FERM domain-containing proteins are active components of the cell nucleus

Abstract

The FERM domain is a conserved and widespread protein module that appeared in the common ancestor of amoebae, fungi, and animals, and is therefore now found in a wide variety of species. The primary function of the FERM domain is localizing to the plasma membrane through binding lipids and proteins of the membrane; thus, for a long time, FERM domain-containing proteins (FDCPs) were considered exclusively cytoskeletal. Although their role in the cytoplasm has been extensively studied, the recent discovery of the presence and importance of cytoskeletal proteins in the nucleus suggests that FDCPs might also play an important role in nuclear function. In this review, we collected data on their nuclear localization, transport, and possible functions, which are still scattered throughout the literature, with special regard to the role of the FERM domain in these processes. With this, we would like to draw attention to the exciting, new dimension of the role of FDCPs, their nuclear activity, which could be an interesting novel direction for future research.

Figure 1.. Conserved structure of the FERM domain.

(A) Crystal structures of the FERM domains of human FERM domain–containing proteins representing all FERM domain–containing protein groups demonstrate the high evolutionary conservation of the FERM domain. PDB accession numbers are in parentheses: 1NI2 (Smith et al, 2003), 1GG3 (Han et al, 2000), 3WA0 (Mori et al, 2014), 7BYJ (Wang et al, 2020), 4NY0 (Brami-Cherrier et al, 2014), 5MV8 (Yu et al, 2017), 5D68 (Zhang et al, 2015), and 6QLY (Martinelli et al, 2020). (B) Cartoon depiction of the general structure of the FERM domain. The main binding sites identified in ERM proteins are labeled. ➀—transient IP3/PIP2-binding site; ➁—stable IP3/PIP2-binding site; and (a–d)—protein–protein interaction sites. Binding partners are for example in the case of (a)—moesin + Crumbs (PDB accession number 4YL8 [Wei et al, 2015]); (b)—radixin + EBP50 (2D10 [Terawaki et al, 2006]); (c)—radixin + MT1-MMP (3X23 [Terawaki et al, 2015]); and (d)—Merlin + Lats1 (4ZRK [Li et al, 2015]).

Figure 2.. Domain structure of FERM domain–containing proteins.

Proteins are grouped according to the phylogenetic relations of the FERM domains (Ali & Khan, 2014). Only the most relevant domains are shown; size is not for scale. Structures are shown in the N- to C-terminal direction. Known NLS, NES, and cytoplasmic retention motifs are highlighted with red, dark blue, and yellow rectangles, respectively.

Figure 3.. Evolutionary history of FERM domain–containing proteins.

A simplified dendrogram below the species images represents evolutionary relationships and shows the origin of selected FERM domain–containing proteins. Major taxonomic groups are shown above the species images. (The former name of Amorphea was Unikonta.) The figure is based on the publication of Ali and Khan (2014).

Figure 4.. Evolutionary conservation of verified NLS and cytoplasmic retention motifs regulating the nuclear import of FERM domain proteins.

(A) Conservation of the NLS (highlighted with gray background) predicted in the FERM domain of ERMs. (B) Conservation of human Merlin NLS (gray background) and the N-terminal end of the FERM domain. (C) Conservation of the cytoplasmic retention (gray background) identified in the FERM domain of human Merlin extends to the protein 4.1 and ERM families. (D) Conservation of the NLS (gray background) of vertebrate protein 4.1R proteins. (E) AlphaFold structure prediction of full-length protein 4.1R based on the sequence of UniProt ID P11171. Under the protein sequences, asterisks (*) indicate positions, which have fully conserved residues (also indicated in red and bold letters); colon (:) indicates conservation between amino acids of strongly similar properties (marked in bold letters); subscript period (.) indicates conservation between residues of weakly similar properties; and no symbol indicates no conservation.