BRIEF COMMUNICATION www.jasn.org
Comparison of Glomerular and Podocyte mRNA Profiles in Streptozotocin-Induced Diabetes
† † † ‡ † Jia Fu,* Chengguo Wei, Kyung Lee, Weijia Zhang, Wu He, Peter Chuang, Zhihong Liu,* † and John Cijiang He §
*National Clinical Research Center of Kidney Diseases, Jinling Hospital, Nanjing University School of Medicine, Nanjing, China; †Division of Nephrology, Department of Medicine, ‡Flow Cytometry Shared Resource Facility, Icahn School of Medicine at Mount Sinai, New York; §Renal Program, James J. Peters VA Medical Center at Bronx, New York.
ABSTRACT Evaluating the mRNA profile of podocytes in the diabetic kidney may indicate STZ induction alone.10 In order to spe- genes involved in the pathogenesis of diabetic nephropathy. To determine if the cifically label and isolate podoctyes, we 2 2 podocyte-specific gene information contained in mRNA profiles of the whole crossed the eNOS / mice with the IRG glomerulus of the diabetic kidney accurately reflects gene expression in the isolated mice11 that ubiquitously express a red 2/2 podocytes, we crossed Nos3 IRG mice with podocin-rtTA and TetON-Cre mice fluorescent protein prior to Cre-mediated for enhanced green fluorescent protein labeling of podocytes before diabetic injury. recombination and an enhanced green Diabetes was induced by streptozotocin, and mRNA profiles of isolated glomeruli fluorescent protein (EGFP) following re- and sorted podocytes from diabetic and control mice were examined 10 weeks combination. These mice were further later. Expression of podocyte-specific markers in glomeruli was downregulated in bred with podocin-rtTA and TetON-Cre diabetic mice compared with controls. However, expression of these markers was (LC1) transgenic mice for inducible not altered in sorted podocytes from diabetic mice. When mRNA levels of glomeruli podocyte-specific EGFP expression. Mice were corrected for podocyte number per glomerulus, the differences in podocyte were fed with doxycycline to induce EGFP marker expression disappeared. Analysis of the differentially expressed genes in expression permanently in podocytes diabetic mice also revealed distinct upregulated pathways in the glomeruli (mito- prior to diabetes induction by at 8 weeks 2 2 chondrial function, oxidative stress) and in podocytes (actin organization). In con- of age (STZ-eNOS / ). Body weight, clusion, our data suggest reduced expression of podocyte markers in glomeruli is blood glucose, and urine excretion of al- a secondary effect of reduced podocyte number, thus podocyte-specific gene ex- bumin were monitored every 2 weeks pression detected in the whole glomerulus may not represent that in podocytes in (Supplemental Figure 1, A–D). Urine the diabetic kidney. albumin-to-creatinine ratio steadily in- J Am Soc Nephrol 27: 1006–1014, 2016. doi: 10.1681/ASN.2015040421 creased in the diabetic mice starting from 2 weeks post-STZ injection, and by 8 weeks a 10-fold increase was observed in comparison to the control mice A large body of evidence suggests that performed to ascertain differential regula- podocyte injury is a key event in diabetic tion of genes involved in DN pathogene- nephropathy (DN). The reduction in sis.6,7 However, due to the heterogeneity Received April 18, 2015. Accepted July 12, 2015. podocyte density is the strongest predictor in cell types, these data provide limited J.F. and C.W. contributed equally to the work. of progressive DN1 and its extent corre- information specifically on podocyte in- fi Published online ahead of print. Publication date lates directly with the magnitude of pro- jury. Recently, mRNA pro les obtained available at www.jasn.org. teinuria.2 Apoptosis, detachment of directly from podocytes have been report- podocytes from the glomerular basement ed,8,9 but such information from the di- Correspondence: Dr. John Cijiang He, Division of Nephrology, Box 1243, Mount Sinai School of membrane, and epithelial-mesenchymal abetic kidney has not been determined. Medicine, One Gustave L. Levy Place, New York NY transition (EMT) are potential mecha- In this study we compared the mRNA 10029, or Dr. Zhihong Liu, National Clinical Re- nisms for podocyte injury and loss in profiles of glomeruli and podocytes be- search Center of Kidney Diseases, Jinling Hospital, Nanjing University School of Medicine, Nanjing, 3–5 DN ; however, its exact mechanism of tween diabetic and control mice. We China, 210002. Email: [email protected] or loss remains unclear. Gene expression employed streptozotocin (STZ)-induced [email protected] fi 2/2 pro ling of glomeruli or cortex of animal diabetes in eNOS mice, resulting in a Copyright © 2016 by the American Society of or human diabetic kidneys have been more pronounced DN phenotype than Nephrology
1006 ISSN : 1046-6673/2704-1006 JAmSocNephrol27: 1006–1014, 2016 www.jasn.org BRIEF COMMUNICATION
(Supplemental Figure 1D). By 10 weeks recognize only the extracellular domain However, whether these processes occur 2 2 STZ-eNOS / mice developed typical of nephrin) and/or to the loss of podo- mainly in podocytes, mesangial cells, or histologic findings of DN, including me- cytes. Immunostaining of diabetic kid- glomerular endothelial cells cannot be sangial expansion and foot process efface- neys for nephrin and podocin in fact concluded. ment (Supplemental Figure 2, A and B). showed a reduction in staining area, Analysis of the DEGs in sorted podo- We obtained whole glomeruli or sor- rather than of staining intensity (Figure cytes showed that many upregulated ted podocytes from diabetic and control 2B). We also did not detect any changes genes were involved in the actin organi- mice at 10 weeks post-STZ injection for in EMT marker expression in podocytes zation (Supplemental Figure 9), suggest- mRNA sequencing (RNA-seq). Sorted by qPCR analysis (Figure 3A) and by im- ing that significant alteration of the actin podocytes were viable and had enriched munostaining (Supplemental Figure 5). cytoskeleton occurs in the early stage of expression of podocyte differentiation RNA-seq data further revealed that the diabetes-induced podocyte injury, markers (Supplemental Figure 3, A and expression of cell death-related genes which is consistent with the foot process B). Each glomerular RNA sample was Bid, Dapk1,andCd40 were significantly effacement (Supplemental Figure 2B). In from an individual mouse (n=4), while increased in diabetic podocytes, which persistent or aggravated injury, the early RNA samples from sorted podocytes was confirmed by qPCR analysis (Figure cytoskeletal changes may eventually lead were pooled from five mice per sample 3B). In addition, a modest increase in to podocyte detachment or death, ensu- (n=3 diabetic, n=2 controls; one sample cleaved Caspase-3 expression was ob- ing in their loss in DN. It also suggests from a control mouse was discarded served by immunostaining in EGFP- that intervention to prevent podocyte de- from analysis due to a technical issue). positive cells, suggesting an increased tachment or death in the early stages of The top 50 differentially expressed genes podocyte apoptosis in the diabetic kid- injury may be an effective therapy against (DEGs) in diabetic glomeruli or podo- ney (Supplemental Figure 6). diabetes-induced podocyte loss. Most of cytes are listed in Supplemental Table 1. The above findings, from an unbiased the downregulated genes in diabetic podo- The principle component analysis and the approach, suggest that the decreased cytes were involved in the RNA processing heatmapofDEGsareshowninFigure1, podocyte marker expression in diabetic and endoplasmic reticulum function A–C. Consistent with the previous stud- glomeruli is likely a secondary effect of (Supplemental Figure 10), suggesting a ies, the expression of podocyte-specific podocyte loss, rather than a direct result possible alteration in the mTOR pathway markers WT-1, nephrin, and synaptopo- of podocyte dedifferentiation or EMT. and autophagy/ER stress response in po- din was significantly downregulated in the This is an important step in better un- docytes, which are known to be involved diabetic glomeruli (Table 1). To our sur- derstanding DN pathogenesis, and it also in podocyte injury in DN.14,15 prise, this downregulation was not ob- indirectly supports the notion that termi- The above pathway analyses of DEGs served in sorted podocytes of diabetic nally differentiated podocytes do not fur- in diabetic mice indicate significant dif- mice (Table 1). The discrepancy in podo- ther dedifferentiate, similar to neurons. ferences in the altered pathways between cyte marker expression level between iso- Nevertheless, we cannot rule out the isolated glomeruli versus podocytes. In- lated glomeruli and sorted podocytes was possibility that podocytes may undergo terestingly, the regulation of actin cyto- further validated by real-time quantitative dedifferentiation or EMT in other diabetic skeleton-related genes differs between PCR (qPCR) (Figure 2A). We confirmed animal models with more advanced DN. glomeruli and podocytes (i.e., some gene by immunostaining that EGFP labeling Analysis of the DEGs in isolated expression changes in opposite direc- co-localizes with WT-1 (Supplemental glomeruli revealed that the upregulated tions), suggesting that changes of actin Figure 4A), and consistent with the pre- genes in diabetic glomeruli were mostly cytoskeleton-related genes in other glo- vious findings3 podocyte number per glo- involved in the regulation of mitochon- merular cells may mask their changes in merulus was significantly lower in the drial function and the oxidative stress podocytes. Our observation that actin diabetic kidneys (Supplemental Figure 4B). pathway, whereas the downregulated genes cytoskeleton and mitochondrial func- When the RNA-seq data were corrected for were in involved in cell-cell signaling/ tion/oxidative stress are major pathways podocyte number per glomerulus, the ap- communication, growth factor receptor- differentially regulated in podocytes and parent downregulation of podocyte marker mediated pathways, and angiogenesis glomeruli, respectively, were further val- expression in diabetic glomeruli also disap- (Supplemental Figures 7 and 8). Consis- idated by qPCR analysis of select genes peared (Table 1), consistent with the obser- tent with the previous observations,13 in each pathway (Figure 3, C and D). vation in the sorted podocytes. these data suggest that mitochondrial 8-Oxoguanine (8-oxoG) immunostaining Asignificant loss of nephrin immu- dysfunction and oxidative stress are further confirmed that the oxidative nostaining has been previously reported key events in the diabetic kidney leading stress in the diabetic kidney is increased in the glomeruli of patients with ad- to glomerular injury. Alteration of mostly in glomerular endothelial cells vanced DN.12 This may be due to the growth factor-mediated pathways may rather than in podocytes (Supplemental cleavage of the extracellular domain of be related to cell survival. Changes in Figure 11, A–C), which is consistent nephrin molecule which becomes unde- genes related to angiogenesis may be re- with a recently published study.16 These tectable (as all available antibodies flective of endothelial cell injury in DN. findings raise an important question as
J Am Soc Nephrol 27: 1006–1014, 2016 Podocyte Gene Expression Profiles in Diabetic Kidney 1007 BRIEF COMMUNICATION www.jasn.org
Figure 1. Heat map and PCA analysis between glomeruli and sorted podocytes from control and diabetic mice. The PCA was performed for transcriptomic data from glomeruli (A) and podocytes (B) between diabetic and non-diabetic mice. Genes with the highest loadings in the first three principal components were plotted in 3D visualization. For both isolated glomeruli and podocytes, PCA revealed that the diabetic and non-diabetic samples form distinct clusters, indicating that samples within each biologic group have more similarity. Each symbol represents each biologic sample (green circles, non-diabetic mice; red circles, diabetic mice). Heatmaps of the top 50 up- or downregulated genes from isolated glomeruli (C) and sorted podocytes (D) between diabetic and control mice (green marks,
1008 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 1006–1014, 2016 www.jasn.org BRIEF COMMUNICATION
Table 1. Relative expression levels of podocyte markers in sorted podocytes and isolated glomeruli between control and diabetic mice Control 1 Control 2 Control 3 Control 4 STZ 1 STZ 2 STZ 3 STZ 4 Fold Change P Value Relative gene expression in sorted podocytes (normalized to Control 1) Nphsl 1.0000 1.4010 1.4496 1.4010 1.0136 1.0729 0.73 Nphs2 1.0000 1.0510 1.1061 1.2399 1.3092 1.1881 0.09 Synpo 1.0000 1.2015 1.2964 0.9701 1.1626 1.0384 0.794 WT1 1.0000 1.2140 1.1817 0.9231 0.8920 0.9023 0.50 Relative gene expression in isolated glomeruli (normalized to Control 1) Nphsl 1.0000 0.9028 0.8721 1.0153 0.4458 0.6045 0.7450 0.5020 0.6061 0.0021 Nphs2 1.0000 0.8765 1.2160 1.2981 0.7768 0.6394 1.2565 0.5793 0.7407 0.17 Synpo 1.0000 0.8482 0.9441 1.2078 0.5581 0.6555 0.9655 0.4299 0.6522 0.04 WT1 1.0000 0.9631 0.9858 1.0115 0.4076 0.5153 0.7412 0.4816 0.5418 0.001 Relative gene expression in isolated glomeruli corrected for podocyte number (normalized to Control 1) Nphsl 1.0000 0.9028 0.8721 1.0153 0.7402 1.0036 1.2370 0.8015 0.9979 0.99 Nphs2 1.0000 0.8765 1.2160 1.2981 1.2898 1.0617 1.0863 0.8298 0.9720 0.83 Synpo 1.0000 0.8482 0.9441 1.2078 0.9266 1.0883 1.3030 0.7138 1.0079 0.96 WT1 1.0000 0.9631 0.9858 1.0115 0.6767 0.8556 1.2306 0.6676 0.8662 0.35 The expression of selected podocyte-specific markers obtained from the RNA-seq data were normalized to a control sample (Control 1) and compared between control and diabetic mice. The top panel shows the selected gene expression in sorted podocytes; the middle panel shows their expression in isolated glomeruli; and the bottom panel shows their expression in isolated glomeruli when first corrected by the number of podocytes. The relative gene expression was calculated by dividing normalized reads value (reads per kilobase per million reads) for each sample by the average value of Control 1. The fold change in gene expression in diabetic mice was calculated by dividing average normalized STZ value by control value. P values were calculated using Student’s unpaired t test. to whether mitochondrial dysfunction excluded.18 Future studies are required compare the two approaches and to de- and increased oxidative stress occur in to determine whether similar observa- termine whether the translating ribosome diabetic podocytes. However, many pre- tions would be made in other animal affinity purification method may yield vious studies, mostly from cultured cells, models of DN (i.e., type 2 diabetes, more information on the mechanism of suggestthatoxidativestressisakey non-toxin–induced diabetes, and in a podocyte injury in DN. Third, although event in podocyte injury in diabetic kid- more susceptible genetic background) each podocyte sample contained pooled ney. In addition, recent studies from and to compare the podocyte gene ex- podocytes sorted from five mice, the Dr. Kumar Sharma’s laboratory suggest pression profiles between mice with number of sample size per group used that mitochondrial-generated superoxide early and late DN. We anticipate that for sequencing was still small. Therefore, is reduced in the diabetic kidney.17 pathways related to actin cytoskeleton it is difficult to control for experimental Therefore, it would be important to disorganization in podocytes might oc- variation that might contribute to the al- further determine whether mitochondrial- cur in early DN. Second, although we tered gene expressions between groups. generated reactive oxygen species and used sorted podocytes from control In conclusion, podocyte mRNA pro- NADPH oxidase (NOX)-generated reac- mice to normalize for possible altered files provide more precise information tive oxygen species are differentially reg- gene expression in sorting and digestion on the mechanism of podocyte injury in ulated in specific glomerular cell types processes, the degree of such change may comparison to the glomerular mRNA and at various stages during DN. not be identical between diabetic and profiles. Because it is not possible to sort This study has a few limitations. First, control mice. A recent approach of podocytes from human diabetic kidney, it is limited to a model of type 1 diabetes translating ribosome affinity purifica- we believe that our data from whole induced by STZ at a single time point. tion which avoids the digestion and glomeruli and sorted podocytes may be Although we used multiple low doses of sorting processes may provide better in- valuable in interpreting the gene expres- STZ to induce diabetes, a direct toxic formation on mRNA profiles of podo- sion profiles obtained from human di- effect of STZ in kidney cells cannot be cytes.9 Future studies are required to abetic glomeruli.
downregulation; red marks, upregulation). There were four glomerular RNA samples in each group, two podocyte RNA samples from control mice and three samples from diabetic mice. Unsupervised hierarchical clustering was performed for gene expression between samples. Four clustersineachheatmaprepresentedupregulatedgenesindiabeticmice(top,leftcluster),upregulatedgenesincontrolmice (top,rightcluster), downregulated genes in diabetic mice (bottom, left cluster), and downregulated genes in control mice (bottom right cluster).
J Am Soc Nephrol 27: 1006–1014, 2016 Podocyte Gene Expression Profiles in Diabetic Kidney 1009 BRIEF COMMUNICATION www.jasn.org
Figure 2. qPCR analysis and immunostaining of podocyte marker between diabetic and control mice. (A) qPCR was performed for podocyte-specific genes, Neph1, Neph2, synaptopodin, and WT-1, in isolated glomeruli and sorted podocytes from both diabetic and non-diabetic mice. (*P,0.05 compared with citrate-treated control glomeruli, n=3; GOI, gene of interest). (B) Immunostaining of podocyte
1010 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 1006–1014, 2016 www.jasn.org BRIEF COMMUNICATION
CONCISE METHODS using a commercial assay from Bethyl Labo- pieces, and digested in 3 ml digestion buffer at ratory Inc. (Houston, TX). Urine albumin 37°C for 15 min on a rotator (100 rpm). Diges- Generation of Transgenic Mice excretion was expressed as the ratio of urine ted tissues were passed through a 100-mmcell Animal studies were performed in accordance albumin to creatinine. strainer and collected by centrifugation. The with the guidelines of and approved by the pellet was resuspended in 2 ml of Hanks’ buff- Institutional Animal Care and Use Committee at Kidney Histology ered salt solution and glomeruli were washed the Icahn School of Medicine at Mount Sinai Harvested kidney samples for histology were three times and collected using a magnet. The (New York, NY). Mice were housed in a specific fixed in 10% formalin, embedded in paraffin, separated glomeruli were resuspended in 2 ml pathogen-free facility with free access to chow and cut into 4 mm sections. Periodic acid- digestion buffer and incubated at 37°C for and water and a 12-hour day/night cycle. Breed- Schiff stained sections were used for assess- 40 min at 1400 rpm/min on a thermomixer. ing and genotyping were performed according to ment of kidney histology. Assessment of the During the second digestion period, the solu- the standard procedures. NPHS2.rtTA19 and IRG mesangial expansion was performed by pixel tion was vortexed every 10 min and sheared mice11 (B6;C3-Tg(CAG-DsRed,-EGFP)5Gae/J) counts on a minimum of 15 glomeruli per with a 27G needle every 15 min. Then, the were purchased from The Jackson Laboratory section in a blinded manner, under 4003 solution was put on a magnetic particle con- (BarHarbor,ME).TheLC1 transgenic mouse magnifications (Zeiss AX10 microscope; centrator and the supernatant was pooled. The was a generous gift from M.J. Moeller (Univer- Carl Zeiss Canada, Toronto, ON, Canada). suspension was then sieved through a 40mm sity of Aachen, Germany).20 To generate diabetic For transmission electron microscopy, kid- cell strainer and centrifuged at 1500 rpm for mice with inducible EGFP labeling in podocytes, ney cortex samples fixed in 2.5% glutaralde- 5 min at 4°C. After a two-step approach for – – we first crossed the IRG mice with eNOS / in hyde were sectioned, mounted on a copper primary cell purification, single cells were re- C57BL/6 background (Jackson Laboratory), and grid, and then images were photographed suspended in 0.5 ml of Hanks’ buffered salt then with NPHS2.rtTA and LC1 mice in order to using a Hitachi H7650 microscope (Tokyo, solution supplemented with 2% fetal bovine generate the experimental NPHS2.rtTA;LC1; Japan) as described previously.22 serum, 25 mM HEPES and 496-diamidino-2- – – IRG;eNOS / mice. For the induction Cre- phenylindole (1 mg/ml). The single-cell sus- mediated EGFP expression, animals received Immunofluorescence Staining pension was then sorted into EGFP-positive doxycycline hydrochloride (Sigma-Aldrich, Frozen sections were used for immunofluores- and EGFP-negative populations with a BD St Louis, MO) via drinking water (2 mg/ml cence staining for WT-1, podocin, and nephrin Aria II cell sorter with a laser excitation at with 5% sucrose) during pregnancy and as described previously23 and images were 488 nm and a sheath pressure of 30 PSI. On nursing up to 4 weeks of age. To increase taken using the Carl Zeiss Axioplan 2 IE mi- average, 450,000 podocytes were sorted per the induction efficiency, the mice received croscope. 8-OxoG staining on formaldehyde- mouse (see Supplementary Figure 1). doxycycline (2 mg/ml) orally from P6 to perfused frozen sections was done as previously P18 three times per week.21 For induction described16 using anti–8-oxoG monoclonal an- mRNA Isolation for RNA of diabetes, male NPHS2.rtTA;LC1;IRG; tibody (N45.1; Japan Institute for the Control Sequencing – – eNOS / mice at 8 weeks of age were injected of Aging). Sectionswere also stained withrabbit Total RNA was isolated from either isolated over 5 consecutive days with low-dose STZ anti-podocin antibody (a gift from Dr. Peter glomeruli or sorted podocytes by using the (50 mg/g per day intraperitoneally; Sigma- Mundel) and antibodies for CD31 (MEC RNeasy mini kit (Qiagen 74104) according to Aldrich). Body weight and hind-limb blood 7.46; Abcam, Inc.), E-cadherin (4065; Cell Sig- the manufacturer’s protocol. RNA concentra- glucose levels were monitored bi-weekly by naling Technology), Alpha SMA (5694; Abcam, tions were quantified using a Nano-drop Spec- glucometer readings. Diabetes was con- Inc.), and Cleaved caspase3 (9664; Cell Signal- trophotometer at a wavelength of 260 nm. firmed by fasting blood glucose level ing Technology). RNA samples were analyzed by Bioanalyzer .300 mg/dl. The age- and sex-matched lit- at a concentration of 100–200 ng/ml to verify 2/2 termates injected with vehicle (CL-eNOS ) Isolation of Glomeruli and Sorting of the concentration and the purity of samples. served as non-diabetic controls. Urine samples Podocytes Only the samples with RNA integrity values of were collected bi-weekly. The mice were sacri- Glomeruli were isolated by Dynabead perfusion .7.0 were used for mRNA sequencing at the ficed at 10 weeks post-STZ injection. and EGFP-labeled podocytes were sorted as Genomic Core Facility at Icahn School of Med- described recently.8 Briefly, animals were per- icine at Mount Sinai School. Measurement of Urinary Albumin- fused with prewarmed 8 ml bead solution and to-Creatinine Ratio 2 ml bead solution with enzymatic digestion Bioinformatics Analysis of mRNA Urine creatinine was quantified using com- buffer (Collagenase type II 300U/ml, Proteinase Sequencing (RNA-seq) Data mercial kits from BioAssay Systems (Hay- E 1 mg/ml and Dnase I 50 U/ml). Kidneys then The RNA-seq data were analyzed by following ward, CA). Urine albumin was determined were removed, decapsulated, minced into 1 mm3 the procedure described below. Briefly, after
2 2 2 2 markers was performed in the kidney of these mice. EGFP was visualized in the podocytes of both STZ-eNOS / and CL-eNOS / mice. Immunofluorescence staining was performed for nephrin (top panel) and podocin (bottom panel) in the kidney of these mice. The repre- sentative images taken from in each group are shown (n=3, original magnification 3400, bar=50 mm).
J Am Soc Nephrol 27: 1006–1014, 2016 Podocyte Gene Expression Profiles in Diabetic Kidney 1011 BRIEF COMMUNICATION www.jasn.org
Figure 3. qPCR validation of gene expression related to the major pathways in isolated glomeruli and podocytes between diabetic and control mice. (A) RT-PCR analysis was performed to assess the levels of mRNA for EMT-related genes expression in podocytes sorted from 2 2 2 2 2 2 STZ-eNOS / and CL-eNOS / mice (*P,0.05 compared with those in CL-eNOS / , n=3). (B) RT-PCR analysis was performed to assess 2 2 2 2 the levels of mRNA for cell-death–related gene expression in podocytes sorted from STZ-eNOS / and CL-eNOS / mice (*P,0.05 2 2 compared with those in CL-eNOS / , n=3). (C) RT-PCR analysis was performed to assess the levels of mRNA for actin cytoskeleton-related 2 2 2 2 gene expression in glomeruli and podocytes isolated from STZ-eNOS / and CL-eNOS / mice (*P,0.05 compared with those in 2 2 CL-eNOS / , n=3). (D) RT-PCR analysis of mitochondria and oxidative stress-related genes in isolated glomeruli and sorted podocytes 2 2 2 2 2 2 from both STZ-eNOS / and CL-eNOS / mice (n=3, *P,0.05 compared with CL-eNOS / ). GOI, gene of interest. sequence quality filtering at a cutoff of a musculus reference genome and transcrip- combined to calculate an expression level minimum quality score Q20 in at least 90% tome (build mm10) using the Burrows- for a corresponding transcript and further bases, the good-quality reads aligned to Reads Wheeler Aligner (bwa).24 The reads that are normalized based on reads per kilobase per were processed and aligned to the University uniquely aligned to the exon and splicing million reads25 in order to compare tran- of California Santa Cruz (UCSC) Mus junction sites for each transcript were scription levels among samples. The
1012 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 1006–1014, 2016 www.jasn.org BRIEF COMMUNICATION transcripts with a low raw read count ,100 in supportedbyNationalInstitutesofHealth(NIH) 10. Yuen DA, Stead BE, Zhang Y, White KE, Kabir all the samples were excluded for downstream 1R01-DK078897, NIH 1R01-DK088541, VA MG, Thai K, Advani SL, Connelly KA, Takano T, Zhu L, Cox AJ, Kelly DJ, Gibson IW, analysis. Gene expression value was trans- Merit Award, and NIH P01-DK56492; P.Y.C. is Takahashi T, Harris RC, Advani A: eNOS de- formedtothelog2basescale.Principle supported by NIH 1R01-DK098126-01A1. ficiency predisposes podocytes to injury in Component Analysis (PCA) was first per- diabetes. JAmSocNephrol23: 1810–1823, formed to assess the sample correlations us- 2012 ing the expression data of all the genes. The DISCLOSURES 11. De Gasperi R, Rocher AB, Sosa MA, Wearne SL, Perez GM, Friedrich VL Jr, Hof PR, Elder differentially expressed genes in STZ mice None. GA: The IRG mouse: a two-color fluorescent fi compared with control mice were identi ed reporter for assessing Cre-mediated re- bytheRpackageDEGseqforsortedpodo- REFERENCES combination and imaging complex cellular cytes26 and we selected the genes based on relationships in situ. Genesis 46: 308–317, DEGseq adjusted P,0.05 and 1.5-fold 2008 1. Steffes MW, Schmidt D, McCrery R, Basgen 12. Jim B, Ghanta M, Qipo A, Fan Y, Chuang PY, change in this 2:3 comparison on which a JM; International Diabetic Nephropathy Cohen HW, Abadi M, Thomas DB, He JC: standard statistic test was not applicable. Study Group: Glomerular cell number in Dysregulated nephrin in diabetic nephropa- The limma test27 was applied for analysis normal subjects and in type 1 diabetic pa- thy of type 2 diabetes: a cross sectional of data in isolated glomeruli. A specific tients. Kidney Int 59: 2104–2113, 2001 study. PLoS One 7: e36041, 2012 gene was considered differentially expressed if 2. Pagtalunan ME, Miller PL, Jumping-Eagle 13. Reidy K, Kang HM, Hostetter T, Susztak K: S, Nelson RG, Myers BD, Rennke HG, Molecular mechanisms of diabetic kidney the P value given by these methods was #0.05. Coplon NS, Sun L, Meyer TW: Podocyte disease. J Clin Invest 124: 2333–2340, 2014 The Gene Ontology and pathway analysis for loss and progressive glomerular injury in 14. Gödel M, Hartleben B, Herbach N, Liu S, the differentially expressed genes were then type II diabetes. J Clin Invest 99: 342–348, Zschiedrich S, Lu S, Debreczeni-Mór A, performed with fold change cutoff of $1.5 us- 1997 Lindenmeyer MT, Rastaldi MP, Hartleben G, ing INGENUITY© IPA (www.ingenuity.com/ 3. Susztak K, Raff AC, Schiffer M, Böttinger EP: Wiech T, Fornoni A, Nelson RG, Kretzler M, Glucose-induced reactive oxygen species Wanke R, Pavenstädt H, Kerjaschki D, Cohen products/ipa) and the online tool Enrichr.28 cause apoptosis of podocytes and podocyte CD, Hall MN, Rüegg MA, Inoki K, Walz G, The read coverage of gene functional elements depletion at the onset of diabetic nephrop- Huber TB: Role of mTOR in podocyte func- was also visualized by the Integrative Genome athy. Diabetes 55: 225–233, 2006 tion and diabetic nephropathy in humans Viewer tool (www.broadinstitute.org/igv/) 4. Petermann AT, Pippin J, Krofft R, Blonski M, and mice. JClinInvest121: 2197–2209, 2011 fi from the genome alignment file. Heatmap Grif n S, Durvasula R, Shankland SJ: Viable 15. Cunard R, Sharma K: The endoplasmic re- podocytes detach in experimental diabetic ticulum stress response and diabetic kidney analysis was performed for the top 50 differen- nephropathy: potential mechanism un- disease. Am J Physiol Renal Physiol 300: tially expressed genes using the DEGseq test derlying glomerulosclerosis. Nephron, Exp F1054–F1061, 2011 after the median center was transformed using Nephrol 98: e114–e123, 2004 16. Daehn I, Casalena G, Zhang T, Shi S, Multi-Experiment Viewer software.29 5. Kang YS, Li Y, Dai C, Kiss LP, Wu C, Liu Y: Fenninger F, Barasch N, Yu L, D’Agati V, Inhibition of integrin-linked kinase blocks Schlondorff D, Kriz W, Haraldsson B, podocyte epithelial-mesenchymal transition Bottinger EP: Endothelial mitochondrial oxi- Real-Time PCR and ameliorates proteinuria. Kidney Int 78: dative stress determines podocyte depletion Primers for RT-PCR were designed by using 363–373, 2010 in segmental glomerulosclerosis. JClinIn- Primer-Blast (NCBI) (Supplemental Table 2). 6. Hodgin JB, Nair V, Zhang H, Randolph A, vest 124: 1608–1621, 2014 Gene expressionwas normalized to Gapdh, and Harris RC, Nelson RG, Weil EJ, Cavalcoli JD, 17. Dugan LL, You YH, Ali SS, Diamond-Stanic Patel JM, Brosius FC 3rd, Kretzler M: Identi- M, Miyamoto S, DeCleves AE, Andreyev A, fold change in expression relative to the control fi –ΔΔCt cation of cross-species shared transcrip- Quach T, Ly S, Shekhtman G, Nguyen W, group was calculated using the 2 method. tional networks of diabetic nephropathy in Chepetan A, Le TP, Wang L, Xu M, Paik KP, Two technical replicates per gene were used. human and mouse glomeruli. Diabetes 62: Fogo A, Viollet B, Murphy A, Brosius F, 299–308, 2013 Naviaux RK, Sharma K: AMPK dysregulation 7. Woroniecka KI, Park AS, Mohtat D, Thomas promotes diabetes-related reduction of su- Statistical Analysis JClin 6 DB, Pullman JM, Susztak K: Transcriptome peroxide and mitochondrial function. Data are expressed as mean SEM. For compar- analysis of human diabetic kidney disease. Invest 123: 4888–4899, 2013 ison of means between three or more groups, Diabetes 60: 2354–2369, 2011 18. Breyer MD, Böttinger E, Brosius FC 3rd, two-way ANOVAwith Bonferroni post-test was 8. Boerries M, Grahammer F, Eiselein S, Buck Coffman TM, Harris RC, Heilig CW, Sharma applied. For comparisons of means between M, Meyer C, Goedel M, Bechtel W, K; AMDCC: Mouse models of diabetic ne- Zschiedrich S, Pfeifer D, Laloë D, Arrondel C, phropathy. J Am Soc Nephrol 16: 27–45, two groups, two-tailed, unpaired t tests were Gonçalves S, Krüger M, Harvey SJ, Busch H, 2005 performed. Prism 5 (GraphPad, La Jolla, CA) Dengjel J, Huber TB: Molecular fingerprint- 19. Shigehara T, Zaragoza C, Kitiyakara C, was used for statistical analyses. ing of the podocyte reveals novel gene and Takahashi H, Lu H, Moeller M, Holzman LB, protein regulatory networks. Kidney Int 83: Kopp JB: Inducible podocyte-specificgene 1052–1064, 2013 expression in transgenic mice. JAmSoc 9. Grgic I, Hofmeister AF, Genovese G, Berhardy AJ, Nephrol 14: 1998–2003, 2003 ACKNOWLEDGMENTS Sun H, Maarouf OH, Bijol V, Pollak MR, 20. Smeets B, Uhlig S, Fuss A, Mooren F, Wetzels Humphreys BD: Discovery of new glomerular JF, Floege J, Moeller MJ: Tracing the origin disease-relevant genes by translational pro- of glomerular extracapillary lesions from pa- J.F., Z.H.L., and J.C.H. are supported by filing of podocytes in vivo. Kidney Int 86: rietal epithelial cells. JAmSocNephrol20: Chinese 973 fund 2012CB517601; J.C.H. is 1116–1129, 2014 2604–2615, 2009
J Am Soc Nephrol 27: 1006–1014, 2016 Podocyte Gene Expression Profiles in Diabetic Kidney 1013 BRIEF COMMUNICATION www.jasn.org
21. Wanner N, Hartleben B, Herbach N, Goedel 24. Li H, Durbin R: Fast and accurate short read interactive and collaborative HTML5 gene M, Stickel N, Zeiser R, Walz G, Moeller MJ, alignment with Burrows-Wheeler transform. list enrichment analysis tool. BMC Bio- Grahammer F, Huber TB: Unraveling the role Bioinformatics 25: 1754–1760, 2009 informatics 14: 128, 2013 of podocyte turnover in glomerular aging and 25. Mortazavi A, Williams BA, McCue K, Schaeffer L, 29. Saeed AI, Sharov V, White J, Li J, Liang W, injury. J Am Soc Nephrol 25: 707–716, 2014 Wold B: Mapping and quantifying mammalian Bhagabati N, Braisted J, Klapa M, Currier T, 22. Mallipattu SK, Liu R, Zhong Y, Chen EY, transcriptomes by RNA-Seq. Nat Methods 5: Thiagarajan M, Sturn A, SnuffinM,Rezantsev D’Agati V, Kaufman L, Ma’ayan A, Klotman 621–628, 2008 A, Popov D, Ryltsov A, Kostukovich E, PE, Chuang PY, He JC: Expression of HIV 26. Wang L, Feng Z, Wang X, Wang X, Zhang X: Borisovsky I, Liu Z, Vinsavich A, Trush V, transgene aggravates kidney injury in di- DEGseq: an R package for identifying dif- Quackenbush J: TM4: a free, open-source abetic mice. Kidney Int 83: 626–634, 2013 ferentially expressed genes from RNA-seq system for microarray data management and 23. Li X, Chuang PY, D’Agati VD, Dai Y, Yacoub data. Bioinformatics 26: 136–138, 2010 analysis. Biotechniques 34: 374–378, 2003 R, Fu J, Xu J, Taku O, Premsrirut PK, Holzman 27. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi LB, He JC: Nephrin Preserves Podocyte Via- W, Smyth GK: limma powers differential ex- bility and Glomerular Structure and Function pression analyses for RNA-sequencing and mi- Nucleic Acids Res in Adult Kidneys [published online ahead of croarray studies. 43: e47, 2015 This article contains supplemental material online print February 2, 2015]. JAmSocNephrol 28. Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, at http://jasn.asnjournals.org/lookup/suppl/doi:10. doi:10.1681/ASN.2014040405 Meirelles GV, Clark NR, Ma’ayan A: Enrichr: 1681/ASN.2015040421/-/DCSupplemental.
1014 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 1006–1014, 2016 Supplementary figure 1 (Figure S1): Body weight, hyperglycemia, and albuminuria in diabe c and control mice A. C.
B. D. Supplementary figure 2 (Figure S2): Histology and morphometric analysis A. Light microscopy CL-eNOS-/- STZ-eNOS-/- Quan fica on of glomerular mesangial area
PAS x200
PAS x400
Quan fica on of foot process width and GBM thickness B. EM CL-eNOS-/- STZ-eNOS-/-
x5K Supplementary figure 3 (Figure S3): Podocyte sor ng process and qPCR valida on of markers in sorted podocytes
A. Podocyte sor ng process
Perfusion with dynabeads Tissue diges on Glomeruli Diges on through renal artery
B. Valida on of podocyte marker expression in the sorted podocytes Supplementary figure 4 (Figure S4): Podocyte number is reduced in diabe c glomeruli
A. WT1 GFP WT1 MergeMerged STZ-eNOS-/- d
GFP WT1 Merged
CL-eNOS-/-
B. Supplementary figure 5 (Figure S5): Immunostaining for EMT-related markers
A. GFP α-SMA Merged
STZ-eNOS-/-
GFP α-SMA Merged
CL-eNOS-/-
B. GFP E-cadherin Merged
STZ-eNOS-/-
GFP E-cadherin Merged
CL-eNOS-/- Supplementary figure 6 (Figure S6): Immunostaining for cleaved caspase 3
Cleaved Caspase 3 GFP Merged
STZ-eNOS-/-
CL-eNOS-/- Supplementary figure 7 (Figure S7): Go terms on up-regulated genes in STZ-glomeruli versus control glomeruli
-log10(p) 0 5 10 15 20
genera on of precursor metabolites and energy (GO:0006091)
respiratory electron transport chain (GO:0022904)
mitochondrial electron transport, NADH to ubiquinone (GO:0006120)
oxida on reduc on (GO:0055114)
excre on (GO:0007588)
cellular respira on (GO:0045333)
ion transport (GO:0006811)
ca on transport (GO:0006812)
energy deriva on by oxida on of organic compounds (GO:0015980)
carboxylic acid catabolic process (GO:0046395)
cellular homeostasis (GO:0019725)
response to oxida ve stress (GO:0006979)
cofactor metabolic process (GO:0051186)
cellular ion homeostasis (GO:0006873)
posi ve regula on of fibroblast prolifera on (GO:0048146)
humoral immune response (GO:0006959) Supplementary figure 8 (Figure S8): Go terms on down-regulated genes in STZ-glomeruli versus control glomeruli
-log10(p) 0 1 2 3 4 5 6
enzyme linked receptor protein signaling pathway (GO:0007167)
cell adhesion (GO:0007155)
cell communica on (GO:0007154)
transmembrane tyrosine kinase signaling(GO:0007169)
signal transduc on (GO:0007165)
epidermal growth factor receptor signaling pathway (GO:0007173)
cell surface receptor linked signal transduc on (GO:0007166)
phosphate metabolic process (GO:0006796)
transforming growth factor beta receptor signaling(GO:0007179)
regula on of ac n filament length (GO:0030832)
nega ve regula on of angiogenesis (GO:0016525)
regula on of cell prolifera on (GO:0042127)
regula on of angiogenesis (GO:0045765)
cell-cell signaling (GO:0007267)
regula on of cell mo on (GO:0051270) Supplementary figure 9 (Figure S9) Go terms on up-regulated genes in STZ-Podocyte versus control podocytes
-log10(p) 0 1 2 3 4 5
microtubule-based process (GO:0007017)
posi ve regula on of cell migra on (GO:0030335)
regula on of cell migra on (GO:0030334)
posi ve regula on of cell mo on (GO:0051272)
regula on of cell mo on (GO:0051270)
microtubule cytoskeleton organiza on (GO:0000226)
cell projec on assembly (GO:0030031)
cell-cell junc on organiza on (GO:0045216)
regula on of cytoskeleton organiza on (GO:0051493)
cytoskeleton organiza on (GO:0007010)
cell-cell junc on assembly (GO:0007043)
locomotory behavior (GO:0007626)
ac va on of protein kinase ac vity (GO:0032147)
cell mo on (GO:0006928)
regula on of GTPase ac vity (GO:0043087) Supplementary figure 10 (Figure S10): Go terms on down-regulated genes in STZ-Podocyte versus control podocytes
-log10(p) 0 10 20 30 40 50 60
transla onal elonga on (GO:0006414) mitochondrial electron transport, NADH to ubiquinone (GO:0006120) respiratory electron transport chain (GO:0022904) genera on of precursor metabolites and energy (GO:0006091) oxida on reduc on (GO:0055114) metabolic process (GO:0008152) rRNA processing (GO:0006364) rRNA metabolic process (GO:0016072) RNA processing (GO:0006396) RNA metabolic process (GO:0016070) proteasomal ubiqui n-dependent protein catabolic process (GO:0043161) transla onal ini a on (GO:0006413) nega ve of ubiqui n-protein ligase ac vity in mito c cell cycle (GO:0051436) proteasomal ubiqui n-dependent protein catabolic process (GO:0031145) posi ve ubiqui n-protein ligase ac vity in mito c cell cycle (GO:0051437) posi ve regula on of ubiqui n-protein ligase ac vity (GO:0051443) nega ve regula on of ubiqui n-protein ligase ac vity (GO:0051444) regula on of ubiqui n-protein ligase ac vity (GO:0051438) Supplementary figure11 (Figure S11): Oxida ve stress was induced mostly in glomerular endothelial cells instead of podocytes in eNOS-/- diabe c mice A. eNOS-/- +Vehicle eNOS-/- + STZ 8-oxoG
Podocin 8-oxoG Merge B.
CD31 8-oxoG Merge C. Supplemenatry table 1 (Table 1S): List of top 50 genes which were differen ated expressed between diabe c and control mice with Fc>2
• Glo_Con_Glo_Con_2fold_downregulated • Gdf5, Steap4, Angptl7, Sox8, Kcng2, Gja3, Gzma, Snx31, Fg p1, Pcdh19, Ptchd4, Cacna2d2, Sstr5, 1810041L15Rik, Lrrc17, Aifm3, Samd5, Thsd7a, Vtcn1, De 47, Astn2, Robo2, Ephb1, Grm7, Eya1, Sulf1, Gm266, Magi2, 6430531B16Rik, Steap3, Cyp26a1, Hs3st3b1, Dpysl5, C1qtnf7, Nebl, Cers6, Fgfr4, Lrrc49, Lmx1b, Gbx1, Amph, Zbtb7c, Arhgap32, Dbx1, C030030A07Rik, 6530402F18Rik, Zbtb8a, Angptl2, Dpp4, Dlgap1 • Glo_Con_Glo_Con_2fold_upregulated • Col10a1, Clec4n, Rgs1, Lcn2, Fpr1, H2-M2, P2ry13, C1qb, C1qc, Ccl2, Lpcat2b, C1qa, Mpeg1, Sostdc1, Spp1, Pigr, Fpr2, Tlr8, C3ar1, Cldn3, Fosb, Pdk4, Fam46b, Gpr65, Clec12a, S100g, Wfdc2, Mreg, Lgals3, Plcd4, Cd68, Acp5, Slc40a1, Sox9, Bcl2a1b, Mmp12, Fcgr1, Ptafr, Ctss, Kcns1, Abcg3, Atp6v0d2, Spic, Emr1, 1700091H14Rik, Calb1, Tlr1, Calml3, Nr0b2, Arhgef38
• Podo_STZ_Podo_Con_2fold_downregulated • S100g, Wfdc15b, De 1, Gpx6, Pvalb, Fxyd2, Cldn8, Ppp1r1a, BB031773, Gm20826, Klk1, Pgam2, Kl, Ptafr, Mal, Klhl38, Tmem52b, Degs2, Paqr5, Calb1, Emx1, E d1, Umod, Ly6a, Sprr1a, Slc16a7, Gdf1, Cers1, Slc12a3, Sostdc1, S100a14, Clcnkb, Oog4, Gm4070, Ttc36, Glod5, Wfdc2, Sfrp1, Pdzk1ip1, 1700074H08Rik, Pkib, Tmem190, Kng2, Tmem213, 1700011H14Rik, Kcnj1, Cd2, Tmem229a, Fetub, Cldn10, Fabp3 • Podo_STZ_Podo_Con_2fold_upregulated • Tlr5, Gabrb1, Nedd9, Nlrp9a, Mgat5, Rassf8, Cdkl5, D10Bwg1379e, Mybpc2, Nt5dc3, Zdhhc11, Tmtc1, Ksr2, Olfr1444, Capsl, Agtpbp1, Fv1, Prx, Nat8l, Lrba, Lrba, Slc26a7, Rpgr, Btbd9, Nrf1, Vwa8, Tcaim, Pstpip2, Col4a6, Lims1, Zfp101, Creb3l2, Topaz1, Stk4, Mrs2, Ncam1, Pogk, Pogk, Pdlim5, Zfp72, Pcdhga10, Ipo7, 2700050L05Rik, Kpna1, Kif18a, Zfp759, Trim44, Pprc1, Amot, Gpld1 Supplementary table 2 (Table S2): Primers for RT-PCR Gene Forward Reverse Gapdh 5’-GCCATCAACGACCCCTTCAT-3’ 5’-ATGATGACCCGTTTGGCTCC-3’ Nphs1 5’-GTGCCCTGAAGGACCCTACT-3’ 5’-CCTGTGGATCCCTTTGACAT-3’ Nphs2 5’-CTTGGCACATCGATCCCTCA-3’ 5’-CGCACTTTGGCCTGTCTTTG-3’ Synpo 5’-CTTTGGGGAAGAGGCCGATTG -3’ 5’-GTTTTCGGTGAAGCTTGTGC-3’ Wt1 5’-GAGAGCCAGCCTACCATCC-3’ 5’-GGGTCCTCGTGTTTGAAGGAA-3’ Pecam-1 5’-AGCCTAGTGTGGAAGCCAAC-3’ 5’-AGCCTTCCGTTCTCTTGGTG-3’ Cdh5 5’-GTCGATGCTAACACAGGGAATG-3’ 5’-AATACCTGGTGCGAAAACACA-3’ Vav1 5’-TGTGAGAAGTTCGGCCTCAAG-3’ 5’-CAGAGCAGACAGGGTGTAGAT-3’ Was 5’-CCAGCCGTTCAGCAGAACAT-3’ 5’-GGTTATCCTTCACGAAGCACA-3’ Itgax 5’-CTGGATAGCCTTTCTTCTGCTG-3’ 5’-GCACACTGTGTCCGAACTCA-3’ Rac2 5’-GACAGTAAGCCGGTGAACCTG-3’ 5’-CTGACTAGCGAGAAGCAGATG-3’ Pdgfa 5’-AGGTATGTATCCACACATGCGT-3’ 5’-AGTTCCTGTTGGTTTCATCTCG-3’ Iqgap2 5’-TACGGCTCAATCGTGGATGAT-3’ 5’-GGTGGCAATTCTTCAACTAAGCA-3’ Rock2 5’-TTGGTTCGTCATAAGGCATCAC-3’ 5’-TGTTGGCAAAGGCCATAATATCT-3’ Ndu 5 5’-CAAGAGACTGTTTGTCGTCAAGC-3’ 5’-TGTTCACCAGTGTTATGCCAAT-3’ Ndu 6 5’-GTCTTTAAGGCGTACCGCTC-3’ 5’-CTGGGCTTCGAGCTAACAATG-3’ Ndufs6 5’-TTCGGGGTTCAAGTGTCGC-3’ 5’-CACAGGCTGTTGTGCTATCAA-3’ Ndufs4 5’-TCCTTTGATGGGTTGGGCAT-3’ 5’-GACTTGGACTTGGGTTTCGG-3’ Pax2 5’-AAGCCCGGAGTGATTGGTG-3’ 5’-CAGGCGAACATAGTCGGGTT-3’ Cyb5rl 5’-TAATCAAGTGCTACCGGACTGG-3’ 5’-CGCCGTACTTCTTTGGTTCAT-3’ Nnt 5’-GGGTCAGTTGTTGTGGATTTAGC-3’ 5’-GCCTTCAGGAGCTTAGTGATGTT-3’ Gene Forward Reverse Bax 5’-TGAAGACAGGGGCCTTTTTG-3’ 5’-AATTCGCCGGAGACACTCG-3’ Cdc40 5’-TGTCATCTGTGAAAAGGTGGTC-3’ 5’-ACTGGAGCAGCGGTGTTATG-3’ Bid 5’-GCCGAGCACATCACAGACC-3’ 5’-TGGCAATGTTGTGGATGATTTCT-3’ Dapk1 5’-ATGACTGTGTTCAGGCAGGAA-3’ 5’-CCGGTACTTTTCTCACGACATTT-3’ Acta1 5’-TACCACCGGCATCGTGTTG-3’ 5’-GCGCACAATCTCACGTTCAG-3’ Col1a1 5’-GCTCCTCTTAGGGGCCACT-3’ 5’-CCACGTCTCACCATTGGGG-3’ Fn1 5’-GATGTCCGAACAGCTATTTACCA-3’ 5’-CCTTGCGACTTCAGCCACT-3’ Cdh1 5’-CAGGTCTCCTCATGGCTTTGC-3’ 5’-CTTCCGAAAAGAAGGCTGTCC-3’