Fig. 7: Tubule remodeling processes in relation to tubule fate and functional recovery.

a Serial in vivo 2-photon microscopy images of representative Mid and IR regions in CycB1-GFP kidneys at indicated days after partial IRI show GFP expression (arrowheads) in flattened tubule epithelium (FL) and granular casts (arrows) in atrophic tubules (ω). Scale bar: 100 µm. b Dynamic assessment of PT-S1 albumin reuptake over time (n = 80, 40, 28, and 22 for sham, uninjured, recovered, and atrophic tubule segments from 3 sham and 4 partial IRI mice); mean ± 95% CI with scatterplot. Statistical test: linear mixed-effect model, p-values from two-sided test (Supplementary Extended Statistics.p). c, d Volumetric quantification of PI+ nuclei (% of total nuclei/segment) at day 0 (from tubules with >1% PI+) (c) and GFP+ nuclei (% of total nuclei/ segment) at day 3/4 (d), in recovered and atrophic tubule segments, respectively (c: n = 89 and 92 segments from 4 and 3 mice; d: n = 124 and 168 segments from 4 mice, respectively); mean ± 95% CI with scatterplot. Statistical tests: linear mixed-effect model, p-values from two-sided test (Supplementary Extended Statistics.q). e Correlative in vivo and ex vivo 2-photon microscopy images of CycB1-GFP kidneys 21 days after partial IRI reveal selective tubular VCAM-1 immunostaining (blue, arrows) in atrophic tubules (ω). Scale bar: 100 µm. f, g Tubular fate distribution (%) across partial IRI areas (f: n = 82,232, and 153 segments for Not-IR, Mid, and IR from 4, 6, and 5 mice, respectively), and days after partial-IRI (g: n = 103 and 106 for recovered and atrophic tubules from 6 mice, respectively). h Binary regression of luminal granular cast area (in % of total segment cross-sectional area) at day 3/4 and respective tubule fate (0 = non-atrophy; 1 = atrophy), (n = 237 segments from 4 mice, respectively). Fit (slope*x + intercept); R², and effect size as odds ratio (OR) for the predictor are reported. *: p < 0.05; **: p < 0.01; ***: p < 0.001.