Animals and cell cultures.

Both WT C57BL/6 (B6) and Clu KO mice in a B6 background (B6-Clu^−/−) were received from the breeding colonies in the animal facility at the Jack Bell Research Centre (Vancouver, BC). All the animals (males, 10–12 wk old) for the experiments were cared for in accordance with the Canadian Council on Animal Care guidelines under the protocols approved by the Animal Use Subcommittee at the University of British Columbia (Vancouver, BC).

Clu null TECs were isolated from the kidney cortex of Clu KO mice as described previously (56) and were immortalized or cloned by transfection with origin-deficient SV40 DNA. The phenotype of cloned murine TECs was confirmed by their expression of E-cadherin and CD13 (alanine aminopeptidase) using fluorescence-activating cell sorting (FACS) analysis and were grown in complete K1 (K1^+/+) medium as described previously (11).

Ectopic expression of Clu in cultured TECs.

Clu null TECs were converted to Clu-expressing TECs (TEC-Clu^hClu) by stable expression of human Clu isoform 1 cDNA using pHEX6300 vector, as described previously (27), whereas Clu null TECs with stable expression of empty pHEX6300 vector were used as Clu-negative control cells (TEC-Clu^−/−). Both cell types were grown and maintained in complete K1 medium in the presence of the selective antibiotic zeocin (up to 500 μg/ml) and were further selected by flow cytometric cell sorting based on the expression of pHEX6300-encoded green fluorescent protein (GFP). More than 95% of cells in both TEC-Clu^hClu and TEC-Clu^−/− cultures were confirmed to have a high level of GFP by either flow cytometry or fluorescence microscopy.

Renal IRI.

Renal IRI was induced in the left kidney of WT vs. Clu KO mice. In brief, mice were anesthetized with a combination of ketamine (100 mg/kg) and xylazine (10 mg/kg), and isoflurane as needed. The left kidneys were exposed through a flank incision, followed by the induction of ischemia in these kidneys through clamping renal pedicles at the body temperature of 32°C for either 28 min (renal repair) or for 38 min (mouse survival). After the clamps were released, reperfusion of the kidneys was confirmed visually. In the examination of renal repair/regeneration in these left kidneys after reperfusion, the nonischemic right kidneys in the same mice were kept for life support. In the examination of renal function or mouse survival after IRI, the right kidneys were removed (unilateral nephrectomy).

Western blotting.

Clu protein in protein extracts of TECs and renal tissues was examined by Western blotting as described previously (56). Briefly, protein extract samples (100–150 μg/sample) were fractionated by 10% SDS-PAGE, then transferred onto nitrocellulose membranes. The levels of Clu protein were identified with goat polyclonal anti-Clu-α antibodies (1:1,000 dilution; C-18, Santa Cruz Biotechnology, Santa Cruz, CA) and visualized by an enhanced chemiluminescence assay (ECL, Amersham Pharmacia Biotech, Buckinghamshire, UK). The β-actin as a protein content determinant in each sample was reprobed using anti–actin IgG on the same blot (Sigma-Aldrich Canada, Oakville, ON). To quantitate the Clu expression in Western blot analysis, the density of Clu (sClu) and reprobed β-actin bands was measured by densitometry. In each blot, the relative level of sClu in each sample was calculated by normalization with its density of β-actin.

Immunohistochemical analysis.

The kidneys harvested from mice were perfused with PBS before formalin fixation, paraffin embedding, and sectioning. The expression of Clu or Ki67 protein in kidney sections was assessed by a standard immunohistochemical method. Briefly, after deparaffin and rehydration buffered-formalin-fixed sections were treated with 3% H₂O₂ in Tris-buffered saline (TBS; pH 7.4) for 30 min at room temperature (RT) to quench endogenous peroxidase, followed by permeabilization with 0.2% Triton X-100 for 10 min at RT. After being washed with TBS containing 0.1% Tween 20 (TBS-T) and blocked with 2% normal serum, the sections were incubated with 1:50 dilution of primary goat polyclonal anti-Clu-α antibody (C-18, Santa Cruz Biotechnology) or rabbit polyclonal anti-Ki67 antibodies (AB9280, EMD Millipore, Billerica, MA) overnight at 4°C. The immune complexes of Clu and anti-Clu antibody or of Ki67 and anti-Ki67 antibody on the tissue section were detected using anti-goat or anti-rabbit Ig antibody conjugated with biotin, the secondary antibody, and were visualized using a 3,3′-diaminobenzidine (DAB) peroxidase substrate kit (Vector Labs, Burlington, ON). The control negative staining included the sections incubated with normal goat or rabbit IgG instead of anti-Clu or anti-Ki67 antibody as the primary antibody.

To quantitate the expression of Clu or Ki67 in the kidney in the immunohistochemical analysis, the number of Clu-expressing tubules or Ki67-expressing cells was counted in each view, which was randomly selected in the region of renal cortex under ×400 magnification (high-power field; hpf). The Clu-expressing tubules or Ki67-positive cells in each kidney were quantitated by averaging at least 20 nonoverlapping fields in two serial sections.

Determination of renal function and semiquantitative assessment of renal injury.

The function of the kidneys was determined using the levels of SCr or BUN. SCr was measured by using the CREA method on the Dimension Vista System with a Flex reagent cartridge in the Chemistry Laboratory at Vancouver Coastal Health Regional Laboratory Medicine (Vancouver, BC), and BUN measurement was performed using a QuantiChrom urea assay kit (BioAssay Systems).

Histological assessment of tubular injury in kidney sections was performed in a blinded fashion. Formalin-fixed and paraffin-embedded sections (5-μm thickness, longitudinal) were stained by periodic acid-Schiff (PAS) methods. The percentage of damaged tubules (combined necrosis and vacuolization) in the total tubules was counted in each view, randomly selected in the region of renal cortex under ×400 magnification, and averaged at least of 20 nonoverlapping fields for each kidney.

PCR array.

The expression of 84 cell cycle-associated genes in the kidneys or cultured TECs was quantitatively examined using PCR Array kits following the manufacturer's instructions (SABiosciences-Qiagen, Valencia, CA). The kidney samples were prepared as follows: after perfusion with saline, one part of renal tissue was snap-frozen in liquid nitrogen and stored at −80°C for RNA extraction. The RNA in TECs was directly extracted from the monolayer after the removal of culture medium. In each group (Clu-expressing vs. Clu null), four different samples (renal tissue samples from 4 different animals, or TEC cultures grown at 4 different times) were randomly selected for the determination of the gene expression profile using PCR arrays. The total RNA from these samples was extracted and purified using an RNeasy Microarray Tissue Mini kit (Qiagen), and converted to cDNA using an RT² First Strand Kit (Qiagen). The expression of selected genes was amplified by real-time PCR using RT² Profile PCR arrays (Qiagen). Data were analyzed using Web-based PCR Array Data Analysis Software (www.SABiosciences.com/pcrarraydataanalysis.php).

Cell proliferation by methylthiazol tetrazolium assay.

Methylthiazol tetrazolium (MTT) assay was used to measure cell proliferation or viable cell number in cultures. Cells were seeded at 2,500/well in 96-well microplates, and each condition was tested at least in quadruplicate. In brief, 10 μl of 0.5 mg/ml of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich Canada) were added to each well and incubated at 37°C for 4 h. The formazan crystals in viable cells were then dissolved in 100 μl/well of DMSO (Sigma-Aldrich Canada). The OD of the color in each well as a proliferation index was quantified at 562-nm wavelength using an ELx808 Ultra Microplate Reader (BioTek, Winooski, VT). The growth rate was calculated as follows: growth rate (OD/day) = (OD^end − OD^begin)/time (day), where time represented the period of examination, OD^end represented the absorbance unit at the end of the examination, and OD^begin represented the absorbance unit at the beginning of the examination.

Cell cycle distribution analysis.

TECs (1 × 10⁶ cells/well) were seeded in six-well plates in complete K1 medium for 24 h and arrested and synchronized cell growth in serum growth factor-free K1 medium overnight. The cells were incubated in complete K1 medium for 24 h, followed by harvesting by a brief trypsinization and centrifugation, and finally washed twice with ice-cold PBS. The staining with propidium iodide (PI; Invitrogen Canada) was performed as follows: cells were fixed with ice-cold 70% ethanol overnight, followed by treatment with 80 μg/ml of DNase-free RNase and staining with 40 μg/ml of PI at 37°C for 30 min. The DNA content was measured using flow cytometry or FACS and then analyzed with flowJo analysis software (Tree Star, Ashland OR).

Cell migration in a wound-healing assay.

The migration of TECs (TEC-Clu^hClu vs. TEC-Clu^−/−) was compared in monolayer cell cultures. Briefly, cells (1 × 10⁶ cells/well) were seeded in six-well plates. After overnight incubation, a wound in a cell monolayer was created by scratching with a pipette tip, followed by a gentle rinse twice with warm PBS to remove the detached cells. Cell migration at the wound edge to the wound space was captured using a Zeiss 200M Axiovert inverted microscope (Carl Zeiss Microscopy, Gottingen, Germany), and the migration distance from 0 to 6 h was calculated using Axiovison Rel 4.8 software (Carl Zeiss Microscopy).

Statistical analysis.

Two-way ANOVA or t-tests (2-tailed distribution) were used as appropriate for comparisons between groups. Data were collected from individual experiments or mice in each study for statistical analysis. A P value of ≤0.05 was considered significant.

Clu expression is required for renal regeneration after renal IRI.

The kidney is able to repair itself after AKI by both intrinsic and extrinsic mechanisms (13), which have not yet been fully understood. To examine the impact of Clu expression on renal repair after AKI, renal function and tissue regeneration after IRI in Clu-expressing WT vs. Clu KO mice were examined.

The kinetics of both renal repair and Clu protein expression were first examined during the recovery phase of IRI. As shown in Fig. 1A, the expression of Clu protein in renal cortical tissue extracts from WT mice was upregulated following renal IRI in a time-dependent manner, reaching the highest level at day 3 postreperfusion, but declining to almost the basal levels at day 7. By immunohistochemical analysis, the expression of Clu protein was localized mainly in the renal tubules (Fig. 1B), while an absence of Clu protein was detected in both extracts and sections of the kidneys from Clu KO mice. To reveal the correlation between renal Clu expression and tissue repair, the levels of tissue damage in the cortex of between these WT and Clu KO kidneys were semiquantitatively compared by histological analysis with PAS staining. As shown in Fig. 2, the basal level of damaged tubules (tubules/hpf) of WT at day 0 (1.26 ± 0.55) was not significantly different from that in KO controls (0.94 ± 0.27), and following IRI, it was increased to 7.77 ± 3.76 at day 1 in WT kidneys, and 13.62 ± 7.33, a maximum level, at day 3, followed by the presence of less tubular damage (4.13 ± 2.45) and tissue repair at day 7. However, in Clu KO kidneys, the tubular damage was more severe than in WT mice (P < 0.0001, WT vs. KO, 2-way ANOVA), and was not repaired at all by day 7, which was indicated by a progressive increase in tubular damage from 8.58 ± 1.12 at day 1 to 22.14 ± 1.7 at day 7. Taken together, these data suggest that IRI-induced Clu expression positively correlates with the levels of tubular damage in WT kidneys, and the lack of Clu expression in Clu KO kidneys impairs the renal repair or regeneration after IRI.

Renal repair deficiency in clu KO mice is associated with renal failure and mortality.

To further confirm renal repair deficiency after IRI in Clu KO mice as shown in Fig. 2, both renal function after 28 min of ischemia and mouse survival after 38 min of ischemia were examined in uninephrectomized Clu KO mice compared with WT mice. As shown in Fig. 3, in the KO group seven of nine Clu KO mice exhibited obvious renal failure, indicated by >100 mg/dl BUN and equal to or >0.94 mg/dl SCr (6 of them: over 1 mg/dl SCr: severe renal failure), and one had mild renal failure (BUN: 101.7 mg/dl; SCr: 0.39 mg/dl). In the WT group, renal insufficiency was seen in only two of nine mice; both BUN (108.74 mg/dl) and SCr (0.63 mg/dl) in one mouse were similar to those in Clu KO mice with renal failure, and the other just had a mild degree of renal dysfunction (BUN: 101.48 mg/dl; SCr: 0.34 mg/dl). The statistical analysis indicated that the levels of both BUN and SCr in Clu KO mice were significantly higher than those in WT mice (BUN: P = 0.0241; SCr: P = 0.0008, n = 9). When the ischemia was prolonged to 38 min in this model as shown in Fig. 4, 7 of 10 Clu mice died in a period of 3 days after IRI, while only one died in the WT group (P = 0.0048, log-rank test, n = 10). Taken together, renal failure and mortality of Clu KO mice reflect the poor renal regeneration or repair after IRI in these mice.

Clu-mediated renal repair is associated with cell proliferation in tubules.

To reveal the mechanisms and pathways by which Clu mediates renal repair during recovery from IRI, proliferating cells specifically labeled by the expression of nuclear protein Ki67 (42) were directly counted in the sections of WT vs. Clu KO kidneys under the microscope. As shown in Fig. 5, the numbers of Ki67-expressing tubular cells in the WT group were changed in parallel with degree of tubular damage (Fig. 2); the number of proliferating cells increased from 10.65 ± 3.81 at day 0 to 32.1 ± 11.66 at day 3, followed by a decrease to 21.57 ± 9.26 at day 7, whereas in Clu KO kidneys, the numbers of Ki67-positive cells did not significantly change following reperfusion (9.63 ± 3.42 at day 0 to 14.63 ± 3.88 at day 3 and 14.6 ± 2.92 at day 7) and were statistically lower than those in WT kidneys (P < 0.0001, WT vs. KO, 2-way ANOVA). These data suggest that tubular damage stimulates the cell proliferation in Clu-expressing kidneys, but not in Clu null kidneys, or the lack of Clu expression disrupts cell proliferation in kidney tubules.

To further examine how Clu influences cell proliferation in the kidneys after IRI, the expression of 84 genes identified as positively or negatively regulating the cell cycle, the transitions between the phases, DNA replication, checkpoints, and arrest was examined in the kidneys after 3 days of reperfusion using a Cell Cycle PCR array. As listed in Supplemental Table S1 (all supplemental material for this article is accessible on the journal Web site), the transcription of 16 genes (Shc1, Pmp22, Skp2, Cks1b, Ccnd1, CCnd2, Mki67, Ran, Aurka, Stag1, Tsg101, Gpr132, Msh2, Hus1, Tfdp1, and Slfn1) were statistically upregulated, whereas genes Notch2 and Gadd45a were downregulated in WT kidneys compared with Clu KO kidneys used as a negative control. Among these predominantly 18 Clu-associated genes, 15 are associated with the promotion of cell cycle progression, growth, and DNA repair, while 3 (Notch2, Pmp22, Slfn1) are associated with cell cycle arrest or apoptosis.

Ectopic expression of clu enhances renal tubular cell proliferation.

To further understand the functions of Clu in renal repair, the effect of ectopic expression of human Clu (hClu) on the proliferation or division and migration of TECs was examined. As shown in Fig. 6A, both cClu (presecreted or cytoplasmic form) and sClu (secreted form) were detected in Clu null TEC lines expressing the pHEX6300 vector containing full-length hClu isoform 2 cDNA, TEC-Clu^hClu, while in control cells the same Clu null TEC lines that were stably expressing empty pHEX6300 (TEC-Clu^−/− cells), no Clu proteins were detected.

The proliferation of TEC-Clu^hClu vs. TEC-Clu^−/− cells was first compared under the same normal culture conditions using an MTT assay. As shown in Fig. 6B, the absorbance (OD) values, reflecting viable cell numbers, in TEC-Clu^hClu cell cultures were significantly higher than those in TEC-Clu^−/− cell cultures even during the equilibrium phase at day 3 (P < 0.0001, 2-way ANOVA). Similar results were also seen when the increase in cell proliferation was calculated by the percentage following the timing of incubation until day 3 (P = 0.0474, 2-way ANOVA) (Fig. 6C). These data suggest that ectopic expression of Clu promotes cell proliferation or growth.

To further confirm this observation, cell cycle distribution in TEC-Clu^hClu vs. TEC-Clu^−/− cells in response to the stimulation of growth factors and serum after overnight starvation was counted using flow cytometric analysis with a PI stain. As shown in Fig. 7 and Table 1, the cell cycle distribution of TEC-Clu^hClu cells was different from that of TEC-Clu^−/− cells; during a period of 24 h-incubation with growth factors and serum, the number of TEC-Clu^hClu cells at the G2/M phase was significantly higher than that of TEC-Clu^−/− cells, and there was a progressive shift of the cells from the G0/G1 to the S and G2/M phases of the cell cycle following the time of incubation, which was not seen in TEC-Clu^−/− cells. These data suggest that ectopic expression of Clu increases the number of G2/M-arrested TECs and normalizes the cell cycle progression for cell proliferation.

Clu regulates cell cycle-related gene expression in TECs.

The effect of ectopic expression of Clu on cell cycle-related genes in TECs was examined using a PCR array. As listed in Supplemental Table S1, the expression of 52 genes was significantly affected by the presence of Clu; 50 genes were upregulated, while 2 genes (Pmp22 and Ccnd2) were downregulated. According to the functions of these genes, the effect of Clu on the cell cycle of cultured TECs was divided into three groups: 1) 29 upregulated genes (Notch2, Shc1, Bcl2, Skp2, Cdk1, Cdk4, Cdk6, Cks1b, Cdc6, Cdc25a, Cdc25c, Chek1, Ccna2, Ccnd3, Ccne1, Ccnf, Ccnb1, Ccnc, Mcm2, Mcm3, Mcm4, Ran, Aurka, Aurkb, Mad2l1, Stag1, Stmn1, Dst, and Tsg101) and downregulated Pmp22 promoted cell survival and cycle progression; 2) 10 upregulated genes (Gpr132, Mre11a, Msh2, Rad9, Rad17, Rad21, Atr, Brca1, Hus1, and Nbn) acted as cell cycle checkpoints in response to DNA damage; and 3) 11 upregulated genes (Casp3, Cdk5rap1, Trp53, Ppm1d, Wee1, Chek2, E2f1, E2f2, E2f3, E2f4, and Tfdp1) and downregulated gene Ccnd2 were associated with cell cycle arrest and apoptosis. Compared with the expression profile of these genes in WT kidneys against Clu KO background controls after IRI (Supplemental Table S1), seven genes (Shc1, Skp2, Cks1b, Ran, Aurka, Stag1, and Tsg101) for cell cycle progression and survival and three genes (Gpr132, Msh2, and Hus1) for DNA repair were commonly upregulated in both in vitro and in vivo systems, while only one gene, Tfdp1 (for stimulation of E2F-dependent transcription), in the group for cell cycle arrest and apoptosis was found to be upregulated in both systems, suggesting that Clu may have a more profound effect on the signaling pathways mediating cell cycle progression and survival rather than cell cycle arrest.

Clu suppresses TEC migration in vitro.

The repair of the kidney after IRI involves both proliferation and differentiation of proximal TECs as well as cell migration (4, 36). To understand the mechanism by which Clu mediates renal repair after IRI, the role of Clu in TEC migration was investigated. As shown in Fig. 8, ectopic expression of Clu in TECs resulted in a decrease in cell migration in an in vitro model of wound healing, indicated by the fact that TEC-Clu^−hClu cells migrated 105 ± 17.34 μm in 6 h compared with 138 ± 26.9 μm in TEC-Clu^−/− cell cultures at the same time point (P = 0.0107), suggesting that restraining TEC migration by Clu may be required for the renal repair and regeneration after injury.