Acute kidney injury (AKI) is a known risk factor for chronic kidney disease (CKD). The mechanisms for AKI-CKD transition may involve failed tubule repair and progressive interstitial fibrosis but are incompletely understood. Which events lead to failed tubule repair? Can injury spread over time? And are these processes steered by myofibroblasts and fibrotic tissue remodeling?Here, we combined serial intravital 2-photon microscopy and a novel model of partial ischemia-reperfusion injury (partial IRI), which allowed us to investigate injured tissue alongside uninjured tissue in the same kidney and over time. Thus, we occluded only one branch of the left renal artery for 21 min, rendering half the kidney ischemic. We then tracked tubule necrosis and remodeling alongside myofibroblast dynamics for up to 3 weeks. Necrotic cell death was detected by in vivo propidium iodide staining. PDGFRβ-CreERT2-Salsa6F mice (N=5) identified main myofibroblast precursor cells, by tdTomato-expression. Myofibroblast-differentiation was confirmed by ex vivo αSMA staining. Proximal tubule (PT) function was assessed by in vivo uptake of fluorescent albumin.Necrotic cell death was solely detectable upon reperfusion. Border areas in between post-ischemic and non-ischemic tissue presented severely necrotic PTs alongside uninjured PTs. From 1 day after partial IRI, we observed formation of granular casts in the lumen of necrotic tubules, which 1-2 days later also appeared in initially non-necrotic PTs downstream. Tubule necrosis and granular cast accumulation preceded thinned epithelium that lacked albumin endocytosis (day 3-7) and recruitment of myofibroblasts (peaking D4-10). Of note, non-necrotic tubules, without granular cast accumulation, didn’t remodel nor recruit myofibroblasts.Among all remodeling tubule segments, 24% turned atrophic. Importantly, preceding necrotic cell death was higher in atrophic than in recovering tubules (55±13% vs. 39±18% necrotic cells/total segment cell number, (mean±SD), p<.0001). Preceding granular casts accounted for 38±15% vs. 12±11% (mean ±SD) of the luminal area in atrophic and recovering tubules, respectively (p<.0001). Ex vivo staining for pro-fibrotic VCAM-1 demonstrated selective positivity in atrophic tubules, which appeared scattered at D8 while at D21, entire atrophic tubules stained positive (n=3, each). Consistently, we detected persistent myofibroblast enclosure of atrophic tubule, while recruitment around recovered PTs resolved by day 14 (50±21% vs. 23±18% of coverage, p<.001).Tracking the same tubules in vivo and over time, we demonstrated early proximal tubule necrosis, which induced luminal granular cast formation and injury propagation. These events preceded myofibroblast accumulation and predicted tubule atrophy. We further documented de novo expression of pro-inflammatory VCAM-1 and persistent myofibroblast recruitment selectively in atrophic tubules, suggesting an active role of this tubule state in AKI-CKD transition. Novo Nordisk Foundation, Augustinus Foundation, Aarhus University Research Foundation (AUFF) This is the full abstract presented at the American Physiology Summit 2023 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.