Fig. 5

Dephosphorylation of RRM2 stimulates RRM2 binding with GSS to promote their corecruitment to the proteasome and subsequent activation of ferroptosis. a, b Reciprocal IP experiments of RRM2 and GSS in RRM2^−/− HepG2 cells reconstituted with RRM2^WT, RRM2^T33A or RRM2^T33E. The amount of proteins in the immunoprecipitates was normalized to that in whole-cell lysates (Input), and was graphed in the lower panel. c, d Reciprocal IP experiments for RRM2 and GSS in HepG2 cells with or without NU6102 (20 μM, 24 h) and SB203580 (10 μM, 24 h) treatment. The amount of protein in the immunoprecipitates was normalized to that in whole-cell lysates (Input), and was graphed in the lower panel. e Proximal protein ligation between endogenous GSS and the indicated exogenous RRM2-Myc, as measured by PLA in HepG2 cells expressing RRM2^WT, RRM2^T33A or RRM2^T33E. Scale bar, 20 μm. The PLA signals were also calculated and graphed, and the data from the “Empty” group were arbitrarily set to 1. f RRM2^WT, RRM2^T33A or RRM2^T33E was expressed in reconstituted RRM2^−/− HepG2 cells. Immunoprecipitations were acquired with anti-PSMB5 antibodies and further analyzed by immunoblotting using anti-RRM2 and anti-GSS antibodies. GSS and RRM2 enrichment in the immunoprecipitates was calculated as the normalization to the levels in whole-cell lysates (input). g–i GSH (g), cell death (h) and 4-HNE (i) were measured in HepG2 and SMMC-7721 cells with or without RRM2 knockdown before they were further treated with Fer-1 (1 μM, 24 h), ZVAD-FMK (20 μM, 24 h), Nec-1 (20 μM, 24 h), or ectopically expressed GSS, RRM2^T33A or RRM2^WT in the presence or absence of NU6102 (20 μM, 24 h). The data are shown as the mean ± SD from three biological replicates (including IB). *P < 0.05, **P < 0.01 indicates statistical significance. The data from a–d and f–i were analyzed using one-way ANOVA