Nitric oxide increases lysine 48-linked ubiquitination following arterial injury

Abstract

Background: Proteins are targeted for degradation by the addition of a polyubiquitin chain. Chains of ubiquitin linked via lysine 48 (K48) are associated with protein degradation while chains linked via lysine 63 (K63) are associated with intracellular signaling. We have previously shown that nitric oxide (NO) inhibits neointimal hyperplasia in association with increasing the ubiquitination and degradation of UbcH10. The aim of this study is to characterize the effect of arterial injury and NO on K48- or K63-linked ubiquitination of cellular proteins.

Methods: The rat carotid artery balloon injury model was performed. Treatment groups included uninjured, injury alone, injury + proline NONOate (PROLI/NO), and PROLI/NO alone. Arteries were harvested at designated time points and sectioned for immunohistochemical analysis of K48- and K63-linked ubiquitination or homogenized for protein analysis. Vascular smooth muscle cells (VSMC) harvested from rat aortae were exposed to the NO donor diethylenetriamine NONOate (DETA/NO). Protein expression was determined by Western blot analysis, or immunoprecipitation and Western blot analysis.

Results: Arterial injury increased K48-linked ubiquitination in vivo. The addition of PROLI/NO following injury caused a further increase in K48-linked ubiquitination at 1 and 3 d, however, levels returned to that of injury alone by 2 wk. Interestingly, treatment with PROLI/NO alone increased K48-linked ubiquitination in vivo to levels similar to injury alone. There were lesser or opposite changes in K63-linked ubiquitination in all three treatment groups. DETA/NO increased K48-linked ubiquitination in VSMC in vitro but had minimal effects on K63-linked ubiquitination. Low doses of DETA/NO decreased K48-linked ubiquitination of cyclin A and B, while high doses of DETA/NO increased K48-linked ubiquitination of cyclin A and B. Minimal changes were seen in K63-linked ubiquitination of cyclin A and B in vitro.

Conclusions: Arterial injury and NO increased K48-linked ubiquitination in vivo and in vitro. Remarkably, minimal changes were seen in K63-linked ubiquitination. These novel findings provide important insights into the vascular biology of arterial injury and suggest that one mechanism by which NO may prevent neointimal hyperplasia is through regulation of protein ubiquitination.

Figure 1

A) Carotid arterial injury and nitric oxide (NO) treatment both increased K48-, but not K63-linked ubiquitination in vivo. Following balloon injury and treatment with or without the NO donor proline NONOate (PROLI/NO, 20 mg, t = 2, 24, or 72 hours), rat carotid arteries were homogenized and subjected to Western blot analysis using antibodies to K48- or K63-linked ubiquitin. Injury increased K48-linked ubiquitination at 24 and 72 hours. Treatment with NO following injury further increased K48-linked ubiquitination at all time points. In contrast, decreases in K63-linked ubiquitination following injury or NO treatment are seen. B) Graphical representation of the densitometry analysis normalized to the control group. n=7–8/treatment group.

Figure 2

Carotid artery injury and nitric oxide (NO) treatment increased K48-, but not K63-linked ubiquitination in vivo. Following balloon arterial injury and treatment with or without PROLI/NO(20 mg, t = 14 days), or treatment with PROLI/NO alone, rat carotid artery cross sections underwent immunohistochemistry staining 2 weeks after injury for K48- and K63-linked ubiquitin (brown). K48-linked ubiquitination increased following arterial injury, injury + NO treatment, or NO alone. Minimal changes were seen following injury or NO treatment in K63-linked ubiquitination. n=3/treatment group. Negative control reflects staining with saline substituted for the primary antibody.

Figure 3

Carotid injury and nitric oxide (NO) treatment increased K48-linked ubiquitination in vivo. Blinded grading of neointimal and adventitial immunohistochemical staining for K48-linked ubiquitin chains shows that injury, injury + NO, and NO alone all significantly increase K48-linked ubiquitin staining relative to control arteries at 2 weeks (* P < 0.001 vs. uninjured group; **P < 0.007 vs. injury alone group). n=3/treatment group.

Figure 4

A) Nitric oxide (NO) increased K48-, but not K63-linked ubiquitination in vitro. Vascular smooth muscle cells (VSMC) were treated with the NO donor diethylenetriamine NONOate (DETA/NO, 0.5–1 mM), then subjected to Western blot analysis using antibodies to K48- or K63-linked ubiquitin. NO caused an increase in K48-linked ubiquitination at both concentrations. Lesser or opposite changes were seen in K63-linked ubiquitination. B) Graphical representation of the densitometry analysis normalized to the control group. t = 24 hours. Images are representative of three separate experiments.

Figure 5

A) Nitric oxide (NO) has a bimodal effect on K48-, but not K63-linked ubiquitination of cyclins A and B. VSMC were treated with DETA/NO (0.5–1 mM). Immunoprecipitation (IP) was performed using antibodies to K48- or K63-linked ubiquitin. Immunoprecipitated proteins were then subjected to Western blot (WB) analysis using antibodies to cyclin A or B. While 0.5 mM DETA/NO decreased K48-linked ubiquitination, 1 mM DETA/NO increased K48-linked ubiquitination. 1 mM DETA/NO decreased K63-linked ubiquitination. B) Graphical representation of the densitometry analysis normalized to the control group. Images are representative of two separate experiments.