- Email: [email protected]
Foxc1 and Foxc2 are necessary to maintain glomerular podocytes
Experimental Cell Research 352 (2017) 265–272
Contents lists available at ScienceDirect
Experimental Cell Research journal homepage: www.elsevier.com/locate/yexcr
Foxc1 and Foxc2 are necessary to maintain glomerular podocytes a,⁎
b
MARK
c
Masaru Motojima , Tsutomu Kume , Taiji Matsusaka a b c
Department of Clinical Pharmacology, Tokai University School of Medicine, Isehara, Kanagawa 259-1193, Japan Feinberg Cardiovascular Research Institute, Feinberg School of Medicine, Northwestern University, Chicago, Illinois 60611, USA Department of Molecular Life Sciences, Tokai University School of Medicine, Isehara, Kanagawa 259-1193, Japan
A R T I C L E I N F O
A BS T RAC T
Keywords: Foxc1 Foxc2 Podocyte Proteinuria FSGS
Foxc1 and Foxc2 (Foxc1/2) are transcription factors involved in many biological processes. In adult kidneys, expression of Foxc1/2 is confined to the glomerular epithelial cells, i.e., podocytes. To bypass embryonic lethality of Foxc1/2 null mice, mice ubiquitously expressing inducible-Cre (ROSA26CreERT2) or mice expressing Cre in podocytes (Nephrin-Cre) were mated with floxed-Foxc1 and floxed-Foxc2 mice. The CreERT2 was activated in adult mice by administrations of tamoxifen. Eight weeks after tamoxifen treatment, ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice developed microalbuminuria, while ROSA26-Cre ERT2; Foxc1flox/flox; Foxc2+/flox mice had no microalbuminuria. The kidneys of conditional-Foxc1/2 null mice showed proteinaceous casts, protein reabsorption droplets in tubules and huge vacuoles in podocytes, indicating severe podocyte injury and massive proteinuria. Comparison of gene expression profiles revealed that Foxc1/2 maintain expression of genes necessary for podocyte function such as podocin and Cxcl12. In addition, mice with an innate podocyte-specific deletion of Foxc1/2 by Nephrin-Cre develop similar podocyte injury. These results demonstrate dose-dependence of Foxc1/2 gene in maintaining the podocyte with a more critical role for Foxc2 than Foxc1 and a critical role of Foxc1/2 in regulating expression of genes that maintain podocyte integrity.
1. Introduction Foxc1 and Foxc2 (Foxc1/2) belong to subgroup C of the Forkhead-box (FOX) transcription factor superfamily which are involved in development of many organs [1,2]. In humans, mutations of FOXC1 are responsible for the Axenfeld-Rieger syndrome and mutations of FOXC2 underlie the lymphedema-distichiasis syndrome. Mice with the Foxc1 hydrocephalus (ch) mutation have skeletal and ocular anomalies and die perinatally with hydrocephalus due to a lack of calvarial bones [3,4]. Deletion of Foxc2 results in embryonic lethality due to defective cardiovascular development, skeletal, and lymphatic anomalies [5,6]. Foxc1 ch mutants often have abnormal kidney and urinary tract development including duplex urinary system, hydronephrosis and megaureter [7,8]. Foxc2 knockout mice (Foxc2-/-) have been reported to have abnormal glomerular capillary tufts [9]. Conditional knockout of Foxc2 in kidneys using Pax2-Cre mice (Pax2-Cre; Foxc2flox/flox) develop glomerular cysts with normal glomerular tufts and kidney hypoplasia due to reduced number of nephrons and insufficient elongation of tubules [10].
⁎
During kidney development, Foxc1/2 are first expressed in mesenchymal cells surrounding the budding site of the Wolffian duct. Later in development, their expression becomes confined to the podocyte (7, 8, 11, Supplemental Fig. 1). Podocytes are highly differentiated cells and constitute a critical physical filtration barrier within the glomerulus for large molecules such as albumin [13]. Completely overlapping expression and mostly identical DNA binding domain of Foxc1/2 indicate functional redundancy [1,2]. Indeed, mice double heterozygous for Foxc1/2 (Foxc1+/-; Foxc2+/-) share phenotype with Foxc1-/- mice (duplex urinary system, hydronephrosis and megaureter), with Foxc2-/- mice (abnormal glomerular tufts) and with Pax2-Cre; Foxc2flox/flox mice (glomerular cysts) [9,10,12]. Because all the above strains of mice are embryonically or perinatally lethal, it is difficult to determine the specific role of Foxc1/2 in podocytes. To overcome this limitation, we generated two types of conditional knockout mice for both Foxc1/2, one is knocked out in adulthood by mating with ROSA26-CreERT2 mice and the other strain is podocyte-specific knockout created by mating with NephrinCre mice.
Corresponding author. E-mail address: [email protected] (M. Motojima).
http://dx.doi.org/10.1016/j.yexcr.2017.02.016 Received 15 December 2016; Received in revised form 7 February 2017; Accepted 11 February 2017 Available online 20 February 2017 0014-4827/ © 2017 Elsevier Inc. All rights reserved.
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
2. Methods 2.1. Mice All animal protocols were approved by the Animal Experimentation Committee of Tokai University. Nephrin-Cre, ROSA26-CreERT2 mice that express CreERT2 ubiquitously, floxed Foxc1 and floxed Foxc2 mice were maintained as described previously [10,12,14–16]. To activate CreERT2, ROSA26-CreERT2: floxed Foxc1/2 mice were injected with tamoxifen i.p. for 3 consecutive days at 8 weeks of age.
Fig. 1. Development of albuminuria after tamoxifen-treatment in ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice. Ten ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice were treated with tamoxifen and urinary albumin/creatinine ratio (mg/mg) was followed for 8 weeks. Each circle represent the highest urinary albumin/creatinine ratio in each mouse. P < 0.01 vs. –Cre control mice.
2.2. Urinalysis Concentrations of albumin and creatinine were determined in 24-h urine samples. For mice before weaning, urine samples were analyzed by the polyacrylamide gel electrophoresis.
Fig. 2. Renal histology of tamoxifen-treated ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice. Control mice (Foxc1flox/flox; Foxc2flox/flox without CreERT2)(A) were normal, whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria (B) showed severe tubular damage with protein reabsorption droplets (arrows) and proteinaceous casts (arrowheads) 8 weeks after tamoxifen-treatment. Control mice showed normal glomerulus (C), whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria often had podocyte vacuolization (D, F) and occasionally segmental (E) and global (F) hyalinosis or sclerosis. Scale bar =50 µm.
266
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 3. Podocyte damage in ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment. Compared to control mice (A, C, E), CreERT2; Foxc1flox/flox; Foxc2flox/ (B, D, F) with massive albuminuria showed segmental sclerosis (B), intense desmin staining in injured podocytes (arrowheads in D), and loss of nephrin staining in the sclerotic lesion (F). A, C, E and E, D, F represent serial sections, respectively. Scale bar =50 µm.
flox
2.3. Histology
2.5. Isolation of glomeruli
Kidney sections were stained with Periodic acid–Schiff (PAS) and hematoxylin. Nephrin or desmin was detected by immunostaining as previously described [17].
Under anesthesia, kidneys were perfused with PBS containing magnetic beads. After digestion of kidney tissues by collagenase A and DNase I, glomeruli trapping magnetic beads were purified by washing them with PBS using a strong magnet. For glomerular RNA analysis, glomeruli were isolated from mice 1 week after the tamoxifentreatment.
2.4. Transmission electron microscopy (TEM)
2.6. Podocyte culture
Ultrathin sections of 50 nm were made at several levels through each glomerulus and analyzed by transmission electron microscopy.
Isolated glomeruli were incubated in DMEM/F12 medium containing 5% FBS and 0.5% ITS-A in a collagen coated 10 cm dish. Podocytes 267
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 4. Transmission electron micrograph of ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment. Control (Foxc1flox/flox; Foxc2flox/flox without CreERT2) mice showed normal glomerular ultrastructure (A), whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice showed vacuolar degeneration, foot process effacement (arrowhead), irregular thickening of the glomerular basement membrane and swelling of glomerular endothelial cells (B). Scale bar =10 µm.
(qPCR) in RNA from cultured podocytes (n=5) and isolated glomeruli (n=4). 2.8. Statistical analysis Results are expressed as mean ± SD. Comparisons between each group were made using the Student's t-test. The results were deemed statistically significant when the P value was < 0.05. Precise protocols are available in supplemental materials 3. Results 3.1. Ubiquitous and inducible Foxc1/2 knockout in adults To induce Foxc1/2 knockout in adult mice, floxed-Foxc1/2 mice were mated with mice which ubiquitously express CreERT2 (ROSA26CreERT2). Mice carrying homozygous floxed-Foxc1, homozygous floxed-Foxc2 and the ROSA26-CreERT2 transgene (ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox) were obtained by mating and, as adults, were treated with tamoxifen to activate CreERT2. Although the level of albuminuria was variable, overt albuminuria developed in all tamoxifen-treated mice by 8 weeks (Fig. 1). Three of 10 mice died before 8 weeks of age, likely due to massive proteinuria. Histological analysis was performed in all surviving mice (Fig. 2). Control mice without CreERT2 had normal renal morphology (Fig. 2A and C), whereas the massively proteinuric ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice showed remarkable tubular damage with protein reabsorption droplets, tubular dilation and proteinaceous casts (Fig. 2B). Tubular morphology was highly heterogeneous and the severity of tubular damage paralleled the quantity of albuminuria. The most common glomerular injury was vacuolar degeneration of podocytes (Fig. 2C-F). Occasionally, glomeruli showed segmental or global sclerosis or hyalinosis (Fig. 2E, F). Podocytes injury was further confirmed by immunostaining for desmin and nephrin (Fig. 3). Strong desmin staining was found in podocytes containing huge vacuoles. On the other hand, staining of nephrin, a podocyte-specific protein, was not seen in the injured podocytes. TEM analysis further confirmed podocyte degeneration evidenced by vacuolar degeneration, microvillous transformation, foot process effacement of podocytes, and irregular thickness of the glomerular basement membrane (Fig. 4). Glomerular endothelial cells were often swollen, which may be secondary to the severe podocyte injury. Despite the extensive injuries in the kidney, other organs/tissues including the eyeballs, cardiovascular system and lymphatic vessels were normal in these mice. To assess the impact of partially deleting Foxc1/2, adult mice carrying ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox or ROSA26CreERT2; Foxc1flox/flox; Foxc2+/flox were also treated with tamoxifen.
Fig. 5. Foxc1/2 dose-dependency of podocyte injury in tamoxifen-treated ROSA26CreERT2; Foxc1/2 mice. Microalbuminuria developed in 3 of 6 ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment but not in ROSA26CreERT2; Foxc1flox/flox; Foxc2+/flox mice (A). The degree of albumin/creatinine was low compared to mice with complete Foxc1/2 deletion shown in Figure1. TEM revealed that glomeruli of ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice showed foot process effacement, vacuolar degeneration and microvillous transformation (B).
from each mouse were divided into two groups, then treated with vehicle or 1 µM 4-hydroxytamoxifen for 4 days. 2.7. Analysis of gene expression Total RNA from isolated cultured podocytes was subjected to comprehensive gene expression analysis (n=2) using DNA microarray. Gene expression levels were further determined by Quantitative PCR 268
Experimental Cell Research 352 (2017) 265–272
FGB;PLAU;PLAT FGB;PLAU;PLAT CBS;DIO3;PLAT;KMO;SOD3 CYP27A1;SULT1A1;CYP26B1;SULT1E1;EPHX2;CYP2S1;MGST3 CLDN11;FZD5;ZBTB16;FHL2;LDB2;PHF8 CYP27A1;CYP26B1;CYP2S1 3.065769 2.830995 2.490912 2.23027 1.792044 1.53552 −1.446 −1.335 −1.653 −1.79 −1.439 −1.233
0.0072
0.0048 0.0069 0.017 0.0268 0.0269 0.0283
5/72
3/19 3/22 5/90 7/178 6/139 3/39
0.1201 0.1201 0.2216 0.2878 0.2878 0.2878
3.460009 −1.632
KIRREL3;NPHS2;ITGB3;PTPRO;CLDN1
6.299399 3.818662 3.543606 −1.857 −1.801 −1.672 0.0008 0.0063 0.0058 6/64 7/132 5/68
0.1201
6.578785 6.552408 6.497944 6.41693 0.0004 0.0005 0.0005 0.0014
0.0337 0.1201 0.1201
Micro-albuminuria developed in 3 of 6 ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice, but not in any of ROSA26-CreERT2; Foxc1flox/flox; Foxc2+/flox mice (Fig. 5A) 8 weeks after tamoxifen-treatment. The degree of albuminuria was much less in ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice than conditional-Foxc1/2 null mice (Fig. 1). Despite microalbuminuria, the kidneys of ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice were normal by light microscopy, although TEM analysis revealed foot process effacement, microvillous transformation and vacuolar degeneration of podocytes (Fig. 5B). ROSA26-CreERT2; Foxc2flox/flox mice did not developed microalbuminuria 8 weeks after the tamoxifen-treatment and showed no microscopic abnormalities in the kidney. To assess the genes regulated by Foxc1/2, we first compared gene expression profiles of glomeruli in vehicle-treated control mice with tamoxifen-treated mice. In this situation, contaminated tubular cells largely affected enrichment analyses. Therefore, we performed primary culture of podocytes from ROSA26- CreERT2; Foxc1flox/flox; Foxc2flox/ flox mice (n=5). Podocytes from each mouse were divided into two groups, i.e. vehicle treatment or treatment with 4-hydroxytamoxifen. After this first passage, almost all of growing cells were podocytes. Quantitative PCR (qPCR) confirmed more than 95% reduction in Foxc1 and Foxc2 mRNA in the tamoxifen group (data not shown). Gene expression profiles were compared between the two groups (n=2). Changes in each gene were deemed significant when fold changes were more than 2 between the two groups in both pairs and both samples of the higher group showed “detected” flags. After 4 days of treatment, 282 genes were down-regulated and 436 genes were up-regulated (Supplemental Tables 2, 3). These genes were subjected to enrichment analysis using Enricher. Wikipathway 2016 returned highly enriched gene sets in down-regulated genes (Table 1). Genes listed in ‘XPodNet protein-protein interactions in the podocyte expanded by STRING’ showed highest enrichment and contained functionally important genes for podocytes. Of these 25 genes, 23 genes were similarly down-regulated at 7 days indicating consistency of these results (Data not shown). We further confirmed changes of expression of podocin (Nphs2), Cxcl12 and Clic5 in cultured podocytes by qPCR (n=5, Supplemental Fig. 2). These genes were selected because loss of these genes is closely related to podocyte integrity. In glomeruli of ROSA26- CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 1 week after tamoxifen-treatment, the expression of podocin, Cxcl2 and Clic5 was reduced to 23%, 20% and 50%, respectively (n=4, Fig. 6). Reduction of Clic5 was not statistically significant probably due to considerable expression in endothelial cells [11]. In these mice, expression of Foxc1 was 15% and Foxc2 was 3% of their littermates without Cre. Enrichment scores of up-regulated gene sets were low (Supplemental Table 4). 3.2. Podocyte-selective Foxc1/2 knockout
Complement and Coagulation Cascades_Mus musculus_WP449 Complement and Coagulation Cascades_Homo sapiens_WP558 Endochondral Ossification_Mus musculus_WP1270 PodNet: protein-protein interactions in the podocyte_Mus musculus_WP2310 Endochondral Ossification_Homo sapiens_WP474 Metapathway biotransformation_Mus musculus_WP1251 Primary Focal Segmental Glomerulosclerosis FSGS_Mus musculus_WP2573 Primary Focal Segmental Glomerulosclerosis FSGS_Homo sapiens_WP2572 Blood Clotting Cascade_Mus musculus_WP460 Blood Clotting Cascade_Homo sapiens_WP272 Selenium Micronutrient Network_Homo sapiens_WP15 Metapathway biotransformation_Homo sapiens_WP702 Ectoderm Differentiation_Homo sapiens_WP2858 Oxidation by Cytochrome P450_Mus musculus_WP1274
6/55 6/59 6/59 13/305
0.032 0.032 0.032 0.0424
−1.911 −1.903 −1.888 −2.031
ROBO2;CLIC5;SH2D4A;CFH;NPR1;ITGB3;PTPRO;FHL2;PTEN;CLDN2;CLDN1;RABGEF1; PLAU;MAP2;KIRREL3;ANGPT1;BAIAP2;BMP6;NET1;SDK1;CXCL12;NPHS2;PDE3A; APBA1;MAPT FGB;CFH;PLAU;PROS1;PLAT;KNG1 FGB;CFH;PLAU;PROS1;PLAT;KNG1 MMP13;PLAU;MGP;ENPP1;PLAT;BMP6 ROBO2;KIRREL3;CLIC5;CFH;ANGPT1;SH2D4A;ITGB3;PTPRO;BAIAP2;RABGEF1;SDK1; CXCL12;NPHS2 MMP13;PLAU;MGP;ENPP1;PLAT;BMP6 CYP27A1;CYP26B1;SULT1E1;EPHX2;CYP2S1;MGST3;INMT KIRREL3;NPHS2;ITGB3;PTPRO;CLDN1 7.195471 −2.122 0.0009 25/808 XPodNet - protein-protein interactions in the podocyte expanded by STRING_Mus musculus_WP2309
0.0337
Combined Score P-value Overlap Term
Table 1 Enrichment analysis of genes down-regulated by Foxc1/2 deletion by Wikipathways 2016.
Adjusted Pvalue
Z-score
Genes
M. Motojima et al.
To knockout Foxc1/2 selectively in podocytes, floxed-Foxc1 and floxed-Foxc2 mice were mated with the Nephrin-Cre mice that express Cre under the control of nephrin promoter. Most mice double-homozygous for floxed-Foxc1, floxed-Foxc2 and expressing Cre in podocytes (Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox) died before weaning (3–4 weeks of age). Urine samples (2 µl) were analyzed by polyacrylamide gel electrophoresis. Six out of 7 Nephrin-Cre; Foxc1flox/flox; Foxc2flox/ flox mice had massive non-selective proteinuria (Fig. 7A). Proteinaceous casts in dilated tubules were present in each of these mice and protein reabsorption droplets were observed in 5 of 6 mice (Fig. 7B, C) suggesting massive proteinuria. The other one had no microscopic anomaly. Podocyte vacuolization were observed less frequently than in ROSA26-Cre; Foxc1flox/flox; Foxc2flox/flox mice which had massive proteinuria (Fig. 2). Glomerular sclerosis and hyalinosis were occasionally seen in glomeruli with vacuolized podocytes. Since nephrin is expressed in maturing podocytes, we investigated Podocyte density using WT1 staining. Podocyte density of Nephrin-Cre; Foxc1flox/flox; 269
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 6. Effects of Foxc1/2 deletion on gene expression of podocytes. Total RNA were prepared from isolated glomeruli of ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 1 week after tamoxifen-treatment. Expression of podocin (Nphs2) and Cxcl12 were assessed by qPCR (n=4). Significant down-regulation of podocin, and Cxcl12 reconfirmed the cultured podocyte data.
Foxc2flox/flox mice was not statistically different from that in their littermates without Cre (Supplemental Fig. 3). Mice carrying at least one intact Foxc1/2 allele survived until 8 weeks of age. At this time point, mild to moderate albuminuria was observed in mice carrying Nephrin-Cre; Foxc1+/flox; Foxc2flox/flox, but not in mice carrying Nephrin-Cre; Foxc1flox/flox; Foxc2+/flox (Fig. 8). These mice showed mild histological abnormality including focal proteinaceous casts, mild dilatation of tubules and occasionally vacuolated podocytes (Supplemental Fig. 4). Their littermates carrying other genotypes including Nephrin-Cre; Foxc1+/+; Foxc2flox/flox had no proteinuria.
ular lesions are fully established due to systemic complications of massive fluid accumulation. Deletion of Foxc1/2 from maturing podocytes by the Nephrin-Cre resulted in early onset of podocyte injury and death before weaning. The results indicate that these transcription factors have important roles in podocyte maturation and maintenance. Gene expression profile analysis showed several hundred genes were differently regulated by Foxc1/2 knockout. Quantitative PCR analysis confirmed that podocin, and Cxcl12 were down-regulated after the deletion, indicating that Foxc1/2 maintain expression of these genes. Podocin is a podocyte specific protein necessary to form slit diaphragm. Deletion of this gene results in massive proteinuria and nephrotic syndrome [13]. Of note, Foxc1/2 binding motif in podocin promoter has been reported to control the promoter activity [21]. Cxcl12 is expressed in podocyte and its receptor Cxcr4 is expressed in glomerular endothelial cells. Deletion of either of these genes results in the same phenotype, i.e., disorganized glomerular tufts [22]. Disorganized glomerular tufts have been reported in Foxc2-/- and Foxc1+/-; Foxc2+/- [9,12]. Interestingly, Foxc1 has been reported to upregulate Cxcl12 in mesenchymal progenitor cells to maintain niches for hematopoietic stem and progenitor cells [23]. Clic5 is an intracellular chloride channel located at the slit diaphragm. Deletion of this gene results in foot process effacement and proteinuria [24,25]. In addition to these evaluated genes, gene expression profiles indicated that other potentially harmful genes, including Cox2, Wnt4, Tgfb3, Ednra and Cnr1 were up-regulated by Foxc1/2 deletion, and that a potentially protective gene, Cfh, was down-regulated by Foxc1/2 deletion (Table 1, Supplemental Table 2 and 3) [26–32]. Thus, Foxc1/2 not only maintain expression of genes indispensable to podocyte integrity but also suppress potentially harmful genes. Disruption of both Foxc2 alleles and one Foxc1 allele led to podocyte injury. The severity of podocyte injury was most extensive in Foxc1-/-Foxc2-/-, then Foxc1-/+Foxc2-/-, then Foxc1-/-Foxc2+/indicating that Foxc2 has a predominant role in maturation and maintenance of podocytes. Such dose-dependency of Foxc1/2 and a predominant role for Foxc2 have been reported in endothelial cell specification in the somite [33], development of lymphatic vasculature [34], mesoderm formation [35] and development of cardiovascular system [36–38]. Foxc2 also plays predominant roles in glomerular development. Foxc2-/- mice and Foxc1+/-; Foxc2+/- mice developed disorganized glomerular tufts [9,12] and Pax2-Cre; Foxc2flox/flox developed glomerular cysts [10] while renal parenchyma of Foxc1 ch mutants was normal [7]. Disorganized glomerular tufts and glomerular
4. Discussion The present study demonstrates that deletion of Foxc1/2 in adulthood causes massive proteinuria. The proteinuria was associated with morphological changes in the kidneys including proteinaceous casts and protein reabsorption droplets in dilated tubules, severe podocyte injury, glomerular hyalinosis and sclerosis. Of note, structural morphology in non-renal tissues was normal despite the recognized critical role of these transcription factors in development of the eye, the cardiovascular system and the lymphatic vessels [3–6]. Huge vacuoles in podocytes observed in the present study have previously been described by us in another proteinuric model, NEP25 transgenic mice treated with immunotoxin [18]. NEP25 mice express human CD25 selectively in podocytes and immunotoxin binding to human CD25 causes damage to podocytes. NEP25 mice treated with moderate to high dose immunotoxin develop massive proteinuria, edema, and die 1 week after the immunotoxin-induced podocyte injury. Notably, tubular anomalies in the present study, including proteinaceous casts and protein reabsorption droplets in dilated tubules are also frequently found in the NEP25 although podocyte adhesion to parietal epithelial cells and proliferation of parietal epithelial cells were less frequent in the present study. We confirmed severe podocyte injury by the TEM and desmin staining which is recognized as a sensitive maker of podocyte damage and phenotypic changes [19,20]. Therefore, Foxc1/ 2 are crucial to maintain the podocyte integrity. Despite severe podocyte damages, glomerular sclerosis or hyalinosis was less frequent in mice in the current study. Although it took several weeks to develop podocyte injury, these histological findings indicate that podocyte degeneration occurred quite rapidly but mice died before development of full-blown glomerular sclerosis. Thus, like NEP25, it is possible that ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice die before glomer270
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 8. Foxc1/2 dose-dependency of podocyte injury in Nephrin-Cre; Foxc1/2 mice. Mild to moderate albuminuria developed in 5 of 6 Nephrin-Cre; Foxc1+/flox; Foxc2flox/flox mice at 8 weeks of age but not in Nephrin-Cre; Foxc1flox/flox; Foxc2+/flox mice.
the tamoxifen-treatment (Supplemental Fig. 5), Foxc1 also contributes to maintain podocytes. However, Foxc2 is predominant in maturation and maintenance of podocytes. Mutations in FOXC2 cause the lymphedema-distichiasis syndrome, an autosomal dominant disease. Four of 6 members in a family of German-Irish descent with lymphedema-distichiasis syndrome have been reported to have renal disease [39]. All affected members carry a frame shift mutation in the FOXC2 gene and homozygous T allele of 512C-T polymorphism in the 5-prime UTR of the FOXC2 gene. The authors speculate that a combination of mutations in FOXC2 is the cause of renal diseases. Alternatively, another genetic or environmental factor may be involved in the pathogenesis of renal disease in this family. Even though the impact of a single gene mutation in Foxc1/2 on renal disease may be subtle, these genes, especially Foxc2, may contribute to deterioration in renal function with aging or primary nephropathies. The current era of routine whole-genome sequencing and diagnostics may soon uncover susceptibility genes that have subtle effects on progression of chronic renal diseases including Foxc2. Our results indicate that Foxc1 and Foxc2 cooperatively maintain podocytes by regulating podocyte gene expression, and Foxc2 plays predominant roles in podocytes. Disclosures None. Acknowledgement We thank Dr. Valentina Kon for helpful discussions during a preparation of this manuscript. The authors wish to acknowledge Support Center for Medical Research and Education, Tokai University for technical support in gene expression profiling and animal care. This study was supported by Grant-in Aid for Scientific Research 25461234, 22590899 (M.M.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We also thank Ms. Shiho Imai and Ms. Chie Sakurai for excellent technical assistance.
Fig. 7. Development of proteinuria in Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice at 2 weeks of age. Urine samples were subjected to SDS-PAGE (A). Compared to control mice without Cre (-Cre), Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice (+Cre) had massive non-selective proteinuria. Arrow head shows the position of albumin. Renal histology of control mice (B) is normal, whereas, Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice (C) showed protein reabsorption droplets and proteinaceous casts and mild tubular dilatation. Degeneration of podocyte was not so obvious compared to tamoxifen-treated ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria. Scale bar =100 µm.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.yexcr.2017.02.016. References [1] B.A. Benayoun, S. Caburet, R.A. Veitia, Forkhead transcription factors: key players in health and disease, Trends Genet. 27 (2011) 224–232. [2] K., H. Kaestner, W. Knochel, D.E. Martinez, Unified nomenclature for the winged helix/forkhead transcription factors, Genes Dev. 14 (2000) 142–146. [3] R. Rice, D.P. Rice, B.R. Olsen, I. Thesleff, Progression of calvarial bone develop-
cysts were not observed in the present study and may reflect the fact that fundamental glomerular structures might already be established when Cre was expressed or activated. Since ROSA26-CreERT2; Foxc1flox/flox; Foxc2+/- mice developed microalbuminuria 1 year after 271
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
[21] B. He, L. Ebarasi, Z. Zhao, J. Guo, J.R. Ojala, K. Hultenby, S. De Val, C. Betsholtz, K. Tryggvason, Lmx1b and FoxC combinatorially regulate podocin expression in podocytes, J. Am. Soc. Nephrol. 25 (2014) 2764–2777. http://dx.doi.org/10.1681/ ASN.2012080823. [22] Y. Takabatake, T. Sugiyama, H. Kohara, T. Matsusaka, H. Kurihara, P.A. Koni, Y. Nagasawa, T. Hamano, I. Matsui, N. Kawada, E. Imai, T. Nagasawa, H. Rakugi, Y. Isaka, The CXCL12 (SDF-1)/CXCR4 axis is essential for the development of renal vasculature, J. Am. Soc. Nephrol. 20 (2009) 1714–1723. http://dx.doi.org/ 10.1681/ASN.2008060640. [23] Y. Omatsu, M. Seike, T. Sugiyama, T. Kume, T. Nagasawa, Foxc1 is a critical regulator of haematopoietic stem/progenitor cell niche formation, Nature 508 (2014) 536–540. http://dx.doi.org/10.1038/nature13071. [24] B.A. Pierchala, M.R. Muñoz, C., C. Tsui, Proteomic analysis of the slit diaphragm complex: CLIC5 is a protein critical for podocyte morphology and function, Kidney Int. 78 (2010) 868–882. http://dx.doi.org/10.1038/ki.2010.212. [25] B. Wegner, A. Al-Momany, S.C. Kulak, K. Kozlowski, M. Obeidat, N. Jahroudi, J. Paes, M. Berryman, B.J. Ballermann, CLIC5A, a component of the ezrinpodocalyxin complex in glomeruli, is a determinant of podocyte integrity, Am. J. Physiol. Ren. Physiol. 298 (2010) F1492–F1503. http://dx.doi.org/10.1152/ ajprenal.00030.2010. [26] S. Agrawal, A.J. Guess, M.A. Chanley, W.E. Smoyer, Albumin-induced podocyte injury and protection are associated with regulation of COX-2, Kidney Int. 86 (2014) 1150–1160. http://dx.doi.org/10.1038/ki.2014.196. [27] H. Cheng, X. Fan, G.W. Moeckel, R.C. Harris, Podocyte COX-2 exacerbates diabetic nephropathy by increasing podocyte (pro)renin receptor expression, J. Am. Soc. Nephrol. 22 (2011) 1240–1251. http://dx.doi.org/10.1681/ASN.2010111149. [28] L. Zhou, Y. Liu, Wnt/β-catenin signalling and podocyte dysfunction in proteinuric kidney disease, Nat. Rev. Nephrol. 11 (2015) 535–545. http://dx.doi.org/10.1038/ nrneph.2015.88. [29] L. Bao, M. Haas, R.J. Quigg, Complement factor H deficiency accelerates development of lupus nephritis, J. Am. Soc. Nephrol. 22 (2011) 285–295. http:// dx.doi.org/10.1681/ASN.2010060647. [30] A.W. Minto, H.M. Wilson, A.J. Rees, R.J. Quigg, P.A. Brown, Selective expression of TGF-beta2 and TGF-beta3 isoforms in early mesangioproliferative glomerulonephritis, Nephron Exp. Nephrol. 96 (e111-e118) (2004) 2004. [31] I. Daehn, G. Casalena, T. Zhang, S. Shi, F. Fenninger, N. Barasch, L. Yu, V. D'Agati, D. Schlondorff, W. Kriz, B. Haraldsson, E.P. Bottinger, Endothelial mitochondrial oxidative stress determines podocyte depletion in segmental glomerulosclerosis, J. Clin. Investig. 124 (2014) 1608–1621. http://dx.doi.org/10.1172/JCI71195. [32] T. Jourdan, G. Szanda, A.Z. Rosenberg, J. Tam, B.J. Earley, G. Godlewski, R. Cinar, Z. Liu, J. Liu, C. Ju, P. Pacher, G. Kunos, Overactive cannabinoid 1 receptor in podocytes drives type 2 diabetic nephropathy, Proc. Natl. Acad. Sci. USA 111 (2014) E5420–E5428. http://dx.doi.org/10.1073/pnas.1419901111. [33] A. Mayeuf-Louchart, D. Montarras, C. Bodin, T. Kume, S.D. Vincent, M. Buckingham, Endothelial cell specification in the somite is compromised in Pax3-positive progenitors of Foxc1/2 conditional mutants, with loss of forelimb myogenesis, Development 143 (2016) 872–879. http://dx.doi.org/10.1242/ dev.128017. [34] A. Fatima, Y. Wang, Y. Uchida, P. Norden, T. Liu, A. Culver, W.H. Dietz, F. Culver, M. Millay, Y.S. Mukouyama, T. Kume, Foxc1 and Foxc2 deletion causes abnormal lymphangiogenesis and correlates with ERK hyperactivation, J. Clin. Investig. 126 (2016) 2437–2451. http://dx.doi.org/10.1172/JCI80465. [35] B. Wilm, R.G. James, T.M. Schultheiss, B.L.M. Hogan, The forkhead genes, Foxc1 and Foxc2, regulate paraxial versus intermediate mesoderm cell fate, Dev. Biol. 271 (2004) 176–189. [36] S. Seo, T. Kume, Forkhead transcription factors, Foxc1 and Foxc2, are required for the morphogenesis of the cardiac outflow tract, Dev. Biol. 296 (2006) 421–436. [37] S. Seo, H. Fujita, A. Nakano, M. Kang, A. Duarte, T. Kume, The forkhead transcription factors, Foxc1 and Foxc2, are required for arterial specification and lymphatic sprouting during vascular development, Dev. Biol. 294 (2006) 458–470. [38] T. Kume, H. Jiang, J.M. Topczewska, B.L. Hogan, The murine winged helix transcription factors, Foxc1 and Foxc2, are both required for cardiovascular development and somitogenesis, Genes Dev. 15 (2001) 2470–2482. [39] C. Yildirim-Toruner, K. Subramanian, L. El Manjra, E. Chen, S. Goldstein, E. Vitale, A novel frameshift mutation of FOXC2 gene in a family with hereditary lymphedema-distichiasis syndrome associated with renal disease and diabetes mellitus, Am. J. Med. Genet. 131A (2004) 281–286.
ment requires Foxc1 regulation of Msx2 and Alx4, Dev. Biol. 262 (2003) 75–87. [4] T. Kume, K.Y. Deng, V. Winfrey, D.B. Gould, M.A. Walter, B.L.M. Hogan, The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus, Cell 93 (1998) 985–996. [5] K. Iida, H. Koseki, H. Kakinuma, N. Kato, Y. Mizutani-Koseki, H. Ohuchi, H. Yoshioka, S. Noji, K. Kawamura, Y. Kataoka, F. Ueno, M. Taniguchi, N. Yoshida, T. Sugiyama, N. Miura, Essential roles of the winged helix transcription factor MFH-1 in aortic arch patterning and skeletogenesis, Development 124 (1997) 4627–4638. [6] G.E. Winnier, L. Hargett, B.L.M. Hogan, The winged helix transcription factor MFH1 is required for proliferation and patterning of paraxial mesoderm in the mouse embryo, Genes Dev. 11 (1997) 926–940. [7] F. Komaki, Y. Miyazaki, F. Niimura, T. Matsusaka, I. Ichikawa, M. Motojima, Foxc1 gene null mutation causes ectopic budding and kidney hypoplasia but not dysplasia, Cells Tissues Organs 198 (2013) 22–27. [8] T. Kume, K. Deng, B.L.M. Hogan, Murine forkhead/winged helix genes Foxc1 (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney and urinary tract, Development 127 (2000) 1387–1395. [9] M. Takemoto, L. He, J. Norlin, J. Patrakka, Z. Xiao, T. Petrova, C. Bondjers, J. Asp, E. Wallgard, Y. Sun, T. Samuelsson, P. Mostad, S. Lundin, N. Miura, Y. Sado, K. Alitalo, S.E. Quaggin, K. Tryggvason, C. Betsholtz, Large-scale identification of genes implicated in kidney glomerulus development and function, EMBO J. 25 (2006) 1160–1174. [10] M. Motojima, S. Ogiwara, T. Matsusaka, S.Y. Kim, N. Sagawa, K. Abe, M. Ohtsuka, Conditional knockout of Foxc2 gene in kidney: efficient generation of conditional alleles of single-exon gene by double-selection system, Mamm. Genome 27 (2016) 62–69. http://dx.doi.org/10.1007/s00335-015-9610-y. [11] A.P. McMahon, B.J. Aronow, D.R. Davidson, J.A. Davies, K.W. Gaido, S. Grimmond, J.L. Lessard, M.H. Little, S.S. Potter, E.L. Wilder, P. Zhang, GUDMAP: the genitourinary developmental molecular anatomy project, J. Am. Soc. Nephrol. 19 (4) (2008) 667–671. http://dx.doi.org/10.1681/ ASN.2007101078. [12] M. Motojima, S. Tanimoto, M. Ohtsuka, T. Matsusaka, T. Kume, K. Abe, Characterization of kidney and skeleton phenotypes of mice double heterozygous for Foxc1 and Foxc2, Cells Tissues Organs 201 (2016) 380–389. http://dx.doi.org/ 10.1159/000445027. [13] F. Grahammer, C. Schell, T.B. Huber, The podocyte slit diaphragm–from a thin grey line to a complex signalling hub, Nat. Rev. Nephrol. 9 (2013) 587–598. http:// dx.doi.org/10.1038/nrneph.2013.169. [14] T. Asano, F. Niimura, I. Pastan, A.B. Fogo, I. Ichikawa, T. Matsusaka, Permanent genetic tagging of podocytes: fate of injured podocytes in a mouse model of glomerular sclerosis, J. Am. Soc. Nephrol. 16 (2005) 2257–2262. [15] J. Seibler, B. Zevnik, B. Küter-Luks, S. Andreas, H. Kern, T. Hennek, A. Rode, C. Heimann, N. Faust, G. Kauselmann, M. Schoor, R. Jaenisch, K. Rajewsky, R. Kühn, F. Schwenk, Rapid generation of inducible mouse mutants, Nucleic Acids Res. 31 (2003) e12. [16] A. Sasman, C. Nassano-Miller, K.S. Shim, H.Y. Koo, T. Liu, K.M. Schultz, M. Millay, A. Nanano, M. Kang, T. Suzuki, T. Kume, Generation of conditional alleles for Foxc1 and Foxc2 in mice, Genesis 50 (2012) 766–774. [17] H. Ueda, Y. Miyazaki, T. Matsusaka, Y. Utsunomiya, T. Kawamura, T. Hosoya, I. Ichikawa, Bmp in podocytes is essential for normal glomerular capillary formation, J. Am. Soc. Nephrol. 19 (2008) 685–694. [18] T. Matsusaka, J. Xin, S. Niwa, K. Kobayashi, A. Akatsuka, H. Hashizume, Q.C. Wang, I. Pastan, A.B. Fogo, I. Ichikawa, Genetic engineering of glomerular sclerosis in the mouse via control of onset and severity of podocyte-specific injury, J. Am. Soc. Nephrol. 16 (2005) 1013–1023. [19] J. Funk, V. Ott, A. Herrmann, W. Rapp, S. Raab, W. Riboulet, A. Vandjour, E. Hainaut, A. Benardeau, T. Singer, B. Jacobsen, Semiautomated quantitative image analysis of glomerular immunohistochemistry markers desmin, vimentin, podocin, synaptopodin and WT-1 in acute and chronic rat kidney disease models, Histochem. Cell Biol. 145 (2016) 315–326. http://dx.doi.org/10.1007/s00418015-1391-6. [20] T. Kakimoto, K. Okada, Y. Hirohashi, R. Relator, M. Kawai, T. Iguchi, K. Fujitaka, M. Nishio, T. Kato, A. Fukunari, H. Utsumi, Automated image analysis of a glomerular injury marker desmin in spontaneously diabetic Torii rats treated with losartan, J. Endocrinol. 222 (2014) 43–51. http://dx.doi.org/10.1530/JOE-140164.
272
Contents lists available at ScienceDirect
Experimental Cell Research journal homepage: www.elsevier.com/locate/yexcr
Foxc1 and Foxc2 are necessary to maintain glomerular podocytes a,⁎
b
MARK
c
Masaru Motojima , Tsutomu Kume , Taiji Matsusaka a b c
Department of Clinical Pharmacology, Tokai University School of Medicine, Isehara, Kanagawa 259-1193, Japan Feinberg Cardiovascular Research Institute, Feinberg School of Medicine, Northwestern University, Chicago, Illinois 60611, USA Department of Molecular Life Sciences, Tokai University School of Medicine, Isehara, Kanagawa 259-1193, Japan
A R T I C L E I N F O
A BS T RAC T
Keywords: Foxc1 Foxc2 Podocyte Proteinuria FSGS
Foxc1 and Foxc2 (Foxc1/2) are transcription factors involved in many biological processes. In adult kidneys, expression of Foxc1/2 is confined to the glomerular epithelial cells, i.e., podocytes. To bypass embryonic lethality of Foxc1/2 null mice, mice ubiquitously expressing inducible-Cre (ROSA26CreERT2) or mice expressing Cre in podocytes (Nephrin-Cre) were mated with floxed-Foxc1 and floxed-Foxc2 mice. The CreERT2 was activated in adult mice by administrations of tamoxifen. Eight weeks after tamoxifen treatment, ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice developed microalbuminuria, while ROSA26-Cre ERT2; Foxc1flox/flox; Foxc2+/flox mice had no microalbuminuria. The kidneys of conditional-Foxc1/2 null mice showed proteinaceous casts, protein reabsorption droplets in tubules and huge vacuoles in podocytes, indicating severe podocyte injury and massive proteinuria. Comparison of gene expression profiles revealed that Foxc1/2 maintain expression of genes necessary for podocyte function such as podocin and Cxcl12. In addition, mice with an innate podocyte-specific deletion of Foxc1/2 by Nephrin-Cre develop similar podocyte injury. These results demonstrate dose-dependence of Foxc1/2 gene in maintaining the podocyte with a more critical role for Foxc2 than Foxc1 and a critical role of Foxc1/2 in regulating expression of genes that maintain podocyte integrity.
1. Introduction Foxc1 and Foxc2 (Foxc1/2) belong to subgroup C of the Forkhead-box (FOX) transcription factor superfamily which are involved in development of many organs [1,2]. In humans, mutations of FOXC1 are responsible for the Axenfeld-Rieger syndrome and mutations of FOXC2 underlie the lymphedema-distichiasis syndrome. Mice with the Foxc1 hydrocephalus (ch) mutation have skeletal and ocular anomalies and die perinatally with hydrocephalus due to a lack of calvarial bones [3,4]. Deletion of Foxc2 results in embryonic lethality due to defective cardiovascular development, skeletal, and lymphatic anomalies [5,6]. Foxc1 ch mutants often have abnormal kidney and urinary tract development including duplex urinary system, hydronephrosis and megaureter [7,8]. Foxc2 knockout mice (Foxc2-/-) have been reported to have abnormal glomerular capillary tufts [9]. Conditional knockout of Foxc2 in kidneys using Pax2-Cre mice (Pax2-Cre; Foxc2flox/flox) develop glomerular cysts with normal glomerular tufts and kidney hypoplasia due to reduced number of nephrons and insufficient elongation of tubules [10].
⁎
During kidney development, Foxc1/2 are first expressed in mesenchymal cells surrounding the budding site of the Wolffian duct. Later in development, their expression becomes confined to the podocyte (7, 8, 11, Supplemental Fig. 1). Podocytes are highly differentiated cells and constitute a critical physical filtration barrier within the glomerulus for large molecules such as albumin [13]. Completely overlapping expression and mostly identical DNA binding domain of Foxc1/2 indicate functional redundancy [1,2]. Indeed, mice double heterozygous for Foxc1/2 (Foxc1+/-; Foxc2+/-) share phenotype with Foxc1-/- mice (duplex urinary system, hydronephrosis and megaureter), with Foxc2-/- mice (abnormal glomerular tufts) and with Pax2-Cre; Foxc2flox/flox mice (glomerular cysts) [9,10,12]. Because all the above strains of mice are embryonically or perinatally lethal, it is difficult to determine the specific role of Foxc1/2 in podocytes. To overcome this limitation, we generated two types of conditional knockout mice for both Foxc1/2, one is knocked out in adulthood by mating with ROSA26-CreERT2 mice and the other strain is podocyte-specific knockout created by mating with NephrinCre mice.
Corresponding author. E-mail address: [email protected] (M. Motojima).
http://dx.doi.org/10.1016/j.yexcr.2017.02.016 Received 15 December 2016; Received in revised form 7 February 2017; Accepted 11 February 2017 Available online 20 February 2017 0014-4827/ © 2017 Elsevier Inc. All rights reserved.
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
2. Methods 2.1. Mice All animal protocols were approved by the Animal Experimentation Committee of Tokai University. Nephrin-Cre, ROSA26-CreERT2 mice that express CreERT2 ubiquitously, floxed Foxc1 and floxed Foxc2 mice were maintained as described previously [10,12,14–16]. To activate CreERT2, ROSA26-CreERT2: floxed Foxc1/2 mice were injected with tamoxifen i.p. for 3 consecutive days at 8 weeks of age.
Fig. 1. Development of albuminuria after tamoxifen-treatment in ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice. Ten ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice were treated with tamoxifen and urinary albumin/creatinine ratio (mg/mg) was followed for 8 weeks. Each circle represent the highest urinary albumin/creatinine ratio in each mouse. P < 0.01 vs. –Cre control mice.
2.2. Urinalysis Concentrations of albumin and creatinine were determined in 24-h urine samples. For mice before weaning, urine samples were analyzed by the polyacrylamide gel electrophoresis.
Fig. 2. Renal histology of tamoxifen-treated ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice. Control mice (Foxc1flox/flox; Foxc2flox/flox without CreERT2)(A) were normal, whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria (B) showed severe tubular damage with protein reabsorption droplets (arrows) and proteinaceous casts (arrowheads) 8 weeks after tamoxifen-treatment. Control mice showed normal glomerulus (C), whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria often had podocyte vacuolization (D, F) and occasionally segmental (E) and global (F) hyalinosis or sclerosis. Scale bar =50 µm.
266
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 3. Podocyte damage in ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment. Compared to control mice (A, C, E), CreERT2; Foxc1flox/flox; Foxc2flox/ (B, D, F) with massive albuminuria showed segmental sclerosis (B), intense desmin staining in injured podocytes (arrowheads in D), and loss of nephrin staining in the sclerotic lesion (F). A, C, E and E, D, F represent serial sections, respectively. Scale bar =50 µm.
flox
2.3. Histology
2.5. Isolation of glomeruli
Kidney sections were stained with Periodic acid–Schiff (PAS) and hematoxylin. Nephrin or desmin was detected by immunostaining as previously described [17].
Under anesthesia, kidneys were perfused with PBS containing magnetic beads. After digestion of kidney tissues by collagenase A and DNase I, glomeruli trapping magnetic beads were purified by washing them with PBS using a strong magnet. For glomerular RNA analysis, glomeruli were isolated from mice 1 week after the tamoxifentreatment.
2.4. Transmission electron microscopy (TEM)
2.6. Podocyte culture
Ultrathin sections of 50 nm were made at several levels through each glomerulus and analyzed by transmission electron microscopy.
Isolated glomeruli were incubated in DMEM/F12 medium containing 5% FBS and 0.5% ITS-A in a collagen coated 10 cm dish. Podocytes 267
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 4. Transmission electron micrograph of ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment. Control (Foxc1flox/flox; Foxc2flox/flox without CreERT2) mice showed normal glomerular ultrastructure (A), whereas CreERT2; Foxc1flox/flox; Foxc2flox/flox mice showed vacuolar degeneration, foot process effacement (arrowhead), irregular thickening of the glomerular basement membrane and swelling of glomerular endothelial cells (B). Scale bar =10 µm.
(qPCR) in RNA from cultured podocytes (n=5) and isolated glomeruli (n=4). 2.8. Statistical analysis Results are expressed as mean ± SD. Comparisons between each group were made using the Student's t-test. The results were deemed statistically significant when the P value was < 0.05. Precise protocols are available in supplemental materials 3. Results 3.1. Ubiquitous and inducible Foxc1/2 knockout in adults To induce Foxc1/2 knockout in adult mice, floxed-Foxc1/2 mice were mated with mice which ubiquitously express CreERT2 (ROSA26CreERT2). Mice carrying homozygous floxed-Foxc1, homozygous floxed-Foxc2 and the ROSA26-CreERT2 transgene (ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox) were obtained by mating and, as adults, were treated with tamoxifen to activate CreERT2. Although the level of albuminuria was variable, overt albuminuria developed in all tamoxifen-treated mice by 8 weeks (Fig. 1). Three of 10 mice died before 8 weeks of age, likely due to massive proteinuria. Histological analysis was performed in all surviving mice (Fig. 2). Control mice without CreERT2 had normal renal morphology (Fig. 2A and C), whereas the massively proteinuric ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice showed remarkable tubular damage with protein reabsorption droplets, tubular dilation and proteinaceous casts (Fig. 2B). Tubular morphology was highly heterogeneous and the severity of tubular damage paralleled the quantity of albuminuria. The most common glomerular injury was vacuolar degeneration of podocytes (Fig. 2C-F). Occasionally, glomeruli showed segmental or global sclerosis or hyalinosis (Fig. 2E, F). Podocytes injury was further confirmed by immunostaining for desmin and nephrin (Fig. 3). Strong desmin staining was found in podocytes containing huge vacuoles. On the other hand, staining of nephrin, a podocyte-specific protein, was not seen in the injured podocytes. TEM analysis further confirmed podocyte degeneration evidenced by vacuolar degeneration, microvillous transformation, foot process effacement of podocytes, and irregular thickness of the glomerular basement membrane (Fig. 4). Glomerular endothelial cells were often swollen, which may be secondary to the severe podocyte injury. Despite the extensive injuries in the kidney, other organs/tissues including the eyeballs, cardiovascular system and lymphatic vessels were normal in these mice. To assess the impact of partially deleting Foxc1/2, adult mice carrying ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox or ROSA26CreERT2; Foxc1flox/flox; Foxc2+/flox were also treated with tamoxifen.
Fig. 5. Foxc1/2 dose-dependency of podocyte injury in tamoxifen-treated ROSA26CreERT2; Foxc1/2 mice. Microalbuminuria developed in 3 of 6 ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice 8 weeks after tamoxifen-treatment but not in ROSA26CreERT2; Foxc1flox/flox; Foxc2+/flox mice (A). The degree of albumin/creatinine was low compared to mice with complete Foxc1/2 deletion shown in Figure1. TEM revealed that glomeruli of ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice showed foot process effacement, vacuolar degeneration and microvillous transformation (B).
from each mouse were divided into two groups, then treated with vehicle or 1 µM 4-hydroxytamoxifen for 4 days. 2.7. Analysis of gene expression Total RNA from isolated cultured podocytes was subjected to comprehensive gene expression analysis (n=2) using DNA microarray. Gene expression levels were further determined by Quantitative PCR 268
Experimental Cell Research 352 (2017) 265–272
FGB;PLAU;PLAT FGB;PLAU;PLAT CBS;DIO3;PLAT;KMO;SOD3 CYP27A1;SULT1A1;CYP26B1;SULT1E1;EPHX2;CYP2S1;MGST3 CLDN11;FZD5;ZBTB16;FHL2;LDB2;PHF8 CYP27A1;CYP26B1;CYP2S1 3.065769 2.830995 2.490912 2.23027 1.792044 1.53552 −1.446 −1.335 −1.653 −1.79 −1.439 −1.233
0.0072
0.0048 0.0069 0.017 0.0268 0.0269 0.0283
5/72
3/19 3/22 5/90 7/178 6/139 3/39
0.1201 0.1201 0.2216 0.2878 0.2878 0.2878
3.460009 −1.632
KIRREL3;NPHS2;ITGB3;PTPRO;CLDN1
6.299399 3.818662 3.543606 −1.857 −1.801 −1.672 0.0008 0.0063 0.0058 6/64 7/132 5/68
0.1201
6.578785 6.552408 6.497944 6.41693 0.0004 0.0005 0.0005 0.0014
0.0337 0.1201 0.1201
Micro-albuminuria developed in 3 of 6 ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice, but not in any of ROSA26-CreERT2; Foxc1flox/flox; Foxc2+/flox mice (Fig. 5A) 8 weeks after tamoxifen-treatment. The degree of albuminuria was much less in ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice than conditional-Foxc1/2 null mice (Fig. 1). Despite microalbuminuria, the kidneys of ROSA26-CreERT2; Foxc1+/flox; Foxc2flox/flox mice were normal by light microscopy, although TEM analysis revealed foot process effacement, microvillous transformation and vacuolar degeneration of podocytes (Fig. 5B). ROSA26-CreERT2; Foxc2flox/flox mice did not developed microalbuminuria 8 weeks after the tamoxifen-treatment and showed no microscopic abnormalities in the kidney. To assess the genes regulated by Foxc1/2, we first compared gene expression profiles of glomeruli in vehicle-treated control mice with tamoxifen-treated mice. In this situation, contaminated tubular cells largely affected enrichment analyses. Therefore, we performed primary culture of podocytes from ROSA26- CreERT2; Foxc1flox/flox; Foxc2flox/ flox mice (n=5). Podocytes from each mouse were divided into two groups, i.e. vehicle treatment or treatment with 4-hydroxytamoxifen. After this first passage, almost all of growing cells were podocytes. Quantitative PCR (qPCR) confirmed more than 95% reduction in Foxc1 and Foxc2 mRNA in the tamoxifen group (data not shown). Gene expression profiles were compared between the two groups (n=2). Changes in each gene were deemed significant when fold changes were more than 2 between the two groups in both pairs and both samples of the higher group showed “detected” flags. After 4 days of treatment, 282 genes were down-regulated and 436 genes were up-regulated (Supplemental Tables 2, 3). These genes were subjected to enrichment analysis using Enricher. Wikipathway 2016 returned highly enriched gene sets in down-regulated genes (Table 1). Genes listed in ‘XPodNet protein-protein interactions in the podocyte expanded by STRING’ showed highest enrichment and contained functionally important genes for podocytes. Of these 25 genes, 23 genes were similarly down-regulated at 7 days indicating consistency of these results (Data not shown). We further confirmed changes of expression of podocin (Nphs2), Cxcl12 and Clic5 in cultured podocytes by qPCR (n=5, Supplemental Fig. 2). These genes were selected because loss of these genes is closely related to podocyte integrity. In glomeruli of ROSA26- CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 1 week after tamoxifen-treatment, the expression of podocin, Cxcl2 and Clic5 was reduced to 23%, 20% and 50%, respectively (n=4, Fig. 6). Reduction of Clic5 was not statistically significant probably due to considerable expression in endothelial cells [11]. In these mice, expression of Foxc1 was 15% and Foxc2 was 3% of their littermates without Cre. Enrichment scores of up-regulated gene sets were low (Supplemental Table 4). 3.2. Podocyte-selective Foxc1/2 knockout
Complement and Coagulation Cascades_Mus musculus_WP449 Complement and Coagulation Cascades_Homo sapiens_WP558 Endochondral Ossification_Mus musculus_WP1270 PodNet: protein-protein interactions in the podocyte_Mus musculus_WP2310 Endochondral Ossification_Homo sapiens_WP474 Metapathway biotransformation_Mus musculus_WP1251 Primary Focal Segmental Glomerulosclerosis FSGS_Mus musculus_WP2573 Primary Focal Segmental Glomerulosclerosis FSGS_Homo sapiens_WP2572 Blood Clotting Cascade_Mus musculus_WP460 Blood Clotting Cascade_Homo sapiens_WP272 Selenium Micronutrient Network_Homo sapiens_WP15 Metapathway biotransformation_Homo sapiens_WP702 Ectoderm Differentiation_Homo sapiens_WP2858 Oxidation by Cytochrome P450_Mus musculus_WP1274
6/55 6/59 6/59 13/305
0.032 0.032 0.032 0.0424
−1.911 −1.903 −1.888 −2.031
ROBO2;CLIC5;SH2D4A;CFH;NPR1;ITGB3;PTPRO;FHL2;PTEN;CLDN2;CLDN1;RABGEF1; PLAU;MAP2;KIRREL3;ANGPT1;BAIAP2;BMP6;NET1;SDK1;CXCL12;NPHS2;PDE3A; APBA1;MAPT FGB;CFH;PLAU;PROS1;PLAT;KNG1 FGB;CFH;PLAU;PROS1;PLAT;KNG1 MMP13;PLAU;MGP;ENPP1;PLAT;BMP6 ROBO2;KIRREL3;CLIC5;CFH;ANGPT1;SH2D4A;ITGB3;PTPRO;BAIAP2;RABGEF1;SDK1; CXCL12;NPHS2 MMP13;PLAU;MGP;ENPP1;PLAT;BMP6 CYP27A1;CYP26B1;SULT1E1;EPHX2;CYP2S1;MGST3;INMT KIRREL3;NPHS2;ITGB3;PTPRO;CLDN1 7.195471 −2.122 0.0009 25/808 XPodNet - protein-protein interactions in the podocyte expanded by STRING_Mus musculus_WP2309
0.0337
Combined Score P-value Overlap Term
Table 1 Enrichment analysis of genes down-regulated by Foxc1/2 deletion by Wikipathways 2016.
Adjusted Pvalue
Z-score
Genes
M. Motojima et al.
To knockout Foxc1/2 selectively in podocytes, floxed-Foxc1 and floxed-Foxc2 mice were mated with the Nephrin-Cre mice that express Cre under the control of nephrin promoter. Most mice double-homozygous for floxed-Foxc1, floxed-Foxc2 and expressing Cre in podocytes (Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox) died before weaning (3–4 weeks of age). Urine samples (2 µl) were analyzed by polyacrylamide gel electrophoresis. Six out of 7 Nephrin-Cre; Foxc1flox/flox; Foxc2flox/ flox mice had massive non-selective proteinuria (Fig. 7A). Proteinaceous casts in dilated tubules were present in each of these mice and protein reabsorption droplets were observed in 5 of 6 mice (Fig. 7B, C) suggesting massive proteinuria. The other one had no microscopic anomaly. Podocyte vacuolization were observed less frequently than in ROSA26-Cre; Foxc1flox/flox; Foxc2flox/flox mice which had massive proteinuria (Fig. 2). Glomerular sclerosis and hyalinosis were occasionally seen in glomeruli with vacuolized podocytes. Since nephrin is expressed in maturing podocytes, we investigated Podocyte density using WT1 staining. Podocyte density of Nephrin-Cre; Foxc1flox/flox; 269
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 6. Effects of Foxc1/2 deletion on gene expression of podocytes. Total RNA were prepared from isolated glomeruli of ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice 1 week after tamoxifen-treatment. Expression of podocin (Nphs2) and Cxcl12 were assessed by qPCR (n=4). Significant down-regulation of podocin, and Cxcl12 reconfirmed the cultured podocyte data.
Foxc2flox/flox mice was not statistically different from that in their littermates without Cre (Supplemental Fig. 3). Mice carrying at least one intact Foxc1/2 allele survived until 8 weeks of age. At this time point, mild to moderate albuminuria was observed in mice carrying Nephrin-Cre; Foxc1+/flox; Foxc2flox/flox, but not in mice carrying Nephrin-Cre; Foxc1flox/flox; Foxc2+/flox (Fig. 8). These mice showed mild histological abnormality including focal proteinaceous casts, mild dilatation of tubules and occasionally vacuolated podocytes (Supplemental Fig. 4). Their littermates carrying other genotypes including Nephrin-Cre; Foxc1+/+; Foxc2flox/flox had no proteinuria.
ular lesions are fully established due to systemic complications of massive fluid accumulation. Deletion of Foxc1/2 from maturing podocytes by the Nephrin-Cre resulted in early onset of podocyte injury and death before weaning. The results indicate that these transcription factors have important roles in podocyte maturation and maintenance. Gene expression profile analysis showed several hundred genes were differently regulated by Foxc1/2 knockout. Quantitative PCR analysis confirmed that podocin, and Cxcl12 were down-regulated after the deletion, indicating that Foxc1/2 maintain expression of these genes. Podocin is a podocyte specific protein necessary to form slit diaphragm. Deletion of this gene results in massive proteinuria and nephrotic syndrome [13]. Of note, Foxc1/2 binding motif in podocin promoter has been reported to control the promoter activity [21]. Cxcl12 is expressed in podocyte and its receptor Cxcr4 is expressed in glomerular endothelial cells. Deletion of either of these genes results in the same phenotype, i.e., disorganized glomerular tufts [22]. Disorganized glomerular tufts have been reported in Foxc2-/- and Foxc1+/-; Foxc2+/- [9,12]. Interestingly, Foxc1 has been reported to upregulate Cxcl12 in mesenchymal progenitor cells to maintain niches for hematopoietic stem and progenitor cells [23]. Clic5 is an intracellular chloride channel located at the slit diaphragm. Deletion of this gene results in foot process effacement and proteinuria [24,25]. In addition to these evaluated genes, gene expression profiles indicated that other potentially harmful genes, including Cox2, Wnt4, Tgfb3, Ednra and Cnr1 were up-regulated by Foxc1/2 deletion, and that a potentially protective gene, Cfh, was down-regulated by Foxc1/2 deletion (Table 1, Supplemental Table 2 and 3) [26–32]. Thus, Foxc1/2 not only maintain expression of genes indispensable to podocyte integrity but also suppress potentially harmful genes. Disruption of both Foxc2 alleles and one Foxc1 allele led to podocyte injury. The severity of podocyte injury was most extensive in Foxc1-/-Foxc2-/-, then Foxc1-/+Foxc2-/-, then Foxc1-/-Foxc2+/indicating that Foxc2 has a predominant role in maturation and maintenance of podocytes. Such dose-dependency of Foxc1/2 and a predominant role for Foxc2 have been reported in endothelial cell specification in the somite [33], development of lymphatic vasculature [34], mesoderm formation [35] and development of cardiovascular system [36–38]. Foxc2 also plays predominant roles in glomerular development. Foxc2-/- mice and Foxc1+/-; Foxc2+/- mice developed disorganized glomerular tufts [9,12] and Pax2-Cre; Foxc2flox/flox developed glomerular cysts [10] while renal parenchyma of Foxc1 ch mutants was normal [7]. Disorganized glomerular tufts and glomerular
4. Discussion The present study demonstrates that deletion of Foxc1/2 in adulthood causes massive proteinuria. The proteinuria was associated with morphological changes in the kidneys including proteinaceous casts and protein reabsorption droplets in dilated tubules, severe podocyte injury, glomerular hyalinosis and sclerosis. Of note, structural morphology in non-renal tissues was normal despite the recognized critical role of these transcription factors in development of the eye, the cardiovascular system and the lymphatic vessels [3–6]. Huge vacuoles in podocytes observed in the present study have previously been described by us in another proteinuric model, NEP25 transgenic mice treated with immunotoxin [18]. NEP25 mice express human CD25 selectively in podocytes and immunotoxin binding to human CD25 causes damage to podocytes. NEP25 mice treated with moderate to high dose immunotoxin develop massive proteinuria, edema, and die 1 week after the immunotoxin-induced podocyte injury. Notably, tubular anomalies in the present study, including proteinaceous casts and protein reabsorption droplets in dilated tubules are also frequently found in the NEP25 although podocyte adhesion to parietal epithelial cells and proliferation of parietal epithelial cells were less frequent in the present study. We confirmed severe podocyte injury by the TEM and desmin staining which is recognized as a sensitive maker of podocyte damage and phenotypic changes [19,20]. Therefore, Foxc1/ 2 are crucial to maintain the podocyte integrity. Despite severe podocyte damages, glomerular sclerosis or hyalinosis was less frequent in mice in the current study. Although it took several weeks to develop podocyte injury, these histological findings indicate that podocyte degeneration occurred quite rapidly but mice died before development of full-blown glomerular sclerosis. Thus, like NEP25, it is possible that ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice die before glomer270
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
Fig. 8. Foxc1/2 dose-dependency of podocyte injury in Nephrin-Cre; Foxc1/2 mice. Mild to moderate albuminuria developed in 5 of 6 Nephrin-Cre; Foxc1+/flox; Foxc2flox/flox mice at 8 weeks of age but not in Nephrin-Cre; Foxc1flox/flox; Foxc2+/flox mice.
the tamoxifen-treatment (Supplemental Fig. 5), Foxc1 also contributes to maintain podocytes. However, Foxc2 is predominant in maturation and maintenance of podocytes. Mutations in FOXC2 cause the lymphedema-distichiasis syndrome, an autosomal dominant disease. Four of 6 members in a family of German-Irish descent with lymphedema-distichiasis syndrome have been reported to have renal disease [39]. All affected members carry a frame shift mutation in the FOXC2 gene and homozygous T allele of 512C-T polymorphism in the 5-prime UTR of the FOXC2 gene. The authors speculate that a combination of mutations in FOXC2 is the cause of renal diseases. Alternatively, another genetic or environmental factor may be involved in the pathogenesis of renal disease in this family. Even though the impact of a single gene mutation in Foxc1/2 on renal disease may be subtle, these genes, especially Foxc2, may contribute to deterioration in renal function with aging or primary nephropathies. The current era of routine whole-genome sequencing and diagnostics may soon uncover susceptibility genes that have subtle effects on progression of chronic renal diseases including Foxc2. Our results indicate that Foxc1 and Foxc2 cooperatively maintain podocytes by regulating podocyte gene expression, and Foxc2 plays predominant roles in podocytes. Disclosures None. Acknowledgement We thank Dr. Valentina Kon for helpful discussions during a preparation of this manuscript. The authors wish to acknowledge Support Center for Medical Research and Education, Tokai University for technical support in gene expression profiling and animal care. This study was supported by Grant-in Aid for Scientific Research 25461234, 22590899 (M.M.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We also thank Ms. Shiho Imai and Ms. Chie Sakurai for excellent technical assistance.
Fig. 7. Development of proteinuria in Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice at 2 weeks of age. Urine samples were subjected to SDS-PAGE (A). Compared to control mice without Cre (-Cre), Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice (+Cre) had massive non-selective proteinuria. Arrow head shows the position of albumin. Renal histology of control mice (B) is normal, whereas, Nephrin-Cre; Foxc1flox/flox; Foxc2flox/flox mice (C) showed protein reabsorption droplets and proteinaceous casts and mild tubular dilatation. Degeneration of podocyte was not so obvious compared to tamoxifen-treated ROSA26-CreERT2; Foxc1flox/flox; Foxc2flox/flox mice with massive albuminuria. Scale bar =100 µm.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.yexcr.2017.02.016. References [1] B.A. Benayoun, S. Caburet, R.A. Veitia, Forkhead transcription factors: key players in health and disease, Trends Genet. 27 (2011) 224–232. [2] K., H. Kaestner, W. Knochel, D.E. Martinez, Unified nomenclature for the winged helix/forkhead transcription factors, Genes Dev. 14 (2000) 142–146. [3] R. Rice, D.P. Rice, B.R. Olsen, I. Thesleff, Progression of calvarial bone develop-
cysts were not observed in the present study and may reflect the fact that fundamental glomerular structures might already be established when Cre was expressed or activated. Since ROSA26-CreERT2; Foxc1flox/flox; Foxc2+/- mice developed microalbuminuria 1 year after 271
Experimental Cell Research 352 (2017) 265–272
M. Motojima et al.
[21] B. He, L. Ebarasi, Z. Zhao, J. Guo, J.R. Ojala, K. Hultenby, S. De Val, C. Betsholtz, K. Tryggvason, Lmx1b and FoxC combinatorially regulate podocin expression in podocytes, J. Am. Soc. Nephrol. 25 (2014) 2764–2777. http://dx.doi.org/10.1681/ ASN.2012080823. [22] Y. Takabatake, T. Sugiyama, H. Kohara, T. Matsusaka, H. Kurihara, P.A. Koni, Y. Nagasawa, T. Hamano, I. Matsui, N. Kawada, E. Imai, T. Nagasawa, H. Rakugi, Y. Isaka, The CXCL12 (SDF-1)/CXCR4 axis is essential for the development of renal vasculature, J. Am. Soc. Nephrol. 20 (2009) 1714–1723. http://dx.doi.org/ 10.1681/ASN.2008060640. [23] Y. Omatsu, M. Seike, T. Sugiyama, T. Kume, T. Nagasawa, Foxc1 is a critical regulator of haematopoietic stem/progenitor cell niche formation, Nature 508 (2014) 536–540. http://dx.doi.org/10.1038/nature13071. [24] B.A. Pierchala, M.R. Muñoz, C., C. Tsui, Proteomic analysis of the slit diaphragm complex: CLIC5 is a protein critical for podocyte morphology and function, Kidney Int. 78 (2010) 868–882. http://dx.doi.org/10.1038/ki.2010.212. [25] B. Wegner, A. Al-Momany, S.C. Kulak, K. Kozlowski, M. Obeidat, N. Jahroudi, J. Paes, M. Berryman, B.J. Ballermann, CLIC5A, a component of the ezrinpodocalyxin complex in glomeruli, is a determinant of podocyte integrity, Am. J. Physiol. Ren. Physiol. 298 (2010) F1492–F1503. http://dx.doi.org/10.1152/ ajprenal.00030.2010. [26] S. Agrawal, A.J. Guess, M.A. Chanley, W.E. Smoyer, Albumin-induced podocyte injury and protection are associated with regulation of COX-2, Kidney Int. 86 (2014) 1150–1160. http://dx.doi.org/10.1038/ki.2014.196. [27] H. Cheng, X. Fan, G.W. Moeckel, R.C. Harris, Podocyte COX-2 exacerbates diabetic nephropathy by increasing podocyte (pro)renin receptor expression, J. Am. Soc. Nephrol. 22 (2011) 1240–1251. http://dx.doi.org/10.1681/ASN.2010111149. [28] L. Zhou, Y. Liu, Wnt/β-catenin signalling and podocyte dysfunction in proteinuric kidney disease, Nat. Rev. Nephrol. 11 (2015) 535–545. http://dx.doi.org/10.1038/ nrneph.2015.88. [29] L. Bao, M. Haas, R.J. Quigg, Complement factor H deficiency accelerates development of lupus nephritis, J. Am. Soc. Nephrol. 22 (2011) 285–295. http:// dx.doi.org/10.1681/ASN.2010060647. [30] A.W. Minto, H.M. Wilson, A.J. Rees, R.J. Quigg, P.A. Brown, Selective expression of TGF-beta2 and TGF-beta3 isoforms in early mesangioproliferative glomerulonephritis, Nephron Exp. Nephrol. 96 (e111-e118) (2004) 2004. [31] I. Daehn, G. Casalena, T. Zhang, S. Shi, F. Fenninger, N. Barasch, L. Yu, V. D'Agati, D. Schlondorff, W. Kriz, B. Haraldsson, E.P. Bottinger, Endothelial mitochondrial oxidative stress determines podocyte depletion in segmental glomerulosclerosis, J. Clin. Investig. 124 (2014) 1608–1621. http://dx.doi.org/10.1172/JCI71195. [32] T. Jourdan, G. Szanda, A.Z. Rosenberg, J. Tam, B.J. Earley, G. Godlewski, R. Cinar, Z. Liu, J. Liu, C. Ju, P. Pacher, G. Kunos, Overactive cannabinoid 1 receptor in podocytes drives type 2 diabetic nephropathy, Proc. Natl. Acad. Sci. USA 111 (2014) E5420–E5428. http://dx.doi.org/10.1073/pnas.1419901111. [33] A. Mayeuf-Louchart, D. Montarras, C. Bodin, T. Kume, S.D. Vincent, M. Buckingham, Endothelial cell specification in the somite is compromised in Pax3-positive progenitors of Foxc1/2 conditional mutants, with loss of forelimb myogenesis, Development 143 (2016) 872–879. http://dx.doi.org/10.1242/ dev.128017. [34] A. Fatima, Y. Wang, Y. Uchida, P. Norden, T. Liu, A. Culver, W.H. Dietz, F. Culver, M. Millay, Y.S. Mukouyama, T. Kume, Foxc1 and Foxc2 deletion causes abnormal lymphangiogenesis and correlates with ERK hyperactivation, J. Clin. Investig. 126 (2016) 2437–2451. http://dx.doi.org/10.1172/JCI80465. [35] B. Wilm, R.G. James, T.M. Schultheiss, B.L.M. Hogan, The forkhead genes, Foxc1 and Foxc2, regulate paraxial versus intermediate mesoderm cell fate, Dev. Biol. 271 (2004) 176–189. [36] S. Seo, T. Kume, Forkhead transcription factors, Foxc1 and Foxc2, are required for the morphogenesis of the cardiac outflow tract, Dev. Biol. 296 (2006) 421–436. [37] S. Seo, H. Fujita, A. Nakano, M. Kang, A. Duarte, T. Kume, The forkhead transcription factors, Foxc1 and Foxc2, are required for arterial specification and lymphatic sprouting during vascular development, Dev. Biol. 294 (2006) 458–470. [38] T. Kume, H. Jiang, J.M. Topczewska, B.L. Hogan, The murine winged helix transcription factors, Foxc1 and Foxc2, are both required for cardiovascular development and somitogenesis, Genes Dev. 15 (2001) 2470–2482. [39] C. Yildirim-Toruner, K. Subramanian, L. El Manjra, E. Chen, S. Goldstein, E. Vitale, A novel frameshift mutation of FOXC2 gene in a family with hereditary lymphedema-distichiasis syndrome associated with renal disease and diabetes mellitus, Am. J. Med. Genet. 131A (2004) 281–286.
ment requires Foxc1 regulation of Msx2 and Alx4, Dev. Biol. 262 (2003) 75–87. [4] T. Kume, K.Y. Deng, V. Winfrey, D.B. Gould, M.A. Walter, B.L.M. Hogan, The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus, Cell 93 (1998) 985–996. [5] K. Iida, H. Koseki, H. Kakinuma, N. Kato, Y. Mizutani-Koseki, H. Ohuchi, H. Yoshioka, S. Noji, K. Kawamura, Y. Kataoka, F. Ueno, M. Taniguchi, N. Yoshida, T. Sugiyama, N. Miura, Essential roles of the winged helix transcription factor MFH-1 in aortic arch patterning and skeletogenesis, Development 124 (1997) 4627–4638. [6] G.E. Winnier, L. Hargett, B.L.M. Hogan, The winged helix transcription factor MFH1 is required for proliferation and patterning of paraxial mesoderm in the mouse embryo, Genes Dev. 11 (1997) 926–940. [7] F. Komaki, Y. Miyazaki, F. Niimura, T. Matsusaka, I. Ichikawa, M. Motojima, Foxc1 gene null mutation causes ectopic budding and kidney hypoplasia but not dysplasia, Cells Tissues Organs 198 (2013) 22–27. [8] T. Kume, K. Deng, B.L.M. Hogan, Murine forkhead/winged helix genes Foxc1 (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney and urinary tract, Development 127 (2000) 1387–1395. [9] M. Takemoto, L. He, J. Norlin, J. Patrakka, Z. Xiao, T. Petrova, C. Bondjers, J. Asp, E. Wallgard, Y. Sun, T. Samuelsson, P. Mostad, S. Lundin, N. Miura, Y. Sado, K. Alitalo, S.E. Quaggin, K. Tryggvason, C. Betsholtz, Large-scale identification of genes implicated in kidney glomerulus development and function, EMBO J. 25 (2006) 1160–1174. [10] M. Motojima, S. Ogiwara, T. Matsusaka, S.Y. Kim, N. Sagawa, K. Abe, M. Ohtsuka, Conditional knockout of Foxc2 gene in kidney: efficient generation of conditional alleles of single-exon gene by double-selection system, Mamm. Genome 27 (2016) 62–69. http://dx.doi.org/10.1007/s00335-015-9610-y. [11] A.P. McMahon, B.J. Aronow, D.R. Davidson, J.A. Davies, K.W. Gaido, S. Grimmond, J.L. Lessard, M.H. Little, S.S. Potter, E.L. Wilder, P. Zhang, GUDMAP: the genitourinary developmental molecular anatomy project, J. Am. Soc. Nephrol. 19 (4) (2008) 667–671. http://dx.doi.org/10.1681/ ASN.2007101078. [12] M. Motojima, S. Tanimoto, M. Ohtsuka, T. Matsusaka, T. Kume, K. Abe, Characterization of kidney and skeleton phenotypes of mice double heterozygous for Foxc1 and Foxc2, Cells Tissues Organs 201 (2016) 380–389. http://dx.doi.org/ 10.1159/000445027. [13] F. Grahammer, C. Schell, T.B. Huber, The podocyte slit diaphragm–from a thin grey line to a complex signalling hub, Nat. Rev. Nephrol. 9 (2013) 587–598. http:// dx.doi.org/10.1038/nrneph.2013.169. [14] T. Asano, F. Niimura, I. Pastan, A.B. Fogo, I. Ichikawa, T. Matsusaka, Permanent genetic tagging of podocytes: fate of injured podocytes in a mouse model of glomerular sclerosis, J. Am. Soc. Nephrol. 16 (2005) 2257–2262. [15] J. Seibler, B. Zevnik, B. Küter-Luks, S. Andreas, H. Kern, T. Hennek, A. Rode, C. Heimann, N. Faust, G. Kauselmann, M. Schoor, R. Jaenisch, K. Rajewsky, R. Kühn, F. Schwenk, Rapid generation of inducible mouse mutants, Nucleic Acids Res. 31 (2003) e12. [16] A. Sasman, C. Nassano-Miller, K.S. Shim, H.Y. Koo, T. Liu, K.M. Schultz, M. Millay, A. Nanano, M. Kang, T. Suzuki, T. Kume, Generation of conditional alleles for Foxc1 and Foxc2 in mice, Genesis 50 (2012) 766–774. [17] H. Ueda, Y. Miyazaki, T. Matsusaka, Y. Utsunomiya, T. Kawamura, T. Hosoya, I. Ichikawa, Bmp in podocytes is essential for normal glomerular capillary formation, J. Am. Soc. Nephrol. 19 (2008) 685–694. [18] T. Matsusaka, J. Xin, S. Niwa, K. Kobayashi, A. Akatsuka, H. Hashizume, Q.C. Wang, I. Pastan, A.B. Fogo, I. Ichikawa, Genetic engineering of glomerular sclerosis in the mouse via control of onset and severity of podocyte-specific injury, J. Am. Soc. Nephrol. 16 (2005) 1013–1023. [19] J. Funk, V. Ott, A. Herrmann, W. Rapp, S. Raab, W. Riboulet, A. Vandjour, E. Hainaut, A. Benardeau, T. Singer, B. Jacobsen, Semiautomated quantitative image analysis of glomerular immunohistochemistry markers desmin, vimentin, podocin, synaptopodin and WT-1 in acute and chronic rat kidney disease models, Histochem. Cell Biol. 145 (2016) 315–326. http://dx.doi.org/10.1007/s00418015-1391-6. [20] T. Kakimoto, K. Okada, Y. Hirohashi, R. Relator, M. Kawai, T. Iguchi, K. Fujitaka, M. Nishio, T. Kato, A. Fukunari, H. Utsumi, Automated image analysis of a glomerular injury marker desmin in spontaneously diabetic Torii rats treated with losartan, J. Endocrinol. 222 (2014) 43–51. http://dx.doi.org/10.1530/JOE-140164.
272