SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

The canonical Wnt/β-catenin pathway was recently identified as a factor in the pathogenesis of several renal diseases. The aim of this study was to evaluate Wnt signaling activity during disease development in a murine model of lupus nephritis.

Methods

Wnt activity and Dkk-1 expression were serially assayed in the serum and kidneys of (NZB × NZW)F1 mice during progression of lupus nephritis. The effects of serum obtained from mice with lupus and serum-equivalent concentrations of Dkk-1 on mesangial cells were assessed in vitro.

Results

Gene expression analyses revealed increased canonical Wnt pathway activity in kidneys during development of lupus nephritis, paralleled by an increase in renal and serum levels of the Wnt inhibitor Dkk-1. Sera obtained from proteinuric-stage (NZB × NZW)F1 mice showed strong Wnt-inhibitory effects in vitro. Dkk-1 concentrations comparable to those observed in lupus-prone mice induced apoptosis in tubular and mesangial cells in vitro, whereas no such effect was seen for the range of concentrations observed in young prediseased mice and control BALB/c mice.

Conclusion

These data demonstrate that renal Wnt signaling activity is increased in lupus and is accompanied by an increase in renal and serum levels of Dkk-1. The Wnt pathway is involved in the turnover of extracellular matrix constituents and represents a potential mediator of the morphologic changes that occur within the glomerulus during the development of nephritis. Furthermore, increased levels of Dkk-1 serve as a potential proapoptotic stimulus in vitro and possibly in vivo and could be an important element in the initiation and progression of systemic and end-organ disease manifestations in systemic lupus erythematosus.

The Wnt signaling pathway is an intracellular signaling cascade that was initially linked to embryogenesis and cancer development (1). Understanding of the roles of Wnt signaling in postembryonic mammals is currently increasing, and during the last few years, interest in this signaling network has increased due to the discovery of its involvement in several important physiologic and pathophysiologic conditions. The classic, so-called canonical Wnt pathway is activated through binding of Wnt agonist proteins to members of the Frizzled family of receptors. This interaction initiates an intracytoplasmic signal transduction cascade that prevents glycogen synthase kinase 3β (GSK3β)–mediated degradation of β-catenin, followed by translocation of β-catenin into the nucleus, where it associates with the T cell factor/lymphocyte enhancement factor (TCF/LEF) family of transcription factors. These transcription factors activate a variety of target genes involved in cellular growth and differentiation. The canonical Wnt/β-catenin pathway is tightly regulated by several secreted antagonists, including the secreted Frizzled-related proteins, Wnt inhibitory factor, and the Dickkopf protein family. Of these, Dkk-1 is a specific inhibitor of canonical Wnt/β-catenin signaling and has been studied in the context of regulation of the Wnt pathway (2).

Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by the development of autoreactivity against nuclear antigens, including double-stranded DNA (dsDNA). In (NZB × NZW)F1 (NZB/NZW) mice, a lupus-like disease develops spontaneously and is complicated by severe, progressive nephritis. The production of circulating anti-dsDNA autoantibodies is accompanied by the deposition of immune complexes within extracellular matrices, particularly the glomerular basement membrane. This process is commonly accompanied by the apparent expansion of extracellular matrices. The initiating mechanisms remain unknown but are thought to involve signaling pathways triggered by the deposited immune complexes.

In the current study, we analyzed whether the Wnt signaling pathway may be involved in progression of lupus nephritis, based on emerging reports indicative of a role of this pathway in renal extracellular matrix homeostasis. Increased renal Wnt signaling was recently observed in various models of renal fibrosis, including unilateral ureteral obstruction (3, 4). In this model, activation of the Wnt/β-catenin pathway caused increased interstitial expression of extracellular matrix constituents, including type I procollagen and fibronectin (4). Moreover, inhibition of Wnt signaling by overexpression of the secreted Wnt antagonist Dkk-1 attenuated interstitial collagen matrix accumulation and the severity of renal fibrosis (4). These data point to Wnt signaling as a potential intervention target in fibrotic and possibly inflammatory renal disease processes (5). In addition, increased Wnt signaling has been shown to induce the expression of matrix metalloproteinases (MMPs) (6, 7), which may be of importance in extracellular matrix remodeling and the loss of membrane integrity that occur in lupus nephritis (8).

Less is known about the roles of Wnt signaling in glomerular cells and diseases affecting the glomerular compartment. A recent study revealed that adriamycin-induced nephropathy caused Wnt/β-catenin signaling in podocytes and demonstrated a crucial role of β-catenin expression in the consequent development of proteinuria (9). Increased β-catenin expression was also demonstrated in biopsy specimens obtained from patients with diabetic nephropathy and focal segmental glomerulosclerosis (9). Although the spectrum of effects of renal Wnt signaling remains incompletely characterized, these data are suggestive of a role of Wnt/β-catenin signaling in several glomerulopathies (9). Based on these results and as part of the ongoing effort to characterize the molecular changes underlying the development and progression of lupus nephritis, we evaluated renal Wnt signaling in NZB/NZW mice at different stages of disease progression.

Various disturbances in the clearance of apoptotic cells have been reported and are thought to play a role in the development of SLE (10, 11), but key elements in the pathogenesis of the disease are still unknown. Conceivably, surpassing the capacity for apoptotic cell clearance could trigger a local inflammatory reaction via the exposure of endogenous danger signals (12). Under such circumstances, apoptotic cell antigens are taken up by antigen-presenting cells (e.g., dendritic cells) that have been activated by proinflammatory signals. This allows for immunogenic presentation of self antigens, which might launch and maintain an autoimmune response (for review, see ref.13). In this respect, an increased load of apoptotic cells has been proposed to inhibit clearance mechanisms (14). An emerging aspect of Wnt signaling seems to be its role in the protection of cells against apoptosis upon exposure to various types of cellular stress, at least in vitro (15, 16). The aim of the current study was to evaluate whether alterations in Wnt pathway activity could influence the apoptotic cell load during the development of lupus in NZB/NZW mice.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Animals, tissue samples, and cell culture.

Female NZB/NZW mice and BALB/c mice were purchased from Harlan. The treatment and care of the mice were performed in accordance with the guidelines of the Norwegian Ethical and Welfare Board for Research Animals, and the study was approved by the Institutional Review Board.

Serum samples from NZB/NZW mice and BALB/c mice were collected every second week and stored at −20°C. Proteinuria in NZB/NZW mice was monitored weekly by semiquantitative dipstick measurements (Bayer). At the age of 10 weeks and 20 weeks or upon development of overt nephritis (proteinuria >3 gm/liter), mice were killed by CO2 suffocation. The kidneys were extirpated, cut, and preserved in RNAlater (Qiagen) for further analysis of gene expression or were snap-frozen in liquid nitrogen.

Mouse mesangial cells were obtained from the American Type Culture Collection and cultured in Dulbecco's modified Eagle's medium/Nutrient Mixture F-12 Ham (Sigma-Aldrich) with 20% fetal bovine serum, 2 mML-glutamine, 50 mM penicillin, and 50 mM streptomycin. Clonetics human renal proximal tubular epithelial cells (Lonza) were maintained in serum-free renal epithelial cell culture medium with supplements provided by the manufacturer. All cells used were from passage 4. Recombinant human and mouse Dkk-1 were obtained from R&D Systems.

RNA extraction and complementary DNA (cDNA) synthesis.

Total RNA was isolated from RNAlater-preserved kidneys (5 mice in each group) using EZ1 RNA Tissue Mini Kit (Qiagen). The concentration of extracted RNA was determined spectrophotometrically with a NanoDrop spectrophotometer. Samples (2 μg RNA in 100-μl reaction mixture) were reverse transcribed into cDNA with random primers, using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems).

Real-time polymerase chain reaction (PCR).

Quantitative PCR was performed using an ABI Prism 7900HT Sequence Detection System (Applied Biosystems). Predesigned FAM-labeled primer/probe sets (Applied Biosystems) specific for the analyzed genes (Axin2, accession no. Mm00443610_m1; Actb, accession no. Mm00607939_s1; Dkk1, accession no. Mm00438422_m1; Myc, accession no. Mm00487803_m1; Mycn, accession no. Mm00627179_m1; Ppard, accession no. Mm00803184_m1; Lef1, accession no. Mm01310389_m1) were used, and TATA box binding protein (accession no. Mm00498583_m1) was used as an endogenous control. FAM-labeled primer/probe sets for additional TCF/LEF-responsive genes were obtained from Integrated DNA Technologies, and the sequences are shown in the legend of Table 1. The relative expression levels were calculated using the ΔΔCt method. Normalization was accomplished by comparing the relative expression for each sample with that for one of the 4-week-old control BALB/c mice. For cell culture experiments, samples were normalized against untreated controls.

Table 1. Expression of selected TCF/LEF-responsive genes*
MiceMycRhoUFoxn1Dvl2Jun
  • *

    Values are the mean ± SEM fold change relative to the average expression level in 10-week-old (NZB × NZW)F1 mice (n = 5) after normalization for the expression level of TATA box binding protein within the same samples. The following forward primer and probe sequences were used: for Tbp, forward AAGAAAGGGAGAATCATGGACC, probe CCTGAGCATAAGGTGGAAGGCTGTT, reverse GAGTAAGTCCTGTGCCGTAAG; for Myc, forward GCTGTTTGAAGGCTGGATTTC, probe CGTAGTCGAGGTCATAGTTCCTGTTGGT, reverse GATGAAATAGGGCTGTACGGAG; for RhoU, forward CCCACCGAGTACATCCCTAC, probe ATGAGTTTGACAAGCTGAGGCCCC, reverse TGTCTGTGTTGGTGTAGCAG; for Foxn1, forward AAGAACAGTAAGACCGGAAGC, probe TTCTGTTCGCCATAACCTGTCCCTC, reverse TCTCCACCTTCTCAAAGCAC; for Dvl2, forward TTTCAAGAGCGTTTTGCAGC, probe TCCGATGACAATGCCCGCCTAC, reverse ACACAAGCCAGGAGACAAC; for Jun, forward TTGTTACAGAAGCAGGGACG, probe AGGCTAACCCCGCGTGAAGT, reverse GTCGTAGAAGGTCGTTTCCATC. TCF/LEF = T cell factor/lymphocyte enhancement factor.

  • P < 0.05 versus BALB/c mice.

BALB/c1.42 ± 0.361.15 ± 0.290.85 ± 0.551.02 ± 0.150.46 ± 0.04
(NZB × NZW)F1     
 10 weeks old1.02 ± 0.251.02 ± 0.281.01 ± 0.111.00 ± 0.041.00 ± 0.08
 20 weeks old1.48 ± 0.261.39 ± 0.281.65 ± 0.451.31 ± 0.140.83 ± 0.18
 Proteinuric8.5 ± 3.525.19 ± 1.2118.38 ± 10.31.77 ± 0.482.81 ± 1.65

Immunofluorescence microscopy.

Four-micrometer–thick OCT compound–embedded cryostat sections of kidney were blocked with 10% calf serum and 1% bovine serum albumin (BSA) in phosphate buffered saline (PBS). The sections were subsequently incubated with a rabbit anti-human Dkk-1 polyclonal antibody, rabbit anti-human phosphorylated glycogen synthase kinase 3β (GSK3β; Santa Cruz Biotechnology), or a rabbit anti-human β-catenin polyclonal antibody (Abcam) followed by incubation with an Alexa 546–conjugated anti-rabbit F(ab′)2 secondary antibody (Invitrogen). Normal rabbit IgG was used as a negative control.

For immunostaining using a mouse anti-human activated β-catenin monoclonal antibody (Millipore), ultrathin zinc-fixated cryosections were incubated with 0.1 mg/ml anti-mouse F(ab′)2 overnight at 4°C to reduce background staining by in vivo–bound antibodies. Sections were then processed as described above. All sections were run in parallel, with the omission of the primary antibody to evaluate background staining for each section.

Anti-dsDNA and anti–Dkk-1 enzyme-linked immunosorbent assay (ELISA).

Serum antibodies against calf thymus dsDNA were detected and titrated by standard indirect ELISA, as previously described (17). A murine monoclonal anti-dsDNA antibody was included as an intraassay control for the cutoff level (18). Detection of serum Dkk-1 was performed using the mouse Dkk-1 DuoSet ELISA kit (R&D Systems) according to the supplied protocol. Serum samples were diluted 1:6 in PBS with 1% BSA.

Protein isolation and Western blot analysis.

Protein extracts were prepared from snap-frozen kidney tissue by homogenization in Tris buffer (0.5 moles/liter Tris HCl, pH 7.5, 150 mmoles/liter NaCl). The samples were then centrifuged for 5 minutes, and the supernatant was collected. Nuclear and nucleus-depleted lysates were prepared from 20-mg pieces of snap-frozen lyophilized kidney tissue, using the Pierce Nuclear and Cytoplasmic Extraction Reagent Kit. Total protein was measured, and the protein content of the samples was normalized using the Pierce BCA Protein Assay.

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis and Western blotting were performed according to standard procedures. Membranes were blocked with 5% (weight/volume) BSA and probed with the relevant primary antibody overnight at 4°C. Membranes were washed and incubated with a horseradish peroxidase–conjugated secondary antibody, and binding was assayed by chemiluminescence detection. The determination of molecular weight was performed using MagicMark XP molecular weight markers (Invitrogen).

Flow cytometry.

Cell samples were analyzed on a FACSCalibur cytometer (BD Biosciences). Apoptosis was detected using an annexin V–fluorescein isothiocyanate and propidium iodide staining kit (BD Biosciences) according to the protocol supplied by the manufacturer. Data were analyzed using CellQuest software (BD Biosciences).

Statistical analysis.

Results are presented as the mean ± SEM. Comparisons between groups were performed by one-way analysis of variance followed by Bonferroni's correction for multiple comparisons, using GraphPad Prism version 5.0. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Increased expression of messenger RNA (mRNA) for β-catenin and TCF/LEF-responsive genes in kidneys from preproteinuric- and proteinuric-stage NZB/NZW mice.

Real-time PCR analysis of mRNA isolated from kidney homogenates revealed an increase in the expression of β-catenin, occurring at ∼20 weeks and further increasing during the progression toward proteinuria (Figure 1A). This increase in gene expression paralleled the appearance of significant titers of anti-dsDNA autoantibodies (see below). In contrast, β-catenin was stably expressed at a lower level in age-matched BALB/c mice (Figure 1A). In order to determine whether the increase in β-catenin expression was a reflection of increased Wnt activity, the mRNA expression of several known TCF/LEF-responsive genes, including Axin2, Ppard, Myc, Lef1, and RhoU (19, 20), was assayed. The results revealed a consistent increase in the expression of these genes in NZB/NZW mice (Figures 1B–D and Table 1) that paralleled the increase in β-catenin mRNA expression, suggesting that canonical Wnt signaling is increased within the kidneys during the development of lupus nephritis. In contrast, a reduction in Wnt pathway activity with increasing age was seen in BALB/c mice (data not shown).

thumbnail image

Figure 1. Wnt/β-catenin pathway activity in (NZB × NZW)F1 (NZB/NZW; B/W) mouse kidneys. Real-time polymerase chain reaction analysis was performed on RNA isolated from NZB/NZW and BALB/c mice of different ages (n = 5 per age group). Gene expression data are shown for β-catenin (A) and the T cell factor/lymphocyte enhancement factor–regulated genes Axin2 (B), Ppard (C), and Lef1 (D). These data demonstrate an increase in gene expression during progression toward proteinuria. Gene expression in BALB/c mice of corresponding ages remained stable or decreased with increasing age. Expression levels were normalized against that of TATA box binding protein and are expressed as fold change values relative to 8-week-old mice, using the ΔΔCt method. Values are the mean ± SEM. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001.

Download figure to PowerPoint

Immunofluorescence studies of kidney biopsy specimens revealed an increase in total β-catenin staining throughout the tubular and glomerular compartments, paralleling the increase in β-catenin mRNA expression (Figure 2B). Glomerular staining was particularly strong in proteinuric-stage mice. These findings suggest that the observed increase in Wnt/β-catenin activity occurs diffusely throughout the kidney. Immunostaining of cryosections gave no detectable signals for active β-catenin in BALB/c mice or 10-week-old NZB/NZW mice, whereas diffuse intracellular staining was seen in 20-week-old NZB/NZW mice (Figure 2C). Similarly, immunoblotting against the phosphorylated, inactive form of GSK3β was increased in 20-week-old NZB/NZW mice (Figure 3C), with corresponding speckled intracytoplasmic immunostaining in cryosections from these mice (Figure 2D). Immunoblotting of nuclear protein extracts revealed increased expression of the active form of β-catenin in 20-week-old NZB/NZW mice (Figure 2A), further supporting activation of Wnt signaling in these mice.

thumbnail image

Figure 2. Increased abundance of active (Act.) β-catenin in NZB/NZW (B/W) mouse kidneys. A, Western blot analysis of nuclear protein extracts demonstrated increased amounts of the active (nonphosphorylated) form of β-catenin in 20-week-old and proteinuric-stage NZB/NZW mice compared to 10-week-old mice. β-actin was used as a loading control. B–D, Immunofluorescence staining using an anti–β-catenin antibody (B) and an antibody specific for the active, nonphosphorylated form of β-catenin (C) demonstrated increased amounts of β-actin immunostaining in 20-week-old and proteinuric-stage NZB/NZW mice compared to 10-week-old mice. Scattered, intracytoplasmic immunostaining against phosphorylated, inactive glycogen synthase kinase 3β (pGSK3β) was detectable in proteinuric-stage NZB/NZW mice but not in 10-week-old mice (D). Images were obtained using identical exposure settings. Tot. = total. Original magnification × 200.

Download figure to PowerPoint

Increased renal expression of Dkk-1 during development of lupus nephritis.

Dkk-1 is a secreted inhibitor of the canonical Wnt pathway and is known to be induced by canonical Wnt signaling (21), probably serving as a negative feedback mechanism. In line with our findings of increased expression of TCF/LEF-responsive genes, mRNA expression of Dkk-1 in kidney homogenates increased beginning at 20 weeks of age (Figure 3A).

thumbnail image

Figure 3. Dkk-1 expression in mouse kidneys. A, Gene expression for Dkk-1 in renal homogenates from NZB/NZW (B/W) mice and BALB/c mice (n = 5 per age group) showed increased expression in 20-week-old and proteinuric-stage NZB/NZW mice. Expression levels were normalized against TATA box binding protein and expressed as the fold change values, using the ΔΔCt method. B, Immunofluorescence staining of kidney cryosections demonstrated a similar increase in Dkk-1 immunostaining in 20-week-old and proteinuric-stage NZB/NZW mice as compared to that in prediseased 10-week-old mice. Immunostaining was increased within both the tubular and glomerular compartments. Images were obtained using identical exposure settings. Original magnification × 200. C, Western blotting against Dkk-1 and the phosphorylated, inactivated form of GSK3β (pGSK3β) in renal protein homogenates demonstrated increased levels of Dkk-1 and pGSK3β in 20-week-old and proteinuric-stage NZB/NZW mice compared to those in 10-week-old NZB/NZW mice and BALB/c mice. β-actin was used as a loading control. ∗∗ = P < 0.01; ∗∗∗ = P < 0.001.

Download figure to PowerPoint

Immunofluorescence staining similarly revealed a strong increase in Dkk-1 expression in both tubular and glomerular cells from NZB/NZW mice, in both the preproteinuric (20 weeks old) and proteinuric groups compared to younger NZB/NZW mice, in which Dkk-1 immunostaining was weaker, at levels comparable to those in BALB/c mice (Figure 3B). Western blotting revealed similar increases in Dkk-1 expression in 20-week-old and proteinuric mice (Figure 3C).

Wnt-inhibitory effects of lupus serum.

The observed increase in renal Wnt/β-catenin signaling could be caused by either intrinsic or circulating Wnt agonists. In order to evaluate the net influence of serum factors on canonical Wnt pathway activity, mesangial cells were exposed to serum from NZB/NZW mice and BALB/c mice of different ages. Messenger RNA expression for the TCF/LEF-responsive genes Axin2 and Ppard was assayed, revealing a dramatic decrease in the expression of these genes upon exposure to sera from 20-week-old and proteinuric-stage NZB/NZW mice compared to sera from 10-week-old mice (Figures 4A and B). This suggests that NZB/NZW mouse sera contain increased levels of ≥1 Wnt inhibitors, and that the regulation of renal Wnt activity in lupus nephritis is not controlled by circulating factors.

thumbnail image

Figure 4. Serum Dkk-1 concentration (Conc.) and in vitro Wnt-inhibitory effects of serum obtained from mice with lupus. A and B, The expression of axin2 (A) and Ppard (B) was determined in primary mouse mesangial cells following 6-hour incubation in the presence of serum (10% by volume) from NZB/NZW mice of various ages. Expression was decreased in cells exposed to serum from 20-week-old and proteinuric-stage NZB/NZW mice (n = 5 per group) compared to that in cells exposed to serum from 10-week-old NZB/NZW mice. Expression levels were normalized against TATA box binding protein and expressed as the fold change, using the ΔΔCt method. C, Serum concentrations of Dkk-1, as determined by enzyme-linked immunosorbent assay, were significantly increased in 20-week-old and proteinuric-stage NZB/NZW mice compared to the concentrations in control BALB/c mice and 10-week-old NZB/NZW mice (n = 5 per group). D, Serum titers of anti–double-stranded DNA (anti-dsDNA) autoantibodies and serum concentrations of Dkk-1 were determined by serial sampling of NZB/NZW mice at different ages. Values are the mean ± SEM. ∗∗ = P < 0.01; ∗∗∗ = P < 0.001. O.D. = optical density.

Download figure to PowerPoint

Based on our findings of increased expression of Dkk-1 in the kidneys of NZB/NZW mice, Dkk-1 levels in sera were determined by ELISA. Serum samples obtained from normal BALB/c mice showed stable concentrations in all age groups (data not shown). In NZB/NZW mice, Dkk-1 was present at levels comparable to those in control BALB/c mice at age 10 weeks (Figure 4C). In 20-week-old and proteinuric-stage NZB/NZW mice, there was a 5-fold increase in the serum level of Dkk-1 compared to that in the younger mice (Figure 4C).

Serial measurements of anti-dsDNA titers and serum Dkk-1 revealed a significant increase in the concentration of Dkk-1 at ∼18 weeks, with an apparent inverse correlation between the pattern of fluctuation in serum Dkk-1 and the anti-dsDNA titer (Figure 4D). To evaluate whether these fluctuations in serum Dkk-1 concentrations could be explained by autoantibodies against Dkk-1, immunoblotting of sera against Dkk-1 was performed under denaturing conditions, to reveal Dkk-1 masked by autoantibodies. The results of these studies were consistent with the ELISA results, suggesting that the variations in serum Dkk-1 concentrations are not explained by the presence of autoantibodies against this protein (data not shown).

Proapoptotic effects of Dkk-1.

Incubation of cultured primary mesangial cells with increasing concentrations of recombinant Dkk-1 led to induction of apoptosis at concentrations comparable to those seen in serum from mice with lupus (12,000–20,000 pg/ml), as evidenced by increased mRNA expression of the apoptosis-related genes casp3 (Figure 5A) and Bax (data not shown). In contrast, expression of the apoptosis-inhibiting factor Bcl2 remained relatively stable, as reflected by an increased Bax:Bcl2 ratio at Dkk-1 concentrations >10,000 pg/ml (Figure 5B), which is considered to be a sign of induction of apoptosis in mesangial cells (22, 23).

thumbnail image

Figure 5. In vitro induction of apoptosis by recombinant Dkk-1. A, The addition of recombinant Dkk-1 to cultured primary mouse mesangial cells (n = 5 per group) revealed increased mRNA expression levels of the proapoptotic caspase 3 gene at high concentrations (Conc.) of Dkk-1 (>10,000 pg/ml) following 6-hour incubation. B, The mRNA expression ratio for Bax and Bcl-2 was significantly increased at Dkk-1 concentrations >10,000 pg/ml after 6-hour incubation. C, Apoptosis was assayed using flow cytometry after fluorescence-based annexin V/propidium iodide staining, revealing an increased percentage of annexin V positivity in both murine mesangial cells and human proximal tubular epithelial cells after 24-hour incubation with Dkk-1 in the concentration range seen in NZB/NZW mice (10,000–20,000 pg/ml). All experiments were performed in triplicate. D, The mRNA expression ratio for Bax and Bcl-2 following 6-hour incubation of mesangial cells (n = 5 per group) with serum (20% by volume) from NZB/NZW mice (10 weeks old, 20 weeks old, and proteinuric-stage) revealed an increased ratio at increasing ages. Values are the mean ± SEM. ∗ = P < 0.05; ∗∗ = P < 0.01; ∗∗∗ = P < 0.001.

Download figure to PowerPoint

Cell surface annexin V staining of mouse mesangial and human proximal tubular cells after 24 hours of incubation with Dkk-1 showed a dramatically increased number of apoptotic cells in the Dkk-1 concentration range observed in sera from mice with lupus, with only a modest increase in apoptosis at concentrations comparable to those in BALB/c mouse sera (Figure 5C). These data indicate a strong proapoptotic effect of the increased circulating levels of Dkk-1 observed in NZB/NZW mice during lupus development.

Consistent with previous reports suggestive of proapoptotic effects of sera from patients with SLE (24), incubation of murine mesangial cells with sera from NZB/NZW mice caused an increased Bax:Bcl2 ratio using sera from 20-week-old and nephritic-stage mice compared to younger mice (Figure 5D). Similar results were obtained using human proximal tubular cells (data not shown).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Our data show evidence of increased renal expression of β-catenin and several TCF/LEF-responsive genes during the preproteinuric phase of the development of lupus nephritis in NZB/NZW mice, reflecting increased renal Wnt/β-catenin signaling during disease progression. This suggests a role of canonical Wnt signaling in the development of nephritis and, based on recent results for other glomerulopathies (9), provides a potential mechanism to explain the extracellular membrane expansion that accompanies progression toward proteinuria and renal failure. Previous work by our group revealed an increase in the glomerular expression of MMP-2 and MMP-9 during the development of lupus nephritis (25), which was accompanied by qualitative changes in the composition of the type IV collagen matrix of the glomerular basement membrane. Interestingly, secretion of both MMP-2 and MMP-9 is induced by canonical Wnt signaling in T cells in vitro (26). Little is known about the effects of increased Wnt pathway activity on glomerular extracellular matrix synthesis, and this issue will need to be addressed by in vivo studies. Such studies are further warranted by recent data implicating increased β-catenin expression as a mediator of podocyte dysfunction and proteinuria in mice (9).

Increased serum and renal levels of the Wnt inhibitor Dkk-1 paralleled the observed increase in renal canonical Wnt pathway activity. Dkk1 is a Wnt-responsive gene in what is presumed to be a negative regulatory feedback mechanism of Wnt signaling (21). Although it is seemingly counterintuitive, increased levels of Dkk-1 within the kidney could thus be seen as a compensatory response to the observed increase in Wnt signaling. The discrepancy between increased Wnt activity in kidney and a Wnt-inhibitory effect of NZB/NZW mouse serum in vitro suggests that circulatory factors are not responsible for the changes occurring within the kidney. Wnt pathway expression has been demonstrated to protect against proapoptotic stimuli by inhibiting expression of the Bcl-2 protein Bax in renal epithelial cells (27). It is therefore plausible that the observed increase in renal Wnt/β-catenin expression could represent a compensatory response to increased cellular stress. Such stress could stem from immune complex–mediated complement activation and other proinflammatory events occurring within the kidney and other tissues.

The increased levels of Dkk-1 could explain the Wnt-suppressive effect of NZB/NZW mouse serum, but the relative contribution of Dkk-1 and other serum factors has not yet been determined and is currently being investigated. The temporal association between increased serum Dkk-1 levels and the emergence of anti-dsDNA autoreactivity is interesting and could reflect processes related to the initiating stages of SLE development.

Reduced apoptotic cell clearance has been suggested as a key element in the pathogenesis of SLE (10, 28). Immune complexes containing chromatin particles have been shown to be deposited within both the glomerular basement membrane and the mesangial extracellular matrix and are thought to serve as a source of inflammatory signaling, but the origin of these chromatin deposits remains unknown. Wnt signaling has been shown to reduce the susceptibility to apoptosis in normal and malignant cells upon exposure to various types of cellular stress (29). Dkk-1 has proapoptotic effects that appear to be mediated by its inhibitory effect on canonical Wnt signaling (30). The involved mechanisms have not been fully characterized but seem to include the p53 pathway (30). The finding that Dkk-1 serves as a proapoptotic factor in mesangial cells (15) offers an attractive explanation for the appearance of chromatin-containing immune complexes within the mesangium. The presence of circulating anti-dsDNA autoantibodies favors immune complex formation in situ (31) or within the circulation (32), and deposition of anti-dsDNA will in turn allow for complement activation and generation of an inflammatory focus. If they occur in the context of tissue damage, proapoptotic factors such as Dkk-1 could promote a sustained source of apoptotic cells within an inflammatory milieu. This could provide substrates for the perpetuation of an autoimmune reaction.

Interestingly, previous studies have demonstrated the presence of ≥1 apoptosis-inducing factors in sera from patients with SLE (24). The identity of these factors remains unknown, although they have been demonstrated to involve death receptor–independent activation of caspase-mediated apoptosis (33). The finding of proapoptotic effects of Dkk-1 in the concentration range observed in older NZB/NZW mice provides a potential explanation for such findings.

The origin of the observed increase in serum Dkk-1 expression in NZB/NZW mice remains unclear. Recent data suggest that both type I and type II interferons (IFNs) can induce Dkk-1 expression both in vitro and in vivo (34–36). Disturbances in IFN signaling have been a topic of interest in SLE research, and increased levels of circulating IFNα are commonly seen, correlating with the level of disease activity and severity (37, 38). The increased levels of Dkk-1 observed in kidneys and possibly in other tissues could also serve as a potential contributor to increased serum levels of the protein. It has been shown that Dkk-1 is induced upon p53-mediated proapoptotic signaling (39). Given the increased amounts of apoptotic cells within the circulation of patients with SLE, it is therefore also possible that serum Dkk-1 could be a reflection of an increased circulating apoptotic cell load. This could also serve as a link between the fluctuations seen in the Dkk-1 concentration and the anti-dsDNA titer.

In this study, we describe the involvement of the canonical Wnt/β-catenin signaling pathway in lupus nephritis. Sera obtained from NZB/NZW mice inhibited canonical Wnt signaling and promoted apoptosis in vitro. Increased circulating levels of the Wnt antagonist Dkk-1 were seen beginning at the onset of anti-dsDNA autoreactivity and could be an important factor in disease pathogenesis by increasing the extracellular load of potentially immunogenic chromatin. Kidneys from NZB/NZW mice showed evidence of increased Wnt/β-catenin signaling, adding to an increasing number of renal diseases in which this pathway seems to have a pathogenetic impact. These results provide a framework for studies of the roles of the Wnt signaling pathway in human SLE and possibly in other autoimmune diseases.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Tveita had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Tveita, Rekvig.

Acquisition of data. Tveita.

Analysis and interpretation of data. Tveita, Rekvig.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES