An Importin Code in neuronal transport from synapse-to-nucleus?

Abstract

The principle cells of the brain- neurons- express more genes than any other cell type. As many rudimentary processes of the brain, such as long-lasting changes in synaptic efficacy, involve alterations to the expression of genes, the characterisation of proteins and processes that influence transcription is a high priority in neuroscience. An emerging system capable of transducing signals that likely influence transcription is synapse-to-nucleus macromolecular protein shuttling, in which synaptic proteins are relayed to the nucleus in a complex with importins (also referred to by their gene name: Karyopherins/KPNs) and molecular motors such as dynein (Jordan and Kreutz, 2009, Figure ​Figure1C).1C). Despite a central position in a neuronal process capable of altering transcription, there is a dearth of research into importin function in neurites. In this opinion paper we summarize neuronal importin understanding to date, present novel ideas relating to their neuronal function and highlight the reductionist nature of the classical description of nuclear import with the specific example that complex composition is likely much more intricate than is being considered. Figure 1 Importin structures, canonical and non-caconocal nuclear import and The Importin Code. (A) The general domain structure of importin-α and importin-β (1) importin-α is composed of an N-terminus importin-β binding domain ... The classical function of importins Whilst proteins smaller than 40 kDa are free to diffuse in and out of the nucleus via the nuclear pore complex (NPC), proteins larger than this require importins. Two related families of importins have been described: Importin-α and importin-β. In humans, there are 7 importin-α isoforms (1–7) and 19 importin-β isoforms (Yasuhara et al., 2009; Xu et al., 2010; Zienkiewicz et al., 2013, Figure ​Figure1A).1A). Importin-α subtypes comprise of an importin-β binding (IBB) domain followed by 10 armadillo (ARM) repeats that organize into 3 alpha helices (Conti et al., 1998; Kelley et al., 2010). Importin-β subtypes are composed of 19–20 HEAT repeats that arrange into a super helicoidal molecule (Xu et al., 2010). The classical pathway best describes importin understanding in nuclear import (Figure 1B1). This starts with the formation of a trimeric complex outside of the nucleus consisting of a cargo protein bound to an importin-α with its nuclear localisation signal (NLS) and the importin-α isoform bound to an importin-β with its N-terminal IBB domain (Goldfarb et al., 2004). Once formed in the cytoplasm, the trimeric complex translocates to the nucleus, either passively via diffusion or actively with the retrograde molecular motor dynein (Goldfarb et al., 2004). At the nuclear envelope, importin-β mediates the passage of the NPC, probably due to mediation with NPC-proteins FG-nucleoporins (Lott and Cingolani, 2011) and following entry into the nucleus the 100-fold higher concentration of Ran-GTP binds to importin-β, liberating the cargo for the second leg of their nuclear mission and the importins for translocation back to the cytoplasm (Conti and Izaurralde, 2001; Goldfarb et al., 2004; Stewart, 2007; Mason and Goldfarb, 2009; Ch'ng and Martin, 2011, Figure ​Figure1C).1C). There are several exceptions to the classical description (to be discussed) but the involvement of importins in the nuclear import of cargo proteins is clear. It is also known that each importin isoform has a unique catalog of interacting partners and distinct subcellular expression (Nadler et al., 1997; Kelley et al., 2010; Schaller et al., 2014). As a result of these findings, a parsimonious hypothesis emerged: Importins function as adaptor proteins, linking distinct cargo proteins to nuclear import complexes. Therefore, by changing the expression of importins contributing to nuclear import—which is induced by activity—the nuclear proteome is shuffled, with likely alterations in transcriptional output and cellular phenotypes. As long-term forms of synaptic plasticity require such alterations in transcriptional output from the nucleus, the possibility that importins are central in this neuronal process is obvious (Thompson et al., 2004; West and Greenberg, 2011).