Additional file 1: Figure S1. Synthetic route of C5A. Figure S2. 1H NMR spectrum of 1 in DMSO-d6, 400 MHz, 25 ��C. Figure S3. (a) 1H NMR spectrum of C5A in DMSO-d6, 400 MHz, 25 ��C. (b) 13C NMR spectrum of C5A in DMSO-d6, 100 MHz, 25 ��C. Figure S4. Absorbance of C5A (10 ��M) at 400 nm as a function of time following addition of SDT (2 mM) in PBS buffer (10 mM), and the corresponding fitting curve according to quasi-first order reaction decay model. Figure S5. Mass spectrum of C5A (QFT-ESI). Figure S6. Zeta potential detection of C5A@EVs and Pc/C5A@EVs upon incubation with or without SDT in PBS buffer. (a) Zeta potential measure of C5A@EVs in the absence (blue) or presence of SDT (red) in PBS buffer (10 mM, pH = 7.4) at 25 ���. (b) Zeta potential measure of Pc/C5A@EVs in the absence (blue) or presence of SDT (red) in PBS buffer (10 mM, pH = 7.4) at 25 ���. Figure S7. Representative TEM images of Pc/C5A@EVs on day 1, 3 and 7. Scale bar, 100 nm. Figure S8. In vitro colocalization of Pc/C5A@EVs and EVs with LysoTracker Red in TECs by CLSM. Figure S9. CLSM images of Pc/C5A under hypoxic or normoxic condition. Figure S10. CLSM images of Pc@EVs under hypoxic or normoxic condition. Figure S11. Hemolysis rate of Pc/C5A and Pc/C5A@EVs after incubation with mouse erythrocytes. Figure S12. Body weight changes of the mice with different treatments (n = 5). Figure S13. In vivo imaging of kidney hypoxia via Pc/C5A@EVs and Pc/C5A. (a) Illustration of the imaging tracing strategy after intravenous injection of Pc/C5A@EVs or Pc/C5A in unilateral hypoxic renal injury mice. (b) In vivo fluorescence images of Pc/C5A@EVs and Pc/C5A in unilateral hypoxic renal injury mice. Figure S14. In vivo imaging of hypoxia-injured mice with intravenous injection of Pc, Pc@EVs or PKH26@EVs at different time intervals (n = 3). Figure S15. (a) Ex vivo images of major organs at the designed time points after Pc, Pc@EV or PKH26@EV administration. (b-d) Time-dependent fluorescence intensity changes in major organs at designated time intervals after sacrificing renal hypoxia-injured mice (n = 5). Figure S16. Representative CLSM images of kidney slices from the renal hypoxia-injured mice after administration of Pc/C5A@EVs or Pc/C5A for 24 h. Figure S17. (a) Representative H&E-staining images of kidney sections on day 1, 3 and 7 after injection. Massive necrosis in the proximal tubules with hyaline cast formation (asterisks) was observed, and administration of Pc/C5A@EVs largely prevented histopathologic alterations after hypoxic injury. Scale bar, 50 ��m. (b-c) Quantitative histological assessment of hyaline cast formation (b) and tubular necrosis (c) on day 3 postinjection. Figure S18. RT-qPCR analysis of the expression of apoptosis-related genes on day 7 postinjection. Figure S19. Representative Western blot showing Caspase 8 expression in kidney tissues with different treatments on day 7 postinjection. (a) Representative Western blot images of Caspase 8 in different groups. (b) Quantitative analysis of Western blot. Figure S20. Representative images of Masson trichrome staining in different groups (n = 8). Figure S21. Representative images of immunofluorescence staining of F4/80 (green) and Pc/C5A@EVs or Pc/C5A (red) in the kidney tissues in different groups. Figure S22. The raw data for Western blot in Fig. 2d. Figure S23. The raw data for Western blot in Fig. 6a. Figure S24. The raw data for Western blot in Fig. 6b. Figure S25. The raw data for Western blot in Fig. 6c. Figure S26. The raw data for Western blot in Fig. 7j. Figure S27. The raw data for Western blot in Fig. 8k-l. Table S1 Primers used in the RT-qPCR assay.