Lysine acetylation of cytoskeletal proteins: Emergence of an actin code

Abstract

Reversible lysine acetylation of nuclear proteins such as histones is a long-established important regulatory mechanism for chromatin remodeling and transcription. In the cytoplasm, acetylation of a number of cytoskeletal proteins, including tubulin, cortactin, and the formin mDia2, regulates both cytoskeletal assembly and stability. More recently, acetylation of actin itself was revealed to regulate cytoplasmic actin polymerization through the formin INF2, with downstream effects on ER-to-mitochondrial calcium transfer, mitochondrial fission, and vesicle transport. This finding raises the possibility that actin acetylation, along with other post-translational modifications to actin, might constitute an "actin code," similar to the "histone code" or "tubulin code," controlling functional shifts to these central cellular proteins. Given the multiple roles of actin in nuclear functions, its modifications might also have important roles in gene expression.

Figure 1.

Lysine acetylation. Enzymatic lysine acetylation occurs through KATs, with the acetyl group donated by acetyl-CoA. KDACs remove the modification. Acetylation has three general effects: neutralizing lysine’s positive charge, blocking other modifications to the lysine, and creating a binding site for bromodomain-containing proteins. BD, bromodomain.

Figure 2.

Tubulin and actin acetylation. (A) Tubulin dimer of α- and β-tubulin. K40 of α-tubulin is the predominant acetylation site. Also shown are the C termini of α- and β-tubulin, which are subject to multiple PTMs. Molecular structure model from PDB accession no. 4U3J. (B) The microtubule, with C termini to the exterior and K40 in the lumen. (C) Acetylated microtubules engage in less inter-protofilament contacts, making them more flexible than nonacetylated microtubules, resulting in more resistance to mechanical compression because they bend instead of breaking, perhaps explaining why stable cellular microtubules are highly acetylated. (D) Ribbon model of actin monomer, with bound ATP (cyan). Zoom is the D-loop of subdomain 2, showing residues subject to PTMs: K50 and K61 (acetylation), M44 and M47 (oxidation), and Y53 (phosphorylation). Molecular structure model from PDB accession no. 4PKG. (E) Model of actin filament of four subunits (ADP-bound), showing the D-loop oriented to exterior. Molecular structure model from PDB accession no. 6DJN. (F) Cartoon of actin polymerization, showing unfavorable dimerization and trimerization steps, with subsequent elongation more favorable. Actin monomers add to the barbed end almost exclusively in cells.

Figure 3.

Acetylation-based regulation of actin dynamics through INF2 and cortactin. (A) INF2 regulation. Top: Bar diagram of INF2, with the FH1 and FH2 domains being involved in actin polymerization, and the DID and DAD domains being regulatory. Bottom: INF2 dimer inhibition by complex between CAP and Ac-actin, with CAP/Ac-actin serving as a bridge between DID and DAD domains. Deacetylation of actin by HDAC6 releases this bridge, allowing the FH2 domain to interact with actin. For simplicity, CAP/actin is shown as a dimer, although evidence suggests that CAP is hexameric. Also, shown here is interaction between CAP/Ac-actin and only one of the two INF2 subunits in the INF2 dimer, again for simplicity. (B) Cortactin regulation. Top: Bar diagram of cortactin, with the NTA interacting with Arp2/3 complex, the ABD consisting of 6.5 repeat regions and interacting with actin filaments, and a C-terminal SH3 domain that interacts with multiple proteins. Each repeat of the ABD is subject to acetylation. Bottom: Speculative model for acetylation-based cortactin regulation. Cortactin bound to both Arp2/3 complex and the mother filament at an Arp2/3 complex–mediated branch. Cortactin acetylation lowers the affinity of the ABD for actin filaments, which could enhance branch disassembly. NTA, N-terminal acidic region.

Figure 4.

Nuclear actin. Examples of the three general types of nuclear actin function. Top: Actin monomer as a tightly bound component of multiprotein complexes. Shown here is the INO80 chromatin remodeling complex. A complex of actin and two Arps (Arp4 and Arp8) forms a module that is connected to the main complex by the helical HSA domain (yellow), and serves to specify the position of INO80 on the nucleosome. Left: Actin monomer as a reversible regulator of nuclear proteins. Shown here is actin bound to MRTF, both in the cytoplasm and in the nucleus. In both compartments, actin monomer inhibits MRTF’s ability to activate transcription through SRF. Actin polymerization in both the cytosol and the nucleus releases MRTF for SRF activation. Here, the nuclear actin polymerization is formin-mediated. Right: Actin filaments as force-producing structures for nuclear dynamics. Shown here is myosin-mediated movement of a heterochromatin double-strand break (bound to MRN) along an Arp2/3 complex–polymerized actin filament to the nuclear periphery for repair. MRTF, myocardin-related transcription factor; NPC, nuclear pore complex; MRN, MRE11-Rad50-NBS1 complex.