Crucial role of the cryptic epitope SLAYGLR within osteopontin in renal crystal formation of mice

Authors


Abstract

Osteopontin plays a crucial role in the formation of renal calcium crystals, which are primarily induced by renal tubular cell injury, especially mitochondrial damage. We have previously shown that the impaired Arg-Gly-Asp (RGD) sequence of osteopontin inhibits renal crystal formation by using OPN-transgenic mice and OPN-knockout (OPN-KO) mice. Here, we investigated the effects of an antimurine osteopontin antibody (35B6-Ab) that specifically reacts with the 162SLAYGLR168 sequence, which is exposed by thrombin cleavage and is located adjacent to the RGD sequence, on renal crystal formation.

Renal crystals induced by daily administration of glyoxylate over 9 days (from days 1 to 9) in a murine model were sporadically detected in the renal tubular cells at the corticomedullary junction, where thrombin-cleaved osteopontin expression was also coincidentally detected. On days 0, 3, 6, and 9, 35B6-Ab administration inhibited renal crystal formation and induced significant morphological changes in a dose-dependent manner (250, 500, and 1000 µg per mouse). Scanning electron microscopy showed that the crystals in 35B6-Ab–treated mice were aberrantly formed and their density was low; in contrast, the crystals in untreated mice that were not administered 35B6-Ab had a radial pattern of growth (rosette petal–like crystals), and their density was high. Microstructure analysis of renal tubular cells by transmission electron microscopy revealed that untreated mice showed collapsed mitochondria in the flattened cytoplasm of renal tubular cells, unlike the corresponding structures in 35B6-Ab–treated mice, in which renal tubular cell injury was inhibited. In vitro, 35B6-Ab was found to inhibit the attachment of 14C-labeled crystals to renal tubular culture cells and reduce morphological damage to these cells.

We conclude that thrombin-cleaved osteopontin plays an important role in formation of renal calcium crystals and that 35B6-Ab contributes to the remarkable inhibition of early-stage renal crystal formation by preventing renal tubular cell injury and crystal-cell attachment. © 2011 American Society for Bone and Mineral Research

Introduction

Renal stone disease is an exceedingly common clinical condition in industrialized countries. Up to 15% of men and 6% of women have at least one renal stone episode during their lifetime;1 about 50% of these individuals experience recurrence.2 However, worldwide, there are no available medical managements to prevent renal stone recurrence.

Renal tubular cell (RTC) injury is regarded as one of the major risk factors for crystal formation in the kidney.3, 4 Recent studies have shown that mitochondrial damage, and oxidative stress induce early-stage calcium oxalate crystal formation in mice,5 and reducing oxidative stress via compounds such as superoxide dismutase, green tea, and vitamin E is associated with decreased RTC injury and crystal deposition in the kidneys.5–7

Renal stones show two phases: a mineral phase and an organic phase known as the matrix. The organic material accounts for at least 5% of the stone's dry weight.8 The presence of any intracrystalline proteins, such as osteopontin (OPN), prothrombin, osteocalcin, or calprotectin, within the stone structure implies a pivotal role of matrix proteins in crystal formation.8–11 We previously reported that OPN is one of the key molecules in the pathogenesis of the formation of renal calcium stones.12, 13 Further studies on OPN-deficient mice show that OPN acts as a promoter during crystal formation.14, 15 OPN is intimately involved in many important biological activities that can be attributed to its characteristic structure, which includes two calcium-binding sites, the Arg-Gly-Asp (RGD) sequence, two putative heparin-binding domains, and a thrombin cleavage site.16 We showed that the loss of calcium-binding sites, which act in a complementary manner in calcification regulation,17 inhibits calcium oxalate crystallization and induces significant morphological changes in the crystals.15 The loss of the RGD sequence, which serves as an adhesion motif that promotes cell attachment by interacting with a number of integrins,17 inhibits crystal attachment to the renal tubular epithelium. Thus, two distinct domains of OPN contribute to crystal formation in different ways.15

Thrombin cleavage of human OPN (Arg168–Ser169) exposes a C-terminal cryptic integrin-binding motif, 162SVVYGLR168 (Fig. 1). The change in OPN function after thrombin cleavage strongly suggests that the coexistence of OPN and thrombin in vivo affects various cellular functions.18, 19 This SVVYGLR sequence specifically binds integrins α4β1 and α9β1, which are preferentially expressed by monocytes and neutrophils.20, 21 This integrin-binding sequence is located adjacent to the RGD sequence. The SLAYGLR motif within murine OPN, which corresponds to the human SVVYGLR motif, plays a significant role in the pathogenesis of rheumatoid arthritis in mice.22 Thrombin receptors are widely expressed by glomerular endothelial, mesangial, and tubular cells within the kidney.23–25 Thrombin initiates both proinflammatory and proliferative responses in human proximal tubular cells.26 Therefore, we hypothesized that thrombin-cleaved OPN plays an important role in early-stage crystal formation, which RTC injury and oxidative stress induce.27–29

Figure 1.

Structure of human osteopontin (OPN). The Arg-Gly-Asp (RGD) domain is located near the center of the OPN gene. The thrombin cleavage site is adjacent to the RGD sequence, and thrombin-cleaved OPN exposes a cryptic integrin-binding motif, 162SVVYGLR168. The human SVVYGLR motif corresponds to the murine 162SLAYGLR168 motif.

Thus, to analyze the effect of the cryptic OPN epitope generated by thrombin cleavage on renal crystal formation, we obtained an antimurine OPN antibody (35B6-Ab) that specifically reacts with the SLAYGLR domain. We evaluated the inhibitory effects of 35B6-Ab on renal crystal formation in vivo and in vitro. This approach may clarify the molecular basis of stone formation and provide a deeper insight into a novel therapeutic target for treating renal stones.

Material and Methods

35B6 antibody (35B6-Ab)

Monoclonal anti-OPN antibody 35B6 (IgG1) was obtained by immunizing mice with the synthetic peptide VDVPNGRGDSLAYGLR, which corresponds to the internal sequence of murine OPN, as described in Fig. 1.30

Animals

C57BL/6 mice (8 weeks old, weighing 18–22 g; Charles River Laboratories Japan Inc., Yokohama, Japan) were used. The ambient temperature was maintained at 23 °C (1) with a 12-hour light/dark cycle. All animals had free access to standard chow (1.12, 0.9, 0.26, and 0.21 g/100 g calcium, phosphorus, magnesium, and sodium, respectively; Oriental Yeast, Tokyo, Japan) and water. OPN-deficient mice (OPN−/−) created by Brigid Hogan (Department of Cell Biology, Duke University Medical Center, Durham, NC, USA)31 were mated with C57BL/6 mice for 25 generations. Within the OPN-KO mouse genome, exons 4–7 of the OPN gene in chromosome 5 were subjected to recombination with a neomycin-resistant (NeoR) gene. All experimental procedures involving animals were performed in accordance with the guidelines in the NIH Guide for the Care and Use of Laboratory Animals. The experimental protocols were approved by the Animal Care and Use Committee and Biological Safety Committee of Nagoya City University.

In vivo experimental protocol

Mice were divided into five groups (n = 20 for each group). WT mice were divided into four groups, including a control group, which was injected with 500 µg mouse IgG1 in phosphate-buffered saline (PBS). The remaining groups were injected with 35B6-Ab in PBS at doses of 250 µg, 500 µg, and 1000 µg per mouse. OPN−/− mice were each injected with 500 µg mouse IgG1 in PBS. Experimental protocol is shown in Fig. 2. Renal crystal formation was experimentally induced in all mice by daily intraperitoneal administration of glyoxylate (100 mg/kg), as described above,32 over a period of 9 days (from days 1 to 9). On days 0, 3, 6, and 9, 35B6-Ab or mouse IgG1 was administered by intraperitoneal injection. All mice were weighed on days 1 and 7. The mice were sacrificed at specific time points (days 1, 4, 7, and 10), and blood samples and their kidneys were extracted for examination (n = 5 at each time point). The serum levels of 35B6-Ab were measured by ELISA as described above.33 In brief, an OPN peptide conjugated with bovine serum albumin (BSA) (BSA-CVDVPNGRGDSLAYGLR; KITAYAMA LABES Co. Ltd., Nagano, Japan) was added to the wells of a 96-well plate. The wells were blocked with PBS containing 1% BSA at room temperature for 1 hour, and subsequently washed three times with washing buffer (PBS containing 0.05% Tween 20). After the blocking buffer was aspirated, plasma samples diluted with washing buffer were added to the wells and incubated at 37 °C for 1 hour. After washing three times with washing buffer, HRP–anti-mouse IgG (Jackson Lab, Sacramento CA, USA) was added to the wells, and the plates were incubated at 37 °C for 30 minutes, followed by the addition of 3,3,5,5-tetramethylbenzidine solution (Sigma, St. Louis, MO, USA). The colorimetric reaction was measured at 450 nm (SPECTRA MAX 340; Molecular Devices, Tokyo, Japan). The 35B6 antibody concentration was calculated using a standard curve.

Figure 2.

Experimental protocol.

Immunohistochemical staining of full-length and thrombin-cleaved OPN

For immunohistochemical staining, sections placed on slides were subjected to microwave treatment for 15 minutes and blocked with 0.5% H2O2 in methanol for 30 minutes. The sections were subsequently washed in 0.01 M PBS and further treated with skim milk in PBS for 1 hour at room temperature. These slides then were incubated with polyclonal anti-mouse OPN rabbit IgG (IBL Co. Ltd., Gunma, Japan) and monoclonal anti-OPN N-Half Mouse IgG (IBL Co. Ltd.). The antibody that reacted with the sections was detected using a Vectastain Elite ABC kit for rabbit IgG (Vector Laboratories Inc., Burlingame, CA, USA) for full-length OPN and a Vector M.O.M Immunodetection Kit (Vector Laboratories Inc.) for thrombin-cleaved OPN according to the manufacturer's instructions.

Detection and quantification of renal crystal deposition

The resected kidney specimens were immediately fixed in 4% paraformaldehyde and embedded in paraffin. Cross sections of the specimens (4-µm thick) were cut and stained using the Pizzolato staining method for crystal detection described above.34 To confirm that the positively stained substances were crystals, the sections were observed by POM (BX51-33-O instrument; Olympus, Tokyo, Japan). For the quantitative analysis of crystal formation, the areas of the positively stained regions were measured and expressed as a percentage of the total cross-sectional area of the kidney tissue using NIH image software (v. 1.61; Scion Inc., Bethesda, MD, USA).

Microstructural observation by SEM

The kidney tissues were cut into 4-µm sections and prepared for SEM. The paraffin-embedded sections were dewaxed and washed with phosphoric acid buffer (PB). Refixation was subsequently performed, first with 2.5% glutaraldehyde and then with 2% osmium tetroxide. Dehydration was performed in a 50% to 100% ethanol series. The samples were embedded in epoxy resin, coated with platinum, and photographed using a scanning electron microscope (S-4800; Hitachi, Tokyo, Japan).

Microstructure and crystal observation in renal tubular cells by TEM

Renal tubular cells were observed using a transmission electron microscopy (TEM) method described above.5 After perfusion fixation was carried out with 20 mL 0.1 M PB, and 20 mL glutaraldehyde, kidney tissue was extracted. Kidney tissues were washed with PB, and the subsequent fixation was carried out with 2% osmium liquid for 2 hours. Dehydration was carried out using a series of ethanol concentrations (50% to 100%). Renal tissue was embedded in epoxy resin, and polymerization was carried out at 60 °C for 48 hours. After preparing a super slice section (0.5 µm), the tissues were double stained with uranium and lead and observed using a JEM-1011 TEM system (JEOL Ltd., Tokyo, Japan).

Quantitative RT-PCR

Total RNA was isolated from frozen sections of mouse kidney samples using an RNeasy Mini Kit (Qiagen, Mississauga, Ontario, Canada) according to the manufacturer's instructions. The TaqMan Gene Expression Assay (Applied Biosystems, Foster City, CA, USA), 20× assay mix of primers, and TaqMan MG probe (FAM dye labeled) for OPN mRNA (Mm 00436767-m1; Applied Biosystems) were utilized for quantitative RT-PCR35 using an ABI PRISM 7700 Sequence Detection System (Applied Biosystems). A one-step RT-PCR reaction using a TaqMan One-Step RT-PCR Master Mix Reagent Kit (4309169, Applied Biosystems) was initiated with an initial reverse transcription step at 48 °C for 30 minutes. After denaturation at 95 °C for 10 minutes, PCR was initiated at 95 °C for 15 seconds and completed at 60 °C with a hold for 1 minute. The PCR reaction was repeated 40 times. The reaction consisted of 25 µL 2× Master Mix, 1.25 µL 40× MultiScribe and RNase Inhibitor Mix, 2.5 µL OPN expression quantitative primer, and 400 ng/mL total RNA sample.

Western blot analysis

For whole-protein extraction from the kidney specimens, we used the cell culture lysis reagent (Promega Corp., Madison, WI, USA), according to the manufacturer's instructions. In brief, frozen kidney tissue was immersed in 1× lysis buffer and sonicated on ice for a sufficient duration. After incubation on ice for 15 minutes, the suspension was centrifuged. The supernatant then was collected by centrifugation, and the total protein concentration was quantified on a spectrophotometer by using the BCA protein assay reagent (Pierce Biotechnology, Rockford, IL, USA). Samples containing 30 µg total protein were mixed with loading buffer (Laemmli sample buffer; Bio-Rad Laboratories, Hercules, CA, USA). The samples then were boiled for 10 minutes at 100 °C, run on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel for protein resolution, and transferred onto polyvinylidene difluoride (PDVF) membranes (ImmobilinTM; Millipore Corp., Bedford, MA, USA). The membranes then were blocked with 5% skim milk in Tris-buffered saline (pH 7.5) containing Tween 20 (TBS-T) for 1 hour at room temperature and sequentially incubated with a mouse anti-OPN antibody (1:500; IBL Co. Ltd., Gunma, Japan) or monoclonal anti-OPN N-Half Mouse IgG (1:500; IBL Co. Ltd) at 4 °C overnight. After the membranes were washed with TBS-T, they were incubated with the corresponding peroxidase-conjugated secondary antibodies (anti-mouse IgG, 1:500; Sigma) at room temperature for 1 hour and subsequently washed again with TBS-T. The protein bands were visualized using enhanced chemiluminescence (ECL) Western blotting analysis kits (Pierce Biotechnology) according to the manufacturer's instructions.

In vitro crystal cell interaction

Radioactive inorganic COM crystals were generated as described above.36 Briefly, after mixing 10 mM CaCl2 with 10 mM 14C-labeled NaOx at room temperature for 10 minutes, the crystals were pelleted by centrifugation at 3,000 g for 15 minutes, and suspended in 160 µL of Dulbecco-Vogt modified Eagle's medium (DMEM). MDCK cells were cultured to a confluent stage in DMEM. Three days later, the cultures were assayed at a density of 4 × 106 cells per dish. Cellular adhesion of crystals was measured as described above.37 Briefly, at the time of the assay, the medium was aspirated and replaced with 5 mL of phosphate-buffered saline (PBS; pH 7.4 at 37 °C). 14C-labeled COM crystals were added to the PBS and incubated at 37 °C for 150 minutes. The culture dishes were gently agitated for 5 seconds to uniformly distribute the crystals, which subsequently settled on the surface of the cell monolayer under the force of gravity. After 2 minutes, the buffer was aspirated, and the cells were washed three times with PBS (5 mL). The cell monolayer then was scraped directly into a scintillation vial containing 6N HCl (2 mL), to which 5 mL of the scintillator (BIOFLUOR; Perkin Elmer Co., Ltd, Waltham, MA, USA) was added. The amount of radioactivity was measured using a Coulter counter (LSC-6100; ALOKA CO. LTD., Tokyo, Japan). For the experiments designed to assess the effects of 35B6-Ab on crystal-cell interactions, we designated four experimental groups as follows: (a), without COM crystals in the medium; group (b), with COM crystals in the medium only; group (c), with COM crystals in the medium containing mouse IgG1; and group (d), with COM crystals in the medium containing 35B6-Ab (10 µg).

Statistical analysis

Comparisons between two groups were performed using unpaired two-tailed Student's t-tests. All values are presented as mean (SD). Differences between groups were considered statistically significant at p < 0.05.

Results

Serum levels of 35B6-Ab in WT, 35B6-Ab–treated, and OPN−/− mice

On days 0, 3, 6, and 9, 35B6-Ab was administered by intraperitoneal injection at doses of 250 µg, 500 µg, or 1000 µg per mouse. The serum concentration levels of 35B6-Ab during the experiments increased in a dose-dependent manner (Fig. 3). In WT and OPN−/− mice, which were injected with 500 µg of mouse IgG1 in PBS, the serum levels of the antibody were below the detectable level (data for OPN−/− mice are not shown).

Figure 3.

The serum concentration levels of 35B6-Ab were measured at days 1, 4, 7, and 10. 35B6-Ab was administered by intraperitoneal administration at doses of 250, 500, and 1000 µg per mouse at days 0, 3, 6, and 9. WT mice were injected with 500 µg mouse IgG1. Data are presented as means ± SD. *p < 0.05, **p < 0.01 compared with 35B6-Ab administration at 250 µg per mouse; p < 0.05, ♯♯p < 0.01 compared with 35B6-Ab administration at 500 µg per mouse.

Renal crystal formation induced in Wild-type, 35B6-Ab–treated, and OPN-deficient mice

To examine whether 35B6-Ab exerts an inhibitory effect on renal crystal formation in mice, we experimentally induced renal crystal formation by daily intraperitoneal administration of glyoxylate from days 1 to 9 among five groups of mice, namely, wild-type (WT), 35B6-Ab–treated (250, 500, and 1000 µg), and OPN-deficient (OPN−/−) mice. On days 0 (before glyoxylate administration), 3, 6, and 9, 35B6-Ab was intraperitoneally administered at doses of 250 µg, 500 µg, or 1000 µg per mouse (Fig. 2). Fig. 4A–E show the renal sections (stained using the Pizzolato method) in all groups of mice. Brown-stained crystals were sporadically detected in the renal tubular cells (RTCs) of the corticomedullary junction in WT and 35B6-Ab–treated mice. Crystal formation was remarkably diminished in 35B6-Ab–treated mice compared with WT mice. In OPN−/− mice, few crystals were detected within the cortex or at the corticomedullary junction.

Figure 4.

Renal crystal formation induced in WT, 35B6-Ab–treated, and OPN−/− mice after glyoxylate administration. The localization and numbers of crystal deposits stained using the Pizzolato method in the kidneys of mice in each group: WT mice (A); mice treated with 250 (B), 500 (C), and 1000 µg 35B6-Ab (D); OPN−/− mice (E). Original magnification: A–E, 30× (inset: high-magnification image of renal tubular cells, 400×). Crystal deposits were subjected to quantitative analysis (F). The number of renal crystals is expressed as a percentage of the total cross-sectional area using the NIH images. *p < 0.05 compared with WT mice.

Crystal deposits were subjected to quantitative analysis (Fig. 4F). Crystal formation increased as the experiment progressed. Throughout the experiment, crystal deposits were most abundant in WT mice, followed by 35B6-Ab–treated (250 µg) mice. The administration of 500 µg or 1000 µg 35B6-Ab markedly reduced the amount of renal crystal formation (p < 0.05 on days 4 and 10 compared with WT mice). However, there was no significant difference in the amount of crystal formation between mice treated with 500 µg and those treated with 1000 µg 35B6-Ab. Few crystals formed in OPN−/− mice throughout the experiment.

Morphological analysis of renal crystals

Crystals detected in the renal specimens were observed by performing polarized light optical microphotography (POM) (Fig. 5A–E). In WT mice, the crystals formed an orderly line resembling a flower petal (Fig. 5A). In 35B6-Ab–treated (250 µg) mice, the crystals formed rosette-like shapes similar to those observed in WT mice (Fig. 5B). In 35B6-Ab–treated (500 µg) mice, each crystal seemed to be smaller than that in WT mice (Fig. 5C). In 35B6-Ab–treated (1000 µg) (Fig. 5D) and OPN−/− mice (Fig. 5E), the crystals tended to be smaller and aggregated in an irregular manner without forming flower petal–like structures.

Figure 5.

Crystal structure analysis by POM and SEM from each group on day 7. WT mice (A, F); mice treated with 250 (B, G), 500 (C, H), and 1000 µg 35B6-Ab (D, I); OPN−/− mice (E, J). Crystals were observed by POM (A–E). Crystal nuclei were observed by SEM (F–J). In 35B6-Ab–treated mice, the crystals were small and aberrantly formed, and their density appeared to be low in contrast to those in WT mice. Original magnification: A–E, 800×; F–J, 6,000×. Bar = 5 µm.

Next, to evaluate the effect of 35B6-Ab on microstructural morphology, crystal nuclei from each experimental group were observed on day 7 by scanning electron microscopy (SEM) (Fig. 5F–J). In WT mice, the nuclei appeared to form a radial pattern in an orderly manner with a high nuclear density (Fig. 5F). In 35B6-Ab–treated (250 µg) mice, the nuclei appeared to form a radial pattern, but the crystals were irregular and their density appeared to be relatively low in contrast to those in WT mice (Fig. 5G). In 35B6-Ab–treated (500 µg) mice, the shape of the crystal nuclei appeared radial, but the crystals were small and irregular (Fig. 5H). In 35B6-Ab–treated (1000 µg) mice, there were some aberrantly formed crystals whose shapes were completely different from the normal radial pattern (Fig. 5I). In OPN−/− mice, small microcrystal grains were visible in the nuclei and appeared disordered. These grains differed markedly from the crystals observed in the nuclei of WT mice specimens (Fig. 5J). Thus, 35B6-Ab administration in mice decreased the amount of crystals and changed the crystal morphology.

Microstructure analysis of RTCs by transmission electron microscopy (TEM)

The microstructure of RTCs was observed by TEM. Before glyoxylate administration, the microvilli of RTCs in control mice were elongated and highly dense, and the internal structures of the mitochondria were arranged regularly (Fig. 6A and 6G). On day 4, the cytoplasm of RTCs in WT mice showed fattened microstructures, collapsed mitochondria, and disappearance of elongated microvilli (Fig. 6B). Structures appearing like injured microvilli or mitochondrial ghosts, as assumed from their bare internal structure, were detected in the renal tubular lumen (Fig. 6H). Furthermore, renal crystals were observed in the renal tubular epithelium on day 7 (Fig. 6M). In mice treated with 250 µg of 35B6-Ab, on day 4, the internal mitochondrial structures became indistinct, the microvilli disappeared (Fig. 6C), and mitochondrial ghostlike structures appeared like crystal nuclei in the tubular lumen (Fig. 6I); similar findings were observed in WT mice. In mice treated with 500 µg and 1000 µg of 35B6-Ab on day 4, some materials, including indistinct mitochondrial, had leaked into the tubular lumen, but the tubular cytoplasm was sufficiently thick and the breakdown of mitochondrial microstructures was inhibited compared with WT mice (Fig. 6D–E and Fig. 6J–K); few crystals were retained in the tubular epithelium in 35B6-Ab–treated mice on day 7 (Fig. 6N). In OPN−/− mice, on day 4, a few mitochondrial ghosts had leaked into the tubular lumen; a few crystals in the tubular epithelia of OPN−/− mice were similar to those in 35B6-Ab–treated mice (500 µg and 1000 µg) (Fig. 6F and 6L).

Figure 6.

Microstructures of RTC by TEM. Transmission electron micrographs showing the microstructures of RTC obtained from each group on days 4 (B–F, H–L) and 7 (M, N) were compared with those of control mice on days 0, prior to glyoxylate administration. Control mice (A, G); WT mice (B, H, M). Materials that appeared to be injured microvilli or mitochondrial ghosts were detected in the RTC tubular lumen (indicated by arrows): mice treated with 250 (C, I), 500 (D, J, N), and 1000 µg 35B6-Ab (E, K); OPN−/− mice (F, L). The bars represent 10 µm (A–F) and 2 µm (G–L). N, nucleus; MV, microvilli; TE, tubular epithelium; MIT, mitochondria; MIT-G, mitochondria ghosts; TL, tubular lumen; Cry, crystal.

Locational relationship between renal crystal formation and thrombin-cleaved OPN expression in WT mice

Renal sections from WT mice on day 7 after glyoxylate administration were subjected to Pizzolato staining (Fig. 7A). Brown-stained crystals were sporadically detected in the RTCs of the corticomedullary junction. These stained crystals exhibited strong double refraction using POM (Fig. 7B). Immunohistochemistry revealed that the expression of full-length OPN was widely distributed throughout the kidney (Fig. 7C). On the other hand, expression of thrombin-cleaved OPN was specifically localized in the renal tubular cells of the corticomedullary junction where crystal deposits were produced as shown in Fig. 5A (Fig. 7D).

Figure 7.

Locational relationship between renal crystal formation and OPN expression. (A) Brown-stained crystals (indicated by arrows), stained using the Pizzolato method, were detected sporadically in the renal tubular cells of the corticomedullary junction. (B) These stained crystals (indicated by arrows) exhibited strong double refraction in polarized light optical microphotography. (C) Immunohistochemistry revealed that the expression of full-length OPN was widely distributed throughout the kidney. (D) The expression of thrombin-cleaved OPN (indicated by arrows) was localized in the renal tubular cells of the corticomedullary junction where crystal deposits were produced. Original magnification: 30× (inset: high-magnification image of renal tubular cells, 400×). C, cortex; M, medulla; P, papilla.

Expression of full-length and thrombin-cleaved OPN proteins in WT and 35B6-Ab–treated mice

To evaluate the expression of full-length and thrombin-cleaved OPN, we performed a Western blot assay and immunohistostaining on the kidney specimens. The expression of full-length OPN was nearly absent in WT and 35B6-Ab–treated (500 µg) mice on day 1, and the expression levels in both groups on day 7 were significantly greater than those on day 1. The expression levels of thrombin-cleaved OPN in WT and 35B6-Ab–treated mice on day 7 were also greater than those on day 1. The expression levels of full-length OPN in 35B6-Ab–treated mice were slightly lower than those in WT mice, whereas the expression of thrombin-cleaved OPN on day 7 was similar in both groups (Fig. 8A). The expression of full-length OPN mRNA in the kidneys was examined by quantitative reverse transcription-PCR on days 1 and 7 (Fig. 8B). The expression of OPN mRNA was significantly higher in all mice at day 7 than at day 1. However, there was no significant difference among all groups at day 7. Figures 8C and 8D show the kidney sections of mice treated with 500 µg 35B6-Ab obtained at day 7. Full-length OPN was widely distributed throughout the kidney, particularly in the renal tubular cells (Fig. 8C). The expression of thrombin-cleaved OPN was localized in the renal distal tubules at the corticomedullary junction, which is where crystal deposits usually are located (Fig. 8D); similar findings were observed in the WT mice. As shown in Figures 6C and 6D, the amount and location of OPN expression in 35B6-Ab–treated mice were not significantly different from those in WT mice.

Figure 8.

Expression of full-length and thrombin-cleaved OPN. (A) The expressions of full-length and thrombin-cleaved OPN were verified by a Western blot assay by using proteins extracted from WT mice and 35B6-Ab–treated mice (500 µg) at days 1 and 7. (B) OPN mRNA expression was assessed in WT and 35B6-Ab–treated mice (250, 500, and 1000 µg) at days 1 and 7 by quantitative RT-PCR. Expression levels are relative to those of GAPDH mRNA. There was no significant difference among all groups. Data are presented as means ± SD. Immunohistochemical staining of full-length OPN (C) and thrombin cleaved OPN (D) of 35B6-Ab–treated mice (500 µg) at day 7. Original magnification: 40× (inset: high-magnification image of renal tubular cells, 400×).

In vitro 35B6-Ab–induced changes in the interactions of crystals with Madin-Darby canine kidney cells

We investigated the ability of 35B6-Ab to alter the adherence of 14C-labeled calcium oxalate monohydrate (COM) crystals to Madin-Darby canine kidney (MDCK) cells in vitro. Thus, we designated four experimental groups as follows: group (a), without COM crystals in the medium; group (b), with COM crystals in the medium; group (c), with COM crystals in the medium containing mouse IgG1 (10 µg); and group (d), with COM crystals in the medium containing 35B6-Ab (10 µg) (Fig. 9A). We compared the counted radioactivity (counts per minute) of 14C-labeled COM crystals attached to MDCK cells after treatment to determine the average adherence of COM crystals to MDCK cells. No radioactivity was detected in group (a). Radioactivity of COM crystals was higher in groups (b) and (c). The radioactivity of COM crystals in group (d) was significantly lower than that in groups (b) and (c). Fewer crystals bound to the MDCK cells in the presence of the 35B6-Ab.

Figure 9.

In vitro interactions of crystals with Madin-Darby canine kidney cells (A) The ability of 35B6-Ab to decrease the adherence of calcium oxalate monohydrate (COM) crystals to MDCK cells in vitro was examined by the comparison of the counted radioactivity of 14C-labeled COM crystals data between four groups: group (a), without COM crystals in the medium; group (b), with COM crystals in the medium;group (c), with COM crystals in the medium containing mouse IgG1; and group (d), with COM crystals in the medium containing 35B6-Ab (10 µg). (B) Morphological differences in MDCK cells after adhesion of COM crystals, as observed by differential interference contrast (DIC) microscopy. Original magnification: 400×. Data are presented as mean ± SD. **p < 0.01, compared with group (d).

Culture cells in groups (a), (b), (c), and (d) were observed by differential interference contrast (DIC) microscopy. Culture cells in groups (b) and (c) were swelled, inducing rupture. Culture cells in group (d) with 35B6-Ab (10 µg) appeared intact and similar to those in group (a) (Fig. 9B).

Discussion

OPN is reported to play a non-redundant role in renal crystal formation, which is a complex multistep process that includes supersaturation, crystal nucleation, growth, and aggregation. We have previously identified two functional domains (the RGD sequence and two calcium-binding domains) of OPN in renal stone formation by using two types of OPN-transgenic mice, as well as OPN-deficient mice.15 However, it is interesting to note that the deletion of the RGD sequence or the two calcium-binding domains did not suppress renal crystal formation as effectively as did the complete deletion of OPN in the OPN−/− mice. We speculate that other functional domains of OPN exist in renal crystal formation. In this study, full-length OPN was found to be widely distributed throughout the kidney, whereas thrombin-cleaved OPN was localized in the RTCs at the corticomedullary junction, where crystal deposits are coincidentally located. Thrombin cleavage of human OPN (Arg168–Ser169) exposes a C-terminal cryptic integrin-binding motif, 162SVVYGLR168, which corresponds to the murine SLAYGLR motif (Fig. 1). Previous reports have established that full-length OPN and thrombin-cleaved OPN are present at inflammatory sites and that a specially cleaved OPN that contains the SLAYGLR sequence contributes to cell-mediated inflammation by acting as a cytokine in T-cell–mediated liver disease and rheumatoid arthritis via the NF-κB pathway.38–40 RTC injury, especially mitochondrial damage, and oxidative stress induce the early-stage calcium oxalate crystal formation27, 29, 41 via OPN expression and by activating NF-κB, which is involved in inflammatory and immune responses.42 Both thrombin25 and OPN13 are widely expressed in renal tubular cells. We therefore hypothesized that thrombin-cleaved OPN plays an important role in crystal formation.

To assess the role of the cryptic OPN epitope (SLAYGLR, which is exposed by thrombin cleavage) on renal crystal formation, we obtained 35B6-Ab, which specifically recognizes the SLAYGLR sequence, and prophylactically administered it to a murine model of experimentally induced renal crystal formation. This mouse model was established as a useful animal model for studying the mechanism underlying renal stone formation.32 We found that the administration of 35B6-Ab inhibits renal crystal formation in a dose-dependent manner. Compared with that in WT mice, 35B6-Ab (500 and 1000 µg) markedly reduced the extent of renal crystal formation. During the experiment, the serum concentration levels of 35B6-Ab increased remarkably in a dose-dependent manner. Thus, it seems that this antibody also serves as a useful tool for investigating OPN function in vivo.

In this study, observations of microstructures in vivo by TEM revealed that crystal absorption into renal epithelium cells is essential during the early stage of crystal formation. We suggest that crystals form in the RTCs and are retained in the tubular lumen, thus resulting in RTC injuries, such as the disruption of mitochondria and microvilli. Materials that appeared to be injured microvilli or mitochondrial ghosts moved from the renal tubular cytoplasm into the renal tubular lumen and became crystal nuclei by being retained in the tubular epithelium. In this study, in 35B6-Ab–treated mice (500 and 1000 µg), some mitochondrial ghosts appeared; meanwhile, breakdown of the RTC microstructure was inhibited, and few crystals were retained in the tubular epithelium, similar to the effects found in OPN−/− mice.

In vitro, 35B6-Ab was found to remarkably decrease the amount of radioactivity of 14C-labeled crystals, indicating that 35B6-Ab inhibits the attachment of COM crystals to the Madin-Darby canine kidney cells. Further, 35B6-Ab reduced the morphological damage, that is, rupture of the culture cells. Some previous studies also report that unless intraluminal crystals adhere to the renal tubular epithelium, they are flushed out of the nephrons within a few minutes along with the flow-through fluid.43, 44 Another in vitro study utilizing OPN-deficient cell lines shows that a lack of OPN suppresses the deposition and adhesion processes in renal crystal formation.45 We hypothesized that renal crystal formation is inhibited by 35B-Ab because 35B6-Ab suppresses RTC injuries and the attachment of the crystals to renal epithelial cells.

The antibody used in this study was designed to bind to the matricryptic SLAYGLR sequence (SVVYGLR in humans) of OPN that is only exposed in thrombin-cleaved OPN. Though it could bind equally to the immunizing peptides CGRGDSLAYGLR and CGRDSLAYG, 35B6-Ab failed to bind to CGRGDS.30 We speculate that OPN functions as an adhesive in anchoring and retaining crystals at the epithelium via the SLAYGLR and RGD motifs and that 35B6-Ab inhibits the interaction between OPN and its specific receptors, including α9 and α4 integrins.21

Analysis of the crystal microstructure revealed that in 35B6-Ab–treated mice, the crystal nuclei appeared cracked and the nuclear density was low compared with that in WT mice. For a calculus to grow and maintain coherence, it must combine with a mineral material that accounts for 90% of the stone's dry weight. OPN is known to be a component of renal crystals that forms a bridge between mineral materials. We assume that the SLAYGLR motif plays an essential role in crystal–crystal interactions, such as at calcium-binding sites15 and that antigen-specific induction of thrombin-cleaved OPN contributes to the dramatic changes in crystal microstructure by inhibiting the binding of crystals.

In this study, 35B6-Ab induced dramatic changes in crystal formation, including crystal number and microstructure. The expression levels of both full-length and thrombin-cleaved OPN among all groups of mice at day 7 were higher than those at day 1. At day 7, the expression of full-length OPN in 35B6-Ab–treated mice kidneys was slightly lower than that in WT mice, whereas the expression of thrombin-cleaved OPN was identical between 35B6-Ab–treated and WT mice. However, there was no significant difference in OPN mRNA expression among all groups. Rather than acting by suppressing OPN expression, 35B6-Ab inhibits OPN function, such as the interaction between OPN and its specific receptors, including α9 and α4 integrins. We speculate that thrombin-cleaved OPN affects the attachment of crystals to renal epithelial cells by attracting those integrins.

In conclusion, we found that a specific anti-OPN antibody contributes to the remarkable inhibition of early stage of renal crystal formation by suppressing RTC injury and the attachment of crystals to renal epithelium cells, as well as by suppressing crystal growth. Our findings suggest that the cryptic SLAVGLR sequence of OPN is an attractive new therapeutic target for treating patients with calcium stones.

Disclosures

All the authors state that they have no conflicts of interest.

Acknowledgements

We thank Ms N Kasuga for her assistance in preparing the manuscript. We are also grateful to Dr Hideo Shimizu from Core Laboratory Division, Nagoya City University Graduate School of Medical Sciences, for technical assistance with the electron microscope. This work was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Science, and Technology (22791479 [SH], 21791520 [YH], 23592375 [TY], and 2179517 [AO]).

Authors' roles: Study design: SH, TY, and KK. Study conduct: TU and KK. Data collection: SH. Data analysis: SH, TY, and AO. Data interpretation: SH, TY, AO, and KK. Drafting manuscript: SH, TY, TU, and KK. Revising manuscript content: TY and KK.

Ancillary