Repeated Cisplatin Injury in Human Kidney Tubuloids Produces a Fibroblast-Activating Secretome
Adult kidney-derived tubuloids developed DNA damage, senescence-associated changes and inflammatory transcripts after repeated cisplatin exposure. Their conditioned medium activated human lung fibroblasts and mouse kidney fibroblasts, whereas direct cisplatin did not reproduce the human fibroblast response.
Adult Renal Cells Supported a Repeated-Injury Experiment
An Institute of Science Tokyo study in Advanced Healthcare Materials used adult human kidney tubuloids to connect repeated cisplatin injury with downstream fibroblast responses. The model was designed to examine more than acute cell death: the investigators asked whether surviving injured tubular cells develop senescence-associated features and release factors that influence a separate recipient-cell assay.
Primary renal proximal tubular epithelial cells were isolated from tumor-free cortex in kidneys removed for renal or urinary tract malignancies. After initial expansion, the cells formed tubuloids over one to two weeks in matrix-supported culture. The structures consisted of epithelial layers around the matrix rather than solid cell spheroids. LTL and LRP2/megalin staining supported proximal tubular identity and polarity.
Deconvolution of bulk RNA profiles from four tubuloid samples estimated that approximately 60–65% of the cells had proximal-tubule-like expression. Other populations were also present. Each biochemical sample pooled multiple tubuloids from a culture well. The model was predominantly tubular, with partial representation of other renal populations and no integrated vascular or filtration compartment.
High-Dose Damage and Repeated Lower-Dose Injury Were Distinct
The initial comparison exposed tubuloids to 0.2, 2.0 or 20.0 μg/mL cisplatin. The highest concentration disrupted the three-dimensional architecture and increased γH2AX, kidney injury molecule 1 and cleaved caspase 3. LRP2 and Na+/K+-ATPase declined, while vimentin increased. These measurements linked structural collapse with DNA damage, epithelial injury and altered differentiation markers.
The repeated-treatment experiment used five cycles at 0.2 or 2.0 μg/mL, with drug removal and recovery between cycles. The 20.0 μg/mL condition was excluded because of extensive death and structural damage. Repeated lower-dose treatment increased senescence-associated β-galactosidase, p16, p21 and p53, reduced lamin B1 and produced G2/M accumulation. Untreated adult-cell cultures also showed some baseline senescence-associated signals.
IL-1β and IL-6 messenger RNA increased by approximately 50–100 fold in the repeated-treatment experiment. These were transcript changes, not equivalent fold increases in secreted protein concentration. Together with the cell-cycle and senescence markers, they supported a senescence-associated secretory phenotype as a working explanation for the next experiment.
Conditioned Medium Separated Tubular Injury from Recipient Fibroblast Effects
After repeated treatment and drug removal, the investigators collected conditioned medium during an additional 48-hour serum-free period. This medium was transferred to human lung fibroblasts and, in a separate arm, mouse primary kidney fibroblasts. The human recipient cells were therefore from a different organ, rather than donor-matched renal fibroblasts in a simultaneous coculture.
Medium from tubuloids treated repeatedly with 2.0 μg/mL cisplatin produced the strongest α-SMA and collagen I responses. Direct exposure of human lung fibroblasts to cisplatin-containing medium did not induce the same fibrotic-marker phenotype. This comparison supports a role for factors released from the injured tubuloids rather than direct drug exposure alone.
Conditioned medium also increased fibroblast-mediated collagen-gel contraction and migration, with the strongest response again associated with the higher repeated tubuloid dose. The marker-area comparisons included eight to ten measurements per condition. Contraction and migration experiments each used n = 2, giving those functional observations a smaller evidential base than the marker-imaging comparisons.
The Model Links Two Assays Without Recreating Chronic Kidney Disease
The experiment connects a repeated drug exposure in adult human tubular epithelium to a transferable fibroblast-activating effect. It does not identify IL-1β, IL-6 or another individual factor as necessary for that effect, because neutralization or equivalent causal tests were not reported. Nor does increased vimentin alone establish a completed lineage conversion.
Adult-cell origin and repeated exposure offer useful features for renal drug-safety research. At the same time, the transferred-medium design omits direct epithelial–stromal contact, vascular interactions and urine flow. The fibroblast origin and small contraction/migration assays are concrete considerations when using this model to test interventions against persistent injury, rather than treating its short experimental course as a complete model of long-term kidney disease.