p66Shc mediates SUMO2-induced ROS production and inflammation in endothelial cells.

(A) Immunoblot for Shc and SUMO2 expression in HUVECs following p66Shc knockdown (si-p66Shc) and SUMO2 overexpression (Ad-HA-SUMO2). (B and C) Quantification of ROS (MitoSOX fluorescence) in HUVECs from A. *P<0.05; n=24-32; student’s t-test. Data represents mean ±SEM. Immunoblots are representative of at least three independent experiments. GC, scrambled siRNA used as control; HUVECs, human umbilical vein endothelial cells.

SUMO2 directly modifies p66Shc.

(A) Immunoblot for sumoylated p66Shc in HEK-293 cells overexpressing SUMO2 (HA-SUMO2) and p66Shc (Flag-p66Shc). (B) Immunoblot for p66Shc sumoylation in HEK-293 cells overexpressing SUMO2 (HA-SUMO2) and p66Shc (Flag-p66Shc) with or without overexpression of Ubc9 (HA-Ubc9) and SENP1 (Flag-SENP1). (C) Immunoblot for sumoylation of p66Shc in HUVECs overexpressing SUMO2 (Ad-HA-SUMO2) or p66Shc (Ad-Flag-p66Shc). (D) Immunoblot for p66Shc sumoylation in HUVEC cells overexpressing SUMO2 (Ad-HA-SUMO2) or p66Shc (Ad-Flag-p66Shc) in the presence or absence of anacardic acid. Quantification of SUMO2-p66Shc from immunoblot of D. One-way ANOVA, followed by Tukey’s post-hoc analysis, n=4 *P<0.05, **P<0.01. (F) Immunoblot for sumoylated recombinant p66Shc in the presence of absence of Ubc9. Immunoblots are representative of at least three independent experiments. Data represents mean ±SEM. Ad-Lac Z was used as adenoviral control to compensate for the difference in MOI among the groups.

SUMO2 modifies p66Shc at lysine-81.

(A) Modular structure of p66Shc showing conserved lysine-81 in the CH2 domain. (B) Tandem mass spectrometry of in vitro SUMO2ylated recombinant p66Shc. (C) Immunoblot showing sumoylation of p66Shc in HEK-293 cells overexpressing p66ShcWT (Flag-p66ShcWT) or p66ShcK81R (Flag-p66ShcK81R) with and without SUMO2 (HA-SUMO2). (D) Immunoblot showing sumoylation of p66Shc in HUVECs overexpressing p66ShcWT (Ad-Flag-p66ShcWT) or p66ShcK81R (Ad-p66ShcK81R) with and without SUMO2 (Ad-HA-SUMO2). (E) Immunoblot for sumoylation of recombinant p66ShcWT or p66ShcK81R. Immunoblots are representative of at least three independent experiments. CH2, collagen homology 2; PTB, phosphotyrosine binding; CH1, Collagen homology 1; SH2, Src homology 2.

SUMO2 promotes p66ShcS36 phosphorylation by sumoylation of K81.

(A) Immunoblot for p66ShcS36 phosphorylation in HEK-293 cells overexpressing SUMO2 (HA-SUMO2) and p66ShcWT (Flag-p66ShcWT). (B) Quantification of p66ShcS36 phosphorylation from A. *P<0.05; n=6; one-way ANOVA. (C) Immunoblot for p66ShcS36 phosphorylation in HUVECs overexpressing SUMO2 (Ad-HA-SUMO2) or p66ShcWT (Ad-Flag-p66ShcWT). (D) Quantification of p66ShcS36 phosphorylation from C. *P<0.05; n=8; one-way ANOVA. (E) Immunoblot for p66ShcS36 phosphorylation in HUVECs expressing SUMO2 (Ad-HA-SUMO2) and p66ShcWT (Ad-Flag-p66ShcWT) or p66ShcK81R (Ad-Flag-p66ShcK81R). (F) Quantification of p66ShcS36 phosphorylation from E. *P<0.05, **P<0.01; n=11; two-way ANOVA. Immunoblots were quantified with iBright image analysis software. Data represents mean ±SEM. Immunoblots are representative of at least three independent experiments.

p66ShcK81 SUMO2ylation of K81 in p66Shc promotes its translocation to the mitochondria.

(A) Immunoblot for p66ShcWT in whole cell lysates or the mitochondrial fraction of HUVECs ectopically expressing SUMO2 (Ad-HA-SUMO2) and p66ShcWT (Ad-Flag-p66ShcWT). (B) Quantification of the relative abundance of p66Shc in the mitochondria from A. **P<0.01, ****P<0.0001; n=5; one-way ANOVA. (C) Immunoblot for p66Shc in whole cell lysates or the mitochondrial fraction of HUVECs ectopically expressing SUMO2 (Ad-HA-SUMO2) and p66ShcWT (Ad-Flag-p66ShcWT or p66ShcK81R (Ad-Flag-p66ShcK81R). (D) Quantification of relative abundance of p66Shc in the mitochondria from C. *P<0.05, **P<0.01; n=7; two-way ANOVA. Data represents mean ±SEM. Immunoblots are representative of at least three independent experiments.

p66ShcK81R knockin (KI) are resistant to SUMO2-induced endothelial dysfunction.

(A) DNA sequencing data from a p66ShcK81R knockin (KI) mouse showing a point mutation (A>G) resulting in an amino acid change (KèR), and insertion of silent mutation to create a digestion site for ApaI endonuclease for genotyping. (B) DNA gel electrophoresis of a PCR-amplified and ApaI-digested amplicon identifying the p66ShcK81R KI allele. (C and D) Representative immunoblots showing SUMO2 expression in aortic rings from wild type and p66ShcK81R KI mice (1.67X108 pfu/ring). (E and F) Graphs showing % contraction following induction of endothelium-dependent relaxation in aortic rings from (E) wild-type mice, n=14-16 from 4 mice) and (F) p66ShcK81R KI mice (n=15-16 rings from 4 mice) expressing a control vector (Ad-Lac Z) or overexpressing SUMO2 (Ad-HA-SUMO2). (G and H) Graphs showing % contraction following induction of endothelium-independent relaxation in aortic rings from (G) wild type (n=14-16 from 4 mice) and (H) p66ShcK81R KI mice (n=14-15 from 4 mice) expressing a control vector (Ad-Lac Z) or overexpressing SUMO2 (Ad-HA-SUMO2). Curves were compared for nonlinear fit. ****P<0.0001. Ach-acetylcholine, SNP-sodium nitroprusside, PE-Phenylephrine, Homo-homozygous, Het-heterozygous, and NTC-No template control. Data represents mean ±SEM.

p66ShcK81R knockin (KI) prevents hyperlipidemia-induced endothelial dysfunction and oxidative stress.

(A) Immunoblot showing oxidized-LDL (o-LDL)-induced changes in SUMO2/3-conjugated proteins in HUVECs with and without SUMO2. (B) Graph showing serum cholesterol-level in wild type, LDLr-/- mice on high fat diet, and p66ShcK81R KI mice on high fat diet (n=4/group). One-way ANOVA, ****P<0.0001. Graphs showing % contraction following induction of endothelium-dependent relaxation (C) and endothelium-independent relaxation (D) in aortic rings from male wild-type mice, (n=14 rings from 5 mice), LDLr-/- mice on high fat diet (n=18 rings from 6 mice) and LDLr-/- Xp66ShcK81R KI mice on high fat diet (n=12 rings from 6 mice). two-way ANOVA. ***P<0.001. (E) Representative images showing the level of 8-OHdG in aortic section of LDLr-/- mice on high fat diet (n=6 mice) and LDLr-/- Xp66ShcK81R KI mice on high fat diet (n=6 mice) and their (F) quantification. One-way ANOVA. *P<0.05. HFD-high fat diet Ach-acetylcholine, SNP-sodium nitroprusside, PE-Phenylephrine, 8-OHdG-8 hydroxy deoxyguanosine. Data represents mean ±SEM.

p66ShcK81 sumoylation affects multiple key regulatory pathways in endothelial cells.

(A) Principal component analysis (PCA) of proteins expressed in endothelial cells expressing SUMO2+p66ShcWT and SUMO2+p66ShcK81R. (B) Bubble chart presentation of key regulatory pathways differentially affected by SUMO2-p66ShcWT and SUMO2-p66ShcK81R in endothelial cells. The protein expression data (fold change) was obtained from HUVECs expressing SUMO2 with p66ShcWT or p66ShcK81R compared to those expressing Ad-Lac Z. Ingenuity Pathway Analysis (IPA) was performed on proteins differentially expressed between SUMO2-p66ShcWT and SUMO2-p66ShcK81R. Data are derived from a sample size of n=6.