Differential superoxide production in phosphorylated neuronal nitric oxide synthase mu and alpha variants

Abstract

Neuronal nitric oxide synthase (nNOS) is regulated by phosphorylation in vivo, yet the underlying biochemical mechanisms remain unclear, primarily due to difficulty in obtaining milligram quantities of phosphorylated nNOS protein; detailed spectroscopic and rapid kinetics investigations require purified protein samples at a concentration in the range of hundreds microM. Moreover, the functional diversity of the nNOS isoform is linked to its splice variants. Also of note is that determination of protein phosphorylation stoichiometry remains as a challenge. To address these issues, this study first expanded a recent genetic code expansion approach to produce phosphorylated rat nNOSμ and nNOSα holoproteins through site-specific incorporation of phosphoserine (pSer) at residues 1446 and 1412, respectively; this site is at the C-terminal tail region, a NOS-unique regulatory element. A quantitative mass spectrometric approach was then developed in-house to analyze unphosphorylated peptides in phosphatase-treated and -untreated phospho-nNOS proteins. The observed pSer-incorporation efficiency consistently exceeded 80%, showing high pSer-incorporation efficiency. Notably, EPR spin trapping results demonstrate that under l-arginine-depleted conditions, pSer1412 nNOSα presented a significant reduction in superoxide generation, whereas pSer1446 nNOSμ exhibited the opposite effect, compared to their unphosphorylated counterparts. This suggests that phosphorylation at the C-terminal tail has a regulatory effect on nNOS uncoupling that may differ between variant forms. Furthermore, the methodologies for incorporating pSer into large, complex protein and quantifying the percentage of phosphorylation in recombinant purified protein should be applicable to other protein systems.

Keywords: EPR spin trapping; Nitric oxide synthase; Phosphorylation; Quantitative mass spectrometry; Superoxide.