The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes

Patients with nail-patella syndrome often suffer from a nephropathy, which ultimately results in chronic renal failure. The finding that this disease is caused by mutations in the transcription factor LMX1B, which in the kidney is expressed exclusively in podocytes, offers the opportunity for a better understanding of the renal pathogenesis. In our analysis of the nephropathy in nail-patella syndrome, we have made use of the Lmx1b knockout mouse. Transmission electron micrographs showed that glomerular development in general and the differentiation of podocytes in particular were severely impaired. The glomerular capillary network was poorly elaborated, fenestrae in the endothelial cells were largely missing, and the glomerular basement membrane was split. In addition podocytes retained a cuboidal shape and did not form foot processes and slit diaphragms. Expression of the α4 chain of collagen IV and of podocin was also severely reduced. Using gel shift assays, we demonstrated that LMX1B bound to two AT-rich sequences in the promoter region of NPHS2, the gene encoding podocin. Our results demonstrate that Lmx1b regulates important steps in glomerular development and establish a link between three hereditary kidney diseases: nail-patella syndrome (Lmx1b), steroid-resistant nephrotic syndrome (podocin), and Alport syndrome (collagen IV α4).

Nail-patella syndrome (also called hereditary onycho-osteodysplasia) is a relatively rare hereditary disease that is primarily characterized by dysplastic nails and missing or hypoplastic patellae. One-third to one-half of the patients, however, also suffer from hematuria and proteinuria and may ultimately even develop chronic renal failure (1). In some affected kidneys, focal glomerulosclerosis, a thickened glomerular basement membrane (GBM), and protein casts can already be seen by light microscopy. Electron microscopic investigations almost invariably have revealed a GBM, which contains both fibrillar inclusions and electron-lucent areas and thus presents with a moth-eaten appearance (e.g., refs. 2, 3). Although one publication reports the increased deposition of fibrinogen (4) and another finds evidence for the decreased cross-linking of collagen IV (5), the pathogenesis of the glomerular phenotype remains unclear.

In 1998 the first mutations that lead to the development of nail-patella syndrome were reported in the LMX1B gene (6–8). The findings in patients are corroborated by the fact that the inactivation of Lmx1b in mice also leads to a phenotype strongly resembling nail-patella syndrome (9). The LMX1B protein contains two zinc-binding LIM domains at the NH₂-terminus and a homeodomain in the middle; while the Lin-11, Is1-1, Mec-3 (LIM) domains are felt to represent the interface for the interaction with other proteins, the homeodomain is responsible for the binding to DNA. The mutations described so far lead to the absence or the inactivation of the homeodomain, so that the mutated protein will no longer be able to recognize its target genes. Since a mutant LMX1B gene has been described that contains a stop codon in the region coding for the first LIM domain and therefore should result in the synthesis of only a very short peptide (8), a dominant negative mechanism for the development of the syndrome is unlikely. One rather has to assume haploinsufficiency, which, however, contrasts with the finding that in the mouse both alleles of the Lmx1b gene have to be inactivated to generate a phenotype (9). We have made use of these Lmx1b knockout mice in order to better understand the glomerular phenotype, and we demonstrate a link between nail-patella syndrome, Alport syndrome, and steroid-resistant nephrotic syndrome.

Morphological characterization of the glomerular phenotype. It could already be noticed by gross inspection that the kidneys from Lmx1b^–/– mice were smaller than those from Lmx1b^+/+ animals, suggesting that renal development in the homozygous knockout animals was lagging behind. An overview of respective kidney sections pointed in the same direction, since there appeared to be less mature glomeruli on kidneys from Lmx1b^–/– mice. When the mean glomerular tuft area was determined, it was significantly smaller in Lmx1b^–/– mice (1,426 μm²) than in Lmx1b^+/+ mice (1756 μm²) and Lmx1b^+/– mice (1782 μm²) (P < 0.05, ANOVA).

Transmission electron microscopy of glomeruli from Lmx1b^–/– mice clearly showed that the glomeruli were not fully differentiated. In contrast to wild-type kidneys, in which a branched capillary network had developed in the juxtamedullary glomeruli (Figure 1a), the capillaries in the glomeruli from homozygous knockout mice had not branched (Figure 1b). Furthermore, we noticed that the glomerular endothelial cells of the knockout mice were less fenestrated (Figure 1, c and d). In order to gather preliminary evidence for why no branching had taken place and why the endothelium contained less fenestrae, we conducted an in situ hybridization and RNase protection assay for VEGF. Although VEGF mRNA was detected in podocytes from both Lmx1b^+/+ and Lmx1b^–/– mice by in situ hybridization (Figure 1, e and f), in many (but not all) cases there was less VEGF mRNA in kidneys from the knockout mice by RNase protection assay (Figure 1g).

Figure 1

Development of the glomerular capillaries in Lmx1b^–/– mice. A comparison between newborn wild-type (a) and homozygous knockout (b) mice demonstrates the severely retarded outgrowth of glomerular capillaries when the Lmx1b gene is inactivated. Furthermore, a fibrillar material can be detected in Bowman’s space and adjoining proximal tubules of Lmx1b^–/– mice (asterisk in b). At a higher magnification, the reduction and sometimes even complete lack of fenestrae in the endothelium can be seen in the knockout mice (d) (arrows in c point to fenestrae in the glomerular endothelium of wild-type mice). Although by in situ hybridization VEGF mRNA can be detected in the podocytes of both wild-type (e) and Lmx1b^–/– (f) mice, an RNase protection assay with 20 μg of total kidney RNA (prepared from five Lmx1b^+/+ and four Lmx1b^–/– mice) demonstrates reduced levels of VEGF mRNA in kidneys from the knockout mice (tRNA served as a negative control). A protection assay with a probe directed against 18S rRNA shows that the RNA concentration was determined correctly (g). Bar: 20 μm (a, b, e, and f), 0.5 μm (c), 4 μm (d).

The podocytes of homozygous knockout mice were still cuboidal and had not flattened out over the GBM (Figure 2a). No foot processes and no slit diaphragms were noticed even in the most advanced glomeruli; neighboring podocytes were rather connected by structures resembling adherens junctions (Figure 2c). In addition, the GBM was split (Figure 2b), and a fibrillar material and even erythrocytes were found in Bowman’s space and the lumen of adjoining portions of proximal tubules (Figures 1b and 2d). The fibrillar material was striated with a periodicity of about 17 nm (Figure 2e), indicating that it represents fibrin (21).

Figure 2

Podocyte differentiation in Lmx1b^–/– mice. Podocytes in newborn homozygous knockout animals do not spread out over the GBM (arrows) but retain a cuboidal shape (a). In many areas the GBM was split (b shows a transition between a one-layered GBM marked by arrows and a split GBM marked by arrowheads). In no glomerulus did we notice foot processes and slit diaphragms, but adjacent podocytes were connected by structures resembling adherens junctions (c; note the cytoplasmic plaques outlined by arrows on either side of the junction). In some glomeruli we also observed erythrocytes (E) and fibrillar material (arrows) in Bowman’s space (d), which upon higher magnification showed a striated pattern reminiscent of fibrin (e). Bar: 2 μm (a), 1 μm (b and d), 100 nm (c and e).

In humans, nail-patella syndrome is inherited in an autosomal-dominant fashion, but even after 1 year no morphological abnormalities were found in the glomeruli of heterozygous Lmx1b knockout mice (data not shown). Additional experiments will have to be performed to determine whether breeding the mutation onto a different background will yield a phenotype in heterozygous mice.

α3 Integrin, podocalyxin, and nephrin, but not podocin, are expressed by podocytes of Lmx1b^–/– mice. The lack of foot processes and slit diaphragms prompted us to investigate the expression pattern of podocyte surface proteins, which have been shown to contribute to the establishment of the glomerular filtration barrier. Both the lack of the α3 integrin subunit (22) and the lack of the highly charged protein podocalyxin (23) in mice lead to the absence of foot processes. By immunohistochemistry, however, α3 integrin and podocalyxin could still be easily detected in the podocytes of homozygous knockout mice (Figure 3, a–d). Mutations in NPHS1, the gene coding for nephrin, are responsible for congenital nephrotic syndrome of the Finnish type (24). Nephrin has been localized to the slit diaphragm (11, 25, 26), and patients with this disease typically lack a slit diaphragm (27). It was therefore somewhat surprising to find out that nephrin also was still present in the glomeruli of Lmx1b^–/– mice (Figure 3, e and f).

Figure 3

Expression of α3 integrin, podocalyxin, and nephrin in newborn Lmx1b^–/– mice. By immunohistochemistry, α3 integrin (a and b), podocalyxin (c and d), and nephrin (e and f) were all detected in the glomeruli of wild-type (a, c, and e) and homozygous knockout mice (b, d, and f). Immunohistochemical detection of the proteins in a and b was done with DAB, in c–f by fluorescence. Bar: 20 μm (a–f).

The result was very different when the expression pattern of podocin, which is mutated in patients with steroid-resistant nephrotic syndrome (28–31), was examined. While podocin protein could be easily detected by immunohistochemistry in the glomeruli of wild-type mice (Figure 4a), it was absent in the glomeruli of Lmx1b^–/– mice (Figure 4b). This difference could have been caused by an increased instability of the protein due to the absence of foot processes. Therefore we carried out an in situ hybridization with podocin antisense RNA to determine whether podocin mRNA was present in podocytes, but we were unable to find any (Figure 4, c and d). In order to confirm those histochemical findings, an RNase protection assay was performed with RNA isolated from newborn mouse kidneys. Again, no podocin mRNA was detected in kidneys of Lmx1b^–/– mice (Figure 4e).

Figure 4

Expression of podocin in newborn Lmx1b^–/– mice. Podocin protein (a) and mRNA (c) were present only in podocytes of wild-type mice, not in those of homozygous knockout mice (b and d). The specificity of these results was corroborated by an RNase protection assay with podocin antisense RNA, where protected bands were observed only with RNA isolated from kidneys of wild-type mice (tRNA served as a negative control). A protection assay with a probe directed against 18S rRNA shows that the RNA concentration was determined correctly (e). Immunohistochemical detection of the proteins was done with DAB, and mRNA was detected by nonradioactive in situ hybridization (in order to demonstrate the negative glomerulus, the picture shown in d was taken using Nomarski optics). Bar: 20 μm (a and b), 40 μm (c and d).

Since the absence of podocin mRNA pointed to a transcriptional defect, the putative promoter region of human NPHS2, the gene encoding podocin, was screened for possible LMX1B binding sites. The hamster Lmx1a protein, whose homeodomain is identical to that of human LMX1B (32), recognizes two related AT-rich motifs called the FLAT-E (5′-TAATTA-3′) and FLAT-F (5′-TTAATAAT-3′) elements (33), and it was therefore reasonable to assume that LMX1B binds to similar motifs. We found one perfect FLAT-F element extending from nucleotides 1079 to 1072 and two closely spaced imperfect FLAT-E elements extending from nucleotides 832 to 809 upstream of the start codon in the NPHS2 promoter. Oligonucleotides containing these elements, but not a GC-rich oligonucleotide closer to the start codon, were recognized from bacterially expressed full-length LMX1B protein in a gel shift assay (Figure 5a). The binding was specific, since the shifts disappeared by competing with an excess of the same unlabeled oligonucleotide but not with an excess of the unlabeled nonspecific oligonucleotide (Figure 5b). Furthermore, the “COL4A4,” “–1087,” and “–837” oligonucleotides were able to compete with one another (data not shown). To rule out the possibility that the bacterially expressed protein was not folded correctly, we also employed the in vitro–translated homeodomain of LMX1B and nuclear extracts from stably transfected HeLa cells in gel shift assays. (We used the homeodomain because a nonspecific shift tended to mask the shift obtained with the in vitro–translated full-length LMX1B protein.) Again we demonstrated binding (Figure 5, c and d). To determine whether the recognition of these binding sites by LMX1B in the putative NPHS2 promoter is sufficient to lead to its activation, we transiently transfected COS-7 cells and a conditionally immortalized murine podocyte cell line (18) with an LMX1B expression plasmid and a luciferase reporter plasmid containing about 4.4 kbp of the putative NPHS2 promoter region (extending from 4,303 bp upstream to 129 bp downstream of the start codon). Even after repeated attempts, however, we could not see a regulatory effect of LMX1B on the reporter plasmid. In addition, we used the stably transfected HeLa cells, which upon the removal of doxycycline inducibly express LMX1B, to determine whether the endogenous podocin gene responded to the induction of LMX1B, but once again no effect was noticed (data not shown).

Figure 5

Gel shift assays to demonstrate binding of LMX1B to FLAT-E and FLAT-F elements in the NPHS2 promoter region. The gel shift assays were carried out with an oligonucleotide from the first intron of the human COL4A4 gene COL4A4, whose recognition by LMX1B had been demonstrated before (38) and that therefore served as a positive control; an oligonucleotide containing a perfect FLAT-F element (–1087); an oligonucleotide containing two imperfect FLAT-E elements (–837); and a GC-rich oligonucleotide containing neither a FLAT-E nor a FLAT-F element (–287). (a) Two hundred fifty and 500 ng of bacterially expressed full-length LMX1B protein recognize not only the binding site in the COL4A4 intron, but also the perfect FLAT-F and imperfect FLAT-E elements, while the GC-rich oligonucleotide is not bound. np, oligonucleotide with no protein added. (b) Competition assays using 500 ng of bacterially expressed full-length LMX1B protein and a 25-fold, 100-fold, and 400-fold molar excess of the indicated unlabeled oligonucleotides demonstrate the specificity of the gel shift. Labeled oligonucleotides are indicated on the top, unlabeled oligonucleotides below. nc, no competing unlabeled oligonucleotide added. (c) Four microliters of the in vitro–translated LMX1B homeodomain were incubated with the oligonucleotides described above. It can be easily seen that the homeodomain (HD) recognizes the binding site in the COL4A4 intron, the perfect FLAT-F element, and the imperfect FLAT-E elements, but not the GC-rich oligonucleotide. Specific bands are indicated by arrows; the other bands can also be seen where no DNA was added to the in vitro transcription/translation reaction (H₂O). (d) Nuclear extracts were prepared from stably transfected HeLa cells inducibly expressing a myc epitope–tagged full-length LMX1B protein, and 6 μg of nuclear proteins from noninduced (“off”) and induced (“on”) cells were used for a gel shift assay. Upon the induction of LMX1B, a clear shift can be recognized with the binding site in the COL4A4 intron and the perfect FLAT-F and imperfect FLAT-E elements (arrows). The specificity of the shift can be appreciated from the supershifted band appearing upon the addition of the anti–myc epitope antibody 9E10 (arrowhead in “on + Ab” lane).

Expression pattern of actin cytoskeleton–associated and basement membrane proteins in the glomeruli of Lmx1b^–/– knockout mice. The absence of podocyte foot processes in the homozygous Lmx1b knockout mice could have also been due to a defect in the cytoskeletal apparatus of the podocytes, and we therefore investigated the expression pattern of proteins associated with the actin cytoskeleton, i.e., synaptopodin and CD2AP. Synaptopodin is a proline-rich protein that has been localized to podocyte foot processes and to actin stress fibers (13). The inactivation of the gene encoding CD2AP leads to foot process effacement, and CD2AP also has been shown to interact with nephrin (34, 35). Both proteins, however, were present in the glomeruli of homozygous knockout mice (Figure 6, a–d), although it appeared as if the signal was weaker on those kidney sections. An RNase protection assay demonstrated that roughly equal amounts of CD2AP (Figure 6e), but decreased levels of synaptopodin (Figure 6f), were present in the kidneys of Lmx1b^–/– mice.

Figure 6

Expression of actin cytoskeleton–associated proteins and GBM components in newborn Lmx1b^–/– mice. The actin cytoskeleton–associated proteins synaptopodin (a and b) and CD2AP (c and d) are both expressed in kidneys of wild-type and homozygous knockout mice. An RNase protection assay with 20 μg of total kidney RNA demonstrates approximately equal levels of CD2AP mRNA (e) but reduced amounts of synaptopodin (Synpo) mRNA (f) in the knockout mice (tRNA served as a negative control). The protection assay with a probe directed against 18S rRNA is shown in Figure 1. The GBM components nidogen/entactin (g and h), fibronectin (i and j), and dystroglycan (k and l) can all be detected in kidneys of wild-type and homozygous knockout mice. Immunohistochemical detection of the proteins was done with DAB (a–d, k and l) and by fluorescence (g–j). Bars: 20 μm.

The morphological alterations in the GBM observed in patients with nail-patella syndrome and in the Lmx1b^–/– mice suggested that important ECM components were lacking. Both nidogen/entactin (Figure 6, g and h) and fibronectin (Figure 6, i and j), however, were present in the GBMs of homozygous knockout mice. We next examined dystroglycan, which has been speculated to be important for the proper arrangement of basement membrane components (36), and again we found no difference between wild-type and homozygous knockout mice (Figure 6, k and l). Another important component of the GBM is represented by collagen IV. While the prospective GBM contains the α1 and α2 chains of collagen IV, an additional network consisting of the α3, α4, and α5 chains is added by the podocyte subsequently (15). Mutations in the latter collagen chains are responsible for Alport syndrome, a hereditary kidney disease eventually leading to chronic renal failure (for review see ref. 37). When the expression of the collagen IV α4 chain was examined in the kidneys of the Lmx1b^–/– mice, a pronounced downregulation of this subunit was observed (data not shown), as is described in a publication that appeared while this manuscript was in preparation (38).

Expression pattern of podocyte-specific transcription factors. Since it was obvious from the morphological characterization and the lack of podocin and the collagen IV α4 chain that the developmental defect in Lmx1b^–/– mice affected the podocyte lineage, we tried to determine whether these defects were only due to the lack of Lmx1b, or whether other transcription factors also played a role. The C₂H₂ zinc finger protein WT1 can already be detected in the condensing metanephrogenic mesenchyme, but later the expression of WT1 is restricted to podocytes (39). No difference, however, was observed between the kidneys of wild-type and homozygous knockout mice (Figure 7, a and b).

Figure 7

Expression of transcription factors in newborn Lmx1b^–/– mice. WT1 (a and b) and Foxc2 (c and d) are both expressed in kidneys of wild-type and homozygous knockout mice. WT1 protein was detected immunohistochemically with DAB, the Foxc2 mRNA with nonradioactive in situ hybridization. Bar: 20 μm.

The inactivation of the Lmx1b gene also leads to defects in eye development. In the presence of Lmx1b, two transcription factors of the winged helix family, Foxc1 (also called Mf1, FREAC3, and FKHL7) and Foxc2 (also called Mfh1), are not found in the differentiating corneal stroma, while in the absence of Lmx1b both proteins can be detected in this structure (40). The Foxc2 mRNA had been localized to podocytes before (41), and so we determined its expression pattern in the Lmx1b knockout mice. In contrast to the situation in the eye, no obvious differences for Foxc2 were observed in the kidneys of Lmx1b^–/– and wild-type mice (Figure 7, c and d).

The combination of the ultrastructural with the histochemical characterization of the glomeruli from Lmx1b^–/– mice provides valuable information on the consequences of the inactivation of Lmx1b in the kidney. It is obvious that the formation of a fully branched glomerular capillary network depends on intact podocytes. Without fully differentiated podocytes, capillaries will invade the glomerulus, but they will not branch. It is not completely understood which signals regulate the formation of the glomerular tuft, but previous results from animal studies (42–44) and organ culture experiments (45) argue that VEGF plays an important role in this process. Furthermore, VEGF also seems to regulate the fenestration of endothelial cells (46, 47). Both the retarded branching of the glomerular capillaries and the decreased number of fenestrae in the glomerular endothelium suggest that Lmx1b-deficient podocytes produce less VEGF, and indeed the quantitation of VEGF mRNA by RNase protection assay indicates that podocytes in Lmx1b^–/– mice secrete less VEGF than do their wild-type counterparts. In this context we wish to point out that the inactivation of a single VEGF allele already leads to embryonic lethality (48, 49), which demonstrates the critical dosage effect of VEGF, so that even a subtle decrease in VEGF levels may result in a developmental defect of the glomerular tuft. Alternatively, other pathways like the one based on the interaction between Notch2 and Jag1 (50) may be affected by the absence of Lmx1b.

Podocytes lacking Lmx1b are of a columnar shape and do not form foot processes. To our knowledge, the α3 integrin (22), podocalyxin (23), and Pod1 (51) knockout animals are the only other mice in which podocytic foot processes are lacking a priori, whereas in the case of the laminin β2 (52) and CD2AP (34) knockout mice, foot processes form first and process effacement takes place later. It is also noteworthy that the podocytes in the homozygous Lmx1b knockout animals are not connected by slit diaphragms. Formation of foot processes may therefore be essential to establish slit diaphragms, although it clearly is not sufficient, since foot processes, but no slit diaphragms, are formed in the absence of nephrin (53). The presence of nephrin in the podocytes of Lmx1b knockout mice also argues that nephrin does not suffice to initiate the formation of a slit diaphragm. So far it is a complete mystery how the interdigitating arrangement of foot processes arises and, once established, how it is regulated. Obviously α3 integrin and podocalyxin are critical components in foot process formation, but these proteins are still present in the Lmx1b^–/– mice. By immunohistochemistry there seemed to be a decreased expression of CD2AP and synaptopodin in the podocytes of Lmx1b-deficient mice, but by RNase protection assay we convincingly found only reduced levels of synaptopodin mRNA in those kidneys. Whether CD2AP is downregulated remains an open question, since total kidney RNA was used in our experiments and CD2AP is also expressed outside the glomerulus in tubules (54), so that a reduction in the glomerulus may go unnoticed in an RNase protection assay. Synaptopodin is an actin-associated protein in the foot processes of podocytes, and therefore its downregulation in the Lmx1b knockout mice is not unexpected. Since synaptopodin knockout mice show no structural abnormalities in the glomerulus (55), the reduced expression of synaptopodin in the Lmx1b knockout mice very likely is a secondary effect and cannot explain the absence of foot processes.

Another strong candidate to regulate the formation of interdigitating foot processes is podocin. Podocin is mutated in at least a subset of patients with an onset of steroid-resistant nephrotic syndrome in early childhood (28–31). Because proteinuria typically is associated with the absence of foot processes, it is reasonable to assume that foot processes are missing in the podocytes of those patients as well. If it could be determined whether foot processes were never present or whether they disappeared after having formed, the function of podocin would be clearer. Since, in the kidney, both Lmx1b and podocin are exclusively expressed in podocytes and the inactivation of Lmx1b results in the absence of podocin, Nphs2, the gene coding for podocin, appears as an obvious target for Lmx1b. Indeed, we were able to demonstrate binding of LMX1B to two AT-rich regions in the putative human NPHS2 promoter region. However, we did not see a regulatory effect of LMX1B on an approximately 4.4-kbp NPHS2 promoter fragment employing transient transfections in COS-7 cells and in conditionally immortalized podocytes, nor was the endogenous NPHS2 gene activated in stably transfected HeLa cells. This may indicate that appropriate cofactors are missing in the cell lines used in our experiments. Lmx1b is expressed in a number of other tissues, such as the eye, the brain, and the limb bud, in which no podocin is found, and therefore it may very well be that additional transcription factors play a role in the expression of podocin in the kidney.

The GBM represents another important component of the filtration barrier. While nidogen/entactin, fibronectin, and dystroglycan were present in the GBMs of Lmx1b^–/– mice, the α4 chain of collagen IV was missing. During the preparation of this manuscript, the diminished expression of the collagen IV α3 and α4 chains was described in the GBM of Lmx1b^–/– mice (38). The absence of an α3α4α5 network of collagen IV can explain the split GBM found in the homozygous Lmx1b knockout mice and the moth-eaten appearance of the GBM in patients with nail-patella syndrome. It is noteworthy that the genesis of collagen fibrils also is disturbed in the corneal stroma of the Lmx1b knockout mice (40).

The severe defects in the podocyte suggest that, in a genetic sense, Lmx1b is located far upstream in the podocyte. WT1, another transcription factor mutated in certain hereditary podocyte diseases (56–58), was present in the podocytes of wild-type and knockout mice, arguing that Lmx1b lies downstream of WT1. We were particularly interested to determine the expression pattern of Foxc2 (also called Mfh1) because of its importance for renal development and previous findings in the eyes of Lmx1b knockout mice. Inactivation of the Foxc2 gene results in only a mild kidney phenotype (59, 60), but compound heterozygotes, in which one allele of the related gene Foxc1 is inactivated as well, have pronounced renal defects (60). Furthermore, Foxc1 and Foxc2 are not expressed in the differentiating corneal stroma of the normal mouse but are present in that of the Lmx1b^–/– mouse (40, 61). No differences, however, were seen in the podocytes of homozygous Lmx1b knockout mice, for which we offer the following explanation. The human LMX1B protein has been demonstrated to act as a transcriptional activator (62); therefore the Foxc2 gene is unlikely to be a direct target of Lmx1b during eye development, since it is upregulated and not downregulated in the absence of Lmx1b. Probably Lmx1b acts through at least one intermediary protein, which represses the Foxc2 gene in the differentiating cornea. This so-far hypothetical intermediary repressor protein may be absent in the podocytes, so that the renal expression of Foxc2 is not affected.

The Lmx1b knockout mouse presents an extremely valuable tool to study the differentiation of the podocyte. It apparently mimics the human syndrome, although it is puzzling that nail-patella syndrome is inherited in a dominant fashion in patients, but only the homozygous Lmx1b knockout mice show a renal phenotype. Furthermore, LMX1B presents a link between nail-patella syndrome, Alport syndrome, and steroid-resistant nephrotic syndrome, since it regulates the expression of collagen IV α4 and podocin.

Antibodies were the generous gifts of Albert Chung (entactin), Larry Holzman (nephrin), David Kershaw (podocalyxin), Jeff Miner (collagen IV α4), Peter Mundel (synaptopodin), and Andrey Shaw (CD2AP), and cDNAs were kindly provided by Georg Breier (VEGF₁₂₀), Brigid Hogan (Foxc2), and Brendan Lee (LMX1B). HtTA-1 cells were a kind gift from Hermann Bujard. Murine CD2AP and synaptopodin cDNAs were purchased from the Deutsches Ressourcenzentrum für Genomforschung GmbH (Ressourcenzentrum und Primärdatenbank) (63). We also acknowledge financial support from the German Research Council (Deutsche Forschungsgemeinschaft, Forschergruppe 406, Mechanismen der Progression des chronischen Nierenversagens) and are grateful for the superb arrangement of the figures by Rolf Nonnenmacher.

1. Bennett, WM, et al. The nephropathy of the nail-patella syndrome. Clinicopathologic analysis of 11 kindred. Am J Med 1973. 54:304-319.
   View this article via: PubMed CrossRef Google Scholar
2. del Pozo, E, Lapp, H. Ultrastructure of the kidney in the nephropathy of the nail-patella syndrome. Am J Clin Pathol 1970. 54:845-851.
   View this article via: PubMed Google Scholar
3. Ben-Bassat, M, Cohen, L, Rosenfeld, J. The glomerular basement membrane in the nail-patella syndrome. Arch Pathol 1971. 92:350-355.
   View this article via: PubMed Google Scholar
4. Drut, RM, Chandra, S, Latorraca, R, Gilbert-Barness, E. Nail-patella syndrome in a spontaneously aborted 18-week fetus: ultrastructural and immunofluorescent study of the kidneys. Am J Med Genet 1992. 43:693-696.
   View this article via: PubMed CrossRef Google Scholar
5. Lubec, B, Arbeiter, K, Ulrich, W, Frauscher, G. Hereditary osteo-onycho-renal dysplasia with excess urinary pyridinoline cross-links and abnormal kidney collagen cross-linking. Nephron 1995. 70:255-259.
   View this article via: PubMed CrossRef Google Scholar
6. Dreyer, SD, et al. Mutations in LMX1B cause abnormal skeletal patterning and renal dysplasia in nail patella syndrome. Nat Genet 1998. 19:47-50.
   View this article via: PubMed CrossRef Google Scholar
7. McIntosh, I, et al. Mutation analysis of LMX1B gene in nail-patella syndrome patients. Am J Hum Genet 1998. 63:1651-1658.
   View this article via: PubMed CrossRef Google Scholar
8. Vollrath, D, et al. Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail-patella syndrome. Hum Mol Genet 1998. 7:1091-1098.
   View this article via: PubMed CrossRef Google Scholar
9. Chen, H, et al. Limb and kidney defects in Lmx1b mutant mice suggest an involvement of LMX1B in human nail patella syndrome. Nat Genet 1998. 19:51-55.
   View this article via: PubMed CrossRef Google Scholar
10. Heidet, L, Cai, Y, Guicharnaud, L, Antignac, C, Gubler, M-C. Glomerular expression of type IV collagen chains in normal and X-linked Alport syndrome kidneys. Am J Pathol 2000. 156:1901-1910.
    View this article via: PubMed Google Scholar
11. Holzman, LB, et al. Nephrin localizes to the slit pore of the glomerular epithelial cell. Kidney Int 1999. 56:1481-1491.
    View this article via: PubMed CrossRef Google Scholar
12. Roselli, S, et al. Podocin localizes in the kidney to the slit diaphragm area. Am J Pathol 2002. 160:131-139.
    View this article via: PubMed Google Scholar
13. Mundel, P, et al. Synaptopodin: an actin-associated protein in telencephalic dendrites and renal podocytes. J Cell Biol 1997. 139:193-204.
    View this article via: PubMed CrossRef Google Scholar
14. Dustin, ML, et al. A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 1998. 94:667-677.
    View this article via: PubMed CrossRef Google Scholar
15. Miner, JH, Sanes, JR. Collagen IV α3, α4, and α5 chains in rodent basal laminae: sequence, distribution, association with laminins, and developmental switches. J Cell Biol 1994. 127:879-891.
    View this article via: PubMed CrossRef Google Scholar
16. Obermüller, N, Gretz, N, Kriz, W, Reilly, RF, Witzgall, R. The swelling-activated chloride channel ClC-2, the chloride channel ClC-3, and ClC-5, a chloride channel mutated in kidney stone disease, are expressed in distinct subpopulations of renal epithelial cells. J Clin Invest 1998. 101:635-642.
    View this article via: JCI PubMed CrossRef Google Scholar
17. Ausubel, F.A., et al. 1996. Current protocols in molecular biology. John Wiley & Sons. New York, New York, USA.
18. Mundel, P, et al. Rearrangements of the cytoskeleton and cell contacts induce process formation during differentiation of conditionally immortalized mouse podocyte cell lines. Exp Cell Res 1997. 236:248-258.
    View this article via: PubMed CrossRef Google Scholar
19. Gossen, M, Bujard, H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 1992. 89:5547-5551.
    View this article via: PubMed CrossRef Google Scholar
20. Dong, Y, Skoultchi, AI, Pollard, JW. Efficient DNA transfection of quiescent mammalian cells using poly-L-ornithine. Nucleic Acids Res 1993. 21:771-772.
    View this article via: PubMed CrossRef Google Scholar
21. Margaretten, W, Csavossy, I, McKay, DG. An electron microscopic study of thrombin-induced disseminated intravascular coagulation. Blood 1967. 29:169-181.
    View this article via: PubMed Google Scholar
22. Kreidberg, JA, et al. Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development 1996. 122:3537-3547.
    View this article via: PubMed Google Scholar
23. Doyonnas, R, et al. Anuria, omphalocele, and perinatal lethality in mice lacking the CD34-related protein podocalyxin. J Exp Med 2001. 194:13-27.
    View this article via: PubMed CrossRef Google Scholar
24. Kestilä, M, et al. Positionally cloned gene for a novel glomerular protein — nephrin — is mutated in congenital nephrotic syndrome. Mol Cell 1998. 1:575-582.
    View this article via: PubMed CrossRef Google Scholar
25. Holthöfer, H, et al. Nephrin localizes at the podocyte filtration slit area and is characteristically spliced in the human kidney. Am J Pathol 1999. 155:1681-1687.
    View this article via: PubMed Google Scholar
26. Ruotsalainen, V, et al. Nephrin is specifically located at the slit diaphragm of glomerular podocytes. Proc Natl Acad Sci USA 1999. 96:7962-7967.
    View this article via: PubMed CrossRef Google Scholar
27. Patrakka, J, et al. Congenital nephrotic syndrome (NPHS1): features resulting from different mutations in Finnish patients. Kidney Int 2000. 58:972-980.
    View this article via: PubMed CrossRef Google Scholar
28. Boute, N, et al. NPHS2, encoding the glomerular protein podocin, is mutated in autosomal recessive steroid-resistant nephrotic syndrome. Nat Genet 2000. 24:349-354.
    View this article via: PubMed CrossRef Google Scholar
29. Caridi, G, et al. Prevalence, genetics, and clinical features of patients carrying podocin mutations in steroid-resistant nonfamilial focal segmental glomerulosclerosis. J Am Soc Nephrol 2001. 12:2742-2746.
    View this article via: PubMed Google Scholar
30. Frishberg, Y, et al. Mutations in NPHS2 encoding podocin are a prevalent cause of steroid-resistant nephrotic syndrome among Israeli-Arab children. J Am Soc Nephrol 2002. 13:400-405.
    View this article via: PubMed Google Scholar
31. Karle, SM, et al. Novel mutations in NPHS2 detected in both familial and sporadic steroid-resistant nephrotic syndrome. J Am Soc Nephrol 2002. 13:388-393.
    View this article via: PubMed Google Scholar
32. Iannotti, CA, et al. Identification of a human LMX1 (LMX1.1)-related gene, LMX1.2: tissue-specific expression and linkage mapping on chromosome 9. Genomics 1997. 46:520-524.
    View this article via: PubMed CrossRef Google Scholar
33. German, MS, Wang, J, Chadwick, RB, Rutter, WJ. Synergistic activation of the insulin gene by a LIM-homeodomain protein and a basic helix-loop-helix protein: building a functional insulin minienhancer complex. Genes Dev 1992. 6:2165-2176.
    View this article via: PubMed CrossRef Google Scholar
34. Shih, N-Y, et al. Congenital nephrotic syndrome in mice lacking CD2-associated protein. Science 1999. 286:312-315.
    View this article via: PubMed CrossRef Google Scholar
35. Shih, N-Y, et al. CD2AP localizes to the slit diaphragm and binds to nephrin via a novel C-terminal domain. Am J Pathol 2001. 159:2303-2308.
    View this article via: PubMed Google Scholar
36. Regele, HM, et al. Glomerular expression of dystroglycans is reduced in minimal change nephrosis but not in focal segmental glomerulosclerosis. J Am Soc Nephrol 2000. 11:403-412.
    View this article via: PubMed Google Scholar
37. Kashtan, CE. Alport syndrome and thin glomerular basement membrane disease. J Am Soc Nephrol 1998. 9:1736-1750.
    View this article via: PubMed Google Scholar
38. Morello, R, et al. Regulation of glomerular basement membrane collagen expression by LMX1B contributes to renal disease in nail patella syndrome. Nat Genet 2001. 27:205-208.
    View this article via: PubMed CrossRef Google Scholar
39. Pritchard-Jones, K, et al. The candidate Wilms’ tumour gene is involved in genitourinary development. Nature 1990. 346:194-197.
    View this article via: PubMed CrossRef Google Scholar
40. Pressman, CL, Chen, H, Johnson, RL. Lmx1b, a LIM homeodomain class transcription factor, is necessary for normal development of multiple tissues in the anterior segment of the murine eye. Genesis 2000. 26:15-25.
    View this article via: PubMed CrossRef Google Scholar
41. Miura, N, Wanaka, A, Tohyama, M, Tanaka, K. MFH-1, a new member of the fork head domain family, is expressed in developing mesenchyme. FEBS Lett 1993. 326:171-176.
    View this article via: PubMed CrossRef Google Scholar
42. Kitamoto, Y, Tokunaga, H, Tomita, K. Vascular endothelial growth factor is an essential molecule for mouse kidney development: glomerulogenesis and nephrogenesis. J Clin Invest 1997. 99:2351-2357.
    View this article via: JCI PubMed CrossRef Google Scholar
43. Carmeliet, P, et al. Impaired myocardial angiogenesis and ischemic cardiomyopathy in mice lacking the vascular endothelial growth factor isoforms VEGF₁₆₄ and VEGF₁₈₈. Nat Med 1999. 5:495-502.
    View this article via: PubMed CrossRef Google Scholar
44. Gerber, H-P, et al. VEGF is required for growth and survival in neonatal mice. Development 1999. 126:1149-1159.
    View this article via: PubMed Google Scholar
45. Tufró, A. VEGF spatially directs angiogenesis during metanephric development in vitro. Dev Biol 2000. 227:558-566.
    View this article via: PubMed CrossRef Google Scholar
46. Roberts, WG, Palade, GE. Increased microvascular permeability and endothelial fenestration induced by vascular endothelial growth factor. J Cell Sci 1995. 108:2369-2379.
    View this article via: PubMed Google Scholar
47. Esser, S, et al. Vascular endothelial growth factor induces endothelial fenestrations in vitro. J Cell Biol 1998. 140:947-959.
    View this article via: PubMed CrossRef Google Scholar
48. Carmeliet, P, et al. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 1996. 380:435-439.
    View this article via: PubMed CrossRef Google Scholar
49. Ferrara, N, et al. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996. 380:439-442.
    View this article via: PubMed CrossRef Google Scholar
50. McCright, B, et al. Defects in development of the kidney, heart and eye vasculature in mice homozygous for a hypomorphic Notch2 mutation. Development 2001. 128:491-502.
    View this article via: PubMed Google Scholar
51. Quaggin, SE, et al. The basic-helix-loop-helix protein Pod1 is critically important for kidney and lung organogenesis. Development 1999. 126:5771-5783.
    View this article via: PubMed Google Scholar
52. Noakes, PG, et al. The renal glomerulus of mice lacking s-laminin/laminin β2: nephrosis despite molecular compensation by laminin β1. Nat Genet 1995. 10:400-405.
    View this article via: PubMed CrossRef Google Scholar
53. Putaala, H, Soininen, R, Kilpeläinen, P, Wartiovaara, J, Tryggvason, K. The murine nephrin gene is specifically expressed in kidney, brain and pancreas: inactivation of the gene leads to massive proteinuria and neonatal death. Hum Mol Genet 2001. 10:1-8.
    View this article via: PubMed CrossRef Google Scholar
54. Lehtonen, S, et al. In vivo interaction of the adapter protein CD2-associated protein with the type 2 polycystic kidney disease protein, polycystin-2. J Biol Chem 2000. 275:32888-32893.
    View this article via: PubMed CrossRef Google Scholar
55. Brogan, M, Chabanis, S, Schwarz, K, Shankland, SJ, Mundel, P. Synaptopodin limits podocyte damage in experimental glomerulonephritis. J Am Soc Nephrol 1991. 12:672a. (Abstr.)
56. Barbaux, S, et al. Donor splice-site mutations in WT1 are responsible for Frasier syndrome. Nat Genet 1997. 17:467-470.
    View this article via: PubMed CrossRef Google Scholar
57. Patek, CE, et al. A zinc finger truncation of murine WT1 results in the characteristic urogenital abnormalities of Denys-Drash syndrome. Proc Natl Acad Sci USA 1999. 96:2931-2936.
    View this article via: PubMed CrossRef Google Scholar
58. Pelletier, J, et al. Germline mutations in the Wilms’ tumor suppressor gene are associated with abnormal urogenital development in Denys-Drash syndrome. Cell 1991. 67:437-447.
    View this article via: PubMed CrossRef Google Scholar
59. Iida, K, et al. Essential roles of the winged helix transcription factor MFH-1 in aortic arch patterning and skeletogenesis. Development 1997. 124:4627-4638.
    View this article via: PubMed Google Scholar
60. Kume, T, Deng, K, Hogan, BLM. Murine forkhead/winged helix genes Foxc1 (Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidney and urinary tract. Development 2000. 127:1387-1395.
    View this article via: PubMed Google Scholar
61. Kidson, SH, Kume, T, Deng, K, Winfrey, V, Hogan, BLM. The forkhead/winged-helix gene, Mf1, is necessary for the normal development of the cornea and formation of the anterior chamber in the mouse eye. Dev Biol 1999. 211:306-322.
    View this article via: PubMed CrossRef Google Scholar
62. Dreyer, SD, et al. LMX1B transactivation and expression in nail-patella syndrome. Hum Mol Genet 2000. 9:1067-1074.
    View this article via: PubMed CrossRef Google Scholar
63. Lennon, G, Auffray, C, Polymeropoulos, M, Soares, MB. The I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and their expression. Genomics 1996. 33:151-152.
    View this article via: PubMed CrossRef Google Scholar