2026 Volume 51 Issue 9 Pages 489-500
Acute kidney injuries and renal dysfunction caused by ethylene glycol poisoning are often associated with nephrolithiasis. Calcium oxalate (CaOx) crystal formation is considered one of the most common events at the early stage of nephrolithiasis. The aim of this study is to investigate molecular mechanisms underlying CaOx crystal formation on renal tubular epithelial cells in vitro. We applied oxalate or potassium oxalate (final concentrations, 200-1200 µM) in the medium culturing NRK-52E rat proximal tubular epithelial cells. Cell death was observed with levels of oxalate or potassium oxalate above 800 µM 48 hr after addition. Crystals, mainly consisting of CaOx as confirmed by polarized light microscopy, were detected on cells treated with 800 µM oxalate or potassium oxalate. Transcriptome analysis showed that a gene (S100A11-like) belonging to the S100 family, which consists of homologous Ca-binding proteins, many of which can form complexes with annexin family proteins, was the gene most upregulated by 800 µM oxalate. Immunoblot as well as immunocytochemical analysis indicated not only upregulation but also alteration of intracellular localization of S100A11. The same altered pattern of localization was also observed for annexin A2, which has been shown to be involved in the formation of CaOx crystals. This study presents the possibility that, in addition to annexin family proteins, S100 family proteins might be involved in the regulation of CaOx crystal formation and/or other events associated with crystal formation.
Ethylene glycol (EG) is a colorless, odorless or sweet-smelling, and sweet-tasting liquid that mixes well with water (Litovitz, 1986; Fowles et al., 2017). EG is used for many industrial purposes such as anti-freezing agents, a component of brake fluids, and a precursor to synthesize chemicals such as polyester. Although accidental as well as intentional ingestion of EG and subsequent human death is constantly reported (Leth and Gregersen, 2005; Lovrić et al., 2007; Brent, 2001), effective antidotes against EG poisoning have not been reported other than that for several chemicals such as ethanol (Lobert, 2000; Grauer et al., 1987) and 4-methylpyrazole (fomepizole), an inhibitor of alcohol dehydrogenase (Iaquinto, 2018). EG is metabolized in the liver via the sequential action of alcohol dehydrogenase, aldehyde dehydrogenase, and hydroxyacid oxidase, which results in most of the EG being converted into oxalate (Fowles et al., 2017). When urine oxalate levels reach their saturation level, crystals form in the urine. Calcium oxalate (CaOx) is a typical crystal that includes oxalate, and is the main component in kidney stones (Khan et al., 2016; Coe et al., 2005). It has been assumed that CaOx attaches to renal tubular epithelial cells due to damage to the cells, or vice versa, at the early phase of acute kidney injury (AKI) (Fong-Ngern et al., 2017; Scherberich et al., 1993; Kunii et al., 2025). Since damage to the brush border membrane (BBM) of the proximal tubular epithelium is one of the earliest lesions in many forms of AKI, including that from EG poisoning, it is postulated that CaOx attaches to damaged BBM where CaOx crystals develop into kidney stones. Retention and growth of CaOx crystals on tubular epithelial cells is important for the formation of kidney stones (Coe et al., 2005; Lieske et al., 1994). Therefore, hyperoxaluria and subsequent kidney stone formation is a common and one of the most notable features of EG intoxication although whether CaOx crystals are really involved in renal dysfunction during EG poisoning remains controversial (McMartin, 2009).
The annexin family of proteins are calcium- and phospholipid-binding proteins comprising 12 proteins in humans (Li et al., 2022; Lin and Hu, 2022). Annexins, which have a phospholipid-binding domain, are involved in the regulation of membrane trafficking events such as endocytosis as well as exocytosis (Li et al., 2022; Lin and Hu, 2022). Among annexin family proteins, annexin A2 has been proposed to play an important role in CaOx crystal formation (Kumar et al., 2003). For example, after screening proteins using a proteomics approach, Fong-Ngern et al. identified annexin A2 as a CaOx-binding protein in tubular epithelial cells (Fong-Ngern et al., 2011). In a following study, they showed that the ectopic expression of wild type annexin A2 enhances CaOx attachment to tubular epithelial cells while forms containing mutations in their Ca-binding sites show no such effect (Yoodee et al., 2025). The S100 family is another family of calcium-binding proteins, many of which can form hetero dimers/tetramers with annexins (Zhang et al., 2021; Kligman and Hilt, 1988; Miwa et al., 2008; Rintala-Dempsey et al., 2008). For example, annexin A2 forms a complex with S100A10 which serves as a tissue plasminogen activator (tPA) receptor on the plasma membrane (Bharadwaj et al., 2021). Annexin A2 has also been shown to form complexes with other members of the S100 family, including S100A4, S100A6, and S100A11 (Lin and Hu, 2022). In contrast to the role of S100A10-annexin A2, S100A11-annexin A2 has been suggested to be involved in the process of plasma membrane repair as well as in the invasion of cancer cells (Jaiswal et al., 2014; Ashraf and Gerke, 2022; Takahashi et al., 2024). Nevertheless, there are few or no reports to date indicating the possible involvement of S100 family proteins in EG intoxication.
In this study, we aimed to examine the mechanism of renal tubular epithelial damage and subsequent nephrolithiasis due to EG ingestion. For this purpose, we administered oxalate to NRK-52E rat proximal tubular epithelial cells in vitro. Transcriptome analysis revealed a S100A11-like gene as the most induced gene upon the administration of oxalate. Furthermore, we observed an enrichment of S100A11 to the cytoplasmic granule structures in the cells after oxalate administration, suggesting the possible involvement of S100A11 in the pathophysiology of hyperoxaluria and subsequent nephrolithiasis during EG poisoning.
NRK-52E proximal tubular epithelial cells derived from rat kidney were cultured in a humidified atmosphere of 5% CO2 at 37°C in Dulbecco’s modified eagle medium (4.5 g/L glucose, FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan; Cat. No. 044-29765) supplemented with non-essential amino acids (FUJIFILM Wako Pure Chemical Corporation; Cat. No. 139-15651), 10% FBS (Cosmo Bio Co., Ltd, Tokyo, Japan; Cat. No. CCP-FBS-BR-500) and 100 U/mL penicillin (FUJIFILM Wako Pure Chemical Corporation; Cat. No. 168-23191). After reaching confluency on 6-cm-diameter dishes, the cells were administered the indicated final concentrations of oxalate or potassium oxalate. Solutions of oxalic acid and potassium oxalate were prepared from oxalic acid dihydrate (FUJIFILM Wako Pure Chemical Corporation; Cat. No. 155-00422) and potassium oxalate monohydrate (Sigma-Aldrich, Burlington, MA, USA; Cat. No. 24-5070-2), respectively. Forty-eight hours after administration, the cells were collected and proteins and/or total RNAs were extracted for analysis.
AnimalsAll protocols were approved by the Institutional Animal Care and Use Committee of the Institute of Science Tokyo (A2023-080C), and all methods are reported in accordance with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines and regulations. Male Wistar rats (8 weeks old) were acclimated to a temperature (25°C) under a 12 hr/12 hr light/dark cycle with free access to water and food, and randomly divided into three groups (3-5 rats in each group). EG group (EG2 and EG5) rats were orally administrated a single dose of EG (8 g/kg, FUJIFILM Wako Pure Chemical Corporation; Cat. No. 058-00986) and euthanized by an overdose of sodium pentobarbital (100 mg/kg body weight, Kyoritsu Seiyaku, Tokyo, Japan; Cat. No. SOM04-YA1612) 2 or 5 days after administration. Control group rats received saline solution in a volume equal to that of the EG solution. Kidneys from the euthanized rats were excised and stored at -80°C until use.
Determination of cell viability and toxicityCell viability was measured using a cell counting kit-8 (CCK-8, Dojindo, Kumamoto, Japan). Cell toxicity was evaluated as the relative levels of LDH activity in the culture medium, as determined using an LDH cytotoxic assay kit (FUJIFILM Wako Pure Chemical Corporation) to assess the relative levels of plasma membrane breakage and subsequent release of intracellular LDH into the medium. In brief, cells were seeded on a 96-well plate, and administered oxalate or potassium oxalate. CCK-8 and LDH assays were performed 24 or 48 hr after the addition of oxalate or potassium oxalate using a microplate reader (ARVO X3, PerkinElmer, Waltham, MA, USA) according to the manufacturer’s instructions.
Transcriptome analysisTranscriptome analysis was performed as described (Unuma et al., 2022) using a DNA microarray (ClariomS, Thermo Fisher Scientific, Waltham, MA, USA) for this purpose. In brief, total RNA, which was extracted and purified using Trizol reagent (Thermo Fisher Scientific) and an RNeasy RNA purification kit (QIAGEN, Hilden, Germany) from rat kidney, was hybridized to ClariomS array. The results were analyzed by TAC software (Thermo Fisher Scientific).
ImmunoblottingNRK-52E cells were washed with ice-cold PBS, precipitated by low-speed centrifugation, and lysed in Laemmli’s buffer (Laemmli, 1970). The rat kidneys were homogenized in buffer [10 mM Tris-HCl (pH 8.0), 320 mM sucrose, 1 mM EDTA, 50 mM NaF, 2 mM Na3VO4, protease inhibitor cocktail (Complete; Roche, Mannheim, Germany)] using a bead shocker (Yasui), and the subsequent cell lysates were denatured in Laemmli’s buffer. Total cellular proteins from the cells or kidneys were subjected to SDS-PAGE, blotted onto PVDF membranes, and incubated with the indicated primary antibodies (Supplemental Table S1), followed by incubation with HRP-conjugated secondary antibodies and detection of antigens by enhanced chemiluminescence (ECL; Revvity, Waltham, MA, USA; Cat. No. NEL104001EA). Results of immunoblot analysis were quantified using Image Capture 4 software (ATTO, Tokyo, Japan).
ImmunocytochemistryNRK-52E cells were fixed in 4% paraformaldehyde (FUJIFILM Wako Pure Chemical Corporation; Cat. No. 163-20145) in PBS for 5 min., permeabilized with 0.5% Triton X-100 (FUJIFILM Wako Pure Chemical Corporation; Cat. No. 168-11805) for 5 min, and blocked with 1% bovine serum albumin (Sigma-Aldrich; Cat. No. A7906-50G) in PBS for 30 min, all at room temperature. Next, the cells were incubated with primary antibodies in a moist box at 4°C overnight, followed by further incubation with Alexa488- and Alexa546-conjugated secondary antibodies (Thermo Fisher Scientific) and DAPI. The specimens were observed under a confocal microscope (CS2, Nikon Imaging, Tokyo, Japan).
Statistical analysisAll statistical analyses were performed using GraphPad Prism (version 9.0.0; GraphPad Software, San Diego, CA, USA). Data were analyzed using one-way analysis of variance (ANOVA), followed by Dunnett’s multiple comparisons test to compare each treatment group with the corresponding control group. For the experiments using culture cells, cells exposed to neither oxalate nor potassium oxalate were served as control cells. For the experiments using rats, the rats that did not receive ethylene glycol, were served as control rats.
There have been many reports in which the toxicity of less than and/or around 1000 µM oxalate or potassium oxalate on renal epithelial cells, including NRK-52E cells, has been examined (Wu et al., 2021; Qian et al., 2022; Koul et al., 2003) . We therefore examined the effects of 200-1200 µM oxalate as well as potassium oxalate on cells. Cellular viability was measured using CCK-8 assay, which estimates total cellular dehydrogenase activity (Ghasemi et al., 2021). Cell toxicity was evaluated as relative amount of extracellular LDH, which reflects plasma membrane break and subsequent cell death. To our surprise, as shown in Fig. 1A, slight but significant increases in cell viabilities were observed in cells cultured with more than 1000 µM oxalate for 24 hr. No such effect was observed in cells cultured with potassium oxalate (Fig. 1A). After 48 hr of incubation, the increasing effect of oxalate on cell viability was no longer observed (Fig. 1A). In contrast, significant increases in cell toxicity (cell death) were observed in cells treated with more than 800 µM oxalate or potassium oxalate for 48 hr (Fig. 1B). Collectively, these results might indicate that more than 1000 µM oxalate have effects to increase the cellular dehydrogenase activity and/or proliferation as well as the death in NRK-52E cells. Although the increasing effect of oxalate on the dehydrogenase activity/proliferation peaked at 24 hr (Fig. 1A), the increasing effect may be masked by the cell death-inducing effect 48 hr after incubation. Since oxalate toxicity was observable in cells treated with more than 800 µM oxalate and potassium oxalate for 48 hr, we next examined the effects of up to 800 µM oxalate and/or potassium oxalate.

Cytotoxicity of oxalate and potassium oxalate on NRK-52E cells. Cells were incubated with the indicated concentrations of oxalate or potassium oxalate for 24 or 48 hr. Cell viability (A) and toxicity (B) were evaluated by CCK-8 and LDH release assays, respectively. Each bar represents the mean and S.D. (n=4). LDH activity in the medium of control cells was set to 1. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.
We next examined the morphology of NRK-52E cells with or without the administration of 200-800 µM oxalate or potassium oxalate for 48 hr. Although we could not observe any apparent increase in dead cells, such as cells floating in the culture medium, an aggregation of damaged cells was observed following the administration of 800 µM oxalate (Fig. 2A). In addition, we observed crystals adhering to the cells treated with 800 µM oxalate or 800 µM potassium oxalate (Fig. 2A). The crystals visible in Fig. 2 show a bipyramidal shape, which is typical of CaOx dihydrate crystals (Simhadri et al., 2025).

Adhesion of crystals to NRK-52E cells treated with oxalate and potassium oxalate. Cells were incubated with the indicated concentrations of oxalate (A) or potassium oxalate (B) for 48 hr and observed under light microscopy. Representative phase contrast images are shown. Red arrows indicate bipyramidal crystals adherent to the cells.
To confirm that the crystals formed on NRK-52E cells treated with 800 µM (Fig. 2A) were CaOx, we observed the cells under a light microscope equipped with a polarized filter. As shown in Fig. 3, we observed not only bipyramidal, but also dumbbell-shape birefringent crystals, the latter of which is a typical shape of CaOx monohydrates (Simhadri et al., 2025). Collectively, the crystals formed on NRK-52E cells treated with 800 µM oxalate include both CaOx monohydrates and dihydrates, the former often predominant in kidney stones (Khan et al., 2016). We also examined calcium oxalate crystal formation under cell-free conditions. Neither culture medium alone nor oxalate solution alone produced crystals when incubated in a culture dish for 30 hr (Supplementary Fig. S1). In contrast, addition of oxalate to the culture medium at final concentrations of 500 and 1000 μM resulted in crystal formation (Supplementary Fig. S1). It should be noted that, during hyperoxaluria, calcium oxalate crystal formation occurs in urine and is followed by adhesion of the crystals to the surface of epithelial cells (Khan et al., 2021).

Calcium oxalate crystal adhesion to NRK-52E cells administered oxalate. Cells were incubated with 800 µM oxalate for 48 hr and observed under a polarized light. Bright field images and dark field images are shown. Blue arrows indicate dumbbell-shape crystals having birefringence.
Given the implication of slight but significant cell death (Fig. 1B), as well as numerous recent publications indicating ferroptosis, autophagy, and ER stress in the pathogenesis of nephrolithiasis (Tang et al., 2025; Khan and Alli, 2025; Yang et al., 2024; Abhishek et al., 2017; Sun et al., 2021), we examined the possible involvement of these processes by evaluating several marker proteins by immunoblot analysis. Although ferroptosis is accompanied by decreases in glutathione peroxidase 4 (Gpx4) and SLC7A11 (Berndt et al., 2024), neither oxalate nor potassium oxalate decreased either of these two markers (Fig. 4). These results suggest that ferroptosis may not be involved. The respective increase and decrease of LC3-II and p62 are a feature of autophagy (Komatsu and Ichimura, 2010); however, no changes in the levels of these markers were observed (Fig. 4). These results rule out the possible involvement of autophagy. Although we also did not observe any change in the level of the ER stress marker Bip, a significant decrease of another ER marker, CHOP, was observed in oxalate-treated cells as compared to non-treated cells (Fig. 4). Thus, oxalate might have decreased ER stress under our experimental conditions, in contrast to the large amount of literature showing an induction of ER stress by oxalate (Albert et al., 2020).

Immunoblot analysis of ferroptosis, autophagy, and ER stress in NRK-52E cells administered oxalate. Cells were incubated with the indicated concentrations of oxalate (A) or potassium oxalate (B) for 48 hr. Then lysates of the cells were extracted and subjected to immunoblot analysis. GAPDH served as an internal standard. Each bar represents the mean and S.D. (n=4). ns, not significant; *, P < 0.05.
We conducted a transcriptome analysis using DNA microarray, which should provide an overview of the cellular status of NRK-52E cells treated with oxalate. Interestingly, the most upregulated gene was that for the S100A11-like protein in oxalate-treated cells (Table 1). A protein-protein BLAST analysis showed 83.67% similarity of amino acid sequences between the S100A11 protein and this S100A11-like protein in rats, suggesting that the S100A11 protein might comprise at least two proteins in rats. It should be noted that a glycolytic enzyme, triosephosphate isomerase, as well as a gene for tubulin, was included in the list (Table 2), implying possible changes in energy production as well as the cytoskeleton. Pathway-enrichment analysis showed that the most and 2nd most significantly affected pathways are involved in cell cycle, while the 3rd and 4th most affected pathways are related to cholesterol (Table 2). The result of pathway analysis suggests that there may be a demand for membrane synthesis in cells treated with oxalate.
| No | Fold Change | Gene Symbol | Description |
|---|---|---|---|
| 1 | 5.5 | LOC100362296 | protein S100-A11-like |
| 2 | 4.89 | RGD1565648 | similar to Chain A Solution Structure Of Rabbit Apo-S100a11 (19 Models) |
| 3 | 4.73 | Tpi1 | triosephosphate isomerase 1 |
| 4 | 4.2 | Tuba3b | tubulin, alpha 3B |
| 5 | 4.1 | RGD1563601 | Triosephosphate isomerase |
| 6 | 4.04 | Selp | selectin P |
| 7 | 3.96 | LOC100911515 | ENCODES a protein that exhibits isomerase activity (ortholog) AND triose-phosphate isomerase activity (ortholog) |
| 8 | 3.96 | LOC500959 | triosephosphate isomerase |
| 9 | 3.86 | Tk1 | thymidine kinase 1, soluble |
| 10 | 3.84 | Car1 | carbonic anhydrase I |
| 10 | 3.84 | Acat2 | acetyl-CoA acetyltransferase 2 |
| No | Pathway | Total | Up | Down | Significance | p-value |
|---|---|---|---|---|---|---|
| 1 | Cell cycle | 10 | 8 | 2 | 4.23 | 0.000059 |
| 2 | G1 to S cell cycle control | 7 | 4 | 3 | 2.94 | 0.001150 |
| 3 | Cholesterol metabolism | 4 | 4 | 0 | 2.66 | 0.002192 |
| 4 | Cholesterol biosynthesis | 3 | 3 | 0 | 2.27 | 0.005367 |
| 5 | Eicosanoid synthesis | 3 | 1 | 2 | 1.97 | 0.010629 |
| 6 | Hypertrophy model | 3 | 0 | 3 | 1.91 | 0.012280 |
| 7 | Spinal cord injury | 7 | 4 | 3 | 1.89 | 0.012880 |
| 8 | DNA replication | 4 | 4 | 0 | 1.79 | 0.016315 |
| 9 | Endochondral ossification | 5 | 4 | 1 | 1.77 | 0.016847 |
| 10 | mRNA processing | 7 | 5 | 2 | 1.56 | 0.027533 |
To confirm the induction of the S100A11 and/or S100A11-like proteins, immunoblot analyses were performed using a polyclonal anti-S100A11 antibody raised against a 3-105 amino acid region of human S100A11 (1-105 amino acids in total), which was expected to recognize rat S100A11 as well as S100A11-like proteins. Immunoblot analysis using this antibody showed that S100A11 and/or S100A11-like protein levels increased in response to oxalate administration in NRK-52E cells (Fig. 5A). Given the indications that S100A11 cooperates with annexin A2 as described in “Introduction”, we also examined annexin A2 and found a significant increase in the protein level (Fig. 5A). We next examined S100A11 as well as annexin A2 in the kidney from EG-administered rats. Kidneys from the same animals previously demonstrated to contain renal calcium oxalate crystals (Unuma et al., 2026) were used for the analysis. However, in contrast to the findings in the cultured cells, EG administration did not alter the levels of annexin A2 or S100A11 in rat kidneys (Fig. 5B). Thus, the increases in S100A11 and annexin A2 might be observed only in epithelial tubular cells, although further investigation is needed to elucidate the discrepancy between the rat kidney and the cultured cells.

Induction of S100A11 in NRK-52E cells administered oxalate and in the kidneys of rats administered ethylene glycol. (A) The cells were incubated with the indicated concentrations of oxalate for 48 hr. (B) Wistar rats (8-week-old males) were orally administered ethylene glycol (EG, 8 g/kg body weight), and the renal tissues were excised 2 or 5 days after the administration. Lysates of the cells (A) and the renal tissues from rats (B) were subjected to immunoblot analysis for annexin A2 and S100A11. GAPDH served as an internal standard. Each bar represents the mean and S.D. ns, not significant; *, P < 0.05, **, P < 0.01.
Finally, we examined the cellular localization of S100A11 as well as annexin A2. Immunofluorescence analysis under confocal microscopy revealed increased localization of annexin A2, S100A11, and E-cadherin (plasma membrane marker) to cytoplasmic granule structures in the cells administered oxalate (Fig. 6). These results suggest plasma membrane damage as well as possible association with S100A11 and annexin A2 in the cytoplasm, in the cells administered oxalate.

Immunocytochemistry of S100A11 and Annexin A2 in NRK-52E cells administered oxalate. Cells were incubated with 800 µM oxalate for 48 hr. Representative immunofluorescence images of S100A11 (green) and annexin A2 (green) were obtained under confocal microscopy. The cells were also stained with DAPI (blue) and E-cadherin (red) to visualize nuclei and plasma membrane.
In conclusions, this is the first report suggesting the induction and possible involvement of a S100A11 protein in renal homeostasis under the status of hyperoxaluria. To the best of our knowledge, there has been only one report showing an involvement of a S100A family protein in renal CaOx stone formation; Wang et al., demonstrated an accelerating effect of S100A9 on CaOx stone formation via TLR4 activation and a subsequent promotion of inflammatory injury (Wang et al., 2024). However, whether S100A11 might be involved in nephrolithiasis due to hyperoxaluria or not was not tested in the current study. Thus, further work is necessary to elucidate the role of S100A11 in the pathogenesis of nephrolithiasis due to ethylene glycol poisoning.
FundingThis work was supported by JSPS KAKENHI Grant Numbers (JP23K09760 to T.A. and JP22K10606 to K.U.).
Conflict of interestThe authors declare no competing interests.
Data availabilityData will be made available on request.
Author contributionsConceptualization: Shintaro Isa, Toshihiko Aki, Kana Unuma
Funding acquisition: Toshihiko Aki, Kana Unuma
Investigation: Shintaro Isa
Supervision: Kana Unuma
Visualization: Shintaro Isa, Toshihiko Aki, Kanako Noritake, Atsushi Yamada
Writing – original draft: Toshihiko Aki
Writing – review & editing: Shintaro Isa, Kanako Noritake, Atsushi Yamada, Kana Unuma
Ethical approval and consent to participateAll protocols were approved by the Institutional Animal Care and Use Committee of the Institute of Science Tokyo (A2023-080C), and all methods are reported in accordance with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines and regulations.
Patient consent for publicationNot applicable.