of 50 measurements per mouse). **P<0.01 by ANOVA. (C) Light microscopy analysis of silver-stained sections from 6-month-old WT and Y3F mice. Diffuse light brown staining highlights the expanded GBM and membrane humps in Y3F mice (lower panel). Scale bars, 2 μm in A; 20 μm in C.

C57BL/6 Nephrin^Y3F/Y3F Mice Develop Mild Albuminuria and a Delayed Glomerular Phenotype

In parallel with our analysis of the nephrin-Y3F mutation on the CD-1 background, we also investigated the effects of reduced nephrin tyrosine phosphorylation on the inbred C57BL/6N background. C57BL/6 nephrin^Y3F/Y3F mice developed a milder and later onset phenotype than CD-1 nephrin^Y3F/Y3F mice, and the phenotype was more pronounced in male versus female mice, both of which are consistent with several reports describing strain- and sex-specific differences in the severity of renal disease in genetically modified mice.^24–30 C57BL/6 nephrin^Y3F/Y3F mice showed significant albuminuria at >6 months of age (Figure 5A).

Figure 5.

C57BL/6 nephrin^Y3F/Y3F mice develop mild proteinuria and delayed glomerular phenotype. (A) Quantification of urine samples from male nephrin^WT/WT and nephrin^Y3F/Y3F animals using ACR (n=3 per genotype). *P<0.05 by ANOVA. (B) Light microscopy analysis of periodic acid–Schiff-stained WT and Y3F animals at 6 and 11 months of age highlighting the presence of dilated capillary loops (arrows). (C). Scanning EMs depicting abnormal appearance of podocytes in Y3F animals, wherein foot processes appear short, broad, and curved (arrows) with significant regions of effacement at 11 months (arrowhead). (D) Transmission EM analysis of GBM morphology in WT and Y3F animals showing progressive segmental changes with regions of foot process preservation (arrowheads) adjacent to effacement (arrows) in addition to increased matrix deposition resulting in spike-like protrusions on the epithelial face of the GBM (^). (E) Box and whisker (5%–95%) plots of individual foot process widths measured via transmission EM of mice within the indicated age groups (n=2–4 mice analyzed per group with a minimum of 50 measurements per mouse). Scale bars, 20 μm in B; 2 μm in C and D. *P<0.05 by ANOVA; **P<0.01 by ANOVA.

Detailed analysis of the kidneys of C57BL/6 nephrin^WT/WT and nephrin^Y3F/Y3F animals was performed at 6 and 11 months of age. Similar to the glomerular phenotype observed on the CD-1 background, C57BL/6 nephrin^Y3F/Y3F mice showed histologic signs of gross capillary dilation, which worsened as the mice aged, when proteinaceous casts could also be observed in the tubules of some nephrin^Y3F/Y3F animals (Figure 5B) (data not shown). Average BP remained unchanged between C57BL/6 nephrin^WT/WT and proteinuric nephrin^Y3F/Y3F animals at 10 months of age (Table 2). Examination of podocyte ultrastructure using scanning EM revealed an array of foot process alterations in nephrin^Y3F/Y3F mice at 6 months of age, which by 11 months of age, appeared as broad, short, and disorganized projections with more extensive regions of effacement (Figure 5C). Transmission EM highlighted segmental regions of foot process effacement in nephrin^Y3F/Y3F mice at 6 months of age that were widespread by 11 months of age (Figure 5, D and E), and these regions tended to be associated with areas of increased GBM thickness (Figure 5D). Interestingly, GBM abnormalities were also observed in aged heterozygous C57BL/6 nephrin^WT/Y3F mice (Supplemental Figure 1, B and C), consistent with the heterozygous phenotype seen on the CD-1 background.

C57BL/6 Nephrin^Y3F/Y3F Mice Show Impaired Recovery from Podocyte Injury

The above findings clearly show the importance of nephrin tyrosine phosphorylation in preservation of glomerular function, and we reasoned that this might reflect a requirement for nephrin phosphorylation in controlling dynamic actin regulation and podocyte remodeling that occur throughout life in response to hemodynamic strain and other insults. To directly test this, we used two models of reversible foot process effacement within young C57BL/6 nephrin^Y3F/Y3F mice. We first used the protamine sulfate model, where perfusion of protamine sulfate causes rapid foot process effacement, which can be reversed through subsequent perfusion with heparin sulfate.^8,31 Kidneys within nephrin^WT/WT and nephrin^Y3F/Y3F mice were perfused with HBSS control buffer, protamine sulfate, or protamine sulfate followed by heparin sulfate, fixed, and analyzed by EM. Compared with buffer-treated animals, mice from both genotypes perfused with protamine sulfate showed extensive foot process spreading (Figure 6, A and B). Interestingly, however, although podocyte foot process architecture was restored to baseline after heparin sulfate treatment in nephrin^WT/WT animals, this recovery was impaired in the majority of nephrin^Y3F/Y3F glomeruli analyzed.

Figure 6.

C57BL/6 nephrin^Y3F/Y3F mice display delayed recovery in podocyte injury models. (A) Transmission EM of foot processes in nephrin^WT/WT and nephrin^Y3F/Y3F mice perfused with HBSS, protamine sulfate (PS), or protamine sulfate followed by heparin sulfate (PS/HS). PS induces foot process spreading in both genotypes. A proportion of foot processes is restored with HS in Y3F mice (region 1), although the majority remains spread (region 2, arrows). Images are representative of two mice of each genotype per condition. Scale bar, 2 μm. (B) Box and whisker (5%–95%) plots of individual foot process widths measured via transmission EM (n=2 kidneys analyzed per treatment with a minimum of 60 measurements per mouse). Comparisons were made between treatments as indicated. (C) Quantification of total urinary protein normalized to urine creatinine shows that Y3F mice have an enhanced response to LPS–induced podocyte injury at 24 and 36 hours compared with control (Y3F: n=8–13 per time point; WT: n=4–6 per time point except at 60 hours, where n=3). (D) Quantification using ACR at 0 and 24 hours (n=4–5 per genotype). *P<0.05 by ANOVA; **P<0.01 by ANOVA.

We next examined the LPS model of acute injury, whereby LPS induces transient mild albuminuria within 24 hours after injection, which correlates with foot process effacement.³¹ Nephrin^WT/WT and nephrin^Y3F/Y3F mice were injected with PBS or LPS, and spot urine samples were collected at regular intervals ≤72 hours after injection. Urine protein was increased at 24 hours post-LPS injection in nephrin^Y3F/Y3F mice compared with control mice, and although proteinuria started to decline >24 hours in both genotypes, nephrin^Y3F/Y3F mice still showed elevated levels at 36 hours post-injection (Figure 6C). Quantitation of the urinary ACRs in mice before (0 hours) and 24 hours after injection showed that LPS treatment induced a significant increase in albuminuria in nephrin^Y3F/Y3F mice compared with nephrin^WT/WT mice and that injection of PBS alone did not alter urinary protein levels in either genotype (Figure 6D). Collectively these findings are consistent with a defect in foot process restoration in nephrin^Y3F/Y3F mice.

Nephrin Tyrosine Phosphorylation and Binding to Nck Are Altered during Podocyte Effacement

The recovery defect in nephrin^Y3F/Y3F mice prompted us to explore the dynamics of nephrin signaling that occur coincident with foot process actin remodeling in these injury models. Glomeruli were isolated from WT mice exposed to protamine sulfate or LPS challenge and immunoblotted with phosphospecific nephrin antibodies recognizing Y1176/Y1193 and Y1217.¹⁶ Within the protamine sulfate model, we observed a significant increase in nephrin phosphorylation on Y1176/Y1193 after protamine sulfate perfusion, which decreased after heparin sulfate recovery (Figure 7, A and B). Relative to protamine sulfate–treated mice, Y1176/Y1193 was weakly phosphorylated in glomeruli perfused with buffer alone. By contrast, Y1217 was highly phosphorylated with buffer alone and after heparin sulfate treatment, and a modest decrease in phosphorylation on this tyrosine was observed after protamine sulfate perfusion (Figure 7, A and C). Strikingly, this transient decrease in phosphorylation of Y1217 after protamine sulfate was paralleled by reduced interaction between nephrin and Nck, which was likewise restored after heparin sulfate treatment (Figure 7A). Similar findings were obtained within the LPS model. At 6 hours post-LPS injection, which precedes marked proteinuria, we detected a significant decrease in nephrin phosphorylation on Y1217 and reduced association of nephrin with Nck, both of which were restored by 24 hours when proteinuria peaks and subsequently resolves (Figure 7, D and F). No changes in Y1176/Y1193 were detected at either time point analyzed (Figure 7, D and E). These results further support a role for nephrin phosphotyrosine–based signaling pathways in preservation and restoration of foot process architecture and filtration barrier function in the face of podocyte injury.

Figure 7.

Immunoblot (IB) analysis of nephrin tyrosine phosphorylation and Nck binding in glomeruli isolated from WT mice exposed to podocyte injury. (A) Nephrin tyrosine phosphorylation and nephrin/Nck coimmunoprecipitation (IP) after injection with HBSS, protamine sulfate (PS), or protamine sulfate followed by heparin sulfate (PS/HS). Interaction of nephrin with Nck is decreased on PS perfusion. (B and C) Densitometric quantitation of results in A shows a significant increase in nephrin tyrosine phosphorylation on Y1176/1193 but not Y1217 in PS-treated samples. (D) Nephrin tyrosine phosphorylation and nephrin/Nck IP 6 and 24 hours after injection with PBS or LPS. Interaction of nephrin with Nck is decreased at 6 hours after LPS injection. (E and F) Densitometric quantitation of results in D shows a significant decrease in nephrin tyrosine phosphorylation on Y1217 but not Y1176/1193 at 6 hours after LPS injection. The fold change in nephrin phosphorylation was normalized to control samples (i.e., HBSS or PBS). Mice were analyzed individually, and results are representative of three to five mice per treatment. *P<0.05; **P<0.01.

Discussion

The potential for the cytoplasmic tail of nephrin to be regulated by tyrosine phosphorylation has been recognized since the initial characterization of nephrin.^6,32 Given the clinical relevance of alterations in nephrin expression and function, numerous efforts have been made to determine the role of this phosphorylation. Our nephrin^Y3F knockin mouse model now provides unequivocal evidence that phosphonephrin signaling does not seem to be involved in the initial assembly of foot processes but is, instead, critical for stabilization of podocyte cytoarchitecture and restoration after injury. Our results further highlight the central role of nephrin-Nck interaction in regulation of foot process morphology.

The progressive phenotype of nephrin^Y3F/Y3F mice highlights the linkage between the slit diaphragm, the podocyte-GBM adhesion complex, and the cytoskeletal backbone of foot processes. The actin cytoskeleton is proposed to serve as a mechanotransduction apparatus to facilitate tight coregulation of podocyte contractility and GBM adhesion in response to fluctuating hemodynamic forces during filtration.^33,34 Crucial intermediates that link the slit diaphragm and GBM together with the cytoskeletal backbone are the PINCH-ILK-parvin complex and α-actinin-4.^35,36 Of note, Nck can complex with PINCH-ILK,³⁷ and it plays a direct role in cell adhesion and spreading.^38,39 In the nephrin-Y3F mutant, which cannot bind Nck, integration of the slit diaphragm with the mechanotransduction machinery seems to be compromised but incompletely severed. As such, tethering of the slit diaphragm to the actin cytoskeleton via mechanisms other than phosphonephrin-Nck1/2 interactions could explain why congenital nephrosis does not manifest in these mice (as it does in Nck-deficient animals⁹). Nevertheless, although this tenuous integration may be sufficient for early podocyte morphogenesis, the weakened linkage between the slit diaphragm, actin cytoskeleton, and GBM adhesion complex in the absence of phosphonephrin signaling renders podocytes more vulnerable and less adaptable to increased hemodynamic stress in mature animals, thereby leading to foot process effacement and albuminuria. The enhanced response to LPS-induced damage in nephrin^Y3F/Y3F mice further supports this notion. With time, the loss of podocyte contractile tone and adhesiveness causes the GBM to characteristically protrude at its external aspect closest to podocytes, and similarly, the capillary loops become more prone to gross dilation.³³ The bulging of the GBM and ballooning of the glomerular tuft are among the most striking pathologies in nephrin^Y3F/Y3F mice, and analogous structures have been reported in mice lacking other podocyte proteins essential for binding to actin or the GBM.^29,40–44

An additional characteristic of nephrin^Y3F/Y3F animals is the disorganized patterning of foot processes. Process formation in podocytes shares biologic features of that within neurons, wherein common cytoskeletal elements and signaling proteins are used to generate elaborate cell–based projections.⁴⁵ Intriguingly, mice deficient in Nck expression in neurons display axon guidance defects.⁴⁶ Although conditional deletion of Nck in podocytes results in profound congenital effacement⁹ (which may reflect the involvement of Nck in alternate pathways involving key structural proteins, such as β1-integrin⁴⁷), it is important to note that foot process formation still occurs, even in the complete absence of nephrin.^48,49 We, therefore, suggest that Nck has a comparable role in pathfinding downstream of phosphonephrin during the maturation of podocyte foot processes and that this altered patterning further compromises the stability of the slit diaphragm in nephrin^Y3F/Y3F mice.

The spectrum of glomerular pathologies associated with the nephrin-Y3F mutation is portrayed on both the CD-1 and C57BL/6 backgrounds, despite the differences in onset and severity. C57BL/6 mice have lower BPs than age–matched CD-1 mice,⁵⁰ which may reflect reduced glomerular strain, although this is unlikely to be the sole determinant, and identification of modifier alleles within these backgrounds remains an area of intense interest. It is noteworthy that the nephrin-Y3F protein seems to have a dominant negative effect in the heterozygous state, and this may be relevant in patients with nephrotic syndrome who carry only a single nephrin mutation and those who present outside the neonatal period.⁵¹ Beyond the kidney, nephrin is expressed and functional in several tissues, including the brain, testis, and pancreas.^48,52 However, nephrin^Y3F/Y3F mice are fertile, and they do not display any overt changes in locomotor activity or coordination (data not shown), implying a unique requirement for nephrin phosphorylation in the mature podocyte.

The delayed phenotype of nephrin^Y3F/Y3F mice on the C57BL/6 background allowed us to use these animals to explore the dynamics and overall necessity of nephrin tyrosine phosphorylation in experimental models of effacement, proteinuria, and recovery, which mimic the podocyte injury associated with acquired glomerulopathies. Nephrin^Y3F/Y3F mice undergo foot process spreading on protamine sulfate perfusion and enhanced LPS–induced proteinuria, clearly showing that tyrosine phosphorylation of nephrin on these sites is not required to actively promote effacement in these models.⁵³ In concordance with our findings, a number of recent works has established a nephrin-independent pathway resulting in Rac activation as an important pathomechanism to drive effacement.^13,54,55 Instead, we find that foot process recovery is impaired in nephrin^Y3F/Y3F animals in both injury models, similar to that observed in mice lacking the actin regulators synaptopodin and cofilin.^56,57 Taken together, our data support a model wherein nephrin tyrosine phosphorylation is required to counteract normal podocyte stress and loss of this phosphorylation, as seen in nephrin^Y3F/Y3F mice and some human diseases, leads to irreversible injury and chronic disease.

The injury phases in the disease models studied herein are coincident with dephosphorylation of nephrin Y1217 and uncoupling from Nck as well as decreased pS473Akt in the LPS model.²⁰ Reductions in nephrin phosphorylation, Nck/p85 binding, and Akt activation have also been detected in the puromycin aminonucleoside model of reversible podocyte injury^12,14,16,18 in parallel with increased expression of the phosphatase PTP1B, which can dephosphorylate nephrin Y1217.⁵⁸ By contrast, phosphorylation of nephrin on Y1176/Y1193 is increased with protamine sulfate as noted previously.^7,8 However, this change does not occur in the LPS model, and it does not seem to be essential for effacement; thus, the significance of this site remains to be elucidated. We posit that differential phosphorylation on particular tyrosines may facilitate preferential recruitment of specific signaling proteins to nephrin, such as Nck, as well as p85/PI3K, β-arrestin, and Cas/Crk, which in turn, control distinct processes that regulate foot process remodeling. In the context of Nck, YDxV phosphopeptides show similar binding affinities for the isolated SH2 domains⁵⁹; however, Y1217 is found within a YDQV motif in contrast to the YDEV motif of both Y1176 and Y1193, and Nck seems to have the highest affinity for Y1217 in vivo.⁹ In summary, because induced loss of Nck in podocytes results in rapid demise of the filtration barrier¹⁶ and the inability to induce phosphorylation on Y1217 and subsequent Nck binding in nephrin^Y3F/Y3F mice impairs the rapid actin remodeling required for reformation of foot processes, we refine the existing model to include Nck as a critical signaling axis between nephrin Y1217 and the actin cytoskeleton to maintain podocyte plasticity. Establishment of this causal relationship supports the development of novel therapeutic approaches for human proteinuric kidney diseases that selectively preserve and/or enhance nephrin phosphorylation and thereby, directly target the podocyte actin cytoskeleton.

Concise Methods

Generation of Nephrin^Y3F Mice

To generate nephrin^WT/Y3F mice (official designation B6;129Nphs1^<Tm1.1>/Njns), three A→T nucleotide substitutions were engineered into a construct containing the region from exon 24 to exon 29 of Nphs1, resulting in Y→F missense mutations at p.Y1191F, p.Y1208F, and p.Y1232F. Homologous recombination of the targeting vector into C57BL/6Jx129S6SvEv ES cells produced two independent clones. We initially examined nephrin^Y3F/Y3F mice derived from both clones and saw similar phenotypes (data not shown), and therefore, we selected one of the clones for additional study. All of the animals used in this study were derived from that clone. After verification of germline transmission, mice were crossed to B6.129S1 Hprt-Cre mice to remove the floxed PGK-neomycin selection cassette. The resulting offspring were crossed once to C57BL/6J and subsequently maintained on the C57BL/6N background (475; Charles River Canada).

Mice were genotyped by PCR using Nphs1^Y3F-FW (5′-GCATATGTGAACGCATGAGG-3′) and Nphs1^Y3F-RV (5′-GAAGGTGGTTGGTTGCAGTT-3′) primers. PCR cycling conditions used were 95°C for 5 minutes and then 35 cycles of 94°C (30 seconds), 59°C (30 seconds), and 72°C (45 seconds) followed by a final extension at 72°C for 5 minutes. DNA products were examined on 2% agarose gels. B6;129 nephrin^Y3F/Y3F mice were bred onto the CD-1 (472; Charles River Canada) background for six generations to generate CD-1.B6;129Nphs1^<Tm1.1>/Njns mice before intercrossing heterozygous animals. Data include results from animals between generations 6 and 8.

There were similar histologic findings in male and female nephrin^Y3F/Y3F mice, and data from both sexes are represented in this study. Because the onset of proteinuria was generally earlier in males, male mice were used for all urine analyses, with the exception of the LPS study. The age and sex of mice used in specific injury models are reported in the description of each model.

Animals were housed on a 12-hour/12-hour artificial light-dark cycle. They had free access to standard chow (2014; Harlan Teklad) and drinking water.

Evaluation of Proteinuria

Mice were placed into a metabolic cage until they urinated freely. Urine samples (2 or 5 μl) were diluted in 1× SDS sample buffer, separated by 10% SDS-PAGE, and stained with Coomassie brilliant blue R. The urinary ACR was determined using the Albuwell M ELISA (Exocell) and Creatinine Companion (Exocell) Kits according to the manufacturer’s instructions. For time course recovery studies, the total urinary protein-to-creatinine ratio (in micrograms per milligram) was determined using standard protocols on the basis of a modified Bradford assay⁶⁰ and the Jaffe reaction.

Histologic and Ultrastructural Analyses

For histology, kidneys were halved and fixed in 10% buffered formalin overnight before being embedded in paraffin; 4-μm sections were cut and stained with periodic acid–Schiff or periodic acid–ammoniacal silver⁶¹ to visualize the basement membrane. Slides were viewed on a Leica DM100 (Leica Microsystems, Buffalo Grove, IL). Images were prepared for presentation using Adobe Photoshop CS5 (Adobe Systems, Inc., San Jose, CA).

For EM, sagittal slices (scanning EM) and small pieces (transmission EM) of kidney tissue were fixed in 0.1 M sodium cacodylate buffer containing 4% paraformaldehyde and 2% gluteraldehyde (Electron Microscopy Sciences), postfixed in 1% OsO₄, and dehydrated through graded ethanols. Transmission EM samples were embedded in Quetol–Spurr resin. Ultrathin sections were cut and stained with uranyl acetate and lead citrate and viewed using an FEI CM100 TEM. For scanning EM, samples were critical point dried and sputter coated with gold. Samples were viewed using an FEI XL30 SEM or Hitachi S-540 SEM (Hitachi, Yokohama, Japan).

LPS Injection

Nonproteinuric female C57BL/6 nephrin^WT/WT and nephrin^Y3F/Y3F mice between the ages of 3 and 6 months were used for this experiment. Mice were injected intraperitoneally with 200 μg LPS (1 mg/ml in PBS; L2630; Sigma-Aldrich, St. Louis, MO) or an equal volume of PBS. Spot urine samples were collected before injection, at 24 hours after injection, and for recovery analysis, every 12 hours until 72 hours after injection. For biochemical analyses, mice were euthanized with CO₂ at 6 and 24 hours after injection; kidneys were removed and then, immediately processed for glomerular isolation.

Protamine Sulfate Injury Model

Nonproteinuric male and female C57BL/6 nephrin^WT/WT and nephrin^Y3F/Y3F mice ages 8–12 weeks old were used in this model. All perfusion solutions were maintained at 37°C in a water bath. Mice were anesthetized with a 2%:98% mix of isoflurane:oxygen and kept warm with a lamp.

The abdomen was opened, and a 27-gauge tube was inserted into the abdominal aorta just below the renal arteries. The inferior vena cava was then nicked below the renal veins, and 10 ml warmed HBSS was infused at a rate of 5 ml/min by hand. The kidneys were then perfused at a rate of 4.5 ml/min using a Pump 11 Elite Infusion Syringe Pump (Harvard Apparatus, Holliston, MA) via the abdominal aorta serially with HBSS for 2 minutes, protamine sulfate (2 mg/ml; Sigma-Aldrich) for 15 minutes, HBSS for 2 minutes, and heparin sulfate (800 μg/ml) for 15 minutes. Kidneys were immediately processed for glomerular isolation or placed in appropriate fixative to be further processed for EM. Foot process widths were measured using ImageJ.

Antibodies

The commercial antibodies used were as follows: mouse anti-GAPDH (G041; Applied Biologic Materials Inc.), guinea pig anti-nephrin (20R-NP002; Fitzgerald Inc.), rabbit anti–pS473 Akt (4060; Cell Signaling Technology, Danvers, MA), rabbit anti-Akt (4691; Cell Signaling Technology), rabbit anti–pY416 (active) Src (2101; Cell Signaling Technology), rabbit anti–pY1217 human nephrin (also recognizes mouse pY1232; 2423–1; Epitomics), rabbit anti–pY1193 human nephrin (also recognizes mouse pY1208),¹⁶ mouse anti–β-actin (A5441; Sigma-Aldrich), rabbit anti-podocin (P0372; Sigma-Aldrich), and rabbit anti-Src (Ab-529; Abcam, Inc., Cambridge, MA). Rabbit antinephrin was described previously.¹⁷ Rabbit anti-Nck⁶² was a gift from Louise Larose. Secondary antibodies for immunofluorescence from Invitrogen (Carlsbad, CA) were goat anti–rabbit Alexa Fluor 488 (A11008) and goat anti–guinea pig Alexa Fluor 594 (A11076). Secondary antibodies for immunoblotting from Bio-Rad (Hercules, CA) were goat anti–mouse HRP (170–6516) and goat anti–rabbit HRP (170–6515).

Glomerular Isolation and Lysis

Glomeruli were isolated from mice of the indicated genotypes (one kidney cortex from each of two mice per group) by differential sieving, pelleted, and resuspended in 600 μl RIPA lysis buffer of 50 mM Tris (pH 7.5), 150 mM NaCl, 10% glycerol, 1% NP-40, 0.25% Na-deoxycholate, 0.1% SDS, and 1 mM EDTA supplemented with fresh inhibitors. Alternatively, glomeruli were isolated from minced kidney cortex incubated in 1 mg/ml type 4 collagenase (Worthington) in a 37°C water bath with slight agitation for 30 minutes. Digested tissue was passed through a 100-μm nylon sieve (Thermo Fisher Scientific, Vernon Hills, IL) and washed with ice-cold PBS. Blood was removed by brief incubation in ack lysis buffer of 150 mM NH₄Cl, 10 mM KHCO₃, and 0.1 mM EDTA, and glomeruli were pelleted by spinning at 350×g for 1 minute. Glomeruli were then resuspended in cold PLC lysis buffer of 10% glycerol, 50 mM Hepes, 150 mM NaCl, 1.5 mM MgCl₂, 1 mM EGTA, 10 mM NaPPi, 100 mM NaF, and 1% Triton X-100 supplemented with fresh inhibitors, sonicated for 10 seconds on ice, incubated on ice for 5 minutes, and then, centrifuged at 14,000×g for 10 minutes at 4°C. Protein concentrations were determined using a BCA Protein Assay (Pierce, Rockford, IL). Supernatants were mixed with appropriate amounts of 5× SDS sample buffer and incubated at 100°C for 5 minutes.

Immunoprecipitation

For immunoprecipitation of Nck, protein A beads (Invitrogen) were first conjugated to Nck antibodies to avoid interference from IgG heavy chain during immunoblotting as described elsewhere⁶³ with the following alterations: 0.5 μl antibody was incubated per 100 μl beads in 500 μl PBS for 1 hour at room temperature. Excess unbound antibody was removed by washing three times in PBS followed by spinning at 3000×g for 3 minutes; 100 μl 1.3 mg/ml DSS (Thermo Fisher Scientific) in DMSO was added to crosslink the antibody and beads in 500 μl PBS and left for an additional 1 hour at room temperature. Conjugated beads were successively washed four times in TBS (0.02 M Tris base and 0.15 M NaCl) and an additional four times in 100 mM glycine to remove unconjugated antibody. Conjugated beads were also washed four times in TBS before use; 100 μl conjugated anti-Nck beads were incubated with 2 mg glomerular lysate for 5 hours at 4°C.

Immunoblotting

Protein samples were resolved on 8% or 12% SDS-PAGE gels with 100 μg protein loaded as total lysate. Proteins were transferred to PVDF membrane (EMD Millipore, Billerica, MA) and blocked in 5% skim milk or 5% BSA in 1× TBST. Membranes were incubated with primary antibodies overnight at 4°C: rabbit: anti-Nck (1:1000), anti-nephrin (1:1000), anti-pY1217 (1:2500), anti-pY1193 (1:2500), anti-pY1176/1193 (1:2500), anti-pS473 Akt (1:2000), anti-Akt (1:1000), anti-pY416 Src (1:1000), and anti-Src (1:1000); mouse: anti-GAPDH (1:1000) and anti–β-actin (1:2000). After washing, membranes were incubated with HRP–conjugated secondary antibody diluted 1:10,000 in TBST for 1 hour at room temperature. Signals were detected using ECL (Pierce) or Luminata Crescendo (EMD Millipore). Blots were imaged using a ChemiDoc XRS+ (Bio-Rad) or exposed to film (Pierce). Densitometry was performed using ImageLab version 2.0 analysis software (Bio-Rad).

Indirect Immunofluorescence of Tissue Sections

Kidneys were flash frozen in Cryomatrix (Fisher Scientific, Waltham, MA); 6-μm sections were dried at room temperature for 10 minutes, fixed, and permeabilized in acetone at −20°C for 10 minutes. When using the phosphonephrin antibodies, phosSTOP Tablets (Roche, Basel, Switzerland) were added to all solutions. All subsequent steps were carried out at room temperature. Slides were blocked for 1 hour in 10% goat serum and then, incubated with primary antibodies for 1 hour (guinea pig anti-nephrin [1:100], rabbit anti–pY1217-nephrin [1:50], or anti-podocin [1:100]). After three washes in PBS, slides were incubated with secondary antibodies for 1 hour (goat anti–rabbit Alexa 488 and goat anti–guinea pig Alexa 594; both 1:400). Slides were washed and mounted using Prolong Gold Antifade Mounting Medium (Invitrogen). Epifluoresence images were obtained using Volocity software version 5.3.2 (Improvision) on a DMIRE2 Microscope (Leica Microsystems) using a 63× oil immersion objective. Stacks were captured at 0.2-μm z intervals, and then, deconvolved using an iterative restoration function (95% confidence interval with 15 iterations) in Volocity.

Quantitative Real–Time PCR

For mRNA isolation, kidney cortex was dissected from nephrin^WT/WT (n=3) and nephrin^Y3F/Y3F (n=3) mice, immediately flash frozen, and stored at −80°C. mRNA was isolated using TriZol (Invitrogen), and mRNA concentrations were measured on a nanodrop (Thermo Fisher Scientific) after DNase I (Invitrogen) treatment. RNA integrity was verified on a Bioanalzyer (Agilent Technologies, Santa Clara, CA); all samples used for analysis had an RIN value of at least seven. cDNA synthesis was performed on 1 μg RNA with the Superscript II Kit (Invitrogen) using random primers. Quantitative RT-PCR reactions were carried out using SsoFast EvaGreen Supermix (Bio-Rad) on a CFX96 Real–Time PCR Detection System (Bio-Rad). Primer sequences used in this study are listed in Table 3. Standard curves were generated for all primer pairs used, and melt curves were examined to verify that there was a single amplicon. No reverse transcription and no template controls were included in all runs. Expression differences were calculated using the Δ^Ct method corrected for primer efficiency. Differences were normalized to the reference genes Gapdh and Hprt. All calculations were performed in CFX Manager 3.0 (Bio-Rad).

Table 3.

Primer sequences used for real–time PCR analysis

Gene	Forward Primer (5′-3′)	Reverse Primer (5′-3′)	Size, bp
Actn4	ATGGTCTCCGACATCAACAA	TCATTCAGCAGCCATTCTTC	77
Nphs1	GGAAGAGGTGTCATATCGCC	CTCAAAGGGCAGAGAACCAG	250
Nphs2	TCCATGAGGTGGTAACCAAA	CTTTGGACACATGGGCTAGA	111
Wt1	ACGTCCTTTCATGTGTGCAT	TTTCTCACCAGTGTGCTTCC	94
Gapdh	AGGTCGGTGTGAACGGATTTG	GGGGTCGTTGATGGCAACA	95
Hprt	CATGGACTGATTATGGACAGGACTG	ATCCAGCAGGTCAGCAAAGAACT	126

BP Measurements

Mice were anesthetized with a 2%:98% mix of isoflurane:oxygen. Body temperature was maintained at 37°C using a battery–operated rectal thermometer combined with a 100-W heating lamp. A small incision was made, and a 1.2-F catheter (FTS-1211B-0018; Scisense Inc.) was inserted into the right carotid artery, through the aortic valve, and into the left ventricle. Pressures were digitized at a sampling rate of 2000 Hz and recorded using iWorx analytic software (Labscribe2, Dover, NH).

Statistical Analyses

Values are presented as means±SEMs. Differences between two groups were analyzed by t test, and differences between more than two groups were analyzed by ANOVA using SAS version 9.4 (SAS Institute Inc., Cary, NC). Statistical analyses of quantitative RT-PCR data were performed in CFX Manager software (Bio-Rad). Graphs were prepared using Graphpad Prism version 5.0 (GraphPad Software, La Jolla, CA). P<0.05 was considered statistically significant.

Study Approval

Animal studies were approved by the University of Guelph Animal Care Committee and carried out in accordance with Canadian Council on Animal Care protocols.

Disclosures

None.