Genome-Wide Analysis of Genes Related to Kidney Stone Formation and Elimination in the Calcium Oxalate Nephrolithiasis Model Mouse: Detection of Stone-Preventive Factors and Involvement of Macrophage Activity


  • The authors state that they have no conflicts of interest.


We previously established a mouse kidney stone formation model and showed that mice have a higher tolerance to stone formation than rats. Furthermore, we showed that the generated calcium oxalate crystal deposits could be eliminated after several days. This study investigated the transcriptome of stone formation and elimination in the mouse kidney based on gene selection using a microarray technique. Eight-week-old male C57BL/6N mice were administered 80 mg/kg glyoxylate for 15 days, and kidney calcium oxalate crystal depositions had increased by day 6; thereafter, depositions decreased gradually and had almost disappeared by day 15. On microarray analysis, mRNA expression in the crystal-formed kidneys showed the significant expression of 18,064 genes. Thirty-one, 21, and 25 genes showed at least a 2-fold increased expression during the experimental course (days 3–15), stone formation phase-specific (days 3–6), and stone elimination phase-specific (days 9–15) stages, respectively. Among these genes, those related to chemotaxis and monocyte/macrophage activation were identified. Gene ontology analysis to identify overexpressed genes highlighted categories related to inflammation, immune reactions and the complement activation pathway. Quantitative PCR of 17 previously reported stone-related genes with a significant expression on microarray analysis showed significantly increased chemokines, stone matrix proteins, and their receptors; the significant decrease of several types of transporters and superoxide dismutase; and the persistently high expression of Tamm-Horsfall protein throughout the experiment. In conclusion, inflammation and immune reactivity through macrophage migration are involved in stone formation and elimination in mouse kidneys.


Urolithiasis is a multifactorial disease involving environmental and genetic factors. Calcium-containing kidney stones, which are >90% of all stones, include various types of recurrent kidney stones, such as hypercalciuria or hyperoxaluria; however, the majority are idiopathic, and urinary calcium oxalate (CaOx) crystals or increased urinary supersaturation is observed even in the urine of healthy individuals,(1,2) and it is difficult to estimate the risk of kidney stone formation from only the amount of urinary excretion or concentration of inorganic substances.

The first step in stone formation is crystal generation in the renal tubules, and crystal aggregation and growth are followed by crystal retention in the tubular lumens(3); however, calculations based on the concentrations of ions in tubular fluid and the maximal rate of CaOx crystal growth suggest that a crystal can not become large enough to occlude the tubular lumen during the time necessary for transit from the proximal tubule to the end of the collecting duct.(4) We have been focusing on the stone matrix, which is an organic component making up several percent of calcium-containing stones, and detected osteopontin (OPN) as the main component.(5) Since this discovery, a new perspective on kidney stones as “genetic products” and “mineral matter” has developed, and biomolecular approaches have been initiated. We previously reported that the expression of OPN in renal tubular cells increased in a hyperoxaluric environment in animal and cellular models(6) and distributed OPN protein in kidney stones,(7) OPN deficiency induced the low adaptation of calcium oxalate crystals to the cell surface and inhibited crystal growth using OPN expression-suppressed cultured cells(8,9) and OPN knockout mice,(10) and haplotypes in the promoter region of the OPN gene were related to the risk of kidney stone formation in calcium stone formers.(11)

On the other hand, several investigators have reported OPN as an inhibitor of crystal growth, aggregation, and cell adhesion.(12–14) There have also been many studies about the influence of expression change or single nucleotide polymorphisms (SNPs) of many genes, which are related to other stone matrix proteins, urinary high molecular substances, extracellular matrix, chemotaxis, oxidative stress, cell adhesion, immune reactions, coagulation, oncogene, cell injury, oxalate and calcium metabolism, tubular transporter, and lipid metabolism(5,7,15–28) on individual kidney stone formation, although the results are insufficient to explain the mechanism of stone formation.

The DNA microarray technique is one of the most useful tools to search for the global mechanism of a reproducible biological phenomenon. In the field of urolithiasis, several studies have been performed using animal models and cultured cells. Katsuma et al.(29) and Chen et al.(30) performed genome-wide analysis using microarrays on ethylene glycol-administered rats and increased the expression of genes involved in inflammation, cell adhesion, fibrosis, mitochondrial change, and oxidative stress. Hoopes et al.(31) reported that one third of the genes with the greatest expression increase by microarray analysis on genetic hypercalciuric stone-forming congenic rats were located on chromosome 1, on which a calcium excretion locus was detected. Using microarray analysis, Liang et al.(32) analyzed the mechanism of human renal epithelial cell injury caused by exposure to calcium oxalate monohydrate (COM) or 2,8-dihydro adenine (DHA) crystals.

Recently, we developed a method to induce calcium oxalate crystal depositions in normal mouse kidneys.(33) Briefly, ethylene glycol-administered rats have been central in animal studies of kidney stones, although a murine kidney stone model using gene recombinant technology has been anticipated. However, there are no previous reports on the induction of kidney stones in normal mouse kidneys, and a mouse kidney stone model by intra-abdominal injection of glyoxylate is expected to became a control for future studies using gene recombinant or mutation animals. In stone model mice, OPN-containing crystals (stones) increased until day 6 and thereafter decreased and disappeared by day 15. Moreover, using glyoxylate-administered OPN knockout mice, we reported that OPN has crucial roles in the morphological conversion of calcium oxalate crystals into stones and a significantly lower amount of kidney stone formation in OPN knockout mice than the wildtype.(10) Our subsequent study to investigate differences in stone formation among oxalate precursors showed that hyperoxaluria induced by ethylene glycol administration could not generate crystal deposits, that glyoxylate induced the destruction of microvilli and mitochondria of renal tubular cells, and that their broken-down materials might become nidi of kidney stones by ultrastructural observation with transmitted electron microscopy.(34)

The results of this series of studies indicated that the formation of calcium oxalate crystal deposits in the stone model mouse could not occur only with the excretion of urinary inorganic substances, and the phenotypical background involving OPN expression must be important; however, the factors involved in kidney stone elimination have not been discussed. Resolving their high tolerance to kidney stone formation and the mechanism of stone elimination in mice might not only identify the stone-preventive mechanism in healthy individuals but also suggest stone-preventive or stone-dissolving drugs.

We performed microarray analysis on the kidneys of stone formation model mice to detect the genes most highly expressed during stone formation and elimination periods to identify candidate genes related to the stone-preventive ability of mouse kidneys.


Animal procedures

All animal studies followed the recommendations of the NIH Guide for the Care and Use of Laboratory Animals. Eight-week-old male C57BL/6N mice (n = 18) were used for this study. Calcium oxalate monohydrate crystal depositions in the kidneys were induced with an intra-abdominal injection of 80 mg/kg glyoxylate for 15 days according to the stone model mouse method,(19) which is the only method for inducing calcium oxalate crystal depositions in normal mouse kidneys. Briefly, the prepared glyoxylate solution was kept at 4°C until administration. Intra-abdominal injection was performed every day with a clean 27-gauge needle according to the weight of each mouse. All animals had free access to standard chow and water. Every 3 days, renal specimens were extracted, the upper one half of the kidneys was fixed in 4% paraformaldehyde and embedded in paraffin, and the lower one half was stored at −80°C after overnight perfusion in RNAlater (Ambion) at 4°C until RNA preparation.

Quantification of the amount of crystal depositions

Dewaxed nonstained 4-μm-thick kidney cross-sections were observed with polarized light optical microphotography (BX51–33P-O; Olympus, Tokyo, Japan). Images of the sections were scanned, and crystal regions with strong birefringence were measured and expressed as percentages of the total tissue area of cross-sections of the kidney with soft NIH image 1.61 (Scion). For statistical evaluation for differences in the generated crystal amount between each administration day, a nonpaired t-test was performed. The crystal components were calcium oxalate monohydrate, as reported previously.(10,33)

Microarray analysis and data mining

For microarray analysis, total RNA of the glyoxylate-treated kidneys was isolated with the RNeasy Midi Kit (Qiagen) according to the manufacturer's instructions. cRNA preparation and microarray analysis were conducted at Bio Matrix Research, using the Affmetrix system (Santa Clara). Isolated total RNA (1 μg) was converted into double-stranded cDNA using the One-Cycle cDNA Synthesis Kit (Affymetrix), which was purified using a GeneChip Sample Cleanup Module (Affymetrix). In vitro transcription reactions were performed using a GeneChip IVT Labeling Kit, which includes T7 RNA polymerase and biotin-labeled ribonucleotides. Biotin-labeled cRNA was purified using a GeneChip Sample Cleanup Module. The concentration of cRNA was calculated from light absorbance at 260 nm using a UV spectrophotometer. Next, cRNA (15 μg) was fragmented at 94°C in the presence of a fragmentation buffer (Affymetrix). The labeled cRNA was purified, fragmented, and spiked with in vitro transcription controls. Fifteen micrograms of cRNA was hybridized using the GeneChip Mouse Genome 430 2.0 Array (Affymetrix). The array was incubated for 16 h at 45°C and automatically washed and stained with the GeneChip Hybridization, Wash and Stain Kit (Affymetrix) on an Affmetrix GeneChip Fluidics station. The probe array was scanned using a GeneChip Scanner 3000 7G. All preparations were run on quality-controlled chips and had 3′/5′ signal ratios of ≤2.

For the GeneChip Mouse Genome 430 2.0 array, 11 probe pairs were prepared per transcript. The probe pair consists of perfect match (PM) and mismatch (MM) probes with 25 mers of oligonucleotide; the MM probe is designed as a single base substitute for the PM probe, which has a complementary sequence of the target transcript.

The expression value of the transcript was computed using GeneChip Operating Software (GCOS) with the MAS5 algorithm. Each MM was used as the value of cross-hybridization of PM, and the expression value of the transcript was calculated with the 11 values of PM and MM using each GCOS, in which the probabilities of the values of each transcript were indicated as the “Flag,” Present (p ≥ 0 to <0.04), Marginal (p ≥ 0.04 to <0.06), and Absent (p ≥ 0.06 to <0.5), using a one-sided Wilcoxon's signed rank test between the values of PM and MM.

For analysis, normalizations, relative signal intensities (RSIs) and fold changes (FCs) between glyoxylate-treated (days 3–15) and control samples (day 0) were calculated using GeneSpring 7.3 (Agilent Technologies) data-mining software. Genes were sorted with an average FC of ≥2 during the experimental course (days 3–15), stone formation period (days 3–6), and stone elimination period (days 9–15) compared with day 0 to select the genes related to stone formation and/or elimination separately. The sorted gene lists were analyzed with the GO ontology browser, one of the functions of GeneSpring 7.3, and the categories of biological process, cellular components, and molecular function were sorted based on the annotation of listed genes. The p value of each category was calculated with the Fisher's exact test.

Confirmation of the expression changes of stone-related genes previously reported

Among 53 genes previously reported in relation with kidney stone formation, 49 genes were recognized on the GeneChip, and their expression changes were evaluated. In particular, the expressions of 19 genes with significant change or adequate expression on microarray analysis were reconfirmed with quantitative PCR.

All RNA samples used in microarray analysis were reverse transcribed into cDNA with a high-capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). According to the annotation information of each gene, the TaqMan Gene Expression Assay Product, which is a 20× assay mix of forward and reverse primer sets, and the TaqMan MGB probe (FAM dye-labeled) with complementary sequences to each mRNA sequence were obtained, and quantitative PCR was performed with the TaqMan Universal PCR Master Mix (404437; Applied Biosystems) using the 7500 FAST Real-time PCR System (Applied Biosystems). After denaturing at 95°C for 10 min, PCR was initiated at 95°C for 15 s and completed at 60°C for 1 min. The PCR reaction was repeated 40 times. cDNA amplification was compared with that of control samples (day 0), and their expression ratios were determined with a standard curve of a ×5 dilution series of control samples and corrected for the amount of total RNA. The statistical significance of differences of the expression ratio compared with the control samples (day 0) was calculated using unpaired t-tests,

Pathological detection of calcium oxalate crystal depositions, macrophage migration, and related protein expression

Calcium oxalate crystal depositions were determined by Pizzolato staining, as described previously.(35) Briefly, sequential paraffin sections were dewaxed and rinsed with distilled water. For Pizzolato staining, 30% hydrogen peroxide (1 ml) was added to the slide with tissue sections (pH 6.0). The slide was exposed to light from a 60-W incandescent lamp at a distance of 15 cm (6 in) for 15–30 min. Numerous gas bubbles developed, and it was necessary to pour off the mixture and add fresh solution. The slide was washed thoroughly with distilled water, counterstained with Kernechtrot (Nuclear first red) solution (Merck), and dehydrated in the usual manner.

To detect renal macrophages, F4/80, a mouse macrophage marker, was detected immunohistochemically, as were monocyte chemoattractant protein-1, which is protein coded by C-C chemokine ligand 2 (Ccl2), osteopontin (OPN), which is protein coded by secreted phosphoprotein 1 (Spp1), and Tamm-Horsfall protein (THP), which is protein coded by Uromodulin (Umod). Immunohistochemical staining of each protein was carried out on 4-μm-thick cross-sections autoclaved for antigen activation at 121°C for 5 min and blocked with 0.5% H2O2 in methanol for 30 min, followed by washing in 0.01 M PBS and further treated with skimmed milk in PBS for 1 h at room temperature. These slides were incubated with anti-mouse F4/80 rat monoclonal antibody (Abcam), anti-mouse MCP-1 rat monoclonal antibody (Monosan, Uden, The Netherlands), anti-mouse OPN rabbit polyclonal antibody (IBL, Gunma, Japan), and anti-mouse THP rabbit polyclonal antibody (Santa Cruz Biotechnology) overnight at 4°C. The reacted antibody was detected using a Histofine Simple Stain Kit for rat or rabbit IgG (Nichirei Biosciences, Tokyo, Japan) according to the manufacturer's instructions.


Kidney crystal formation and elimination

After 3 days of glyoxylate administration, kidney crystal depositions were detected in renal tubules mainly located at the border between the renal cortex and medulla. The crystals increased until day 6 of administration, thereafter decreased, and almost disappeared by day 15 (Fig. 1A). The generated crystals had morphological features of a rosette shape and specific birefringence as previously reported.(10) Quantification of the crystal amount by NIHimage software indicated a significant increase of crystals on day 6 compared with day 0 and a decrease on day 15 compared with day 6 (Fig. 1B). There were no significant differences among other time points.

Figure Figure 1.

Detection and quantification of kidney crystal depositions in glyoxylate-administered mouse kidneys. (A) Four-micrometer sections of nonstained kidney sections observed with a polarized optical photomicroscope (×40). (B) Crystallization in each kidney section was quantified by calculating the ratio of crystal regions to the kidney section using NIH image 1.61. Data are presented as the mean ± SD. *p < 0.05.

Microarray analysis

As subjects for microarray analysis, 18,064 genes were nominated with the expression of Present or Marginal on any chips. Compared with the raw values, the expression indexes of the control (day 0), the number of genes with more than two times higher expression during the experimental course (days 3–15), stone formation period (days 3–6), and stone elimination period (days 9–15) were 31, 21, and 25, respectively, and 198, 59, and 321 genes had more than two times decreased expression. Fugure 2 shows the gene trees of extracted genes demonstrating >100 raw values. The top 30 of the most varied genes in the six groups are shown in Tables 1–6.

Table Table 1.. Thirty-One Genes With Highest Expression Increase During Experimental Course (Days 3–15)
original image
Table Table 2.. Twenty-One Genes With Highest Expression Increase During Stone Formation Period (Days 3–6)
original image
Table Table 3.. Twenty-One Genes With Highest Expression Increase During Stone Elimination Period (Days 9–15)
original image
Table Table 4.. Genes With Greatest Expression Decrease (Top 30) During Experimental Course (Days 3–15)
original image
Table Table 5.. Genes With Greatest Expression Decrease (Top 30) During Stone Formation Period (Days 3–6)
original image
Table Table 6.. Genes With Greatest Expression Decrease (Top 30) During Stone Elimination Period (Days 9–15)
original image
Figure Figure 2.

Dendrogram of expression changes in glyoxylate-induced crystal formation kidneys. Among 18,064 genes with a Present or Marginal expression on any chip, 778 genes showed adequate high expression (Raw > 100) and expression changes with >2 times increase (90 genes) or decrease (688 genes). These nominated genes were divided into six groups: three increased expression groups (throughout the experimental course [day 3–15], stone formation period [day 3–6], and stone elimination period [day 9–15]) and three decreased- expression groups.

Throughout the experimental course (days 3–15), the most expressed genes were complement component 3 (C3), and an increased expression of other complement components, Cfi, C4b, C1qb, and C1qa, was detected. Monocyte/macrophage-related genes were increased, such as lysozyme families, which are hydrolytic enzymes, reported as a stone matrix component, glypcan 3, which is related to macrophage maturation; macrophage-expressed gene 1 (Mpeg1), which is a macrophage marker; lectin galactose binding soluble 3 (Lgals3), which is an essential factor in the acquisition of macrophage function; cathepsin S (Ctss), which is a peptidase that breaks down proteins for antigen presentation; histocompatibility 2, a class II antigen A, β 1 (H2-Ab1), which is a component of MHC class II; and myristoylated alanine-rich kinase C substrate (Marcks), which is related to phagocytosis and exocytosis. Other genes were increased: coagulation-related genes, fibrinogen, α and γ polypeptide (Fga and Fgg) tissue sclerosis-related genes, procollagen type III, α 1 (Col3a1), procollagen type I, α 2 (Col1a2), cadherin 11 (Cad11), cell-adhesion factor, vascular cell adhesion molecule 1 (Vcam1), and lipid metabolism-related genes, cytochrome P450, family a, polypeptide 14 (Cyp4a14), and annexin A3 (Anxa3).

Specifically, during the stone formation period (days 3–6), the most increased gene was chemokine (C-C motif) ligand 6 (Ccl6), an agent of monocyte/macrophage chemotaxis. In addition, macrophage proliferation-related genes, such as vimentin (Vim), CD 14 antigen (Cd14), and cytochrome P450, family 1, subfamily b, polypeptide 1 (Cyp1b1), as well as macrophage activation-related genes, such as histocompatibility 2, class II antigen A, β 1 (M2-Ab1) and moesin (Msn), increased. Cell proliferation and DNA synthesis-related genes, ribonucleotide reductase M2 (Rrm2), extracellular matrix components, fibronectin 1 (Fn1), Matrix Gla protein (Mgp), and biglycan (Bgn), also increased during this phase.

Specifically, during the stone elimination period (days 9–15), the most increased gene was immunoglobulin kappa chain, constant region (Igκ-C). As above, monocyte/macrophage-related genes were apolipoprotein E (Apoe), which is involved in macrophage maturation; histocompatibility 2, class II antigen A, α (H2-Aa) for antigen presentation; and LPS-induced TN factor (Litaf), which is specifically expressed in macrophages and is related to the production of TNF-α and chemokines. In addition, increased expression of nuclear protein 1 (Nupr1) for anti-inflammation and tumorigenesis; cytochrome P450, family 27, subfamily b, polypeptide 1 (Cyp27b1), which is an enzyme for vitamin D3 activation; aldehyde dehydrogenase family 1, subfamily A1 (Aldh1a1), which is an anti-oxidant against ischemia; L1 cell-adhesion molecule (L1cam), a marker of distal tubular cell injury; and the serine (or cysteine) peptidase inhibitor (Serpin) family for the regulation of tissue injury caused by neutrophils and several types of transmembranous transporter or channels were detected.

Gene ontology analysis

Using the 77 overexpressed genes, gene ontology analysis was performed from three aspects of biological process, cellular component, and molecular function, and the 10 selected categories are shown in Table 7. In the biological process, 9 of 10 categories were related to inflammation or immune reactions and indicated the involvement of fluid immunology through the production of antibodies. In the cellular component, categories of extracellular matrix constituents accounted for the majority of the list. In addition, in the molecular function, extracellular matrix constituents and cell protection factors of protease activity or oxidative stress were shown.

Table Table 7.. GO Analysis of Genes Overexpressed in Stone-Forming Mouse Kidneys
original image

The gene ontology of the 578 downregulated genes is shown in Table 8. In the biological process, the sorted categories were metabolic process, which is biased toward acid, lipid, and fatty acid metabolism. Mitochondrion-related categories in cellular components and oxide reductase-related activities in molecular functions were prominent.

Table Table 8.. GO Analysis of Genes With Suppressed Expression in Stone-Forming Mouse Kidneys
original image

Expression change of stone-related genes

Among genes previously reported as being involved in urolithiasis, kidney stones, hyperoxaluria, or hypercalciuria, the expressions of 49 genes could be detected on the microarray (Table 9). Fourteen genes that showed significant changes, Positive, and were more than two times increased or decreased, and three genes with a saturated raw value (>10,000), for which accurate quantification was difficult by only microarray analysis (Slc34a1, Spp1, and Umod), had their expression reconfirmed by quantitative PCR (Fig. 3).

Table Table 9.. Expression Profiles of Stone-Related Genes Previously Reported
original image
Figure Figure 3.

Quantitative PCR on stone-related genes extracted from microarray analysis. Expression values of each gene were assayed by quantitative PCR using TaqMan assays. The value of each graph indicates fold changes. Control values are the average data on day 0. Data are presented as mean ± SE. *p < 0.05 and **p < 0.01 vs. control (day 0). TaqMan gene expression assay probe kits were used for c-c chemokine 2 (Ccl2; Mm00441242_m1), CD44 antigen (Cd44; Mm01277163_m1), colony stimulation factor 1 (Csf1; Mm00432688_m1), fibronectin 1 (Fn1; Mm01256744m1), glyoxylate reductase/hydroxypyruvate reductase (Grhpr; Mm0519119m1), Jun oncogene (Jun; Mm00495062_s1), lysozyme (Lyzs; Mm00727183_s1), matrix Gla protein (Mgp; Mm00485009 _m1), myelocytomatosis oncogene (Myc; Mm00487804_m1), S100 calcium-binding protein A8 (S100a9; Mm01220132_g1), solute carrier family 22, member 12 (Slc22a12; Mm00486206_m1), solute carrier family 22, member 18 (Slc22a18; Mm00485426_m1), solute carrier family 34, member 1 (Slc34a1; Mm00485426_m1), superoxide dismutase 2 (Sod2; Mm00449726_m1), secreted phosphoprotein 1 (Spp1; Mm00436767_m1), TGFβ1 (Tgfb1; Mm01178820_m1), and uromodulin (Umod; Mm00447649_m1).

Chemokine (C-C motif) ligand 2 (Ccl2), CD44 antigen (Cd44), colony-stimulating factor 1 (Csf1), fibronectin 1 (Fn1), lysozyme (Lyzs), matrix Gla protein (Mgp), myelocytomatosis oncogene (Myc), secreted phosphoprotein 1 (Spp1), and TGFβ1 (Tgfb1) significantly increased, and Jun oncogene (Jun,) solute carrier family 22, member 12 (Slc22a12), solute carrier family 22, member 18 (Slc22a18), solute carrier family 34, member 1 (Slc34a1), and superoxide dismutase 2 (Sod2) significantly decreased.

Immunohistochemical findings of calcium oxalate crystal depositions and macrophage-related protein expression

Glyoxylate-induced crystal depositions were stained by Pizzolato staining, and the component was detected as calcium oxalate; the number increased on day 6 and decreased on day 12. The renal interstitial macrophages indicated by F4/80 were not observed on day 0, peaked on day 6, and were slightly decreased on day 12, especially macrophages aggregated around the stone formation region between the renal cortex and medulla. MCP-1 expression of tubular cells was increased on days 6 and 12 not only around the stone formation area but also the renal cortex. OPN expression of tubular cells around the stone formation area between the cortex and medulla was increased on day 6 and decreased on day 12. THP expression of renal tubules throughout the cortex to medulla did not change and was detected on days 0, 6, and 12 (Fig. 4).

Figure Figure 4.

Calcium oxalate crystal depositions and macrophage-related protein expression. Pizzolato staining to detect calcium oxalate depositions and immunohistochemical staining to detect related protein expression on days 0, 6, and 12 after initiation of glyoxylate administration using sequential tissue sections (magnification, ×200 [inset indicates the region around crystal depositions; magnification, ×400]). F4/80, anti-murine monocyte/macrophage; MCP-1, monocyte chemoattractant protein-1; OPN, osteopontin; THP, Tamm-Horsfall protein.


The results of all expression data of microarray analysis in the mouse kidney stone model were evaluated from three aspects: the expression ratio, gene ontology, and stone-related genes. As key words common to all evaluations of upregulated genes, chemotaxis, proliferation, phagocytosis, intracellular degradation, and antigen presentation were extracted by GO analysis. On the other hand, downregulated genes belonged to categories of acid, lipid, and fatty acid metabolism such as adiponectin (Adipoq) and lipoprotein lipase (Lpl), and it was clear that direct or indirect effects of glyoxylate administration markedly suppressed lipid metabolism. GO of downregulated genes showed a decreased expression of genes related to lipid metabolism, and the categories of mitochondrial components and oxidoreductase activity indicated oxidative stress during the experiment. Oxidative stress is an important process during stone formation,(20) and we have reported that the differences in mouse kidney stone formation among oxalate precursors resulted in oxidative stress induction(34) and antioxidants could prevent stone formation.(36,37) Oxidative stress plays a key role in the inflammation reaction and lipid metabolism competes with inflammation,(38) in which the NF-κB pathway (inflammation) is induced in response to oxidative stress and the peroxisome proliferator-activated receptor (PPAR) pathway (lipid metabolism) is suppressed. Under oxidative stress, upregulation of metallothionein 2 (Mt2) is a marker of NF-κB pathway signaling(39) and downregulation of stearoyl-coenzyme A desaturase 1(Scd1) is a marker of PPAR pathway signaling,(40) and our results showed both gene expression change and indicated that suppressed lipid metabolism could be related to the response to oxidative stress.

The genes in this study were compared with those in three previous microarray reports: two reports about ethylene glycol model rats(29,30) and a COM crystal-exposed tubular cell model.(32) Among the overexpressed genes, osteopontin (Spp1), lysozyme, MHC-class II,(29) α 1 microglobulin/bikunin, vimentin, Vcam1,(30) zinc finger protein 207, MCP-1 (Ccl2), and plasminogen activator (urokinase)(32) were in common with this study; there were no downregulating common genes in the other reports. TGFβ-stimulated clone-22 (Tsc22) showed decreased expression during the stone elimination period in our study, whereas previous reports indicated an increased expression of Tsc22. Uromodulin and THP showed low expression under normal conditions, but the administration of ethylene glycol induced a strong expression of THP; however, in the mouse model, THP showed a high expression level in the control (day 0), and glyoxylate administration could not induce a significant change of THP expression, which persisted at a high expression level on quantitative PCR and immunohistochemistry throughout the experimental course. Mo et al.(41) reported that THP knockout generated calcium phosphate depositions spontaneously at the interstitium of renal papilla regions. We reported that the administration of ethylene glycol or glycolate, which are easy methods in the rat stone model, could not induce kidney crystal depositions in mice, although glyoxylate could, and mice had strong resistance to kidney stone formation. The differences in the THP expression pattern between rats and mice suggested that THP might play a central role in the susceptibility to kidney stone formation.

Spp1 and OPN could not be detected as a two times overexpressed gene on microarray analysis because of highly saturated raw values of Spp1 (>10,000); however quantitative PCR using the same samples showed a significantly higher expression on day 6 (about four times). Thus, there might be a reluctance to omit extremely highly expressed genes from screening by microarray analysis, indicating that microarray analysis is not a perfect tool for expression analysis; however, combining data from previous reports could lead to more accurate expression profiles. OPN is the main constituent of the stone matrix of calcium-containing stones, distributed in a stratified and radial manner.(7) Our study of OPN knockout mice using the glyoxylate administration method showed that OPN deficiency reduced the number of crystal depositions and caused insufficient growth of crystal morphology(10) when OPN was a promoter of stone formation. The kidney stone formation model(33) was established in normal mice because, when candidate genes related to kidney stones were detected using model mice with a microarray or other assays, consecutive studies using gene recombinant or mutation mice involving individual genes could be developed as the next step. Data from this study might provide important information to elucidate the mechanism of kidney stone formation. Moreover, the model mouse raises a new problem, the stone elimination phenomenon, in reproducibility, and might provide suggestions for the prevention of kidney stones.

To date, the involvement of cell injury and inflammation in kidney stone formation has been suggested. Umekawa et al.(15) reported the increased expression of MCP-1 (Ccl2), a c-c chemokine, in renal tubular cells exposed to COM crystals. MCP-1 could induce the recruitment and migration of immunocytes, especially monocytes/macrophages, to inflammation sites caused by tissue injury.(42) In the rat experimental stone model, the migration of inflammatory cells, such as interstitial monocytes, macrophages, and polymorpho-nuclear leukocytes, could be observed.(43) De Water et al.(44) reported that the amount of kidney-associated oxalate diminished with time, which may have been caused by the removal of interstitial crystals through the endocytosis of multinucleate giant cells, from the findings of immunohistochemistry and electron microscopy. The accumulation and inhibition of macrophage recirculation required chemokine and OPN activities.(45) In this study, the results of microarray and quantitative PCR showed an increased expression of chemokines (Ccl2, Ccl6); OPN (Spp1); CD14, which is a cell surface marker of monocytes/macrophages; Csf1, which is essential for the differentiation and maturation of macrophages; lysozyme, which is a hydrolase of macrophages; and members of the MHC class II subunit family, which are antigen-presenting molecules on the cell surface that could be strong contributors to in monocytes/macrophages in the mouse model. On establishment of the mouse model, we already observed interstitial multinuclear cells with phagocytosis of COM crystals in the mouse kidneys by TEM.(33) Ryall et al.(46) suggested the possible degradation of COM crystals by protease treatment of intracrystalline proteins. From this, migrated macrophages through the expression of chemokines or OPN induced by stone formation stresses might have the ability to phagocytose and degrade COM crystals. The stone elimination phenomenon observed in mouse kidneys might depend on such macrophage abilities.

On the other hand, macrophages play key roles in atherosclerotic formation.(47) We are paying attention to the relationship between urolithiasis and atherosclerosis, because they have many similarities, such as onset age (middle-aged to older men and postmenopausal women), relation with a high-fat diet, calcified components, and involvement of matrix proteins. Dyerberg et al.(48) reported a lower incidence of ischemic heart disease and urolithiasis in Eskimos than Danes and concluded that the reasons were a higher intake of n-3 multivalent saturated fatty acid. We scored the aortic calcifications of stone formers and compared with healthy individuals, and identified a significantly larger number of aortic calcifications in stone formers among young men and older women.(49) The administration of eicosapentaenoic acid, an anti-hyperlipidemia agent, could reduce the recurrence rate of urolithiasis.(50) From this series of studies, high similarities between urolithiasis and atherosclerosis formation were recognized. The subendothelial recruitment of foam cells in atherosclerotic regions was thought to be the result of insufficient treatment or saturation of englobed lipids in macrophages. At stone formation sites, similar processes were speculated, in which renal macrophages could englobe crystals and might digest them.

In this study, we performed genome-wide microarray analysis of kidney stone formation and elimination in the stone model mouse. Compared with previous reports of stone-related genes or microarray in rat models or cultured cells, the ability to prevent stone formation in mouse kidneys might depend on the expression of THP. The strong relationship of stone formation and/or the elimination process with the recruitment of monocytes/macrophages was reported, as well as the similarity to the atherosclerosis formation process; however, this study was unable to distinguish between changes caused by stone formation and changes causing stone formation. We expect that subsequent morphological observation and functional analysis of renal macrophages in stone model mice could lead to the development of preventive medicines for clinical kidney stones.


We thank N. Kasuga, T. Miwa, and T. Iwama for secretarial assistance. This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (20591887, 20591885, 19791122, and 19791124) and the Japan Urological Association (Young Research Grant, 2008).