PMC

(A) Representative images of mitochondrial morphology in U2OS transfected with the indicated siRNAs after 9 hr treatment with vehicle or doxorubicin (DR). Insets are shown in the lower row. (B) Cells were transfected with the indicated siRNAs and treated with DR. Apoptosis was assayed by caspase 3 activation. Knockdowns of Huwe1 and Mfn2 were confirmed by immunoblotting. β-actin is shown as a loading control. (C) After transfection of the indicated siRNAs, cells were treated with DR for 20 hr and cell death monitored by DAPI. Data represent mean ± standard deviation (SD) for three independent experiments (**p< 0.01).

(A) U2OS cells were treated with doxorubicin (DR) for 2 hr. Immunoprecipitated endogenous Mfn2 was immunoblotted with antibody recognizing either S*PXR/K (S* = phospho-serine) or Mfn2. (B) Cells transfected with Mfn2HA or Mfn2HA^S27A and treated with DR for 2 hr. HA immunoprecipitates were immunoblotted for Mfn2 phospho-Ser27 (P-Mfn2) or HA. (C) Cells transfected with Mfn2HA and with siRNAs [si_JNK, si_p38 or si_CTL (control)] were treated with DR for 2 hr. Mfn2HA was immunoprecipitated with HA antibody and immunoblotted for P-Mfn2 or HA. Cell lysates were immunoblotted for activated (P) or total JNK and p38. (D) Lysates from cells transfected with Mfn2HA, FLAG-JNK2α2 and GFP were immunoprecipitated with FLAG antibody and immunoblotted as indicated. (E) HA immunoprecipitates from U2OS cells stably expressing Mfn2HA were subject to an in vitro kinase reaction with purified recombinant 6xHis-JNK2α2 and assessed by immunoblotting. IP from untransfected U2OS cells served as a negative control (first lane).

(A) Schematic of Huwe1. (B) In vitro binding of ³⁵S-labelled Mfn2HA, Mfn2HA^S27D and Mfn2HA^S27A to GST-Huwe1^BH3. WT and mutant Mfn2HA were translated in rabbit reticulocyte lysate; GST-Huwe1^BH3 was expressed in bacteria. (C) In vitro binding of ³⁵S-labeled Mfn2HA^S27D to GST-Huwe1^BH3m. Four most conserved residues in BH3 domains (indicated in bold below) were mutated to generate BH3m. (D) Co-immunoprecipitation of Mfn2HA, Mfn2HA^S27D and Mfn2HA^S27A with catalytically inactive FLAG-Huwe1^C4341A in cells pre-treated with MG132. (E) Semi-quantitative evaluation of the binding relative to the total amount of expressed FLAG-Huwe1^C4341A was determined by ECL and imaging with a CCD camera (G:box; Syngene). Data are from two independent experiments including (D). Data represent mean ± standard deviation (SD) for two independent experiments including (D).

(A) Degradation of Mfn2HA was assessed by ³⁵S pulse-chase in U2OS cells stably transfected with doxycycline-inducible Huwe1 shRNA (U2OS^shHuwe1) treated with doxorubicin (DR, □); doxycycline (sh_Huwe1, ▲); doxycycline + doxorubicin (sh_Huwe1 + DR, ■); or untreated (○). Data represent mean ± standard deviation (SD) for three independent experiments. (B) Ubiquitination of Mfn2 in U2OS^shHuwe1 in which Huwe1 was or was not silenced was assessed after 1 hr treatment with DR. (C) Reduction of Huwe1 in U2OS after 72 hr by transient transfection of Huwe1 siRNAs (si_Huwe1) compared to control (si_CTL). (D) Quantification of Mfn2HA by ³⁵S pulse-chase experiments following Huwe1 or control knockdown. Data represent mean ± standard deviation (SD) for three experiments. (E) Representative experiment from (D). (F) Re-expression of WT or inactive Huwe1 in U2OS after shRNA knockdown. (G) Quantification of ³⁵ S pulse-chase experiments evaluating the effect of re-expressing WT Huwe1 or Huwe1^C4341A (as in F) on Mfn2HA degradation. Data represent mean ± standard deviation (SD) for three experiments. (H) Representative experiment from (G).

(A) Quantification of ³⁵S pulse-chase experiments evaluating effect of doxorubicin (DR) on Mfn2HA degradation after JNK knockdown in U2OS using siRNAs from . Data are mean ± standard deviation (SD) for three experiments. (B) Representative experiment from (A). (C) IPs from U2OS expressing Mfn2HA or Ser mutants were immunoblotted as indicated. (D–F) Degradation of Mfn2 and mutants assessed by ³⁵S pulse-chase metabolic labeling in transiently transfected U2OS cells without treatment (D), with doxorubicin during the chase (E), or with 10 μM lactacystin treatment for 30 min before and during the chase (F). Data represent mean ± SD for three experiments. For (E) and (F), at least one pulse-chase experiment with untreated cells was performed concurrently to validate comparison to (D). Empirical half-lives are shown below.

(A) Cells transfected with mito-dsRED and the indicated plasmids were treated with doxorubicin (DR) for 9 hr beginning 48 hr after transfection and mitochondrial morphology assessed (see Experimental Procedures). In the absence of DR in all groups <5% of cells were found to be fragmented. Two hundred cells were counted in each group in triplicate. Data represent mean ± standard deviation (SD). (B) Representative images from (A). (C and D) Cells transfected as indicated were treated with DR for 20 hr or 2 μM thapsigargin for 18 hr and cell death was monitored by Annexin V staining using fluorescence microscopy. Data represent mean ± standard deviation (SD) for three independent experiments. (E) U2OS cells were transfected with Mfn2 siRNAs targeting 3′UTR of Mfn2 and with WT or S27 mutants of Mfn2HA. Caspase cleavage was assessed by immunoblotting after DR treatment.

(A)Representative images of mitochondrial morphology in U2OS cells transfected with mito-dsRED after 9 hr treatment with vehicle or 10 μM doxorubicin (DR). (B) Protein levels with DR treatment. (C) Mfn2HA turnover was analyzed by ³⁵S pulse-chase metabolic labeling in transiently transfected U2OS cells treated with DMSO or DR. Data represent mean ± standard deviation (SD) for three independent experiments relative to initial amount at the beginning of the chase. (D) Representative experiment from data quantified in (C). (E) Ubiquitination (Ub) of endogenous Mfn2 from U2OS cells was determined by immunoprecipitation with anti-Mfn2 or control (CTL) antisera and immunoblot for ubiquitin after 1 hr treatment with doxorubicin in the absence of proteasome inhibitor. (F, G) U2OS cells in which Mfn2 was knocked down by siRNA (si_Mfn2) were compared to cells transfected with control siRNA (si_CTL) and assessed for caspase activation/cleavage in response to doxorubicin by either immunoblotting (F) or Caspase Glo assay (G); β-actin served as a loading control in (F), data represent mean ± standard deviation (SD) for three independent experiments (G).