CD13 (Sorting and IF, #ab52461; Abcam, Cambridge, UK), lotus tetragonolobus lectin (LTL) (IF, #FL-1321; Vector Labs, Newark, CA, USA), aquaporin 1 (IF, #sc32737; Santa Cruz, Dallas, TX, USA), Dulbecco's modified Eagle’s medium with nutrient mixture F-12 (DMEM/F12) (Gibco, Waltham, MA, USA), FBS (Gibco), epidermal growth factor (EGF) (PeproTech, Waltham, MA, USA), basic fibroblast growth factor (PeproTech), insulin-transferrin-selenium (ITS, Gibco), 1α,25-dihydroxyvitamin D3 (Sigma, St. Louis, MO, USA), all-trans-retinoic acid (ATRA) (Sigma), and the RNAeasy kit (Qiagen, Valencia, CA, USA).

Human kidneys were collected from patients who underwent surgery for nephrectomy at Chungnam National University Human Resources Bank (Daejeon, Korea) in accordance with the relevant guidelines and regulations. Informed consent was obtained from patients for the use of specimens for research purposes only. We selected healthy and normal kidneys (glomerular filtration rate above 65 mL/min/1.73 m²) for cell isolation. Cells were isolated by 2 mg/mL collagenase I digestion for 30 minutes at 37°C with gentle stirring. The cells were then filtered through a 100-μm mesh to isolate single cells. Cell suspensions were cultured in DMEM/F12 supplemented with 10% fetal bovine serum, 20 ng/mL EGF, and 1% penicillin/streptomycin at 5% CO₂ and 37°C. After 2 to 6 hours, cells were incubated with human telomerase reverse transcriptase and SV40 lentiviruses and 4 μg/mL polybrene. To isolate proximal tubule cells, cells were sorted using a fluorescein isothiocyanate-conjugated anti-CD13 antibody. Cells were detached using 0.1% trypsin-ethylenediaminetetraacetic acid at 80% confluence and passaged. To determine the half maximal inhibitory concentration, human immortalized primary cells (2×10³ cells) were seeded into 96-well plates and incubated in DMEM/F12 containing serially diluted drugs for 24 hours. Cell viability was measured using the CytoX cell viability assay kit (LPS Solution, Daejeon, Korea) according to the manufacturer’s instructions.

RNA was isolated using the RNAeasy Mini kit, 1 μg was reverse transcribed using a cDNA archival kit (Life Technologies, Gaithersburg, MD, USA), and quantitative polymerase chain reaction was performed according to the manufacturer’s instructions (Applied Biosystems, Waltham, MA, USA, and Agilent Technologies, Santa Clara, CA, USA) using the SYBRGreen Master Mix. The data were normalized and analyzed using the ΔΔCT method. The primers used are listed in Table S1.

Immortalized renal cells were suspended in a 2:1 mixture of extracellular matrix (ECM) gel (Sigma) and Matrigel, and 50-μL droplets at 1 to 2×10⁴ cells/μL were deposited on the inverted lid of a culture dish. The lid was then placed onto a phosphate-buffered saline-filled dish, and after gelation (3-6 hours later), cells in the matrix were transferred to a spheroid-forming unit filled with DMEM/F12 supplemented with 10% FBS, 1× ITS, 20 ng/mL EGF, 100 nM dexamethasone, 20 μM 1α,25-dihydroxyvitamin D3 and 5 μM ATRA for 6 hours. Cells were then cultured for 2 to 5 additional days without ATRA. For nephrotoxicity tests, kidney spheroids were treated with various concentrations of drugs in serum-free DMEM/F12 for 24 hours, and for transporter activity assays, spheroids were pretreated overnight with cimetidine or verapamil before drug treatment.

Spheroids embedded in 2% agar were fixed with 10% formalin, embedded in paraffin, and cut to a 5-μm thickness. The sections were incubated in 2% bovine serum albumin with 0.2% fish skin gelatin at room temperature for 1 hour to block nonspecific binding. The sections were incubated with primary antibodies overnight at 4°C, and then secondary antibodies at room temperature for 1 hour. Nuclei were counterstained with 4',6-diamidino-2-phenylindole.

Gamma-glutamyl transferase (GGT) activity was measured by following the release of para-nitroanilide from gamma-glutamyl-p-nitroanilide using a GGT activity colorimetric assay kit (BioVision, Milpitas, CA, USA). Cell and kidney spheroids were homogenized in 200 μL of ice-cold GGT assay buffer, and 10-μL aliquots were combined with 90 μL of GGT substrate solution and added to the assay plate for a 5-hour incubation. Absorbance changes at 418 nm were measured every 30 minutes at 37°C.

Parathyroid hormone (PTH) was obtained from Prospec (New Brunswick, NJ, USA). After overnight incubation with 0.1 mM 3-isobutyl-1-methylxanthine, a phosphodiesterase inhibitor, cells were treated with 100 nM PTH for 30 minutes. Intracellular cyclic adenosine monophosphate (cAMP) was measured using a cAMP direct enzyme immunoassay kit.

Data are displayed as the mean ± standard error of the mean and were analyzed with Microsoft Excel (Microsoft Corp., Redmond, WA, USA). Statistical significance was determined using the two-tailed Student t-test. A p-value <0.05 was considered significant.

To isolate the proximal tubule cells from mixed renal primary cells, we used CD13, specific markers with extracellular epitopes expressed in the brush border of proximal tubule cells (Fig. 1A). About 70% of the mixed primary cells expressed CD13, and when cultured after sorting, it was confirmed that almost all cells expressed CD13 (Fig. 1B). We established three different cell lines using primary cells from three different donors, and these cell lines maintained epithelial cell morphology throughout the culture period (Fig. 1C). Moreover, they showed higher GGT activity than hPTECs (Fig. 1D). To generate in vitro functional proximal tubule spheroids, we used a combination of the hanging-drop seeding and rotation methods with ECM gel, as previously described [11]. iRPTC-spheroids showed a tubular structure with AQP1 (a proximal tubule marker) and LTL (an apical glycoprotein in the brush border of the proximal tubule) expression (Fig. 2A). In contrast, hPTEC-spheroids cultured under the same conditions did not show a tubular structure or LTL expression, even though the cells expressed AQP1 (Fig. 2A). The expression levels of major uptake and efflux drug transporters were increased in iRPTC-spheroids generated with the proximal tubule cell lines described in this study compared to hPTEC-spheroids (Fig. 2B). GGT activity and the magnitude of the cAMP response to PTH were significantly higher in iRPTC-spheroids than in hPTEC-spheroids, suggesting that spheroids generated with primary proximal tubule cells have normal functions related to drug and xenobiotic detoxification and a mature PTH receptor expression and cellular response (Fig. 2C and 2D). These data suggest that the iRPTC-spheroids were structurally and functionally matured compared to those formed from hPTECs, which is the most commonly used in vitro renal cell model.

To examine the response to the toxic effects of nephrotoxic drugs, a cell viability assay was conducted in iRPTCs and hPTECs treated with four different compounds: cisplatin, cyclosporin A, digoxin, and doxorubicin (Fig. 3). iRPTCs were exposed to these drugs for 48 hours at different concentrations, and the IC₅₀ values were 18.3 μM for cisplatin, 24.1 μM for cyclosporin A, and 2.8 μM for doxorubicin, respectively (Fig. 3). Cell toxicity was not detected in the spheroids treated with digoxin. However, the IC₅₀ values of hPTECs were 58.4 μM for cisplatin, 99.81 μM for cyclosporin A, 97.3 μM for digoxin and 7.8 μM for doxorubin, respectively (Fig. 3). These data suggest that the iRPTCs were more sensitive to the nephrotoxic drugs than the hPTECs. Moreover, the absence of cytotoxic effects from digoxin indicated that the efflux transporter properly worked in the iRPTCs. Next, we investigated the apoptosis profile in spheroids generated with iRPTCs and hPTECs after treatment with 50 μM cisplatin and 5 μM doxorubicin for 48 hours using the MUSE analyzer. The apoptosis analysis showed that ~75% of cells were apoptotic (annexin V positive cells) in 50 μM cisplatin, and ~62% of cells were apoptotic cells in 5 μM doxorubicin in the spheroids generated with iRPTCs (Fig. 4A). However, in hPTEC-spheroids, the number of dead cells was dramatically decreased after the same concentration of cisplatin or doxorubicin treatment, indicating that spheroids using iRPTCs showed greater sensitivity to nephrotoxic compounds, as shown by the IC₅₀ results. Kidney injury markers such as the expression of HAVCR1, LCN2, CLUSTERIN, and CASPASE-3 were significantly higher in iRPTC-spheroids than in hPTEC-spheroids (Fig. 4B). These results demonstrate that the proximal tubule spheroids described in this study are a more reliable and sensitive in vitro kidney model for nephrotoxicity assessments.