NMT1 and NMT2 are lysine myristoyltransferases regulating the ARF6 GTPase cycle

Tatsiana Kosciuk,Ian R Price,Garrison P Komaniecki,Hening Lin,J Christopher Fromme,Kayla N Johnson,Xiaoyu Zhang,Caroline J Dehart,Paul M Thomas,Shuai Zhang,Steve L Halaby,Min Yang,Chengliang Zhu,Neil L Kelleher

doi:10.1038/s41467-020-14893-x

Tatsiana Kosciuk, Ian R Price + Show 12 more

Open Access

https://doi.org/10.1038/s41467-020-14893-x

Copy DOI

Abstract

Lysine fatty acylation in mammalian cells was discovered nearly three decades ago, yet the enzymes catalyzing it remain unknown. Unexpectedly, we find that human N-terminal glycine myristoyltransferases (NMT) 1 and 2 can efficiently myristoylate specific lysine residues. They modify ADP-ribosylation factor 6 (ARF6) on lysine 3 allowing it to remain on membranes during the GTPase cycle. We demonstrate that the NAD+-dependent deacylase SIRT2 removes the myristoyl group, and our evidence suggests that NMT prefers the GTP-bound while SIRT2 prefers the GDP-bound ARF6. This allows the lysine myrisotylation-demyristoylation cycle to couple to and promote the GTPase cycle of ARF6. Our study provides an explanation for the puzzling dissimilarity of ARF6 to other ARFs and suggests the existence of other substrates regulated by this previously unknown function of NMT. Furthermore, we identified a NMT/SIRT2-ARF6 regulatory axis, which may offer new ways to treat human diseases.

Highlights

Lysine fatty acylation in mammalian cells was discovered nearly three decades ago, yet the enzymes catalyzing it remain unknown
It has been shown that the selectivity for glycine is due to the ability of the α-amine of the N-terminal glycine to rotate and attack the carbonyl carbon of myristoyl-CoA in the active site of N-terminal glycine myristoyltransferases (NMT) without the steric hindrance that would be experienced by other residues[26]
Our work uncovered an ADP-ribosylation factor 6 (ARF6) lysine myristoylation–demyristoylation cycle, which is intimately coupled to its GTPase cycle and controlled by SIRT2, and a previously unknown activity of NMT

Summary

Introduction

Lysine fatty acylation in mammalian cells was discovered nearly three decades ago, yet the enzymes catalyzing it remain unknown. We find that human N-terminal glycine myristoyltransferases (NMT) 1 and 2 can efficiently myristoylate specific lysine residues They modify ADP-ribosylation factor 6 (ARF6) on lysine 3 allowing it to remain on membranes during the GTPase cycle. We demonstrate that the NAD+-dependent deacylase SIRT2 removes the myristoyl group, and our evidence suggests that NMT prefers the GTPbound while SIRT2 prefers the GDP-bound ARF6 This allows the lysine myrisotylationdemyristoylation cycle to couple to and promote the GTPase cycle of ARF6. ARF6, localizes to the plasma membrane and the endocytic system and, unlike other ARFs, tends to remain membrane bound even in the inactive state[21,22] This has been a puzzle as ARF6 has a high structural similarity to ARF1 and follows the same nucleotide-dependent dynamics of the amphipathic helix[23,24]. We found that human NMT1 and NMT2 can catalyze lysine myristoylation of ARF6 providing an explanation for its unusual membrane association

Methods

Results

Conclusion