Diabetes Page 2 of 51
Dissection of glomerular transcriptional profile in patients with diabetic nephropathy:
SRGAP2a protects podocyte structure and function
Yu Pan1#, Song Jiang1#, Qing Hou1#, Dandan Qiu1, Jingsong Shi1, Ling Wang1,Zhaohong Chen1, Mingchao
Zhang1, Aiping Duan1, Weisong Qin1, Ke Zen2*, Zhihong Liu1*
1National Clinical Research Center of Kidney Diseases, Jinling Hospital, Nanjing University School of
Medicine, Nanjing, Jiangsu 210002, China; 2State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing, Jiangsu 210093, China.
Running title: SRGAP2a protects podocyte structure and function
#These authors contributed equally to this work.
*Corresponding authors:
Zhihong Liu, M.D.
National Clinical Research Center of Kidney Diseases,
Jinling Hospital, Nanjing University School of Medicine
Nanjing, Jiangsu 210002, China
Email: [email protected]
Tel.:86 025 84801992
Ke Zen, Ph.D.
State Key Laboratory of Pharmaceutical Biotechnology,
School of Life Sciences, Nanjing University,
Nanjing, Jiangsu 210093, China.
Email: [email protected]
Diabetes Publish Ahead of Print, published online December 14, 2017 Page 3 of 51 Diabetes
Abstract
Podocytes play a pivotal role in maintaining glomerular filtration function through their interdigitated foot
processes. However, the mechanisms that govern the podocyte cytoskeletal rearrangement still remain unclear.
Through analyzing transcriptional profile of renal biopsy from diabetic nephropathy (DN) patients and control
donors, we identify Slit Robo GTP activating protein 2a (SRGAP2a) as one of the main ‘hub’ genes that are
strongly associated with proteinuria and glomerular filtration in type 2 DN patients. Immunofluorescnce
staining and western blot analysis reveal that human and mouse SRGAP2a is primarily localized at podocytes
and largely co localized with synaptopodin. Moreover, podocyte SRGAP2a is downregulated in DN patients
and db/db mice at both mRNA and protein level. SRGAP2a reduction is also observed in cultured podocytes
treated with TGF β or high concentration of glucose. Functional and mechanistic studies show that SRGAP2a
suppresses podocyte motility through inactivating RhoA/Cdc42 but not Rac1. The protective role of
SRGAP2a in podocyte function is also confirmed in zebrafish, in which knockdown of SRGAP2a, a SRGAP2
ortholog in zebrafish, recapitulated podocyte foot process effacement. Finally, increasing podocyte SRGAP2a
level in db/db mice via administration of adenovirus expressing SRGAP2a significantly mitigates podocyte
injury and proteinuria. Our results demonstrate that SRGAP2a protects podocytes via suppressing podocyte
migration.
Keywords: SRGAP2a, transcriptome, proteinuria, podocyte, diabetic nephropathy
2
Diabetes Page 4 of 51
Introduction
Diabetic nephropathy (DN) is one of the leading causes of chronic kidney diseases (CKD) (1 3).
Hypothesis driven studies based on the DN animal model had led to many insights about the development and progression of DN; however, the findings may not to be analogous to human patients. In the last decade, the transcriptomics studies in DN utilizing broad based approach with human kidney tissue component had been performed to uncover the pathogenesis mechanisms of DN (4 6). Several previous studies had revealed different expressed genes, the upstream regulatory factors and enriched signaling pathways altered in DN.
However, the causative effects of these genes and signal pathways in DN have not been well characterized.
Further integrating gene expression data with matching clinical features and identifying possible gene network responsible for the phenotype under disease conditions is necessary.
It has been well documented that glomerular podocytes play a pivotal role in the pathogenesis of diabetic nephropathy (DN) (7; 8). Podocyte depletion and loss is generally found in the early stages of DN (9; 10), and such podocyte injuries are accompanied by gradual decline of glomerular filtration rate (GFR) (11; 12) and the initiation of proteinuria (13). As terminally differentiated cells residing on the outer surface of the glomerular basement membrane (GBM), podocytes play a critical role in maintaining the structure and function of the glomerular filtration barrier through their interdigitated foot processes. This specialized function of podocytes depends on their unique cytoarchitecture, particularly the interdigitating foot like actin rich processes that arise from podocyte cell bodies and surround glomerular capillary walls. At the interface of the interdigitating foot processes and the capillary wall, the unique junction, also termed the slit diaphragm, allows ultrafiltration of serum. To withstand high pressure in the capillaries and to maintain intact and exact filtration properties, the podocyte must possess a dynamic contractile apparatus, and precisely arrange their cytoskeleton spatially and temporally (14). The mechanisms that govern the dynamic arrangement of podocyte cytoskeleton, however, remain unclear.
Accumulating evidence suggests that actin filaments, associated with a unique assembly of linker and adaptor molecules (15), are the predominant cytoskeletal components of podocyte foot processes (16). Besides acting as a scaffold for sub membrane protein complexes, the cortical actin cytoskeleton also provides a tensile 3
Page 5 of 51 Diabetes
architectural support for podocyte cellular extensions. When podocytes undergo foot process spreading and
retraction via remodeling their cytoskeletal architecture and intercellular junctions, abnormal filter barrier
function can occur (17). Buvall et al. have reported that EGFR/Src mediated tyrosine phosphorylation of actin
organizing protein synaptopodin in podocytes promotes binding to the serine/threonine phosphatase
calcineurin, leading to the enhanced Rac1 signaling and ultimate loss of stress fibers in podocytes (18). Other
important molecular switches that regulate the podocyte actin cytoskeleton are Rho family GTPases, the
prototypical members of which are Cdc42, Rac1 and RhoA (19 23). Previous studies showed that aberrant
Rho GTPase signaling was associated with podocyte mobility, leading to proteinuria (24; 25). As a member of
SRGAPs (26), SRGAP2 belongs to the large family of Rho GTPase. It has been shown to display distinct
expression patterns in the central nervous system where it regulates neuronal cell migration (27 29). However,
although SRGAP2 was recently predicted as an enriched protein in podocytes (30), little information about
expression pattern and function of SRGAP2 in the kidney is available. It remains unclear how SRGAP2
modulates the integrity of podocyte actin cytoskeleton when podocytes undergo foot process effacement under
diabetic condition.
In the present study, we employed multiple strategies to reveal the key molecule(s) that change the dynamics
of podocyte cytoskeleton under diabetic condition. By analyzing the gene co expression network and its
association with baseline proteinuria and estimated glomerular filtration rate (eGFR), we have demonstrated
that SRGAP2a, an important component of the Slit/Robo signaling pathway during neuronal development, is
primarily located at podocytes and functions as a central ‘hub’ gene that is tightly associated with the
proteinuria and eGFR in DN patients. Utilizing both in vitro and in vivo systems, we have characterized the
protective role of SRGAP2a in podocyte structure and function. Our results show that podocyte SRGAP2a is
downregulated in DN patients and db/db mice, while increasing podocyte SRGAP2a level in db/db mice can
reverse podocyte cytoskeleton arrangement and thus mitigate the podocyte injury and proteinuria.
Furthermore, the mechanistic studies demonstrate that podocyte SRGAP2a maintains the normal structure and
function of podocytes through suppressing RhoA/Cdc42 activities.
4
Diabetes Page 6 of 51
Materials and Methods
Diabetic nephropathy patient enrollment
41 patients diagnosed with type 2 DN by renal biopsy at National Clinical Research Center of Kidney
Diseases, Jinling Hospital, Nanjing University were enrolled. The clinical characteristics of DN were detailed in Table S1. 20 healthy control glomerular samples were obtained from surgical nephrectomies. The protocol about using human samples was approved by the Human Subjects Committee of Jinling Hospital, Nanjing
University (2013KLY 013 01) and signed consent form was obtained from each patient and control donor.
Glomerulus genome wide gene expression profiling and gene network/function analysis
Microdissection of glomeruli was performed at 4°C. The isolated glomeruli were subject to RNA extraction, followed by cDNA synthesis and qPCR assay (Qiagen, Valencia, CA). Genome wide gene expression profiling was performed using the Affymetrix® microarray platform (HTA 2.0). For identifying expression pattern of different groups, WGCNA was utilized to cluster co expressed genes (Gene module). KEGG pathway analysis (KEGGEST and GeneAnswers Package in R) was performed to identify enriched pathways in gene module. A weighted gene co expression network analysis was constructed using WGCNA in R language (31; 32).
Murine model
The use of animals was approved by the Institutional Animal Care and Use Committee at Jinling Hospital.
The db/db diabetic mice in C57BL6 background and littermate db/m mice were obtained from the Jackson
Laboratory. Body weight and fasting blood glucose levels were monitored weekly. Mouse urinary albumin and creatinine levels were measured using Albuwell M (Exocell Laboratories) and Creatinine Companion Kits
(BioAssay).
SRGAP2a expressing adenovirus
To investigate the effect of SRGAP2a in mouse kidney, mice were transfected with aSRGAP2a expressing adenovirus (Ad SRGAP2a GFP). An adenovirus expressing GFP (Ad GFP) served as a Mock control.
Ad SRGAP2a GFP and Ad GFP were purchased from HanBio (Shanghai, China). Briefly, 50µl of adenovirus 5
Page 7 of 51 Diabetes
(~1011 PFU/ml) expressing GFP alone (Mock) or SRGAP2a GFP were injected into mice via tail vein.
Podocyte count, stable transfection with lentivirus based SRGAP2a shRNAs or SRGAP2a R527A
Podocytes of human and mice were counted as previously described (33). Human podocytes (34) were
initially cultured in RPMI 1640 medium containing 10% FBS and Insulin Transferrin Selenium (ITS) (Gibco)
at 33℃, and then at 37℃ for 10 14 days to allow cell differentiation. To knockdown endogenous srGAP2,
lentivirus shRNA srGAP2a was used to transfect podocyte. The paired oligonucleotides targeting human
srGAP2a gene was synthesized and annealed into pHBLV U6 Scramble ZsGreen Puro vector via digestion
sites of BamH I and EcoR I. Proliferative podocytes (33°C) stably transfected with SRGAP2a shRNA (> 90%
proliferative podocytes were SRGAP2a shRNA/GFP positive) was generate through multiple rounds of
selection against puromycin treatment. Proliferative podocytes were then induced to differentiated podocytes
at 37°C. The primers and for shRNA SRGAP2a and SRGAP2a R527A were listed in Table S2.
Pull down assay for small GTPases activities and immunoprecipitation
RhoA, Rac1 and Cdc42 activities were determined by measuring Rhoketin or PAK1 pulled down by GTP Rho
GTP Rac1 and GTP Cdc42, respectively (35). The small GTPases were separated on 12% SDS PAGE
following the manufacturer’s instructions (BK030, Cytoskeleton Inc. Denver, CO). Purified His tagged
RhoA/Cdc42/Rac1 was also from BK030 kit. Protein A/G agarose (Santa Cruz), His tagged Dynabead
(Invitrogen) and anti SRGAP2a antibodies (Ab121977, Abcam) were used in co immunoprecipitation assay.
Statistical Analysis
Data were presented as the mean ± SD. Comparisons between groups were made using a two tailed unpaired
Student’s t test or one way ANOVA with Bonferroni’s posthoc test. Mann Whitney nonparametric U test was
used to analyze data in abnormal distribution. P<0.05 was considered statistically significant. GraphPad Prism
statistical software version 6 (GraphPad Software, Inc.) and SPSS statistical software version 22 (SPSS, Inc.)
was used for data analysis.
6
Diabetes Page 8 of 51
Result
Reduction of SRGAP2a accounts for proteinuria and the aberrant eGFR observed in DN patients
Considering that animal models of DN do not completely mimic the histological and functional changes of human DN patients (36; 37), we performed the present study directly using kidney tissues from DN patients.
To identify key genes associated with podocyte injury and proteinuria in DN patients, a sequential strategy from transcriptomic analysis to validation study was used (Figure S1). First, we performed genome wide gene expression profiling using the Affymetrix® microarray platform to determine differentially expressed genes in glomeruli between DN patients and control donors (GEO database, the accession number: GSE96804).
Weighted gene co expression network analysis (WGCNA) of glomeruli identified 18 gene co expression modules (Figure S2, A and B). Among these gene co expression modules, the turquoise module, which includes 1810 transcribed genes, exhibited the highest correlation with proteinuria (R= 0.79, P=10 13) and baseline eGFR (R=0.63, P=10 7) (Figure S2C). Gene function analysis further indicated that the differential expression of genes in the turquoise module significantly involved in cytoskeletal protein binding
(P=4.37×10 13) and cytoskeleton structure (P=9.29×10 12) (Table S3). As shown, turquoise module contained
30 “hub genes”, and among these genes, the majority was involved in cell cytoskeletal organization (Figure
S2D). By analyzing gene enriched pathway, we also found that proteins involved in the Axon Guidance Signal pathway were associated with genes encoded by the turquoise module (Figure S2E). The analysis identified that SRGAP2 was one of the “hub genes” that had the strongest association with baseline proteinuria (R=
0.81, P=9.61×10 15) and eGFR (R=0.58, P=1.60×10 6) in DN patients (Table S4), suggesting that the absence of SRGAP2 mediates aberrant renal function in diabetic patients. As shown in Figure 1, whole genome gene expression profiling of glomeruli from DN patients and control donors revealed differential expression of
SRGAP family proteins in the glomerular from DN patients (Figure 1A). The qPCR analysis of glomeruli from DN patients (N=20) and db/db mice (N=6) showed a significant downregulation of SRGAP2 in the glomerular from both DN patients and db/db mice (Figure 1B). The level of SRGAPs was tightly correlated with proteinuria (g/24h), plasma creatinine (mg/dl) and eGFR declining (ml/min/1.73m2/year) (Figure 1C). To preclude the possibility that the SRGAP2a reduction was due to podocyte loss in these patients, we measured the SRGAP2a mRNA levels between DN patients and normal controls after adjusting the podocyte density.
The level of SRGAP2a mRNA level had 31% reduction (P<0.05) in glomeruli of DN patients compared with 7
Page 9 of 51 Diabetes
normal controls after normalizing against the podocytes number (Figure 1D). Our observation of negative
correlation between the levels of SRGAP2 and proteinuria in Chinese DN patients is in agreement with
NEPHROSEQ (http://www.nephroseq.org) data, which showed SRGAP2 downregulation in glomeruli of
Woroniecka diabetes and Ju CKD (Figure S3).
SRGAP2, a GTPase activating protein, was originally identified in neurons where it regulates cell migration
and neurite outgrowth (26; 38). However, SRGAP2 expression is not limited to neuronal cells. The expression
of SRGAP2 has been predicted to express in podocytes (30). Based on our transcriptomic analysis results and
previous findings, we postulated that SRGAP2 is a key regulator of the podocyte cytoskeleton and is essential
for glomerular filtration function. Given that human SRGAP2 contains three paralogs, SRGAP2a, SRGAP2b
and SRGAP2c, we further analyzed which SRGAP2 paralog(s) are associated with development of proteinuria
in DN patients. In agreement with previous report by Higgins Normal Tissue Panel
(http://www.nephroseq.org), our microarray result showed expression of SRGAP2a and SRGAP2c but not
SRGAP2b in human glomeruli. This result excluded the role of SRGAP2b in regulating human glomerular
function. Moreover, we found no significant reduction of SRGAP2c in glomerular from DN patients compare
with control glomerular (Figure S4). Structural analysis also revealed that SRGAP2c only contain F bar
domain but no Rho GAP and SH3 domain (39). Based on these findings, the present study only focused on the
expression and function of SRGAP2a.
Podocyte specific SRGAP2a was downregulated in DN patients and db/db mice
To explore the function of SRGAP2a protein, we first determined where SRGAP2a expression is located in
the kidney. Immunofluorescence staining of SRGAP2a in human kidney tissue revealed that SRGAP2a was
localized in the glomeruli, and co localized with synaptopodin but not the glomerular basement membrane
marker collagen IV (Figure 2, A and B), suggesting that SRGAP2a is a podocyte specific protein. The
enrichment of SRGAP2a in glomeruli was also confirmed in mice. By trapping magnetic particles within the
glomeruli followed by magnetic separation to isolate glomeruli (40), we obtained glomeruli from 8 weeks
control and diabetic db/db mice with purity more than 95%. Western blot results also showed that SRGAP2a
was predominantly expressed in the isolated glomerular fraction (Figure S5). Next, we determined srGPA2a 8
Diabetes Page 10 of 51
level in DN patients. In agreement with microarray data, the reduction of SRGPA2a in glomeruli of DN patients was confirmed by RT qPCR assay using the samples from an independent group of DN patients
(Figure 2D).
Downregulation of podocyte SRGAP2a by hyperglycemia or TGF β1 disrupted podocyte cytoskeleton
Given that both hyperglycemia and TGFβ1 contribute to the development of DN (41), we next determined whether these factors might be involved in downregulating podocyte SRGAP2a. As shown in Figure 3, A and
B, after treatment with high concentration of glucose (30mM) or TGF β1 (5ng/ml) for 24 h, Podocyte cytoskeletons staining by fluorescein conjugated phalloidin showed that high concentration of glucose or
TGF β1 significantly disrupted podocyte cytoskeleton in a time dependent manner. The same treatment also significantly reduced the SRGAP2a expression in a similar time dependent manner (Figure 3C). To test whether stable expression of SRGAP2a can rescue podocyte cytoskeleton disruption by TGF β1 or high concentration of glucose, we examined the fluorescein conjugated phalloidin staining in podocytes stably transfected with Adenovirus expressing SRGAP2a or Mock. As shown in Figure 3D F, although overexpression of SRGAP2a did not significantly alter the distribution of F actin stress fiber in untreated podocytes, SRGAP2a overexpression largely rescued the disruption of F actin stress fiber in podocytes caused by 24 h treatment with TGF β1 or high concentration of glucose.
The role of SRGAP2a in controlling podocyte migration
To determine the effect of SRGAP2a downregulation on podocyte function, we knocked down SRGAP2a in cultured podocytes using lentivirus delivered shRNA (Figure 4A). As shown in Figure 4, B and E, the number of stress fibers in podocytes was significantly reduced by SRGAP2a knockdown. We next examined the effect of SRGAP2a knockdown on podocyte migration using wound closure assay. As shown in Figure 4, C and D, when confluent monolayers of differentiated podocytes were scratched to initiate wound healing process, podocytes with SRGAP2a knockdown showed a significantly accelerated wound closure kinetics compared to untreated podocytes. The migration of podocytes was remarkably enhanced by SRGAP2a knockdown (Figure
4F). These results suggest that SRGAP2a is involved in maintaining podocyte membrane stability.
9
Page 11 of 51 Diabetes
SRGAP2a inactivates RhoA/Cdc42 but not Rac1
Previous study has shown that SRGAP2 regulates the dynamics of the cytoskeleton/membranes through the
larger F BAR/RhoGAP/SH3 SRGAP module (27). Here we also sought to identify the small GTPases that
interact with SRGAP2a in podocytes. We performed proximity ligation assay to detect in situ protein protein
interactions of SRGAP2a with RhoA, Rac1 or Cdc42 in human renal biopsies. As shown in Figure S6, the
interactions of SRGAP2a with RhoA and Cdc42 but not Rac1 were detected in renal biopsies of both DN
patients and control groups. However, a marked reduction of SRGAP2a RhoA and SRGAP2a Cdc42 protein
complexes was observed in renal biopsies from DN patients compared to control group. These results
confirmed an impairment of SRGAP2a Rho GTPase interaction in renal biopsies from DN patients.
We next examined the interactions of SRGAP2a with RhoA and Cdc42 in cultured human podocytes. As
shown in Figure 5A and 5B, downregulation of SRGAP2a via LV SRGAP2a shRNA reduced the binding of
RhoA and Cdc42 with SRGAP2a, suggesting that the role of SRGAP2a in stabilizing podocyte cytoskeleton is
likely through its interaction with RhoA or Cdc42. The co immunoprecipitation assay in cultured podocyte
also confirmed such SRGAP2a RhoA and SRGAP2a Cdc42 interactions (Figure 5, C E). As expected,
SRGAP2a knockdown in cultured human podocytes increased the GTP bound forms of Cdc42 and RhoA but
not Rac1 (Figure 5, C and D). By contrast, SRGAP2a overexpression in podocytes decreased the GTP bound
forms of Cdc42 and RhoA. To further test the specific binding of SRGAP2a with Cdc42 and RhoA, we
mutated the potential Cdc42/RhoA binding site on SRGAP2a to generate SRGAP2a R527A mutant (42).
After overexpressing SRGAP2a R527A mutant in podocytes in which SRGAP2a was knocked down, we
found that SRGAP2a R527A displayed significantly less binding to RhoA and Cdc42 (Figure 5F). In
agreement with this, exogenous purified His RhoA and His Cdc42 also showed little interaction with
SRGAP2a R527A expressed in podocytes (Figure 5G). Taken together, the interactions of SRGAP2a with
RhoA/Cdc42 play a critical role in regulating podocyte cytoskeleton stability: when SRGAP2a is available to
bind to RhoA or Cdc42, these small GTPases remain in inactive states binding to GDP instead of GTP.
SRGAP2a knockdown in zebrafish leads to podocyte developmental defects
Zebrafish kidney shares many features with human kidney and thus studies of zebrafish pronephros 10
Diabetes Page 12 of 51
development and function can provide valuable knowledge for understanding human renal biology (43).
Therefore, morpholino oligonucleotide (MO) embryo injection for podocyte function study has been well recognized (44; 45). Although zebrafish have two SRGAP2 orthologs, SRGAP2a and SRGAP2b, only
SRGAP2a shares high homology with human SRGAP2a (Figure S7 8). The present study thus focused on zebrafish SRGAP2a function.
To reduce the cell death induced by off target effects of MO, p53 MO were co injected with SRGAP2a splicing MO targeting the exon1 and intron1 (e1i1). As shown in Figure 6, A C, we observed 85.3% totally pericardial, periorbital and body edema of zebrafish larvae at 3.5 days post fertilization (dpf), and injection of p53 MO alone did not result in any phenotype. Next, we performed TEM imaging of zebrafish pronephric glomeruli, comparing with p53 MO alone, knockdown of SRGAP2a lead to podocyte foot process effacement, disrupted slit diaphram and disorganized glomerular filtration barrier (Figure 6 D and F). We also investigated the knockdown effect of SRGAP2a in transgenic embryos Tg (pod:GFP). The confocal imaging showed that knockdown of SRGAP2a resulted in diminished expression of GFP (Figure 6E). Finally, to confirm the
SRGAP2a MO induced specific phenotype, zebrafish embryo were injected with SRGAP2a mRNA following the MO treatment. The result clearly showed that injection of SRGAP2a mRNA rescued the MO induced phenotype and reduced the total edema down to 45.1% (Figure 6G). The results show that loss of SRGAP2a results in zebrafish podocyte foot process effacement and glomerular filtration barrier (GFB) disruption.
Exogenous SRGAP2a attenuates podocyte injury in db/db mice
We next utilized db/db mice to explore the protective role of SRGAP2a in podocyte structure and function. In this experiment, the glomeruli were isolated from 6 20 weeks db/db mice. As shown in Figure 7, db/db mice after 16 weeks developed extracellular matrix deposition and massive foot process effacement (Figure 7, A and D), elevated blood glucose level and body weight (Figure 7B) and significant albuminuria (Figure 7C).
Immunofluorescence staining also showed that podocyte SRGAP2a expression was markedly decreased in db/db mice after 12 week old, followed by reduction of other podocyte specific proteins synaptopodin and
WT 1 around 20 week old (Figure 7A). Immunoblot analysis confirmed that reduction of SRGAP2a in glomeruli from db/db mice in 12 week old, which was significantly earlier than the reduction of synaptopodin 11
Page 13 of 51 Diabetes
and WT 1 (Figure 7E). Furthermore, the levels of GTP bound RhoA and Cdc42 in isolated glomeruli from
db/db mice were markedly increased after 16 week (Figure 7F).
To test whether stabilizing levels of podocyte SRGAP2a in db/db mice can protect mice from proteinuria, we
injected 12 week old db/db mice with adenovirus expressing SRGAP2a GFP (Ad SRGAP2a) (Figure 8A). As
shown in Figure 8B and Figure S9, SRGAP2a levels in the glomeruli and tubular cells of db/db mice was
significantly increased following injection with Ad SRGAP2a. The db/db mice treated with Ad SRGAP2a
exhibited a decreased foot process effacement (Figure 8C), extracellular matrix deposition (Figure 8D) and
albuminuria (Figure 8E) compared to the same db/db mice treated with adenovirus expressing GFP (Ad GFP).
The db/db mice administered with Ad SRGAP2a also displayed more stable synaptopodin expression and
more WT 1 positive cells in the kidney (Figure 8, F and G). Furthermore, the db/db mice administered with
Ad SRGAP2a can significantly inhibit the activities of RhoA and Cdc42 (Figure 8, H and I). These results
indicate that SRGAP2a can protect podocyte from foot process effacement and detachment, thus reducing
glomerular abnormalities including sclerosis in db/db mice.
12
Diabetes Page 14 of 51
Discussion
In the present study, we integrated renal biopsy derived transcriptional profiles from type 2 DN patients with matching clinical information to identify critical gene responsible for proteinuria and eGFR of DN. Through analyzing genome wide transcriptional profiles from type 2 DN patients and healthy control donors, as well as functional assays in both in vitro and in vivo systems, we identified SRGAP2a as a major player in protecting podocytes from foot process effacement under chronic diabetic conditions. Mechanistically, SRGAP2a decreases podocyte motility via suppressing Cdc42/RhoA activities.
Through comparing the renal transcriptional profiles between DN patients and control group, most previous studies have identified many genes that are differentially expressed during podocyte dysfunction (5; 6). As matching gene expression data with clinical features and identifying possible gene network responsible for the phenotype under disease conditions provides an ideal approach to reveal disease mechanisms and potential drug targets of DN, the present study has analyzed the gene networks that are responsible for the observed clinical features of DN patients. To achieve this, gene analysis based on genome wide transcriptional profiles from both DN patients and controls was first performed to obtain the gene co expression module and identified the ‘hub genes’ that serve as key connection for various signaling pathways. Next, we correlated expression or loss of these ‘hub’ genes with clinical features of DN patients particularly proteinuria and eGFR.
Through these analyses, we identified loss of expression of SRGAP2a as a pivotal factor in renal dysfunction in diabetic patients.
Our study showed that SRGAP2a is significantly downregulated in DN. Given that SRGAP2a knockdown resulted in an increase of podocyte motility and developmental defects, while exogenous SRGAP2a attenuated podocyte injury, our results strongly argue that SRGAP2a is a key molecule required for maintaining the proper function of podocyte. Using intravital and kidney slice two photon imaging of the three dimensional structure of mouse podocytes, Brahler et al. recently showed that normal podocytes remained non motile state in vivo but shifted to a dynamic state during foot process effacement (46). Our results of specific expression of
SRGAP2a at podocytes and its role in suppressing podocyte migration are in agreement with the notion that decrease the motility of podocyte is critical for maintaining normal podocyte function. 13
Page 15 of 51 Diabetes
The function of SRGAP2 has been extensively studied in neurons revealing that it interacts with members of
the Rho family small GTPases. Rho family of GTPases, particularly Cdc42, RhoA and Rac1, play a critical
role in dynamics of podocyte actin cytoskeleton (47). On one hand, deficiency of Cdc42 or RhoA resulted in
heavy proteinuria, foot process effacement and glomerulosclerosis, accompanied by podocyte cytoskeleton
injury (20; 22; 48). On the other hand, the exceeding activation of RhoA in adult podocytes facilitated foot
process effacement and proteinuria through disturbing actin cytoskeletons and focal adhesion (47; 49).
Therefore, tightly controlled Rho GTPase activities are essential for maintaining stable podocyte actin
cytoskeletons and normal podocyte function. Previous studies have also shown that podocytes could switch
from a RhoA dependent stable state to a Cdc42 and Rac1 dependent migratory state in response to stress (50).
In certain states, RhoA and Rac1 were mutually antagonistic (47; 51). To identify which small Rho GTPase
interacts with SRGAP2a in podocytes, we analyzed the interactions of SRGAP2a with RhoA, Rac1 and Cdc42
in human renal biopsies. Given that significant reduction of SRGAP2a RhoA and SRGAP2a Cdc42 protein
complexes was detected in renal biopsies from DN patients compared to that from control donors, it suggests
that SRGAP2a executes its function through modulating RhoA and Cdc42 activity. In line with this, we also
observed that knockdown of SRGAP2a increased the active forms of Cdc42 and RhoA but not Rac1 in
cultured human podocyte, while upregulating podocyte SRGAP2a in vitro also reduced the activity of RhoA
and Cdc42. Given that SRGAP2a could inhibit RhoA/Cdc42 activation and DN patients displayed an impaired
interaction of SRGAP2a with RhoA/Cdc42 in renal tissues, our results support the notion that, under the stress
conditions of hyperglytcemia and TGF β stimulation, in which podocyte RhoA and Cdc42 were over activated,
suppressing RhoA and Cdc42 activity by SRGAP2a is critical to stabilize podocyte cytoskeleton and maintain
normal podocyte function. In other words, loss of SRGAP2a in podocytes under such injury cues
(hyperglycemia or TGF β) would lead to a greater podocyte migration and renal injury.
Although our data showed that increasing SRGAP2a level protected podocyte F actin stress fiber from
disruption by hyperglycemia or TGF β, overexpression of SRGAP2a under normal condition did not enhance
the level of F actin stress fiber in untreated podocytes (Figure 3). This may suggest that SRGAP2a only
affects RhoA and Cdc42 that are over activated by hyperglycemia or TGF β. In other words, SRGAP2a plays
an important role in balancing the activity of Rho family of GTPases and thus stabilizing podocyte 14
Diabetes Page 16 of 51
cytoskeleton network. In contradiction with our result that podocyte SRGAP2a inactivates RhoA and Cdc42 under hyperglycemia condition or when cells are treated with TGF β, previous study by Guerrier et al. (28) showed that neuronal cell SRGAP2a inhibits the Rac1 activity but induces Cdc42 activity. Given that the alteration of cellular F actin by SRGAP2 is through its interaction with Rho family of GTPases, these results suggest that the interactions of SRGAP2a with Cdc42, RhoA or Rac1 and their functions may vary in different cell types and under different pathological conditions. Recent study by Fan and co workers showed that
SRGAP1, another SRGAP family member, was expressed on the basal surface of podocytes and promoted podocyte detachment through interacting with myosin II regulatory light chain (52; 53). Different from the function of SRGAP2a we observed in stabilizing podocyte F actin fiber, they showed that SRGAP1 actually decreased polymerization of F actin. There are two possible explanations for this discord between two
SRGAP members in term of podocyte F actin polymerization: First, SRGAP2a and SRGAP1 may bind to different Rho family of small GTPases under various conditions thus mediate different signal downstream.
For example, Coutinho Budd et al. showed that the F BAR domains from SRGAP1, SRGAP2 and SRGAP3 regulate cell membrane deformation differently (54). In addition, the study by Yamazaki et al. showed that, instead of inactivating Cdc42 or RhoA, SRGAP1 inactivates Rac1, and that depletion of SRGAP1 leads to
Rac1 over activation but RhoA inactivation (55). In combination with our observation that SRGAP2a inactivates RhoA and Cdc42 but not Rac1, these results may support the notion that SRGAP1 and SRGAP2 do play an opposite role in term of activating/inactivating Rho family of GTPases. Second, role of Robo2 signaling mediated by the same SRGAP protein in modulating F actin polymerization/de polymerization can be completely different in different biological setups, and this notion is supported by our data that rescue of
F actin disruption by overexpression of SRGAP2a only occurred in injured podocyte treated with hyperglycemia or TGF β instead of normal podocytes.
In summary, our studies have identified SRGAP2a mediated Slit Robo signal pathway as a novel mechanism regulating the adhesion and motility of podocytes. Through this signal pathway, SRGAP2a inactivates RhoA and Cdc42 in podocytes and thus decreases podocyte motility and foot process effacement. Given that podocyte SRGAP2a level was reduced in DN patients and enhancing podocyte SRGAP2a expression could rescue the impairment of podocyte function in db/db mice, our findings suggest podocyte SRGAP2a as 15
Page 17 of 51 Diabetes
potential therapeutic target in mitigating podocyte injury and proteinuria in DN patients.
16
Diabetes Page 18 of 51
ACKNOWLEDGMENTS
We would like to thank Jill Littrell (Georgia State University) for critical reading and editing the manuscript.
This work is supported by grants from National Key Research and Development Program of China
(2016YFC0904100), National Key Technology R&D Program (2015BAI12B02), Major International
(Regional) Joint Research Project (81320108007), Key Research and Development Program of Jiangsu
Province (BE2016747) and National Natural Science Foundation of China (No. 81500548, 81500556 and
81300652).
Author Contributions: Y.P., S.J., K.Z., Z.L., designed the study; S.J., Y.P., J.S. participated in data bioinformatics analysis; S.J., Q.H., D.Q., J.S., L.W., Z.C., M.Z. and A.D. performed the experiments; S.J., Y.P., participated in figure preparation; Y. P., S.J., K.Z., Z.L., wrote the manuscript. Z.L is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final version of manuscript.
CONFLICTS OF INTEREST
The authors declare no conflicts of interest.
17
Page 19 of 51 Diabetes
References
1. Fineberg D, Jandeleit-Dahm KA, Cooper ME: Diabetic nephropathy: diagnosis and treatment. Nat Rev Endocrinol 2013;9:713-723 2. Reutens AT, Atkins RC: Epidemiology of diabetic nephropathy. Contrib Nephrol 2011;170:1-7 3. Molitch ME, DeFronzo RA, Franz MJ, Keane WF, Mogensen CE, Parving HH, Steffes MW, American Diabetes A: Nephropathy in diabetes. Diabetes Care 2004;27 Suppl 1:S79-83 4. Hodgin JB, Nair V, Zhang H, Randolph A, Harris RC, Nelson RG, Weil EJ, Cavalcoli JD, Patel JM, Brosius FC, 3rd, Kretzler M: Identification of cross-species shared transcriptional networks of diabetic nephropathy in human and mouse glomeruli. Diabetes 2013;62:299-308 5. Baelde HJ, Eikmans M, Doran PP, Lappin DW, de Heer E, Bruijn JA: Gene expression profiling in glomeruli from human kidneys with diabetic nephropathy. Am J Kidney Dis 2004;43:636-650 6. Woroniecka KI, Park AS, Mohtat D, Thomas DB, Pullman JM, Susztak K: Transcriptome analysis of human diabetic kidney disease. Diabetes 2011;60:2354-2369 7. Wolf G, Chen S, Ziyadeh FN: From the periphery of the glomerular capillary wall toward the center of disease: podocyte injury comes of age in diabetic nephropathy. Diabetes 2005;54:1626-1634 8. Li JJ, Kwak SJ, Jung DS, Kim JJ, Yoo TH, Ryu DR, Han SH, Choi HY, Lee JE, Moon SJ, Kim DK, Han DS, Kang SW: Podocyte biology in diabetic nephropathy. Kidney Int Suppl 2007:S36-42 9. Pagtalunan ME, Miller PL, Jumping-Eagle S, Nelson RG, Myers BD, Rennke HG, Coplon NS, Sun L, Meyer TW: Podocyte loss and progressive glomerular injury in type II diabetes. The Journal of clinical investigation 1997;99:342-348 10. White KE, Bilous RW, Marshall SM, El Nahas M, Remuzzi G, Piras G, De Cosmo S, Viberti G: Podocyte number in normotensive type 1 diabetic patients with albuminuria. Diabetes 2002;51:3083-3089 11. Berthier CC, Zhang H, Schin M, Henger A, Nelson RG, Yee B, Boucherot A, Neusser MA, Cohen CD, Carter-Su C, Argetsinger LS, Rastaldi MP, Brosius FC, Kretzler M: Enhanced expression of Janus kinase-signal transducer and activator of transcription pathway members in human diabetic nephropathy. Diabetes 2009;58:469-477 12. Susztak K, Bottinger EP: Diabetic nephropathy: a frontier for personalized medicine. Journal of the American Society of Nephrology : JASN 2006;17:361-367 13. Wharram BL, Goyal M, Wiggins JE, Sanden SK, Hussain S, Filipiak WE, Saunders TL, Dysko RC, Kohno K, Holzman LB, Wiggins RC: Podocyte depletion causes glomerulosclerosis: diphtheria toxin-induced podocyte depletion in rats expressing human diphtheria toxin receptor transgene. Journal of the American Society of Nephrology : JASN 2005;16:2941-2952 14. Welsh GI, Saleem MA: The podocyte cytoskeleton--key to a functioning glomeruli in health and disease. Nat Rev Nephrol 2011;8:14-21 15. Takeda T: Podocyte cytoskeleton is connected to the integral membrane protein podocalyxin through Na+/H+-exchanger regulatory factor 2 and ezrin. Clin Exp Nephrol 2003;7:260-269 16. Pavenstadt H, Kriz W, Kretzler M: Cell biology of the glomerular podocyte. Physiol Rev 2003;83:253-307 17. George B, Verma R, Soofi AA, Garg P, Zhang J, Park TJ, Giardino L, Ryzhova L, Johnstone DB, Wong H, Nihalani D, Salant DJ, Hanks SK, Curran T, Rastaldi MP, Holzman LB: Crk1/2-dependent signaling is necessary for podocyte foot process spreading in mouse models of glomerular disease. The Journal of clinical investigation 2012;122:674-692 18. Buvall L, Wallentin H, Sieber J, Andreeva S, Choi HY, Mundel P, Greka A: Synaptopodin Is a Coincidence Detector of Tyrosine versus Serine/Threonine Phosphorylation for the Modulation of Rho Protein Crosstalk in Podocytes. Journal of the American Society of Nephrology : JASN 2017;28:837-851 19. Burridge K, Wennerberg K: Rho and Rac take center stage. Cell 2004;116:167-179 20. Scott RP, Hawley SP, Ruston J, Du J, Brakebusch C, Jones N, Pawson T: Podocyte-specific loss of Cdc42 leads to congenital nephropathy. Journal of the American Society of Nephrology : JASN 2012;23:1149-1154 18
Diabetes Page 20 of 51
21. Gee HY, Saisawat P, Ashraf S, Hurd TW, Vega-Warner V, Fang H, Beck BB, Gribouval O, Zhou W, Diaz KA, Natarajan S, Wiggins RC, Lovric S, Chernin G, Schoeb DS, Ovunc B, Frishberg Y, Soliman NA, Fathy HM, Goebel H, Hoefele J, Weber LT, Innis JW, Faul C, Han Z, Washburn J, Antignac C, Levy S, Otto EA, Hildebrandt F: ARHGDIA mutations cause nephrotic syndrome via defective RHO GTPase signaling. The Journal of clinical investigation 2013;123:3243-3253 22. Blattner SM, Hodgin JB, Nishio M, Wylie SA, Saha J, Soofi AA, Vining C, Randolph A, Herbach N, Wanke R, Atkins KB, Gyung Kang H, Henger A, Brakebusch C, Holzman LB, Kretzler M: Divergent functions of the Rho GTPases Rac1 and Cdc42 in podocyte injury. Kidney Int 2013;84:920-930 23. Asanuma K, Yanagida-Asanuma E, Faul C, Tomino Y, Kim K, Mundel P: Synaptopodin orchestrates actin organization and cell motility via regulation of RhoA signalling. Nat Cell Biol 2006;8:485-491 24. Mouawad F, Tsui H, Takano T: Role of Rho-GTPases and their regulatory proteins in glomerular podocyte function. Can J Physiol Pharmacol 2013;91:773-782 25. Scott RP, Quaggin SE: Review series: The cell biology of renal filtration. J Cell Biol 2015;209:199-210 26. Wong K, Ren XR, Huang YZ, Xie Y, Liu G, Saito H, Tang H, Wen L, Brady-Kalnay SM, Mei L, Wu JY, Xiong WC, Rao Y: Signal transduction in neuronal migration: roles of GTPase activating proteins and the small GTPase Cdc42 in the Slit-Robo pathway. Cell 2001;107:209-221 27. Guez-Haddad J, Sporny M, Sasson Y, Gevorkyan-Airapetov L, Lahav-Mankovski N, Margulies D, Radzimanowski J, Opatowsky Y: The Neuronal Migration Factor srGAP2 Achieves Specificity in Ligand Binding through a Two-Component Molecular Mechanism. Structure 2015;23:1989-2000 28. Guerrier S, Coutinho-Budd J, Sassa T, Gresset A, Jordan NV, Chen K, Jin WL, Frost A, Polleux F: The F-BAR domain of srGAP2 induces membrane protrusions required for neuronal migration and morphogenesis. Cell 2009;138:990-1004 29. Bacon C, Endris V, Rappold G: Dynamic expression of the Slit-Robo GTPase activating protein genes during development of the murine nervous system. J Comp Neurol 2009;513:224-236 30. Ju W, Greene CS, Eichinger F, Nair V, Hodgin JB, Bitzer M, Lee YS, Zhu Q, Kehata M, Li M, Jiang S, Rastaldi MP, Cohen CD, Troyanskaya OG, Kretzler M: Defining cell-type specificity at the transcriptional level in human disease. Genome Res 2013;23:1862-1873 31. Zhang B, Horvath S: A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol 2005;4:Article17 32. Langfelder P, Horvath S: WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008;9:559 33. Venkatareddy M, Wang S, Yang Y, Patel S, Wickman L, Nishizono R, Chowdhury M, Hodgin J, Wiggins PA, Wiggins RC: Estimating podocyte number and density using a single histologic section. Journal of the American Society of Nephrology : JASN 2014;25:1118-1129 34. Saleem MA, O'Hare MJ, Reiser J, Coward RJ, Inward CD, Farren T, Xing CY, Ni L, Mathieson PW, Mundel P: A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression. Journal of the American Society of Nephrology : JASN 2002;13:630-638 35. Garcia-Mata R, Wennerberg K, Arthur WT, Noren NK, Ellerbroek SM, Burridge K: Analysis of activated GAPs and GEFs in cell lysates. Methods in enzymology 2006;406:425-437 36. Betz B, Conway BR: An Update on the Use of Animal Models in Diabetic Nephropathy Research. Curr Diab Rep 2016;16:18 37. Brosius FC, 3rd, Alpers CE, Bottinger EP, Breyer MD, Coffman TM, Gurley SB, Harris RC, Kakoki M, Kretzler M, Leiter EH, Levi M, McIndoe RA, Sharma K, Smithies O, Susztak K, Takahashi N, Takahashi T, Animal Models of Diabetic Complications C: Mouse models of diabetic nephropathy. Journal of the American Society of Nephrology : JASN 2009;20:2503-2512 38. Charrier C, Joshi K, Coutinho-Budd J, Kim JE, Lambert N, de Marchena J, Jin WL, Vanderhaeghen P, Ghosh A, Sassa T,
19
Page 21 of 51 Diabetes
Polleux F: Inhibition of SRGAP2 function by its human-specific paralogs induces neoteny during spine maturation. Cell 2012;149:923-935 39. Sporny M, Guez-Haddad J, Kreusch A, Shakartzi S, Neznansky A, Cross A, Isupov MN, Qualmann B, Kessels MM, Opatowsky Y: Structural History of Human SRGAP2 Proteins. Molecular biology and evolution 2017;34:1463-1478 40. Takemoto M, Asker N, Gerhardt H, Lundkvist A, Johansson BR, Saito Y, Betsholtz C: A new method for large scale isolation of kidney glomeruli from mice. Am J Pathol 2002;161:799-805 41. Herman-Edelstein M, Thomas MC, Thallas-Bonke V, Saleem M, Cooper ME, Kantharidis P: Dedifferentiation of immortalized human podocytes in response to transforming growth factor-beta: a model for diabetic podocytopathy. Diabetes 2011;60:1779-1788 42. Ma Y, Mi YJ, Dai YK, Fu HL, Cui DX, Jin WL: The inverse F-BAR domain protein srGAP2 acts through srGAP3 to modulate neuronal differentiation and neurite outgrowth of mouse neuroblastoma cells. PloS one 2013;8:e57865 43. Drummond IA, Davidson AJ: Zebrafish kidney development. Methods Cell Biol 2016;134:391-429 44. Gbadegesin RA, Hall G, Adeyemo A, Hanke N, Tossidou I, Burchette J, Wu G, Homstad A, Sparks MA, Gomez J, Jiang R, Alonso A, Lavin P, Conlon P, Korstanje R, Stander MC, Shamsan G, Barua M, Spurney R, Singhal PC, Kopp JB, Haller H, Howell D, Pollak MR, Shaw AS, Schiffer M, Winn MP: Mutations in the gene that encodes the F-actin binding protein anillin cause FSGS. Journal of the American Society of Nephrology : JASN 2014;25:1991-2002 45. Potla U, Ni J, Vadaparampil J, Yang G, Leventhal JS, Campbell KN, Chuang PY, Morozov A, He JC, D'Agati VD, Klotman PE, Kaufman L: Podocyte-specific RAP1GAP expression contributes to focal segmental glomerulosclerosis-associated glomerular injury. The Journal of clinical investigation 2014;124:1757-1769 46. Brahler S, Yu H, Suleiman H, Krishnan GM, Saunders BT, Kopp JB, Miner JH, Zinselmeyer BH, Shaw AS: Intravital and Kidney Slice Imaging of Podocyte Membrane Dynamics. Journal of the American Society of Nephrology : JASN 2016;27:3285-3290 47. Wang L, Ellis MJ, Gomez JA, Eisner W, Fennell W, Howell DN, Ruiz P, Fields TA, Spurney RF: Mechanisms of the proteinuria induced by Rho GTPases. Kidney Int 2012;81:1075-1085 48. Huang Z, Zhang L, Chen Y, Zhang H, Zhang Q, Li R, Ma J, Li Z, Yu C, Lai Y, Lin T, Zhao X, Zhang B, Ye Z, Liu S, Wang W, Liang X, Liao R, Shi W: Cdc42 deficiency induces podocyte apoptosis by inhibiting the Nwasp/stress fibers/YAP pathway. Cell death & disease 2016;7:e2142 49. Zhu L, Jiang R, Aoudjit L, Jones N, Takano T: Activation of RhoA in podocytes induces focal segmental glomerulosclerosis. Journal of the American Society of Nephrology : JASN 2011;22:1621-1630 50. Mundel P, Reiser J: Proteinuria: an enzymatic disease of the podocyte? Kidney Int 2010;77:571-580 51. Akilesh S, Suleiman H, Yu H, Stander MC, Lavin P, Gbadegesin R, Antignac C, Pollak M, Kopp JB, Winn MP, Shaw AS: Arhgap24 inactivates Rac1 in mouse podocytes, and a mutant form is associated with familial focal segmental glomerulosclerosis. The Journal of clinical investigation 2011;121:4127-4137 52. Fan X, Li Q, Pisarek-Horowitz A, Rasouly HM, Wang X, Bonegio RG, Wang H, McLaughlin M, Mangos S, Kalluri R, Holzman LB, Drummond IA, Brown D, Salant DJ, Lu W: Inhibitory effects of Robo2 on nephrin: a crosstalk between positive and negative signals regulating podocyte structure. Cell Rep 2012;2:52-61 53. Fan X, Yang H, Kumar S, Tumelty KE, Pisarek-Horowitz A, Rasouly HM, Sharma R, Chan S, Tyminski E, Shamashkin M, Belghasem M, Henderson JM, Coyle AJ, Salant DJ, Berasi SP, Lu W: SLIT2/ROBO2 signaling pathway inhibits nonmuscle myosin IIA activity and destabilizes kidney podocyte adhesion. JCI Insight 2016;1:e86934 54. Coutinho-Budd J, Ghukasyan V, Zylka MJ, Polleux F: The F-BAR domains from srGAP1, srGAP2 and srGAP3 regulate membrane deformation differently. Journal of cell science 2012;125:3390-3401 55. Yamazaki D, Itoh T, Miki H, Takenawa T: srGAP1 regulates lamellipodial dynamics and cell migratory behavior by modulating Rac1 activity. Molecular biology of the cell 2013;24:3393-3405
20
Diabetes Page 22 of 51
Figure legends
Figure 1. Transcriptomic analysis of glomeruli from DN patients and control donors. A) Whole genome gene expression profile revealed SRGAP family proteins in the glomerular from DN patients. B) The qPCR analysis of glomeruli from healthy control (h CLT) and DN patients (h DN), db/m (m db/m) and db/db mice
(m db/db). C) The correlation between SRGAPs and the key clinical features including proteinuria (g/24h), plasma creatinine (mg/dl), eGFR (ml/min/1.73m2) and eGFR declining (ml/min/1.73m2 /year). D) SRGAP2 mRNA level in DN patients with or without correction against podocyte density. Results in panel B and D were presented as mean ± SD. *, P <0.05, **, P < 0.01.
Figure 2. Decrease of SRGAP2a level in the glomeruli from diabetic nephropathy patients. A) IF staining demonstrated the protein levels of SRGAP2 (green) and its co localization with synaptopodin (red) in podocytes and downregulation of podocyte SRGAP2a in the glomeruli from DN patients. B) IF staining demonstrated the protein levels of SRGAP2a (red) was not co localization with type IV collagen (Col IV, green) in human glomeruli. C) Quantification of results in (A) by the ratio of integrated optical density (IOD) to area. D) The qPCR analysis of isolated glomeruli demonstrated the significant decrease of SRGAP2 mRNA level in DN patients. Results are presented as mean ± SD. Scale bar, 50 m. *, P <0.05
Figure 3. Exogenous SRGAP2a protected podocyte cytoskeletal damage induced by high concentration of glucose or TGFβ. A) Fluorescein conjugated phalloidin staining of podocyte after treatment with TGFβ1 (5 ng/ml) or high concentration of glucose (30 mM) for 0, 6, 12 and 24 h. B) Quantification of results in panel A.
C) Immunoblotting of SRGAP2a in the podocyte after TGFβ1 or high glucose treatment for the indicated time.
D) Immunoblotting of SRGAP2a overexpression in podocytes after transfecting with the Adenovirus expressing SRGAP2a. E) Fluorescein conjugated phalloidin staining of stably transfected Mock or SRGAP2a podocyte after treatment with TGFβ1 (5 ng/ml) or high concentration of glucose (30 mM) 24 h. (F)
Quantification of results in (E). Data were presented as the mean ± SD from three independent experiments.
*P <0.05, ** P < 0.01.
21
Page 23 of 51 Diabetes
Figure 4. SRGAP2a knockdown re arranged podocyte cytoskeleton and increased podocyte motility. A)
SRGAP2a level in scramble or SRGAP2a shRNA treated podocytes (scramble or shRNA SRGAP2a). B)
Podocytes with SRGAP2a knockdown had decreased number of stress fibers compared to podocytes treated
with vehicle or scramble shRNA). C) Podocytes with SRGAP2a knockdown migrated faster than podocytes
infected with vehicle or scramble shRNA. The images were recorded immediately (0 h) and at 24 h after
scratch. D) Quantification of the cell migration area. N=20 areas from three independent experiments. E)
Quantification of the stress fibers in panel B. ** P<0.01. F) Migration assay using the xCELLigence system
(detailed in Methods). The shRNA SRGAP2a podocytes displayed a significantly higher migratory capacity
(red).
Figure 5. SRGAP2a knockdown impaired SRGAP2a RhoA and SRGAP2a Cdc42 interaction in
cultured podocytes. A) Representative images showing protein complexes (red) of SRGAP2a RhoA,
SRGAP2a Cdc42 and SRGAP2a Rac in SRGAP2a knockdown differentiated podocytes (LV shRNA
SRGAP2) and the control knockdown differentiated podocytes (scramble). The data were analyzed by
proximity ligation assay (PLA). Scale bar, 20 m. B) Quantification of data in panel A. *, P < 0.05. C)
Immunoprecipitation of SRGAP2a bound Rho GTPases. SRGAP2a knockdown abrogated the interaction of
SRGAP2a with RhoA and Cdc42. D) Active GTP bound forms of RhoA, Rac1 and Cdc42 precipitated from
SRGAP2a knockdown and overexpression podocyte using a GST pulldown assay. Cells transfected with
SRGAP2 shRNA resulted in an increased RhoA and Cdc42 but not Rac1 activity compared with scramble
cells. LV shRNA SRGAP2a blocked the RhoA and Cdc42 activity stimulated by high concentration of glucose
or TGF β. E) Quantification of results in panel D. **, P < 0.01. F) Immunoprecipitation of SRGAP2a bound
Rho GTPases. Note that SRGAP2a R527A does not bind to RhoA and Cdc42. G) Immunoprecipitation of
podocyte SRGAP2a or SRGAP2a R527A by purified His Rho GTPases. Note that SRGAP2a R527A does not
interact with ectogenous RhoA/Cdc42.
Figure 6. Functional analysis of SRGAP2a in zebrafish. (A) Exon structure of SRGAP2a around binding
sites of SRGAP2a splice morpholino (i1e1). Primers for reverse transcription–PCR are indicated (arrow). (B)
RT PCR (Lift) of SRGAP2a mRNA from SRGAP2a i1e1 MO injected 3dpf larvae yields extra bands (red 22
Diabetes Page 24 of 51
arrows) in addition to a reduced band from SRGAP2a wild type (WT) mRNA (black arrow). WB (Right) of
SRGAP2a in glomerular from SRGAP2a i1e1 MO injected. (C) Zebrafish co injected with SRGAP2a and p53
MOs displayed periorbital edema (red arrow) and body edema (black arrow), while p53 MO did not produce any phenotype. (D) TEM imaging. Note that the podocyte foot process, slit diaphram and GFB in control zebrafish are intact, while the SRGAP2a morphant zebrafish display foot process effacement, disrupted slit diaphram and disorganized GFB. SRGAP2a overexpression, however, can rescue the GFB structure partially.
(E) A glomeruli in the control and a developmental defected glomeruli in Tg (pod:GFP) under confocal microscope imaging. F) Quantification of foot process width by electron micrographs (G) SRGAP2a overexpression rescued the i1e1 MO induced phenotype. *P<0.05; dpf, days post fertilization.
Figure 7. Decrease of SRGAP2a level in the glomeruli in db/db mice. A) Kidney sections of db/db mice showed glomerular sclerosis (Periodic acid Schiff), focal foot process effacement (electron micrographs) after
16 week. Meanwhile, levels of SRGAP2a (green) and podocyte marker proteins, synaptopodin (red) and WT 1
(green) were decreased. B C) Weight and blood glucose level (B) and albuminuria/creatinine ratio (C) of db/db mice were increased with the increase of age. D) Quantification of foot process width by electron micrographs. E) Levels of WT1, synaptopodin and SRGAP2a in isolated glomeruli from db/db mice were decreased along diabetic nephropathy condition development. F) Increased levels of GTP bound RhoA and
Cdc42 in glomeruli isolated from db/db mice along diabetic nephropathy condition development. Scale bar,
50 m. Results were presented as mean ± SD. *, P <0.05, **, P < 0.01.
Figure 8. Increase mouse renal SRGAP2a level mitigates db/db mouse podocyte dysfunction. A)
Schematic diagram of experimental procedure. Mice were divided into three groups: without treatment
(untreated), with tail vein injection with control adenovirus (Ad GFP) or adenovirus expressing SRGAP2a
(Ad SRGAP2a). B) Upper: Periodic acid Schiff staining of glomeruli from mice infected with Ad GFP or
Ad SRGAP2a. Lower: Representative electron microscopy images of focal foot process effacement. C)
Quantification of mesangial matrix index from mice infected with Ad GFP or Ad SRGAP2a. D) Foot process analyzed by electron micrographs. E) Albuminuria/creatine levels in mice on day 3, 14 and 28 post injection with Ad GFP or Ad SRGAP2a. F) Immunofluorescence label of WT1 and synaptopodin in mouse glomeruli 23
Page 25 of 51 Diabetes
on day 28 post injection with Ad GFP or Ad SRGAP2a. G) Western blot analysis of SRGAP2a, synaptopodin
and WT1 in isolated mouse glomeruli on day 14 post injection with Ad GFP or Ad SRGAP2a. H) GTP bound
forms of RhoA and Cdc42 in isolated glomerular of db/db mice with or without injection of Ad SRGAP2a.
Scale bar, 50 m. Data are presented as mean ± SD, **, P < 0.01.
24
Diabetes Page 26 of 51
Figure 1. Transcriptomic analysis of glomeruli from DN patients and control donors. A) Whole-genome gene expression profile revealed SRGAP family proteins in the glomerular from DN patients. B) The qPCR analysis of glomeruli from healthy control (h-CLT) and DN patients (h-DN), db/m (m-db/m) and db/db mice (m- db/db). C) The correlation between SRGAPs and the key clinical features including proteinuria (g/24h), plasma creatinine (mg/dl), eGFR (ml/min/1.73m2) and eGFR declining (ml/min/1.73m2 /year). D) SRGAP2 mRNA level in DN patients with or without correction against podocyte density. Results in panel B and D were presented as mean ± SD. *, P <0.05, **, P < 0.01.
180x115mm (300 x 300 DPI)
Page 27 of 51 Diabetes
Figure 2. Decrease of SRGAP2a level in the glomeruli from diabetic nephropathy patients. A) IF staining demonstrated the protein levels of SRGAP2 (green) and its co-localization with synaptopodin (red) in podocytes and downregulation of podocyte SRGAP2a in the glomeruli from DN patients. B) IF staining demonstrated the protein levels of SRGAP2a (red) was not co-localization with type IV collagen (Col IV, green) in human glomeruli. C) Quantification of results in (A) by the ratio of integrated optical density (IOD) to area. D) The qPCR analysis of isolated glomeruli demonstrated the significant decrease of SRGAP2 mRNA level in DN patients. Results are presented as mean ± SD. Scale bar, 50µm. *, P <0.05
180x163mm (300 x 300 DPI)
Diabetes Page 28 of 51
Figure 3. Exogenous SRGAP2a protected podocyte cytoskeletal damage induced by high concentration of glucose or TGFβ. A) Fluorescein conjugated phalloidin staining of podocyte after treatment with TGFβ1 (5 ng/ml) or high concentration of glucose (30 mM) for 0, 6, 12 and 24 h. B) Quantification of results in panel A. C) Immunoblotting of SRGAP2a in the podocyte after TGFβ1 or high glucose treatment for the indicated time. D) Immunoblotting of SRGAP2a overexpression in podocytes after transfecting with the Adenovirus expressing SRGAP2a. E) Fluorescein conjugated phalloidin staining of stably transfected Mock or SRGAP2a podocyte after treatment with TGFβ1 (5 ng/ml) or high concentration of glucose (30 mM) 24 h. (F) Quantification of results in (E). Data were presented as the mean ± SD from three independent experiments. *P <0.05, ** P < 0.01.
180x199mm (300 x 300 DPI)
Page 29 of 51 Diabetes
Figure 4. SRGAP2a knockdown re arranged podocyte cytoskeleton and increased podocyte motility. A) SRGAP2a level in scramble or SRGAP2a shRNA treated podocytes (scramble or shRNA SRGAP2a). B) Podocytes with SRGAP2a knockdown had decreased number of stress fibers compared to podocytes treated with vehicle or scramble shRNA). C) Podocytes with SRGAP2a knockdown migrated faster than podocytes infected with vehicle or scramble shRNA. The images were recorded immediately (0 h) and at 24 h after scratch. D) Quantification of the cell migration area. N=20 areas from three independent experiments. E) Quantification of the stress fibers in panel B. ** P<0.01. F) Migration assay using the xCELLigence system (detailed in Methods). The shRNA SRGAP2a podocytes displayed a significantly higher migratory capacity (red).
180x199mm (300 x 300 DPI)
Diabetes Page 30 of 51
Figure 5. SRGAP2a knockdown impaired SRGAP2a RhoA and SRGAP2a Cdc42 interaction in cultured podocytes. A) Representative images showing protein complexes (red) of SRGAP2a RhoA, SRGAP2a Cdc42 and SRGAP2a Rac in SRGAP2a knockdown differentiated podocytes (LV shRNA SRGAP2) and the control knockdown differentiated podocytes (scramble). The data were analyzed by proximity ligation assay (PLA). Scale bar, 20 m. B) Quantification of data in panel A. *, P < 0.05. C) Immunoprecipitation of SRGAP2a bound Rho GTPases. SRGAP2a knockdown abrogated the interaction of SRGAP2a with RhoA and Cdc42. D) Active GTP bound forms of RhoA, Rac1 and Cdc42 precipitated from SRGAP2a knockdown and overexpression podocyte using a GST pulldown assay. Cells transfected with SRGAP2 shRNA resulted in an increased RhoA and Cdc42 but not Rac1 activity compared with scramble cells. LV shRNA SRGAP2a blocked the RhoA and Cdc42 activity stimulated by high concentration of glucose or TGF β. E) Quantification of results in panel D. **, P < 0.01. F) Immunoprecipitation of SRGAP2a bound Rho GTPases. Note that SRGAP2a R527A does not bind to RhoA and Cdc42. G) Immunoprecipitation of podocyte SRGAP2a or SRGAP2a R527A by purified His Rho GTPases. Note that SRGAP2a R527A does not interact with ectogenous RhoA/Cdc42.
180x199mm (300 x 300 DPI) Page 31 of 51 Diabetes
Diabetes Page 32 of 51
Figure 6. Functional analysis of SRGAP2a in zebrafish. (A) Exon structure of SRGAP2a around binding sites of SRGAP2a splice morpholino (i1e1). Primers for reverse transcription–PCR are indicated (arrow). (B) RT- PCR (Lift) of SRGAP2a mRNA from SRGAP2a i1e1 MO-injected 3dpf larvae yields extra bands (red arrows) in addition to a reduced band from SRGAP2a wild-type (WT) mRNA (black arrow). WB (Right) of SRGAP2a in glomerular from SRGAP2a i1e1 MO-injected. (C) Zebrafish co-injected with SRGAP2a and p53 MOs displayed periorbital edema (red arrow) and body edema (black arrow), while p53 MO did not produce any phenotype. (D) TEM imaging. Note that the podocyte foot process, slit diaphram and GFB in control zebrafish are intact, while the SRGAP2a morphant zebrafish display foot process effacement, disrupted slit diaphram and disorganized GFB. SRGAP2a overexpression, however, can rescue the GFB structure partially. (E) A glomeruli in the control and a developmental-defected glomeruli in Tg (pod:GFP) under confocal microscope imaging. F) Quantification of foot process width by electron micrographs (G) SRGAP2a overexpression rescued the i1e1 MO-induced phenotype. *P<0.05; dpf, days post-fertilization.
180x199mm (300 x 300 DPI)
Page 33 of 51 Diabetes
Figure 7. Decrease of SRGAP2a level in the glomeruli in db/db mice. A) Kidney sections of db/db mice showed glomerular sclerosis (Periodic acid-Schiff), focal foot process effacement (electron micrographs) after 16 week. Meanwhile, levels of SRGAP2a (green) and podocyte marker proteins, synaptopodin (red) and WT-1 (green) were decreased. B-C) Weight and blood glucose level (B) and albuminuria/creatinine ratio (C) of db/db mice were increased with the increase of age. D) Quantification of foot process width by electron micrographs. E) Levels of WT1, synaptopodin and SRGAP2a in isolated glomeruli from db/db mice were decreased along diabetic nephropathy condition development. F) Increased levels of GTP-bound RhoA and Cdc42 in glomeruli isolated from db/db mice along diabetic nephropathy condition development. Scale bar, 50µm. Results were presented as mean ± SD. *, P <0.05, **, P < 0.01.
180x199mm (300 x 300 DPI)
Diabetes Page 34 of 51
Figure 8. Increase mouse renal SRGAP2a level mitigates db/db mouse podocyte dysfunction. A) Schematic diagram of experimental procedure. Mice were divided into three groups: without treatment (untreated), with tail vein injection with control adenovirus (Ad-GFP) or adenovirus expressing SRGAP2a (Ad-SRGAP2a). B) Upper: Periodic acid-Schiff staining of glomeruli from mice infected with Ad-GFP or Ad-SRGAP2a. Lower: Representative electron microscopy images of focal foot process effacement. C) Quantification of mesangial matrix index from mice infected with Ad-GFP or Ad-SRGAP2a. D) Foot process analyzed by electron micrographs. E) Albuminuria/creatine levels in mice on day 3, 14 and 28 post-injection with Ad-GFP or Ad- SRGAP2a. F) Immunofluorescence label of WT1 and synaptopodin in mouse glomeruli on day 28 post- injection with Ad-GFP or Ad-SRGAP2a. G) Western blot analysis of SRGAP2a, synaptopodin and WT1 in isolated mouse glomeruli on day 14 post-injection with Ad-GFP or Ad-SRGAP2a. H) GTP-bound forms of RhoA and Cdc42 in isolated glomerular of db/db mice with or without injection of Ad-SRGAP2a. Scale bar, 50µm. Data are presented as mean ± SD, **, P < 0.01.
180x199mm (300 x 300 DPI)
Page 35 of 51 Diabetes
Dissection of glomerular transcriptional profile in patients with diabetic nephropathy: SRGAP2a
protects podocyte structure and function
Yu Pan#, Song Jiang#, Qing Hou#, Dandan Qiu, Jingsong Shi, Ling Wang, Zhaohong Chen, Mingchao Zhang,
Aiping Duan, Ke Zen*, Zhihong Liu*
1. Supplemental Experimental Procedures
2. Four Supplemental Tables
Table S1. Clinical characteristics of DN patients and the healthy control.
Table S2. The qPCR primers and sequences.
Table S3. Gene function analysis the genes in turquoise module.
Table S4. The correlation of the “hub genes” in turquoise module with proteinuria and eGFR
3. Nine Supplemental Figures
Figure S1. Schematic description of experimental design.
Figure S2. The gene expression profile patterns in the glomeruli of diabetic nephropathy patients.
Figure S3 The expression of SRGAP2 in glomeruli in Woronieck diabetes glom data and Ju CKD Glom data
(healthy living donor vs. diabetic nephropathy).
Figure S4. SRGAP2a and SRGAP2c expression in microarray.
Figure S5. SRGAP2a was enriched in the isolated glomerular fraction.
Figure S6. Impaired SRGAP2a RhoGTPase interaction in DN patients.
Figure S7. Alignment of human SRGAP2a, zebrafish SRGAP2a and SRGAP2b amino acids.
Figure S8. Expression of SRGAP2a in zebrafish podocytes.
Figure S9. Preferential expression of SRGAP2a/GFP in db/db mouse glomeruli and liver following tail vein
injection of SRGAP2a/GFP expressing adenovirus (Ad SRGAP2a).
1 / 18
Diabetes Page 36 of 51
Supplemental Experimental Procedures
Periodic Acid-Schiff staining and immunofluorescence
Kidney tissues from db/db mice or patients renal biopsy were fixed in 4% buffered paraformaldehyde for two days, embedded in paraffin and processed for sectioning. Tissue sections (2µm) were examined by Periodic
Acid Schiff staining. For immunofluorescence staining, 5 µm sections of frozen tissues were blocked with bovine serum albumin and incubated with primary antibodies against the srGAP2 (Ab121977; Abcam), synaptopodin proteins (sc 21537; Santa Cruz), WT 1 (sc 192; Santa Cruz), type IV collagen (EURO diagnostic). The sections were then incubated with a Cy3 conjugated anti goat secondary antibody (Dako) or an FITC/or Cy3 conjugated anti rabbit antibody (Dako), and the sections were mounted using Fluoromount.
The slides were examined using a Zeiss LSM710 confocal microscope or a Leica microscope (DM5000B).
Quantitation of the actin cytoskeleton
Rhodamine labeled phalloidin was used to stain F actin in the podocyte and the images were obtained by confocal microscopy and digitized. The rhodamine stained areas of the actin fibers were quantified using
Image J version 1.48 (NIH, Bethesda, MD). The mean podocyte actin content per pixel and the total actin content per cell were calculated and expressed as arbitrary units.
In Situ hybridization
Duolink In Situ kits were used to detect the close proximity of srGAP2 with RhoA, Cdc42, and Rac1 as previously described by Sundqvist (3). The cells were washed with PBS, fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X100. 5 µm sections of frozen tissues were blocked with bovine serum albumin. Subsequently, the blocking solution in the Duolink In Situ kit was used for 30 min at 37°C, after which the cells were incubated in primary antibody including goat anti RhoA antibody (Santa Cruz), goat anti Rac1 antibody (Santa Cruz) or goat anti Cdc42 antibody (Santa Cruz)/rabbit polyclonal anti srGAP2
(HPA 028191, Sigma and Ab121977, Abcam), at 4°C overnight. After treating the cells with two PLA probes for 1 h at 37°C, Ligation Ligase solution was added for 30 min at 37°C, and Amplification Polymerase solution was added for 100 min at 37°C. The slides were then mounted with DAPI and analyzed by fluorescence microscopy. For each slide, the original magnification of × 400 pictures was obtained from 6 2 / 18
Page 37 of 51 Diabetes
different areas, and quantification was performed.
Zebrafish studies
Zebrafish and embryos were maintained in the zebrafish facility of National Clinical Research Center of
Kidney Disease at Jinling Hospital (Nanjing, China). The procedures were approved by Institutional Animal
Care and Use Committee at Jinling Hospital (Nanjing, China). Zebrafish were maintained on a 14/10 hour
light dark cycle at 28.5℃. Eggs were raised in embryo medium at 28.5℃. 1.2 mM morpholino was injected in
to AB strain and Tg (pod:EGFP) for phenotype analysis. For capped mRNA rescue in zebrafish, the srgap2a
coding region was subcloned into pEASY Blunt Zero cloning kit (Transgene, China). Capped mRNA was
synthesized from linearized vectors using the T7 mMESSAGE mMACHINE Kit (Ambion). The mRNA
products were purified with MEGAclearTM Kit (Ambion) prior to injection at 100 ng/ l into 1 cell stage
zebrafish embryos separately with MO. TEM was carried out as described previously (4).
Podocyte migration assays
Confluent monolayers of differentiated podocyte were scraped with a pipette after different treatments. Images
of the same area were acquired at indicated time points using an inverted microscope and were analyzed using
the ImageJ version 1.48 (NIH, Bethesda, MD) image processing program. The percentage of cell migration
area was calculated as Area0 hour Areaindicated time/Area0 hour. Real time migration assay was performed
using the xCELLigence system (Roche Applied Science) in CIM plate 16 according to the manufacturer’s
instructions. Briefly, 24 h after transfection, 4×104 cells were plated in serum free media in the upper chamber.
The lower chambers were filled with 10% FBS for chemoattraction or with serum free media. The data were
analyzed using RTCA software. Results are presented as the time versus cell index curve.
Western blot
Samples of total protein extracted from untreated and treated tissues or cells were using RIPA buffer
containing a protease inhibitor cocktail (Roche) and a phosphatase inhibitor. The membranes were incubated
with primary antibodies against human, mouse or zebryfish srGAP2 (Abcam, Ab121977; Santa Cruz,
3 / 18
Diabetes Page 38 of 51
sc 103497 or Avivasysbio, OAEB00824), synaptopodin (Santa Cruz, sc 21537), and WT1 (Santa Cruz, sc 192).
RNA extraction and quantitative real-time PCR analysis
Total RNA from kidney tissue and cells was isolated using Trizol reagent (Invitrogen) according to manufacturer’s instruction. Total RNA was quantified by absorbance at 260 nm. The mRNA was extracted using the RNAeasy mini kit (QIAGEN Science, Germantown, MD) and cDNA was synthesized from 1.5 g
RNA with the first strand cDNA synthesis kit (Amersham, Buckmgahamshire, UK). Each sample was run and analyzed in triplicate. Real time PCR data were analyzed using the 2 CT method with the SDS Software package (Applied Biosystems). The qPCR primers and sequences were detailed in Supplemental Table S2.
4 / 18
Page 39 of 51 Diabetes
Reference 1. Zhang B, Horvath S: A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol 2005;4:Article17 2. Langfelder P, Horvath S: WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008;9:559 3. Sundqvist A, Zieba A, Vasilaki E, Herrera Hidalgo C, Soderberg O, Koinuma D, Miyazono K, Heldin CH, Landegren U, Ten Dijke P, van Dam H: Specific interactions between Smad proteins and AP-1 components determine TGFbeta-induced breast cancer cell invasion. Oncogene 2013;32:3606-3615 4. Chen Z, Wan X, Hou Q, Shi S, Wang L, Chen P, Zhu X, Zeng C, Qin W, Zhou W, Liu Z: GADD45B mediates podocyte injury in zebrafish by activating the ROS-GADD45B-p38 pathway. Cell death & disease 2016;7:e2068
5 / 18
Diabetes Page 40 of 51
Supplemental Table 1. Clinical characteristics of DN patients and the healthy control.
Healthy Controls (N=20) DN patients (N=41) P value
Age (years) 43.2±6.52 46.9±7.70 0.881 SBP (mmHg) 123.2±18.2 134.2±15.8 0.215
DBP (mmHg) 75.3±10.2 84.8±8.2 0.185 BMI (kg/m2) 21.6±2.52 24.8±1.82 0.112
eGFR (ml/min/1.73m2) 100.23±10.5 63.16±22.12 0.000
Proteinuria (g/24h) － 2.53±1.20 0.000 HbA1C (%) 5.21±0.58 6.78±1.82 0.052
HbA1c (mmol/mol) 32.45±5.07 50.60±13.58 0.056 podocyte density (pods/x106 um3) 220±30 147±41 0.000
6 / 18
Page 41 of 51 Diabetes
Supplemental Table 2. The qPCR primers and sequences used in the present study.
primers and sequences Human SRGAP1(h) F AAAATCAATTAAGGCACGGAAC SRGAP1(h) R TTCAAGGTTGTACTCCGCAGA SRGAP2(h) F CCCAAACCTGCCATAGACACA SRGAP2(h) R ACATTTCTCGGGTTCCTACCTT SRGAP3(h) F CTATACCGCTTGTAGTCGAGA SRGAP3(h) R TTTTAAAACACCAGCGACTGA shRNA SRGAP2a (1) GCAAATTGGTAAATCGGTAAA, (2) CCAAGGACATCTTTCATGACCTGAT, (3) CCACA GTCTACTGACAAGTCTTGTA. SRGAP2a R527A F: CTACAGCATGAAGGAATTTTCGCGGTGTCAGGATCCCAGG; SRGAP2a R527A R: CCTGGGATCCTGACACCGCGAAAATTCCTTCATGCTGTAG. Mouse SRGAP1(m) F AACAATGTCATCATGCGGTTC SRGAP1(m) R GTTTTCATCACCGTGTAGAGC SRGAP2(m) F TGCAGTCTCGCTTATCCAC SRGAP2(m) R TTCGCTCTCCTCTTAGCAA SRGAP3(m) F ACTTGTAGTTGAGAGCTGTATTCG SRGAP3(m) R CCAGCAACTGAATTGATATCTCG Zebrafish nphs2(z) F AGGACCGAAACAGAAACATCTC nphs2(z) R AGGTCCCTCAGTCTCCAATAA enpep(z) F CCTGACTACATTCTTCCGTTCC enpep(z) R GCCAACCAGAGAGCCATTTA fli1a(z) F TGAACGTCAAGCGAGAGTATG fli1a(z) R GAGTCTCAGGAAGTCGTCTTTG SRGAP2a(z) F GTGTTTGGACCAGCAATGTG SRGAP2a(z) R ATTTGAAGACGCAGCTGTTTG SRGAP2a MO GCGACACATTTCACTTACCTTTAAC
7 / 18
Diabetes Page 42 of 51
Supplemental Table 3. Gene function analysis the genes in turquoise module. Rank Enrichment P Term Name 1 4.37E 13 cytoskeletal protein binding 2 9.29E 12 cytoskeleton 3 1.52E 10 protein binding 4 3.90E 09 endomembrane system 5 1.39E 08 cytoplasm 6 1.53E 08 actin cytoskeleton 7 2.32E 08 cell development 8 3.62E 08 cytoskeleton organization 9 1.06E 07 actin filament based process 10 1.29E 07 microtubule cytoskeleton
8 / 18
Page 43 of 51 Diabetes
Supplemental Table 4. Correlation of the “hub genes” in turquoise module with proteinuria and eGFR.
Correlation with proteinuria Correlation with eGFR
R P value R P value CCDC91 0.83 6.54225E 16 0.65 3.59723E 08 SRGAP2 0.81 9.61427E 15 0.58 1.59784E 06 TNNI1 0.81 2.11751E 14 0.67 1.00751E 08 EXPH5 0.81 2.39497E 14 0.65 3.45844E 08 MPP5 0.80 3.25391E 14 0.63 1.43825E 07 USP46 0.80 4.99332E 14 0.53 1.56032E 05 DPP6 0.80 6.50331E 14 0.58 2.08328E 06 PRKAR2B 0.80 6.83201E 14 0.56 5.60258E 06 NFASC 0.80 7.54211E 14 0.60 5.67888E 07 MAGI2 0.79 1.83439E 13 0.61 4.47891E 07 FRMD3 0.79 2.29284E 13 0.67 8.45947E 09 FMN2 0.79 2.75898E 13 0.59 1.02743E 06 SPTB 0.78 4.08825E 13 0.58 1.72764E 06 FRY 0.78 4.35173E 13 0.62 2.38651E 07 ARHGAP19 0.78 7.72538E 13 0.65 4.17016E 08 TCF21 0.78 8.69644E 13 0.61 3.55385E 07 PCOLCE2 0.77 1.06849E 12 0.63 1.02592E 07 CERS6 0.77 2.01344E 12 0.62 2.44366E 07 PLCE1 0.76 2.76927E 12 0.60 7.74222E 07 ARHGEF26 0.76 4.70137E 12 0.68 4.60409E 09 ZDHHC6 0.76 7.56477E 12 0.59 1.40809E 06 CR1 0.75 1.19661E 11 0.63 1.32364E 07 VASN 0.75 1.29902E 11 0.60 7.04573E 07 TYRO3 0.74 2.71659E 11 0.58 2.19617E 06 C14orf37 0.74 3.39427E 11 0.61 4.65852E 07 NPHS1 0.72 2.86836E 10 0.56 4.68138E 06 CPNE8 0.71 3.21045E 10 0.63 1.11699E 07 DACH1 0.71 5.19893E 10 0.63 1.27166E 07 LRRC2 0.69 2.6261E 09 0.62 2.63567E 07 PTPRO 0.67 6.54906E 09 0.59 9.14152E 07
9 / 18
Diabetes Page 44 of 51
Supplemental Figure 1. Schematic description of experimental design.
10 / 18
Page 45 of 51 Diabetes
Supplemental Figure 2. The gene expression profile patterns in the glomerular of diabetic nephropathy patients. A) PCA analysis indicated that the gene expression patterns were difference among DN (red) and healthy controls (blue). B) Weighted gene co expression network analysis (WGCNA) was performed and each module of cluster tree was assigned a unique color label. C) Turquoise module is significantly associated with eGFR (r=0.63, p=10 7) and proteinuria (r= 0.79, p=10 13). D) The hub genes (red) of Turquoise module. E) KEGG analysis the enriched pathway of turquoise module.
11 / 18
Diabetes Page 46 of 51
Supplemental Figure 3 The expression of SRGAP2 in glomeruli in Woronieck diabetes glom data and Ju CKD Glom data (healthy living donor vs. diabetic nephropathy).
12 / 18
Page 47 of 51 Diabetes
Supplemental Figure 4. SRGAP2a and SRGAP2c expression in microarray. **, P < 0.01.
13 / 18
Diabetes Page 48 of 51
Supplemental Figure 5. SRGAP2a was enriched in the glomerular fraction. A) Using magnetic particles perfusion to isolate whole glomeruli from mice to more than 95% purity. B) Enrichment of SRGAP2a in the glomerular fraction.
14 / 18
Page 49 of 51 Diabetes
Supplemental Figure 6. Impaired SRGAP2a RhoGTPase interaction in DN patients. Representative images showed protein complexes (red) of SRGAP2a RhoA, SRGAP2a Cdc42 and SRGAP2a Rac1 analyzed by proximity ligation assay (PLA) in renal biopsies obtained from DN patients and control. Frequency of nuclear and cytoplasmic protein complexes of five samples per group was summarized as bar graphs. Scale bar, 50 m. ** P < 0.01.
15 / 18
Diabetes Page 50 of 51
Supplemental Figure 7. Alignment of human SRGAP2a, zebrafish SRGAP2a and SRGAP2b amino acids. Box residues match the consensus exactly.
16 / 18
Page 51 of 51 Diabetes
Supplemental Figure 8. Expression of SRGAP2a in zebrafish podocytes. A) PCR validation in the sorted podocytes from transgenic zebrafish Tg (pod:GFP), nphs2 was positive in the sorted podocytes, tubular epithelial cells marker enpep and endothelial cells marker fli1a were both negative in the sorted podocytes. B) PCR amplification showed that SRGAP2a was expressed in zebrafish podocytes.
17 / 18
Diabetes Page 52 of 51
Supplemental Figure 9. Preferential expression of SRGAP2a/GFP in db/db mouse glomeruli and liver following tail vein injection of SRGAP2a/GFP expressing adenovirus (Ad SRGAP2a). Adenovirus expressing GFP served as a control. Bright GFP fluorescence in the glomeruli of mice indicated SRGAP2a expression at 3 days after intravenous injection. Scale bar, 100 m.
18 / 18
Page 53 of 51 Diabetes
Supplemental Figure (QC). The quality control of microarray of 61 DN (n=41) and health control (n=20). A)
RNA degradation plot. Each line corresponds to an array. In RNA degradation plot, all arrays have curve with
a regular slope, indicating no unexpected degradation in glomerular RNA samples. B) Density histogram of
log-intensities. In microarray data, raw signal distributions are similar after normalization. C) Box plots of
normalized unscaled standard errors (NUSE) of probe set summaries for each chip. We summarize the batches
of NUSEs for each chip by the median, and this value can be used as a chip data quality index. The NUSE
values of chip data fluctuate around 1.0. D) Relative Log Expression (RLE) Plot. The arrays are centered
around RLE=0 with approximately equal box sizes, and present no quality control problem.