Translation from the Ribosome to the Clinic: Implication in Neurological Disorders and New Perspectives from Recent Advances

Abstract

De novo protein synthesis by the ribosome and its multitude of co-factors must occur in a tightly regulated manner to ensure that the correct proteins are produced accurately at the right time and, in some cases, also in the proper location. With novel techniques such as ribosome profiling and cryogenic electron microscopy, our understanding of this basic biological process is better than ever and continues to grow. Concurrently, increasing attention is focused on how translational regulation in the brain may be disrupted during the progression of various neurological disorders. In fact, translational dysregulation is now recognized as the de facto pathogenic cause for some disorders. Novel mechanisms including ribosome stalling, ribosome-associated quality control, and liquid-liquid phase separation are closely linked to translational regulation, and may thus be involved in the pathogenic process. The relationships between translational dysregulation and neurological disorders, as well as the ways through which we may be able to reverse those detrimental effects, will be examined in this review.

Keywords: mRNA translational regulation; neurological disorders; phase separation; ribosome stalling; ribosome-associated quality control; tRNA dynamics.

Figure 1

Regulation of Cap-dependent translation initiation. Translation begins with the recruitment of the preinitiation complex (40S ribosome, eIF1, eIF1A, eIF3, eIF5, and the eIF2-GTP-Met-tRNA_i complex) to the 5′cap of mRNAs by the eIF4F complex (eIF4A eIF4E, and eIF4G) and eIF4B. This preinitiation complex scans the mRNA for a start codon (AUG) in a 5′ to 3′ manner. Upon recognition of the start codon, eIF5B mediates the release of initiation factors eIF1, eIF2-GDP, and eIF5, allowing the 60S ribosome to join and form the elongation complex 80S ribosome. eIF2B facilitates the recycling of eIF2-GDP to eIF2-GTP, but is inhibited by the phosphorylated form of eIF2α. In many monogenic forms of syndromic autism, mTOR hyperactivation occurs as a result of disturbances in upstream signalling pathways, and in turn enhances translation initiation by direct or indirect phosphorylation of 4E-BP, eIF4B, and eIF4G. A pharmacologic inhibitor of the interaction between eIF4E and eIF4G (4EGI-1) has been shown to have therapeutic benefits for multiple ASD models that have increase translation translational dysregulation. Conversely, a small compound known as ISRIB (integrated stress response inhibitor) can nullify the inhibitory effects of phosphorylated eIF2α on eIF2B [20]. Although it was observed to enhance spatial and fear-associated learning in mice and rats, it remains to be seen whether it can also prevent the decline in cognitive functions in AD and other neurodegenerative diseases.

Figure 2

ER stress, unfolded protein response, and the integrated stress response. Multiple pathways, collectively known as the unfolded protein response, are activated by the detection of misfolded proteins in the ER. This ER stress can be sensed by ATF6, PERK, and IRE1, which act via distinct mechanisms to help alleviate the stress by increasing the protein folding capacity of the ER or decreasing the ER protein folding load. Whereas ATF6 and IRE1 mediate a direct and indirect (via mRNA splicing) transcriptional response, respectively; PERK acts to reduce global protein synthesis by enhancing eIF2α phosphorylation. While global translation is reduced by phosphorylated eIF2α, the translation of a small number of transcripts including ATF4 are preferentially induced, which in turn transcriptionally activate genes to promote survival under stress conditions or induce apoptosis. Together with PERK, GCN2, PKR, and HRI are three other kinases known to phosphorylate eIF2α at serine 51 in response to different types of stress, forming the integrated stress response. Aside from HRI, which is not highly expressed in the brain, the ISR kinases have been shown to be activated in various neurodegenerative diseases and may contribute to the pathology by chronically depressing global protein synthesis.

Figure 3

Ribosome-associated quality control. The RQC pathway is initiated after the sensing and splitting of stalled ribosomes by proteins including GTPBP2, HBS1L, Pelota, and ABCE1. Whereas the 40S-associated mRNA is degraded by exonuclease Xrn1 and the exosome complex, RQC mediates the ubiquitination, CATylation, and extraction of the nascent chain polypeptide for its eventual degradation by the proteasome. The RQC complex is mainly consisted of Ltn1 and NEMF, which recognizes aspects of a stalled 60S subunit, including a protruding tRNA and a surface which would otherwise be interacting with the 40S subunit. In this complex with the 60S ribosome, the RING domain of Ltn1 is perfectly situated such that it sits near the ribosome exit tunnel, thus allowing it to ubiquitinate the nascent chain polypeptide. Conversely, NEMF mediates the mRNA- and 40S-independent addition of alanine and threonine residues (CAT tails) to the emerging polypeptide. Finally, NEMF dissociates and is replaced by ANKZF1 to mediate the tRNA cleavage so that the nascent polypeptide chain can be extracted by p97/VCP and its cofactors. The loss of RQC activity due to genetic removal of its principal components have been shown to result in toxicity in yeast and an ENU-induced Ltn1 mutant was found to cause neurodegeneration in mice. Much remains to be examined to determine how disruptions in the RQC pathway may affect brain functions and whether it has a role in the pathogenic process of various neurological disorders.

Figure 4

Aberrant LLPS and neurodegeneration. LLPS is now known to play a critical role in a growing number of biological processes and regulates the dynamics of various membraneless organelles in the cell. However, it has been demonstrated recently that aberrant LLPS dynamics caused by either disease-associated mutations of RNA-binding proteins or RAN translation products involved in numerous neurodegenerative diseases contribute significantly to the neurotoxicity via several different mechanisms. LLPS formed by mutant proteins or RAN translation products can: (1) disrupt existing membraneless organelles including the nuclear pore complex, stress and RNA granules; (2) sequester RNAs and proteins which are not normally part of the LLPS assemblies or disturb their exchange dynamics such that their normal functions are disrupted (e.g., mRNA translation may be decreased by the reduced availability of certain transcripts or translation factors); and (3) LLPS assemblies can further undergo conformational changes and ultimately lead to the formation of fibrillar aggregates, further disturbing proteostasis in affected neurons.