Loading AI tools
Protein-coding gene in the species Homo sapiens From Wikipedia, the free encyclopedia
Eukaryotic translation initiation factor 4E, also known as eIF4E, is a protein that in humans is encoded by the EIF4E gene.[5][6]
Most eukaryotic cellular mRNAs are blocked at their 5'-ends with the 7-methyl-guanosine five-prime cap structure, m7GpppX (where X is any nucleotide). This structure is involved in several cellular processes including enhanced translational efficiency, splicing, mRNA stability, and RNA nuclear export. eIF4E is a eukaryotic translation initiation factor involved in directing ribosomes to the cap structure of mRNAs as well as other steps in RNA metabolism that require cap-binding. It is a 24-kD polypeptide that exists as both a free form and as part of the eIF4F pre-initiation complex.[7] Many cellular mRNAs require eIF4E in order to be translated into protein. The eIF4E polypeptide is considered by some to be the rate-limiting component of the eukaryotic translation apparatus and is involved in the mRNA-ribosome binding step of eukaryotic protein synthesis.
The other subunits of eIF4F are a 47-kD polypeptide, termed eIF4A,[8] that possesses ATPase and RNA helicase activities, and a 220-kD scaffolding polypeptide, eIF4G.[9][10][11]
Some viruses cut eIF4G in such a way that the eIF4E binding site is removed and the virus is able to translate its proteins without eIF4E. Also some cellular proteins, the most notable being heat shock proteins, do not require eIF4E in order to be translated. Both viruses and cellular proteins achieve this through an internal ribosome entry site in the RNA or through other RNA translation mechanisms such as those going through eIF3d.[12][13]
eIF4E plays roles outside of translation and other cap-binding proteins can engage in cap-dependent translation in an eIF4E-independent manner including factors such as eIF3D, eIF3I, PARN, the nuclear cap-binding complex CBC.[14][15][12][16][17][13][18] Many of these appear to be dependent on both specific features of transcripts as well as cellular context.
eIF4E is found in the nucleus of many mammalian cell types as well as in other species including yeast, drosophila and humans.[19][20][21][22][23][24][25][26] eIF4E is found in nuclear bodies a subset of which colocalize with PML nuclear bodies, and eIF4E is additionally found diffusely in parts of the nucleoplasm in mammalian.[23][21][24][25][27][28][26] In the nucleus, eIF4E plays well defined roles in the export of selected RNAs which contributes to its oncogenic phenotypes.[29][21][24][25][27][28][30][31][32][33] This relies on the ability of eIF4E to bind the m7G cap of RNAs and the presence of the 50 nucleotide eIF4E sensitivity element (4ESE) in the 3’UTR of sensitive transcripts; although other elements may also play a role. This form of export relies on the CRM1/XPO1 pathway.[21][24][27][34][35][26] Nuclear eIF4E has been shown to play other roles in RNA processing including in m7G capping, alternative polyadenylation and splicing.[36][37][38]
Increased nuclear accumulation of eIF4E as well as increased eIF4E-dependent RNA export, m7G capping and splicing of selected transcripts is characteristic of high-eIF4E AML patient samples.[25][30][38][37] RNAs are selected based on USER codes, or cis-acting elements, within their RNAs for specific levels of RNA processing; thus not all transcripts are sensitive to all levels of regulation (including translation).[35][39][18] For its RNA export function, eIF4E directly binds to the leucine rich pentatricopeptide repeat protein (LRPPRC) which directly binds the dorsal surface of eIF4E and simultaneously to the 4ESE RNA thereby acting as a platform for assembly for the RNA export complex.[26][35] The current model is then LRPPRC binds to CRM1/XPO1 to engage the nuclear pore and traffic the 4ESE RNA to the cytoplasm.[28][26][35] In all, the nuclear functions of eIF4E can have potent impacts on the proteome allowing eIF4E to both re-write the message as well as to increase production of proteins based on increased accumulation in the cytoplasm due to increased export as well as to increased number of ribosomes per transcript in some cases. Its multiple roles in RNA processing require its association of RNAs through the m7G cap, and thus eIF4E can be considered a cap-chaperone protein.
Since eIF4E is an initiation factor that is relatively low in abundance, eIF4E can be controlled at multiple levels.[40][18] Regulation of eIF4E may be achieved at the levels of transcription, RNA stability phosphorylation, subcellular localization and partner proteins.[41]
a. Regulation of eIF4E by Gene Expression and RNA stability
The mechanisms responsible for eIF4E transcriptional regulation are not entirely understood. However, several reports suggest a correlation between myc levels and eIF4E mRNA levels during the cell cycle.[42] The basis of this relationship was further established by the characterization of two myc-binding sites (CACGTG E box repeats) in the promoter region of the eIF4E gene.[43] This sequence motif is shared with other in vivo targets for myc and mutations in the E box repeats of eIF4E inactivated the promoter region, thereby diminishing its expression.
Recent studies shown that eIF4E levels can be regulated at transcriptional level by NFkB and C/EBP.[44][45] Transduction of primary AML cells with IkB-SR resulted not only in reduction of eIF4E mRNA levels, but also re-localization of eIF4E protein.[25] eIF4E mRNA stability are also regulated by HuR and TIAR proteins.[46][47] eIF4E gene amplification has been observed in subset of head and neck and breast cancer specimens.[48]
b. Regulation of eIF4E by Phosphorylation
Stimuli such as hormones, growth factors, and mitogens that promote cell proliferation also enhance translation rates by phosphorylating eIF4E.[49] Although eIF4E phosphorylation and translation rates are not always correlated, consistent patterns of eIF4E phosphorylation are observed throughout the cell cycle; wherein low phosphorylation is seen during G0 and M phase and wherein high phosphorylation is seen during G1 and S phase.[50] This evidence is further supported by the crystal structure of eIF4E which suggests that phosphorylation on serine residue 209 may increase the affinity of eIF4E for capped mRNA.
eIF4E phosphorylation is also related to its ability to suppress RNA export and its oncogenic potential as first shown in cell lines.[51]
c. Regulation of eIF4E by Partner Proteins
Assembly of the eIF4F complex is inhibited by proteins known as eIF4E-binding proteins (4E-BPs), which are small heat-stable proteins that block cap-dependent translation.[41] Non-phosphorylated 4E-BPs interact strongly with eIF4E thereby preventing translation; whereas phosphorylated 4E-BPs bind weakly to eIF4E and thus do not interfere with the process of translation.[52] Furthermore, binding of the 4E-BPs inhibits phosphorylation of Ser209 on eIF4E.[53] Of note, 4E-BP1 is found in both the nucleus and the cytoplasm, indicating that it likely modulates nuclear eIF4Es functions of eIF4E as well.[54] A recent study showed that 4E-BP3 regulated eIF4E dependent mRNA nucleo-cytoplasmic export.[55] There are also many cytoplasmic regulators of eIF4E that bind to the same site as 4E-BP1.
Many other partner proteins has been found that can both stimulate or repress eIF4E activity, such as homeodomain containing proteins, including HoxA9, Hex/PRH, Hox 11, Bicoid, Emx-2 and Engrailed 2.[56][24][57][58][59] While HoxA9 promotes mRNA export and translation activities of eIF4E, Hex/PRH inhibits nuclear functions of eIF4E.[25][60][61] The RNA helicase DDX3 directly binds with eIF4E, modulates translation, and has potential functions in P-bodies and mRNA export.[62][26]
RING domains also bind eIF4E. The promyelocytic leukemia protein PML is a potent suppressor of both the nuclear RNA export and oncogenic activities of eIF4E whereby the RING domain of PML directly binds eIF4E on its dorsal surface suppressing eIF4E's oncogenic activity; and moreover a subset of PML and eIF4E nuclear bodies co-localize.[63][21][64][23][24][28] RNA-eIF4E complexes are never observed in PML bodies consistent with the observation that PML suppresses the m7G cap binding function of eIF4E.[21][64][28] Structural studies show that a related arenavirus RING finger protein, Lassa Fever Z protein, can similarly bind eIF4E on the dorsal surface.[64][65][66]
eIF4E nuclear entry is mediated by its direct interactions with Importin 8 where Importin 8 associates with the m7G cap-binding site of eIF4E.[32] Indeed, reduction in Importin 8 levels reduce the oncogenic potential of eIF4E overexpressing cells and its RNA export function. Importin 8 binds to the cap-binding site of eIF4E and is competed by excess m7G cap analogues as observed by NMR. eIF4E also stimulates the RNA export of Importin 8 RNA thereby producing more Importin 8 protein. There may be additional importins that play this role depending on cell type. Although an initial study suggested that the eIF4E transporter protein 4E-T (eIF4ENIF1) facilitated nuclear entry, later studies showed that this factor rather alters the localization of eIF4E to cytoplasmic processing bodies (P-bodies) and repress translation.[67]
Potyvirus viral protein genome linked (VPg) were found to directly bind eIF4E in its cap-binding site. VPg is covalently linked to its genomic RNA and this interaction allows VPg to act as a "cap."[68][69][16][70] The potyvirus VPg has no sequence or structural homology to other VPg's such as those from poliovirus. In vitro, VPg-RNA conjugates were translated with similar efficiency to m7G-capped RNAs indicating that VPg binds eIF4E and engages the translation machinery; while free VPg (in the absence of conjugated RNA) successfully competes for all the cap-dependent activities of eIF4E in the cell inhibiting translation and RNA export.[70]
d. Regulation of eIF4E cellular localization
Several factors that regulate eIF4E functions also modulate the subcellular localization of eIF4E. For instance, overexpression of PRH/Hex leads to cytoplasmic retention of eIF4E, and thus loss of its mRNA export activity and suppression of transformation.[24] PML overexpression leads to sequestration of eIF4E to nuclear bodies with PML and decrease of eIF4E nuclear bodies containing RNA, which correlates to repressed eIF4E dependent mRNA export and can be modulated by stress.[21][23][25] Overexpression of LRPPRC reduces eIF4E’s co-localization with PML in the nucleus and leads to increased mRNA export activity of eIF4E. As discussed above, Importin 8 brings eIF4E into the nucleus and its overexpression stimulates the RNA export and oncogenic tranformation activities of eIF4E in cell lines. Transduction of primary AML cells with IkB-SR resulted not only in reduction of eIF4E mRNA levels, but also re-localization of eIF4E protein.[25]
The role of eIF4E in cancer was established after Lazaris-Karatzas et al. made the discovery that over-expressing eIF4E causes tumorigenic transformation of fibroblasts.[71] Since this initial observation, numerous groups have recapitulated these results in different cell lines.[72] As a result, eIF4E activity is implicated in several cancers including cancers of the breast, lung, and prostate. In fact, transcriptional profiling of metastatic human tumors has revealed a distinct metabolic signature wherein eIF4E is known to be consistently up-regulated.[73]
eIF4E levels are increased in many cancers including acute myeloid leukemia (AML), multiple myeloma, infant ALL, diffuse large B-cell lymphoma, breast cancer, prostate cancer, head and neck cancer and its elevation generally correlates with poor prognosis.[30][74][75][76][77][78][79][80] In many of these cancers such as AML, eIF4E is enriched in nuclei and several of eIF4E’s activities are found to be elevated in primary patient specimens, including capping, splicing, RNA export, and translation.
In the first clinical trials targeting eIF4E, old antiviral drug ribavirin was used as a m7G cap competitor which had substantial activity in cancer cell lines and animal models associated with dysregulated eIF4E.[81][82][75][31][83][84][85][78][86][87][88][89][90][80] In the first trial to ever target eIF4E, ribavirin monotherapy was demonstrated to inhibit eIF4E activity leading to objective clinical responses including complete remissions in AML patients.[30] Interestingly, relocalization of eIF4E from the nucleus to the cytoplasm correlated with clinical remissions indicative of the relevance of its nuclear activities to disease progression.[30] Subsequent ribavirin trials in AML in combination with antileukemic drugs again showed objective clinical responses including remissions and molecular targeting of eIF4E.[76][91] Clinical responses correlated with reduced nuclear eIF4E and clinical relapse with re-emergence of eIF4E nuclear eIF4E and its RNA export activity in these AML studies. Other studies used ribavirin in combination showed similar promising results in head and neck cancer.[79] Ribavirin impairs all of the activities of eIF4E examined to date (splicing, capping, RNA export and translation). Thus, eIF4E has been successfully therapeutically targetable in humans; however drug resistance to ribavirin is an emergent problem to long term disease control.[84][76][91]
eIF4E has also been targeted by antisense oligonucleotides which were very potent in mouse models of prostate cancer,[92] but in monotherapy trials in humans did not provide clinical benefit likely due to the inefficiency of reducing eIF4E levels in humans compared to mice.[93] There is also an allosteric inhibitor of eIF4E which binds between the cap-binding site and the dorsal surface that is used experimentally.[94]
Fragile X mental retardation protein (FMR1) acts to regulate translation of specific mRNAs through its binding of eIF4E. FMRP acts by binding CYFIP1, which directly binds eIF4e at a domain that is structurally similar to those found in 4E-BPs including EIF4EBP3, EIF4EBP1, and EIF4EBP2. The FMRP/CYFIP1 complex binds in such a way as to prevent the eIF4E-eIF4G interaction, which is necessary for translation to occur. The FMRP/CYFIP1/eIF4E interaction is strengthened by the presence of mRNA(s). In particular, BC1 RNA allows for an optimal interaction between FMRP and CYFIP1.[95] RNA-BC1 is a non-translatable, dendritic mRNA, which binds FMRP to allow for its association with a specific target mRNA. BC1 may function to regulate FMRP and mRNA interactions at synapse(s) through its recruitment of FMRP to the appropriate mRNA.[96]
In addition, FMRP may recruit CYFIP1 to specific mRNAs in order to repress translation. The FMRP-CYFIP1 translational inhibitor is regulated by stimulation of neuron(s). Increased synaptic stimulation resulted in the dissociation of eIF4E and CYFIP1, allowing for the initiation of translation.[95]
EIF4E has been shown to interact with:
. Other direct interactors: PML;[21][64] arenavirus Z protein;[64][63][65][66] Importin 8;[32] potyvirus VPg protein,[70] LRPPRC,[35][26] RNMT[117] and others.