Different Oligomeric Properties and Stability of Highly Homologous A1 and Proto-Oncogenic A2 Variants of Mammalian Translation Elongation Factor eEF1

Face-valid phenotypes in a mouse model of the most common mutation in EEF1A2-related neurodevelopmental disorder

De novo heterozygous missense mutations in EEF1A2, encoding neuromuscular translation-elongation factor eEF1A2, are associated with developmental and epileptic encephalopathies. We used CRISPR/Cas9 to recapitulate the most common mutation, E122K, in mice. Although E122K heterozygotes were not observed to have convulsive seizures, they exhibited frequent electrographic seizures and EEG abnormalities, transient early motor deficits and growth defects. Both E122K homozygotes and Eef1a2-null mice developed progressive motor abnormalities, with E122K homozygotes reaching humane endpoints by P31. The null phenotype is driven by progressive spinal neurodegeneration; however, no signs of neurodegeneration were observed in E122K homozygotes. The E122K protein was relatively stable in neurons yet highly unstable in skeletal myocytes, suggesting that the E122K/E122K phenotype is instead driven by loss of function in muscle. Nevertheless, motor abnormalities emerged far earlier in E122K homozygotes than in nulls, suggesting a toxic gain of function and/or a possible dominant-negative effect. This mouse model represents the first animal model of an EEF1A2 missense mutation with face-valid phenotypes and has provided mechanistic insights needed to inform rational treatment design.

The eEF1A protein in cancer: Clinical significance, oncogenic mechanisms, and targeted therapeutic strategies

Eukaryotic elongation factor 1A (eEF1A) is among the most abundant proteins in eukaryotic cells. Evolutionarily conserved across species, eEF1A is in charge of translation elongation for protein biosynthesis as well as a plethora of non-translational moonlighting functions for cellular homeostasis. In malignant cells, however, eEF1A becomes a pleiotropic driver of cancer progression via a broad diversity of pathways, which are not limited to hyperactive translational output. In the past decades, mounting studies have demonstrated the causal link between eEF1A and carcinogenesis, gaining deeper insights into its multifaceted mechanisms and corroborating its value as a prognostic marker in various cancers. On the other hand, an increasing number of natural and synthetic compounds were discovered as anticancer eEF1A-targeting inhibitors. Among them, plitidepsin was approved for the treatment of multiple myeloma whereas metarrestin was currently under clinical development. Despite significant achievements in these two interrelated fields, hitherto there lacks a systematic examination of the eEF1A protein in the context of cancer research. Therefore, the present work aims to delineate its clinical implications, molecular oncogenic mechanisms, and targeted therapeutic strategies as reflected in the ever expanding body of literature, so as to deepen mechanistic understanding of eEF1A-involved tumorigenesis and inspire the development of eEF1A-targeted chemotherapeutics and biologics.

Understanding functions of eEF1 translation elongation factors beyond translation. A proteomic approach

Mammalian translation elongation factors eEF1A1 and eEF1A2 are 92% homologous isoforms whose mutually exclusive tissue-specific expression is regulated during development. The isoforms have similar translation functionality, but show differences in spatial organization and participation in various processes, such as oncogenesis and virus reproduction. The differences may be due to their ability to interact with isoform-specific partner proteins. We used the identified sets of eEF1A1 or eEF1A2 partner proteins to identify cell complexes and/or processes specific to one particular isoform. As a result, we found isoform-specific interactions reflecting the involvement of different eEF1A isoforms in different cellular processes, including actin-related, chromatin-remodeling, ribonuclease H2, adenylyl cyclase, and Cul3-RING ubiquitin ligase complexes as well as initiation of mitochondrial transcription. An essential by-product of our analysis is the elucidation of a number of cellular processes beyond protein biosynthesis, where both isoforms appear to participate such as large ribosomal subunit biogenesis, mRNA splicing, DNA mismatch repair, 26S proteasome activity, P-body and exosomes formation, protein targeting to the membrane. This information suggests that a relatively high content of eEF1A in the cell may be necessary not only to maintain efficient translation, but also to ensure its participation in various cellular processes, where some roles of eEF1A have not yet been described. We believe that the data presented here will be useful for deciphering new auxiliary functions of eEF1A and its isoforms, and provide a new look at the known non-canonical functions of this main component of the human translation-elongation machinery.

The eEF1 family of mammalian translation elongation factors

The eEF1 family of mammalian translation elongation factors is comprised of the two variants of eEF1A (eEF1A1 and eEF1A2), and the eEF1B complex. The latter consists of eEF1Bα, eEF1Bβ, and eEF1Bγ subunits. The two eEF1A variants have similar translation activity but may differ with respect to their secondary, "moonlighting" functions. This variability is underlined by the difference in the spatial organization of eEF1A1 and eEF1A2, and also possibly by the differences in their post-translational modifications. Here, we review the data on the spatial organization and post-translation modifications of eEF1A1 and eEF1A2, and provide examples of their involvement in various processes in addition to translation. We also describe the structural models of eEF1B subunits, their organization in the subcomplexes, and the trimeric model of the entire eEF1B complex. We discuss the functional consequences of such an assembly into a complex as well as the involvement of individual subunits in non-translational processes.

Comparison of the ability of mammalian eEF1A1 and its oncogenic variant eEF1A2 to interact with actin and calmodulin

The question as to why a protein exerts oncogenic properties is answered mainly by well-established ideas that these proteins interfere with cellular signaling pathways. However, the knowledge about structural and functional peculiarities of the oncoproteins causing these effects is far from comprehensive. The 97.5% homologous tissue-specific A1 and A2 isoforms of mammalian translation elongation factor eEF1A represent an interesting model to study a difference between protein variants of a family that differ in oncogenic potential. We propose that the different oncogenic impact of A1 and A2 might be explained by differences in their ability to communicate with their respective cellular partners. Here we probed this hypothesis by studying the interaction of eEF1A with two known partners - calmodulin and actin. Indeed, an inability of the A2 isoform to interact with calmodulin is shown, while calmodulin is capable of binding A1 and interferes with its tRNA-binding and actin-bundling activities in vitro. Both A1 and A2 variants revealed actin-bundling activity; however, the form of bundles formed in the presence of A1 or A2 was distinctly different. Thus, a potential inability of A2 to be controlled by Ca2+-mediated regulatory systems is revealed.

A non-catalytic N-terminal domain negatively influences the nucleotide exchange activity of translation elongation factor 1Bα

Eukaryotic translation elongation factor 1Bα (eEF1Bα) is a functional homolog of the bacterial factor EF-Ts, and is a component of the macromolecular eEF1B complex. eEF1Bα functions as a catalyst of guanine nucleotide exchange on translation elongation factor 1A (eEF1A). The C-terminal domain of eEF1Bα is necessary and sufficient for its catalytic activity, whereas the N-terminal domain interacts with eukaryotic translation elongation factor 1Bγ (eEF1Bγ) to form a tight complex. However, eEF1Bγ has been shown to enhance the catalytic activity of eEF1Bα attributed to the C-terminal domain of eEF1Bα. This suggests that the N-terminal domain of eEF1Bα may in some way influence the guanine nucleotide exchange process. We have shown that full-length recombinant eEF1Bα and its truncated forms are non-globular proteins with elongated shapes. Truncation of the N-terminal domain of eEF1Bα, which is dispensable for catalytic activity, resulted in acceleration of the rate of guanine nucleotide exchange on eEF1A compared to full-length eEF1Bα. A similar effect on the catalytic activity of eEF1Bα was observed after its interaction with eEF1Bγ. We suggest that the non-catalytic N-terminal domain of eEF1Bα may interfere with eEF1A binding to the C-terminal catalytic domain, resulting in a decrease in the overall rate of the guanine nucleotide exchange reaction. Formation of a tight complex between the eEF1Bγ and eEF1Bα N-terminal domains abolishes this inhibitory effect.

Human eukaryotic elongation factor 1A forms oligomers through specific cysteine residues

Eukaryotic elongation factor 1A (eEF1A) is a multifunctional protein involved in bundling actin, severing microtubule, activating the phosphoinositol-4 kinase, and recruiting aminoacyl-tRNAs to ribosomes during protein biosynthesis. Although evidence has shown the presence of the isoform eEF1A1 oligomers, the substantial mechanism of the self-association remains unclear. Herein, we found that human eEF1A1 could spontaneously form oligomers. Specifically, mutagenesis screen on cysteine residues demonstrated that Cys(234) was essential for eEF1A1 oligomerization. In addition, we also found that hydrogen peroxide treatment could induce the formation of eEF1A oligomers in cells. By cysteine replacement, eEF1A2 isoform displayed the ability to oligomerize in cells under the oxidative environment. In summary, in this study we characterized eEF1A1 oligomerization and demonstrated that specific cysteine residues are required for this oligomerization activity.

Characterisation of translation elongation factor eEF1B subunit expression in mammalian cells and tissues and co-localisation with eEF1A2

Translation elongation is the stage of protein synthesis in which the translation factor eEF1A plays a pivotal role that is dependent on GTP exchange. In vertebrates, eEF1A can exist as two separately encoded tissue-specific isoforms, eEF1A1, which is almost ubiquitously expressed, and eEF1A2, which is confined to neurons and muscle. The GTP exchange factor for eEF1A1 is a complex called eEF1B made up of subunits eEF1Bα, eEF1Bδ and eEF1Bγ. Previous studies have cast doubt on the ability of eEF1B to interact with eEF1A2, suggesting that this isoform might use a different GTP exchange factor. We show that eEF1B subunits are all widely expressed to varying degrees in different cell lines and tissues, and at different stages of development. We show that ablation of any of the subunits in human cell lines has a small but significant impact on cell viability and cycling. Finally, we show that both eEF1A1 and eEF1A2 colocalise with all eEF1B subunits, in such close proximity that they are highly likely to be in a complex.

Highly homologous eEF1A1 and eEF1A2 exhibit differential post-translational modification with significant enrichment around localised sites of sequence variation

REVIEWERS

This article was reviewed by Frank Eisenhaber and Ramanathan Sowdhamini.Translation elongation factors eEF1A1 and eEF1A2 are 92% identical but exhibit non-overlapping expression patterns. While the two proteins are predicted to have similar tertiary structures, it is notable that the minor variations between their sequences are highly localised within their modelled structures. We used recently available high-throughput "omics" data to assess the spatial location of post-translational modifications and discovered that they are highly enriched on those surface regions of the protein that correspond to the clusters of sequence variation. This observation suggests how these two isoforms could be differentially regulated allowing them to perform distinct functions.