Abstract
The glomerular filtration barrier separates the blood from the urinary space. Nephrin is a transmembrane protein that belongs to the immunoglobulin superfamily and is localized to the slit diaphragms that are a critical component of this filtration barrier. Mutations in the nephrin gene (NPHS1) lead to congenital Finnish nephropathy, whereas alterations in the level of nephrin expression have been identified in a wide range of acquired glomerular diseases. A 186-bp fragment from the human NPHS1 promoter is capable of directing podocyte-specific expression of a β-galactosidase transgene when placed in front of a heterologous minimal promoter in transgenic mice. The Wilms tumor suppressor gene (WT1) is a zinc-finger–containing transcription factor that is coexpressed with NPHS1 in differentiated podocytes; gel shift binding assays demonstrate that a recombinant WT1 protein can bind and activate the 186-bp NPHS1 fragment in a sequence-specific manner. Taken together, these results suggest that WT1 may be required for regulation of the NPHS1 gene in vivo.
Kidney disease affects over 20 million individuals in the United States alone and is a major cause of morbidity and mortality worldwide. Although the causes of renal failure are diverse, the glomerulus and renal filtration barrier are often the target of injury (1,2⇓). Despite the significant burden of disease, treatment options are limited. In the majority of cases, damage to the renal filtration barrier leads to progressive glomerular scarring and may result in end-stage kidney failure that requires dialysis or transplantation for survival. The glomerular filtration barrier consists of two individual cell types: glomerular visceral epithelial cells (podocytes), and fenestrated endothelial cells that are separated by a glomerular basement membrane. Approximately 180 L of blood per day passes through the glomerular capillaries in the average adult human kidney. Excess solutes and water are filtered into the urine, while critical blood proteins such as albumin and blood cells remain within the capillaries. Damage to this filtration barrier leads to the loss of protein into the urine (proteinuria).
Nephrin belongs to the immunoglobulin superfamily and is localized to the slit diaphragm, a specialized intercellular connection found between podocyte foot processes (3,4⇓). Nephrin was originally identified as the gene that is mutated in congenital Finnish nephropathy (5), a rare autosomal recessive disease that results in severe proteinuria and death in infants unless they receive a renal transplant. Mutant mice that are missing the nephrin protein fail to form slit diaphragms and die at birth with massive proteinuria and kidney failure (6). Alterations in the level of nephrin expression have also been documented in a variety of acquired renal diseases that include diabetic nephropathy, focal segmental glomerulonephritis, and minimal change nephropathy (7,8⇓). Previous studies have shown that 8.3, 5.4, or 4.125 kb of the murine proximal nephrin promoter and 1.125 kb of the human proximal nephrin promoter are capable of directing expression of a β-galactosidase transgene to podocytes in vivo (9–11⇓⇓). Although the human fragment was exclusively expressed in podocytes, the larger fragments directed expression to the developing hindbrain as well. Recently, an alternate first exon that is required for neural expression of the nephrin gene has been identified (12).
By using the NCBI Pairwise-Blast sequence alignment program, we identified an 83-bp region of homology between the 4.125-kb murine and 1.125-kb human promoters. We found that a 186-bp fragment of the human promoter that contains this 83-bp region of homology is capable of directing podocyte-specific expression in transgenic mice. Furthermore, we demonstrate that the Wilms tumor suppressor protein can bind to a putative WT1 consensus sequence from this fragment in vitro and activates a nephrin-luciferase reporter transgene. By using this enhancer, it will be possible to identify additional trans-acting factors that are required for podocyte-specific gene regulation with the goal of finding new molecular targets for the treatment of renal patients.
Materials and Methods
Identification of Homology between the Human and Murine Promoters
We have previously shown that 4.125-kb of the murine nephrin promoter and 1.125-kb of the human nephrin promoter are capable of directing expression of a reporter transgene (β-galactosidase) specifically to podocytes in vivo. To identify regions of homology within these DNA fragments, the sequences from these two promoters were aligned by the NCBI program (Pairwise BLAST; available at http://www.ncbi.com).
Generation of the Nephrin-LacZ Transgene
A total of 186 bp of the human nephrin promoter encoding the 83-bp core region of homology was amplified from the 1.125-kb human nephrin promoter fragment with the following primers: 5′-GGCCCTGGGGTCACGGAGGCTGGGGA-3′ and 5′-CGGGTCGACCTCGTATTTTAGGGGA-3′ (Sigma); and under the following PCR conditions: 95°C × 30 s; 95°C × 1′, 68°C × 4′ for 30 cycles followed by 72°C for 4′. This 186-bp fragment was subcloned into the PCR 2.1 vector (Invitrogen) according to manufacturer’s directions. After digestion with EcoR1, either four copies or one copy of this fragment were subcloned into the hsp-lacZ vector that has been previously described (13). Both one- and four-copy transgenes were constructed to determine whether tandem repeats or a single copy of the enhancer alone were required for expression. This plasmid encodes the minimal promoter from the heat shock protein upstream of a β-galactosidase transgene (Figure 1). Both one-copy and four-copy transgenes were injected into one cell ICR (Charles River) murine embryos as described (14). To identify founder lines, genomic DNA was purified from tails as described (15), and Southern blot analysis was performed. The lacZ gene was used as the hybridization probe as described previously (10).
Figure 1. Four copies or one copy of the 186-bp human NPHS1 fragment (En) were subcloned into EcoR1 sites upstream of a heterologous minimal promoter from the heat shock promoter (hsp) and a β-galactosidase (lacZ) transgene. The hsp-lacZ transgene has previously been described (13). All of the nephrin enhancer elements were subcloned from the 5′ to 3′ direction.
β-Galactosidase Staining of Kidneys and Extrarenal Tissue
Kidneys or tissues from newborn pups were fixed and stained as described previously (10,16⇓). In addition, 10 μm cryosections of each of these tissues were prepared and stained as described previously (17).
Electrophoretic Mobility Shift Assays
A double-stranded DNA probe corresponding to the 186-bp human nephrin promoter fragment was released from pKS4 × 191 plasmid by EcoRI digestion and end-labeled with [∝-32P] dATP by Superscript II (GibcoBRL). Cloning and expression of GST fusion proteins GST-WT1 (A isoform) and GST-ΔZF (WT1 lacking DNA-binding zinc fingers) was previously described (18). DNA binding reactions were performed with 1 μg of purified protein in a total volume of 25 μl containing radiolabeled probe (50,000 cpm), 1 μg poly (dI-dC), 1 mg/ml BSA, and binding buffer (50 mM Tris (pH 8.8), 60 mM NaCl, 5 mM MgCl2, 100 μM ZnCl2, 12% glycerol, 0.5 mM DTT). The reaction mixture was preincubated at room temperature for 10 min before addition of probe. Then protein/DNA complexes were allowed to form for 15 min at room temperature and were resolved on a 5% nondenaturing polyacrylamide gel in 0.5× TBE at 300 V for 1.5 h. Gel was dried and exposed to film. For competition experiments, 1.2 μg of anti-WT1 antibody (C-19, Santa Cruz) or rabbit IgG (Zymed) and a 100- to 1000-fold excess of unlabeled Zif3 (TCGAGCCCGGCGCGGGGGCGAGGGTCGA) or NFκB (AGCTT GGCATAGGTCCTCGGC) double-stranded oligonucleotides were added during the preincubation period.
Plasmid Construction
Construction of WT1-A and WT1-A (F112Y) expression vectors was previously described (18). To make the human nephrin luciferase reporter construct (pGL2-neph-luc) containing four copies of the human nephrin enhancer, pKS-4 × 191 was digested with KpnI/ BamHI and the insert cloned into a promoterless luciferase vector pGL2-Basic (Promega) digested with KpnI/BglII. The plasmid was confirmed by sequencing.
Luciferase Assay
For transient cotransfection assays, WT1-A wild-type or F112Y mutant expression vectors (150, 495, 1600 ng), pGL2-neph-luc (150 ng), and an internal Renilla control (5 ng) were transfected into 293T cells with FuGENE6 reagent (Roche) according to manufacturer’s instructions. Cells were harvested 48 h later and assayed for luciferase activity with the Dual-Luciferase Reporter Assay System (Promega).
Results
Identification of a Podocyte-Specific Enhancer
To identify a putative glomerular enhancer element, the human and murine promoters that have been shown to direct podocyte-specific expression within transgenic kidneys were compared. By using the NCBI Pairwise BLAST program, an 83-bp region of homology was identified (Figure 2A). At the nucleotide level, there was 83% identity. There were no other conserved regions identified. This region was 743 bp upstream of the glomerular transcription initiation site for the human nephrin gene and 444 bp upstream of the predicted initiation site for the murine nephrin gene (Figure 2B). The TF TRANSFAC database was used to identify putative consensus binding sites for transcription factors, an E-box site and retinoic acid receptor sites were identified, and a putative WT1-binding site was identified by visual inspection (Figure 3).
Figure 2. (A) An 83-bp region of homology was identified between the human and mouse NPHS1 promoters. There is 83% identity between these two regions. (B) The homologous region resides 743 bp and 444 bp upstream of the initiating ATG in the human (accession no. AC002133) and murine (accession no. AF190638.1) NPHS1 genes, respectively.
Figure 3. Consensus binding sites for transacting factors. The TF TRANSFAC database was used to identify putative consensus binding sites for transcription factors. An E-box (single underline) and retinoic acid receptor binding sites (double underline) were identified. In addition, a putative WT1-binding site was identified (bold letters). The core 83-bp region of homology is shown in larger type; the flanking nucleotides were present in the transgene.
Transgenic Founder Lines Exhibit Podocyte-Specific Expression
Four independent founder mice that carried four copies of the nephrin “enhancer” and the hsp-lacZ transgene were identified. LacZ staining confirmed that 50% of these independent founders expressed β-galactosidase specifically in podocytes within the kidney from the late S-shape stage of glomerulogenesis onward (see Figure 4). The other two founder mice did not demonstrate any lacZ staining. In addition, two of four founder mice that carried one copy of the nephrin enhancer demonstrated podocyte-specific expression. One of these founder mice exhibited mosaic expression as only a subset of podocytes demonstrated β-galactosidase activity (Figure 4C). Transgene copy number was estimated by dot blot analysis. There was no correlation between copy number and presence or absence of lacZ staining (data not shown).
Figure 4. LacZ staining was specific to podocytes within transgenic kidneys. (A) Whole-mount kidney demonstrates lacZ expression within glomeruli (blue “balls”) of a transgenic newborn mouse kidney (original magnification, ×25). (B) Cryosection of a newborn kidney demonstrates lacZ staining in a late S-shape stage glomerulus in the nephrogenic zone (original magnification, ×250). The blue cells are podocyte precursors. (C) Cryosection of a mature glomerulus from a newborn transgenic kidney demonstrates lacZ staining specifically within podocytes (po) (original magnification, ×200). Notably, only a subset of podocytes are staining in this mosaic transgenic line.
To determine whether there was extrarenal expression of the transgene, all tissues from the newborn founder pups were stained with lacZ. Notably, there was no neural expression in any of the mice. In one of the lines that carried four copies of the transgene, there was also staining in whisker follicles and in the epicardium of the heart.
WT1 Binds to the Enhancer Region In Vitro
To demonstrate a direct interaction between WT1 protein and the nephrin promoter, an electrophoretic mobility shift assay was performed (Figure 5). A DNA probe corresponding to the 186-bp nephrin “enhancer” was incubated with bacterial expressed and purified GST proteins, and subjected to electrophoresis. No complex was formed with GST alone or GSTΔZF (lanes 1 and 3); however, a specific protein-DNA complex was formed when the DNA probe was incubated with GSTWT1 (lane 2). The complex was abolished by incubation with increasing amounts of excess cold competitor probe containing a canonical WT1 binding site (lanes 6 to 7) but was unaffected by a probe that did not contain WT1 binding sites (lanes 4 to 5). Formation of the Neph186-WT1 complex was specifically reduced by a WT1 antibody (lane 9) but unaffected by equal amounts of control IgG (lane 8).
Figure 5. Gel shift binding assay. WT1 binds specifically to the nephrin promoter fragment that directs podocyte-specific expression in transgenic mice. A DNA probe corresponding to a 186-bp fragment of the human nephrin promoter was incubated with GST (lane 1), GSTWT1 (lanes 2 and 4 to 9) or GSTΔZF (lane 3). For competition experiments WT1 was preincubated with unlabeled competitor probes NFκB (lanes 4 to 5) and Zif3 (lanes 6 to 7) and rabbit IgG (lane 8) or a WT1 antibody (lane 9).
WT1 Activates a Nephrin-Luciferase Reporter Transgene
To determine whether WT1 directly activates nephrin expression, four copies of the 186-bp nephrin enhancer were inserted into a promoter-less luciferase reporter (pGL2-neph-luc). The nephrin luciferase reporter was transfected into 293T cells along with expression vectors for wild-type and mutant WT1-A isoform (WTA) (Figure 6). WTA activated the promoter up to 30-fold compared with empty vector control. As expected, WTA-112, a tumor associated missense mutant that is deficient for transcriptional activation, did not activate the nephrin enhancer (18).
Figure 6. Transcriptional activation of the human nephrin “enhancer” by WT1. The pGL2-neph-luc reporter plasmid (150 ng) was transiently cotransfected into 293T cells along with increasing concentrations of empty vector, vector expressing WTA or WTA-112, and a Renilla internal control (5 ng). At 48 h after transfection, a dual luciferase assay was performed. Luciferase data are expressed as fold activation (± SD) relative to empty expression vector and is representative of four experiments each performed in triplicate.
Discussion
Glomerular injury is a leading cause of end-stage kidney failure worldwide. Unfortunately, few successful therapies exist to treat damaged glomeruli. Increased understanding of glomerular-specific gene regulation should provide insight into the molecular basis of renal disease. We demonstrate here that 186 bp of the human nephrin promoter is capable of directing gene expression specifically to the developing and mature podocyte in transgenic mice. We chose to examine this fragment because it is highly conserved between the human and mouse nephrin promoters, with over 80% identity at the nucleotide level. Although mutations have been reported in the nephrin promoter region in three patients with congenital Finnish nephropathy (19,20⇓), they do not occur within this 186-bp region. After the identification of the nephrin gene, another podocyte-restricted gene named podocin was identified as the gene responsible for some cases of autosomal recessive focal segmental glomerulosclerosis (21). Moeller, et al. (22) have previously identified a 2.5-kb fragment of the podocin gene that is also capable of podocyte-specific expression. By using the human genome database and the pairwise BLAST program at NCBI, we also compared the 186-bp fragment to the podocin promoter but were unable to identify a homologous region and suggests that regulation of these two genes may differ. Accordingly, Miner, et al. (23) have shown that podocin expression disappears in lmx1b (a lim-homeodomain containing transcription factor) null mice but nephrin levels remain unaffected.
Only 50% of the founder mice expressed the β-galactosidase transgene, and this did not correlate with transgene copy number. The most likely explanation for lack of expression in some founder mice is the site of integration of the transgene within the genome that is not controlled in these experiments. Of note, lacZ expression was detected as early as late S-shape stage glomeruli when podocytes are not fully differentiated. In contrast, both the 1.125-kb human and the 4.125-kb mouse promoter are incapable of driving expression in podocyte precursors (S-shape stage). Why this discrepancy in temporal expression exists is not clear, but endogenous nephrin is expressed in S-shape stage podocytes. Tryggvason and coworkers have reported that kidney regulatory elements reside within the fragment between −2.1 to −4 kb of the mouse nephrin gene, which does not contain the 186-bp fragment reported here (12,22⇓). In a separate paper, the team of Moeller, et al. (22), by using 1.25 kb of the proximal murine nephrin promoter, were able to show podocyte-specific expression. It is likely that endogenous regulation of nephrin expression requires more than one regulatory element.
To date, a number of transcription factors have been identified that are expressed within the developing and mature podocyte (for review, see (24,25⇓)). Null mutations in most of these genes have been generated by using homologous recombination and gene targeting in embryonic stem cells in mice. However, no transcription factor has been shown to be essential for nephrin expression, although its expression is reduced in DNA-binding mutants of the Kreisler (Krml1/MafB) gene (26) and retinoic acid signaling leads to activation of the murine NPHS1 promoter in cultured podocytes (27). The Wilms tumor suppressor protein is a zinc-finger containing transcription factor that is required for early nephrogenesis; null mutant mice die during midgestation with multiple defects and fail to form kidneys or gonads, precluding the analysis of nephrin expression in glomeruli (28). In addition to its expression in the metanephric mesenchyme, WT1 becomes restricted to differentiating and mature podocytes later in nephrogenesis. Rescue experiments that used a human WT1-containing YAC in WT1 null embryos results in mesangial sclerosis or crescentic glomerulonephritis depending on the gene dose (29,30⇓), suggesting that WT1 is required for normal podocyte differentiation and function. Furthermore, patients with point mutations in the WT1 gene develop Denys-Drash syndrome, which is characterized by glomerular defects and nephrotic syndrome, genital anomalies, and Wilms tumor. Also, mice carrying a common Denys-Drash syndrome mutation develop glomerulosclerosis (31). Finally, WT1 mutations have been identified in patients with nephrotic syndrome and sporadic glomerulosclerosis. Although WT1 transcriptional targets were identified in vitro and in vivo (32,33⇓), downstream targets for WT1 in the podocyte have remained elusive. However, NPHS1 expression was dramatically decreased in mice with reduced levels of the WT1 protein (29), suggesting that nephrin is downstream of WT1 and in Drosophila, the WT1 and nephrin orthologues—Klumpfuss (34,35⇓) and IrreC (36)—were shown to interact genetically. By using the TRANSFAC database and visual inspection of the 83-bp core fragment of homology in the human and murine nephrin promoters, we identified a putative WT1 binding element. Gel shift assays demonstrated that a recombinant WT1 (-KTS) (A isoform) (37), does bind specifically to the nephrin enhancer because it can be competed by excess cold oligonucleotide and can be reduced with antisera to WT1. Furthermore, WT1 is able to transactivate a nephrin-enhancer luciferase reporter gene in vitro.
In summary, we report the first glomerular specific enhancer and demonstrate that WT1 can bind and activate this enhancer in vitro. Identification of this enhancer should facilitate the identification of other cis elements required for glomerular-specific expression through standard approaches such as footprinting and consensus sequence analysis. This element may also be used as the “bait” in yeast-one-hybrid assays or in DNA-Sepharose column purification experiments to pull out additional transacting factors (38) required for nephrin expression. Finally, future studies will determine whether NPHS1 is a direct target for WT1 in vivo.
Acknowledgments
Gordon Guo and Debra J. Morrison contributed equally to this article. We thank Lois Schwartz for excellent technical help and Dragana Vukasovic for expert secretarial support. We also thank Vera Eremina for helpful comments. S.E. Quaggin is the recipient of a Canada Research Chair Tier II and a Premier’s Research of Excellence Award. This work was supported by a CIHR grant to S.E. Quaggin and NIH grants CA102270 and CA599968 to JDL.
- © 2004 American Society of Nephrology
References
Jump to section
More in this TOC Section
Cited By...
- A KDM6A-KLF10 reinforcing feedback mechanism aggravates diabetic podocyte dysfunction
- Wilms' tumour 1 (WT1) in development, homeostasis and disease
- Human Induced Pluripotent Stem Cell-Derived Podocytes Mature into Vascularized Glomeruli upon Experimental Transplantation
- Essential Role of X-Box Binding Protein-1 during Endoplasmic Reticulum Stress in Podocytes
- Integration of Cistromic and Transcriptomic Analyses Identifies Nphs2, Mafb, and Magi2 as Wilms' Tumor 1 Target Genes in Podocyte Differentiation and Maintenance
- Genome-Wide Analysis of Wilms' Tumor 1-Controlled Gene Expression in Podocytes Reveals Key Regulatory Mechanisms
- A Novel Missense Mutation of Wilms' Tumor 1 Causes Autosomal Dominant FSGS
- Mutual Antagonism of Wilms' Tumor 1 and {beta}-Catenin Dictates Podocyte Health and Disease
- Overexpression of Mafb in Podocytes Protects against Diabetic Nephropathy
- TGF-{beta}-Activated Kinase 1 Is Crucial in Podocyte Differentiation and Glomerular Capillary Formation
- Mechanisms of transcriptional regulation by WT1 (Wilms' tumour 1)
- Functions of the podocyte proteins nephrin and Neph3 and the transcriptional regulation of their genes
- Semaphorin3a regulates endothelial cell number and podocyte differentiation during glomerular development
- WT1 Induction of Mitogen-Activated Protein Kinase Phosphatase 3 Represents a Novel Mechanism of Growth Suppression
- Development of the renal glomerulus: good neighbors and good fences
- Identification of Novel Wilms' Tumor Suppressor Gene Target Genes Implicated in Kidney Development
- Nephrotic Syndrome in the First Year of Life: Two Thirds of Cases Are Caused by Mutations in 4 Genes (NPHS1, NPHS2, WT1, and LAMB2)
- Parietal Podocytes in Normal Human Glomeruli
- The Notch-effector HRT1 gene plays a role in glomerular development and patterning of the Xenopus pronephros anlagen
- Comparative promoter analysis allows de novo identification of specialized cell junction-associated proteins
- WT1 Induces Apoptosis through Transcriptional Regulation of the Proapoptotic Bcl-2 Family Member Bak