Cux-1 is a murine homeobox gene that is structurally related to Drosophila cut. Mammalian homologues of cut function primarily as transcriptional repressors of many different genes, including genes encoding the cyclin kinase inhibitors (CKI) p21 and p27 (Valarche et al., 1993; Dufort and Nepveu, 1994; Higgy et al., 1997; Coqueret et al., 1998a; Ledford et al., 2002). The binding of Cux-1 to the promoters of these genes appears to be limited to tissues or stages of development where the target genes are not expressed (Nepveu, 2001). At later stages of development, Cux-1 proteins are down-regulated or lose their ability to bind DNA, and transcription of the target genes is permitted. Cut proteins repress transcription by two different mechanisms: (1) passive repression, whereby cut proteins compete with transcriptional activators for the same binding site (known targets of passive repression by Cux-1 are Sp1 [GC box] and NF-Y [CCAAT box]), or (2) active repression, by means of a carboxy-terminal domain after binding to DNA at a distance from the transcription start site (Mailly et al., 1996).
During kidney development, Cux-1 is highly expressed in the nephrogenic zone, in uninduced and condensing mesenchyme, comma- and S-shaped bodies, and branching ureteric buds. At later stages of development, Cux-1 is down-regulated such that expression is minimal in maturing glomeruli and tubules (Vanden Heuvel et al., 1996). This finding is associated with the up-regulation of the cyclin kinase inhibitor p27 in maturing glomeruli and tubules, which induces cell cycle arrest and terminal differentiation. Previously, we generated transgenic mice ectopically expressing Cux-1 under the direction of the CMV immediate early gene promoter. These mice express Cux-1 in maturing glomeruli and tubules in developing kidneys, resulting in renal hyperplasia, associated with the down-regulation of p27 (Ledford et al., 2002).
Within the cut repeats, there are evolutionarily conserved consensus phosphorylation sites for protein kinase C (PKC) and casein kinase II (Coqueret et al., 1996, 1998b). Moreover, PKC and casein kinase II were shown to phosphorylate Cux-1 on specific serine and threonine residues in the cut repeats. Phosphorylation of Cux-1 resulted in an inhibition of DNA binding activity, and a concomitant inhibition of repression, while treatment with alkaline phosphatase restored DNA binding. Analysis of Cux-1 binding activity in vivo demonstrated that Cux-1 bound DNA when in a dephosphorylated state, but did not bind DNA when phosphorylated. These results demonstrated that Cux-1 activity is regulated by phosphorylation and suggests that specific phosphatases may play a role in Cux-1 activity.
Calcineurin is a SER/THR phosphatase that exists as a heterodimer of a catalytic subunit, called calcineurin A (CnA), and a regulatory subunit, called calcineurin B (CnB; for review see Rusnak and Mertz, 2000). Calcineurin phosphatase activity is initiated only when CnB is bound to both calmodulin and calcium. There are three isoforms of CnA, α, β, and γ, that display some tissue specificity. CnA-α and CnA-β are widely expressed, while the γ isoform appears to be restricted to testis and brain (Muramatsu et al., 1992). Mice carrying a targeted deletion of CnA-α exhibit severe defects in postnatal kidney development, including reduced cell proliferation and increased apoptosis in the nephrogenic zone, contributing to an overall reduction of the nephrogenic zone (Gooch et al., 2004). Additional changes include an absence of mesangial cells in the glomeruli that develop during postnatal kidney development. The cell cycle defect is thought to result from the ectopic expression of the cyclin kinase inhibitor p27 (Gooch et al., 2004). In contrast, mice carrying a targeted deletion of CnA-β have normal kidneys (Gooch et al., 2004). To determine whether the ectopic expression of p27 in the nephrogenic zone of CnA α (−/−) kidneys might result from the down-regulation of Cux-1, we examined Cux-1 expression in kidneys isolated from CnA-α null mice. To determine whether increased Cux-1 expression would rescue the cell proliferation defect observed in the kidneys of CnA-α null mice, we evaluated cell proliferation in kidney organ cultures from Cux-1 transgenic mice treated with cyclosporin A, a specific inhibitor of calcineurin.
To determine whether the reduced nephrogenic zone and increased expression of p27 in CnA α (−/−) mice resulted from alterations in Cux-1 expression or activity, we examined Cux-1 expression in kidneys isolated from CnA α (−/−) mice. In kidneys isolated from 4-day-old wild-type mice, Cux-1 was highly expressed in the nephrogenic zone where it colocalized with proliferating cell nuclear antigen (PCNA; Fig. 1B–D). In kidneys isolated from CnA α (−/−) mice, Cux-1 continued to be expressed in the nephrogenic zone (Fig. 1F). However, in contrast to wild-type, PCNA was expressed in far fewer cells in CnA α (−/−) kidneys and did not colocalize with Cux-1 (Fig. 1G,H). Previous studies have shown that the DNA binding activity of Cux-1 is regulated by PKC and casein kinase II (Coqueret et al., 1996, 1998b). DNA binding activity was reduced in the presence of PMA, an activator of PKC, but mutation of the PKC sites resulted in DNA binding that was not reduced by PMA treatment. One possibility is that the absence of CnA α results in Cux-1 being maintained in the phosphorylated inactive form and suggests that CnA α may positively regulate Cux-1 activity (see model in Fig. 9).
The antibiotic cyclosporin A (CsA) competitively binds calcineurin to inactivate its phosphatase activity (Wiederrecht et al., 1993). Previous studies have shown that inhibiting CnA with CsA leads to nephron deficit in rat metanephric organ cultures (Tendron et al., 2003). Figure 2 shows that metanephroi grown in the presence of CsA continued to express Cux-1, but the expression of PCNA was markedly reduced. We next examined the expression of Pax-2, a marker for condensing mesenchyme and early nephric structures in the nephrogenic zone, and p27, a marker for maturing glomeruli and tubules, in the vehicle (control) and CsA-treated kidney organ cultures. Figure 3 shows that Pax-2 was expressed in the nephrogenic zone of vehicle-treated metanephroi, while p27 was expressed in maturing glomeruli and tubules in a region characterized by the absence of Pax-2. In contrast to the control kidney, the region of Pax-2 expression was significantly smaller in the CsA-treated metanephroi (Fig. 3F). Moreover, p27-positive cells were found at the periphery of the developing kidney culture and within the Pax-2–positive cells (Fig. 3G,H). These results indicate that CsA treatment results in a reduction in the nephrogenic zone and in ectopic expression of p27 and demonstrate that CsA treatment of metanephroi in organ culture phenocopied the CnA α (−/−) phenotype. Thus to begin to determine whether the ectopic expression of p27 in the nephrogenic zone of CnA α (−/−) kidneys results from the down-regulation of Cux-1 activity, metanephric kidneys isolated from transgenic and wild-type mice were treated with cyclosporin A. CsA- or vehicle-treated metanephroi were examined daily by light microscopy. Initially, all wild-type metanephroi were matched for gestational age and had a similar appearance (Fig. 4A,G). By 24 hr, the CsA-treated metanephroi were visibly smaller than the vehicle-treated metanephroi (Fig. 4B,H). While the vehicle-treated metanephroi continued to increase in surface area, the CsA-treated metanephroi were growth inhibited (Fig. 4A–L). Similar to wild-type kidneys, all Cux-1 transgenic metanephroi were matched for gestational age and had a similar appearance (Fig. 4M,S). However, in contrast to wild-type kidneys, metanephroi isolated from Cux-1 transgenic mice continued to grow when treated with CsA (Fig. 4M–X).
The surface areas of 15 paired wild-type or Cux-1 transgenic metanephroi treated with or without CsA from six separate experiments were compared. The percentage change in mean size of vehicle and CsA-treated metanephroi from wild-type or transgenic animals were paired to control for differences in gestational age. Figure 5 shows that vehicle-treated wild-type metanephroi showed an increase in net percentage growth, while CsA treatment resulted in a decrease in net percentage growth. In contrast, both vehicle- and CsA-treated transgenic metanephroi showed an increase in net percentage growth, although the extent of growth was reduced with CsA treatment (Fig. 5). The differences between percentage net growth of vehicle- and CsA-treated metanephroi were statistically significant.
To determine whether inhibition of calcineurin by cyclosporin changed the phosphorylation state of Cux-1, we evaluated Cux-1 isolated from vehicle- or cyclosporin-treated metanephric kidney cultures. Figure 6A shows that the levels of total Cux-1 protein were similar between vehicle- or CsA-treated kidney cultures. However, when phosphorylated Cux-1 protein was assessed, we observed a greater than fivefold increase in phosphorylated Cux-1 in the CsA-treated kidney cultures compared with the control cultures (Fig. 6B). Pretreatment of kidneys with both CsA and PMA did not significantly increase the levels of phosphorylated Cux-1 (Fig. 6B). However, no phosphorylated Cux-1 was detected after treatment with calf intestinal phosphatase (Fig. 6B).
To determine the basis for the rescued growth of Cux-1 transgenic metanephroi in the presence of CsA, we examined Cux-1 expression and cell proliferation in vehicle- or CsA-treated transgenic kidney cultures. Figure 7 shows that, similar to wild-type, Cux-1 was expressed throughout the nephrogenic zone in both vehicle- and CsA-treated metanephroi (Fig. 7B,F). However, in contrast to wild-type kidney cultures, PCNA was not down-regulated in CsA-treated Cux-1 transgenic metanphroi, but was colocalized with Cux-1 (Fig. 7G,H). We next examined p27 expression in vehicle- and CsA-treated wild-type and Cux-1 transgenic metanephroi. p27 is normally expressed in maturing glomeruli and tubules, after the down-regulation of Cux-1. While p27 was absent from the nephrogenic zone of vehicle-treated kidneys (Fig. 8B,F), p27 was ectopically expressed in the nephrogenic zone of CsA-treated wild-type kidney organ cultures (Fig. 8D), similar to CnA α (−/−) kidneys. In contrast, p27 was not expressed in the nephrogenic zone of the CsA-treated Cux-1 transgenic kidney organ cultures (Fig. 8H).
Cux-1 is a murine homeobox gene that is related to the Drosophila cut gene. Mammalian Cut protein expression appears to be restricted to proliferating cells in many different tissues (Nepveu, 2001) where they function as cell cycle-dependent transcriptional repressors. Targets of Cux-1 repression include the cyclin kinase inhibitors p21 and p27 (Coqueret et al., 1998a; Ledford et al., 2002).
During nephrogenesis, maturing nephrons proceed through an orderly sequence of developmental stages that can be distinguished morphologically. These stages are renal vesicle (stage I), comma- and S-shaped body (stage II), developing capillary loop (stage III), and maturing glomerulus (stage IV). The developing kidney displays a spatial gradient of differentiation in which nephrons at the earliest stages of development (stage I and II) are restricted to the nephrogenic zone immediately beneath the renal capsule, while progressively more mature nephrons (stage III and IV) are located toward the center of the kidney (Reeves et al., 1978). We have previously shown that Cux-1 is highly expressed in the nephrogenic zone of developing kidneys, where it is associated with cell proliferation (Vanden Heuvel et al., 1996). In addition, transgenic mice ectopically expressing Cux-1 in the developing kidney exhibit renal hyperplasia, resulting from aberrant repression of p27 in maturing glomeruli and tubules.
Previous studies have demonstrated that loss of CnA α results in disruption of the cell cycle in kidney development (Gooch et al., 2004). The nephrogenic zone of developing kidneys isolated from CnA α(−/−) mice showed decreased proliferation and increased cell death that was associated with increased levels of p27. In addition, glomeruli present in the CnA α(−/−) mice were devoid of mesangial cells. In contrast, Cux-1 transgenic mice exhibit mesangial cell hyperplasia. Moreover, mesangial cells isolated from Cux-1 transgenic mice are not growth restricted in the absence of serum, but progress through the cell cycle (Sharma et al., 2004).
To determine whether the decreased cell proliferation in kidneys from CnA α(−/−) mice resulted from the down-regulation of Cux-1, we examined Cux-1 expression in CnA α(−/−) kidneys and in kidney cultures treated with the antibiotic CsA. CsA specifically inhibits calcineurin, and previous studies have shown that administration of CsA to pregnant rabbits during the period when nephrogenesis is initiated results in defects in the nephrogenic zone (Tendron et al., 2003). Both in CnA α(−/−) kidneys and in kidneys treated with CsA, our results showed that Cux-1 was expressed throughout the nephrogenic zone, similar to the expression in wild-type or vehicle-treated kidneys. However, in contrast to wild-type or vehicle-treated kidneys, Cux-1 expression was not associated with cell proliferation in CnA α(−/−) kidneys or in kidneys treated with CsA. Moreover, kidney cultures treated with CsA showed increased levels of phosphorylated Cux-1, compared with vehicle-treated kidneys, although the total amount of Cux-1 was not different between CsA and vehicle-treated kidneys. Because the ability of Cux-1 to bind DNA is negatively regulated by phosphorylation, these results suggest that Cux-1 activity may be regulated by calcineurin. The loss of calcineurin, either in the CnA α(−/−) mice or by pharmacologic inhibition, would then result in the maintenance of Cux-1 in an inactive state. A potential model of Cux-1 regulation by calcineurin is shown in Figure 9.
To determine whether increased Cux-1 expression could overcome the loss of calcineurin, we cultured kidneys isolated from Cux-1 transgenic mice with cyclosporin. In contrast to wild-type kidneys, the transgenic kidneys were not growth inhibited, but continued to increase in size. Moreover, CsA-treated Cux-1 transgenic kidneys showed extensive cell proliferation in the nephrogenic zone, with reduced expression of p27. We have previously shown that Cux-1 expression in kidneys isolated from Cux-1 transgenic mice is significantly elevated when compared with wild-type mice (Ledford et al., 2002). One possibility is that the amount of Cux-1 protein produced by the transgenic kidneys exceeds the rate of phosphorylation, resulting in significant levels of active Cux-1 protein that is able to bind DNA and repress target genes, including p27. Taken together, these results demonstrate that the ectopic expression of Cux-1 rescued the cell proliferation defects in the nephrogenic zone induced by cyclosporin, and suggest that Cux-1 may be regulated by calcineurin during kidney development in vivo.
Commercial reagents used were rabbit anti-CDP (Cux-1; Santa Cruz # sc-13024), rabbit anti-p27 (Santa Cruz #sc-528), mouse anti-PCNA (SIGMA # P-8825), and goat anti-Pax-2 (Santa Cruz #sc-7747).
Organotypic Kidney Cultures
Metanephric organ cultures were established from mouse embryos as described (Yeger et al., 1996). Embryos were dissected from timed pregnant mice at 12.5 or 14.5 days post coitum (dpc). Embryonic age was verified according to Theiler (1972). Fifteen paired metanephric kidneys and associated ureteric buds were microdissected from wild-type or Cux-1 transgenic mice and placed in a 24-well tissue culture plate containing medium (50/50 DMEM/F12, 2 mM L-glutamine, 10 mM Hepes, 5 μg/ml insulin, 5 μg/ml transferrin, 2.8 nM selenium, 25 ng/ml prostaglandin E, 32 pg/ml T3, and 250 μg/ml). Kidneys were placed on 0.4 μM PET track-etched membranes (Beckton Dickinson) in medium for 24 hr and then replaced with medium containing either 500 ng/ml CsA or vehicle (100% ethanol). Organ cultures were replenished with fresh treatment medium every 24 hr and photographed using a Leica M240 microscope and captured with an Optronics Magnafire digital camera.
Preparation of Phosphoproteins and Western Blot Analysis
Total protein lysates from 15 paired wild-type metanephric kidney cultures grown in media containing either 500 ng/ml cyclosporin A or vehicle (100% ethanol) were applied to PhosphoProtein purification columns using the QIAGEN PhosphoProtein Purification Kit, according to the manufacturer's directions. In some cases, lysates were first incubated with calf intestinal phosphatase (5 μg protein/2 units). Total protein lysates or purified phosphorylated proteins (42 μg) were solubilized in sodium dodecyl sulfate-polyacrylamide gel electrophoresis sample buffer and electrophoresed on 4–15% gradient polyacrylamide gels. Phosphoproteins were transferred to a polyvinylidene difluoride membrane as described previously (Ledford et al., 2002). The immunoblot was blocked in 5% bovine serum albumin in TBST buffer (10 mM Tris-Cl; 150 mM NaCl, pH 7.5; 0.1% Tween 20). Primary antibodies were added at a dilution of 1:100 for Cux-1, PhosphoSerine, and PhosphoThreonine in blocking solution. After overnight incubation at 4°C, filters were washed twice at room temperature with TBS-Tween/Triton buffer (20 mM Tris-Cl; 500 mM NaCl; 0.05% Tween 20; 0.2% Triton X-100, pH 7.5), washed twice at room temperature in TBS buffer, and incubated with secondary antibodies (1:10,000 dilution in TBST) for 1 hr at room temperature. After four additional washes in TBS-Tween/Triton buffer, bound antibody was detected by Super Signal West Pico chemiluminescent Substrate (Pierce) according to manufacturer's directions, followed by exposure to autoradiography film. Total Cux-1 and phospho-Cux-1 protein levels were scanned and quantitated using Gelpro 4.5 software (Media Cybernetics).
For quantitation of metanephric size, kidney organ cultures from wild-type or Cux-1 transgenic mice, treated with cyclosporin A or vehicle, were photographed under light microscopy at identical magnification. Organ culture pictures were quantified using the NIH image J program. Metanephroi were outlined manually, and the surface areas were calculated by pixel counting. Net growth was determined as the mean percentage of net surface area before treatment (baseline) subtracted from net surface area after treatment. Collective repeated measures analysis of variance (ANOVA) was performed during treatment with one-way ANOVA performed on each treatment day to determine a difference between treatment groups. P ≤ 0.05 was considered as a difference in treatment and was followed with post hoc analysis by the Least Significant Difference test, where P ≤ 0.05 again was considered statistically significant.
Metanephroi were fixed in 4% paraformaldehyde and embedded in paraffin. Five-micrometer-thick tissue sections were deparaffinized with xylene and rehydrated with graded ethanols. To obtain adequate signal, the slides were treated with antigen unmasking solution (Vector) according to the manufacturer's protocol. To block endogenous autofluorescence, sections were incubated with 1 M NH4Cl for 30 min. Sections were washed in phosphate buffered saline (PBS) and blocked with 10% normal serum (in PBS from the species the secondary antibody was raised in) for 1 hr. After washing in PBS, the slides were incubated for 1 hr with primary antibodies in a humidified chamber. Antibody dilutions were 1:50 for CDP Ab, 1:3,000 for PCNA Ab, and 1:100 for p27 Ab, in 2% blocking serum in PBS. Slides were incubated at room temperature with 100 μl of primary antibody in a humid chamber and then washed 4 times in PBST. For Cux-1 and PCNA double-labeling experiments, sections were washed with PBST and incubated simultaneously with biotin-conjugated horse anti-rabbit and Texas-Red conjugated horse anti-mouse secondary antibodies (Vector). Sections were then washed with PBST and incubated with fluorescein isothiocyanate–conjugated avidin (Vector). Sections were then washed, mounted with Vectashield (Vector), and viewed with a fluorescence microscope. Images were captured with an Optronics Magnafire digital camera.
We thank Rosetta Barkley, Stanton Fernald, Eileen Roach, and Dianne Vassmer for expert technical assistance. We thank Drs. Dale Abrahamson, James Calvet, and Robin Maser and members of the Abrahamson, Calvet, and Maser laboratories for many helpful discussions and for sharing reagents. G.B.V.H. and J.L.G. were funded by the NIH. Part of this work was previously published in abstract form (J Am Soc Nephrol 15:675A, 2004).