4E-BP1, a multifactor regulated multifunctional protein

ABSTRACT

Eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1 (4E-BP1) is a member of a family of translation repressor proteins, and a well-known substrate of mechanistic target of rapamycin (mTOR) signaling pathway. Phosphorylation of 4E-BP1 causes its release from eIF4E to allow cap-dependent translation to proceed. Recently, 4E-BP1 was shown to be phosphorylated by other kinases besides mTOR, and overexpression of 4E-BP1 was found in different human carcinomas. In this review, we summarize the novel findings on mTOR independent 4E-BP1 phosphorylation in carcinomas. The implications of overexpression and possible multi-function of 4E-BP1 are also discussed.

KEYWORDS:

• cancer
• eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1
• kinase
• mTOR
• overexpression
• phosphorylation

This article is referred to by:

Yin and yang of 4E-BP1 in cancer

mTOR/4E-BP1 signaling pathway

The mechanistic target of rapamycin (mTOR) is a central controller of organismal growth and survival in response to growth factors, cytokines, hormones and nutrients.¹ It is a PI3K-related kinase that forms two distinct protein complexes called mTOR complex 1 or mTORC1 and mTOR complex 2 or mTORC2. In response to upstream stimuli, mTORC1 phosphorylates S6 kinase 1 (S6K1) and eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1 (4E-BP1) to stimulate protein synthesis,² while mTORC2 phosphorylates AKT to promote cell survival.³ The mTOR pathway regulates major cellular processes and is implicated in several pathological conditions, including cancer, obesity, type 2 diabetes, and neurodegeneration.⁴

Typically, translation is regulated at the initiation phase, when a ribosome is recruited to the 5′-cap of an mRNA. 4E-BP1 interdicts translation initiation by binding to the translation factor eIF4E, which is a limiting component of the multi-subunit complex that recruits 40S ribosomal subunits to the 5′-cap of mRNAs. Interaction of hypophosphorylated 4E-BP1 with eIF4E prevents its interaction with eIF4G, inhibits complex assembly and represses translation. Upon activation, mTORC1 remains at the mRNA 5′-cap, where it is well positioned to phosphorylate 4E-BP1 resulting in its dissociation from eIF4E and the recruitment of eIF4G to the 5′-cap, thereby allowing cap-dependent translation initiation to proceed.² As one of the two most established mTOR substrates, 4E-BP1 plays a pivotal role in mTORC1 signaling to control translation and cell proliferation.^2,5,6

4E-BP1 phosphorylation: conventional and novel findings

Phosphorylated 4E-BP1 has been generally accepted as a marker of activated mTOR signaling. Seven phosphorylation sites have been identified in 4E-BP1: Thr 37, Thr 46, Ser 65, Thr 70, Ser 83, Ser 101, and Ser 112 (numbering based on human 4E-BP1; in rodents, the numbers are lower by one).⁷ The first five phosphorylation sites are phylogenetically conserved among all species, and phosphorylation of Thr 37 and Thr 46 serves as a priming event, which is followed by Thr 70 phosphorylation and finally Ser 65 phosphorylation.^8,9 Although mTOR immunoprecipitates phosphorylated Thr 37 and Thr 46 in vitro, whether mTOR phosphorylates 4E-BP1 on all or some of these sites remains unknown. A kinase activity specific to Ser 65 and Thr 70 was reported to be liberated from an mTOR immunoprecipitate, suggesting the presence of a second kinase.¹⁰

Possible kinases involved in 4E-BP1 phosphorylation

Recently, several kinases have been shown to phosphorylate 4E-BP1 dependent or independent of mTOR, indicating that mTOR may not be the only kinase that phosphorylates 4E-BP1, challenging the conventional mTOR/4E-BP1 cascades.¹¹ mTOR inhibitor resistance (for allosteric inhibitors, e.g. rapamycin and rapalogs, and the more potent ATP-competitive mTOR kinase inhibitors) is correlated with the inability of these inhibitors to block 4E-BP1 phosphorylation in some cancer cells.^12-15 Phosphorylation of mTORC1 substrates is subject to multiple regulatory mechanisms and feedback loops that extend far beyond the simple sequence motif of each phosphorylation site.¹⁶ In the case of 4E-BP1, which is regarded as a point of convergence of various signaling pathways, it is possible that other kinases cooperate with or independent of mTORC1 regulate its phosphorylation.^17,18 We have shown that mTOR kinase inhibitors sustained on-target inhibition of mTOR in colorectal cancer (CRC) cells, but only transiently attenuated 4E-BP1 phosphorylation in drug-resistant CRC cells because 4E-BP1 was re-phosphorylated in an mTOR-independent manner (Fig. 1).¹² Several candidates were recently suggested to be responsible for 4E-BP1 phosphorylation. Since GSK3β phosphorylates 4E-BP1 at Thr 37 and Thr 46 and thereby decreases its association with eIF4E, it is also possible that GSK-3 could regulate the phosphorylation of other sites (S 65 and/or T 70) in some cancer cells.¹⁹ MAP kinase p38 may increase 4E-BP1 protein by hyperphosphorylation in response to several cellular stresses such as viral infection, osmotic shock, UV irradiation, and TNFα/CHX-mediated apoptosis.^20,21 UVB-induced phosphorylation of 4E-BP1 at multiple sites including Thr 36, Thr 45, Ser 64 and Thr 69 in the mouse epidermal JB6 cell line was mediated by p38MAPK and its downstream kinase MSK1, which provides a novel mechanism for UV-promoted tumorigenesis.²² ERK phosphorylates 4E-BP1 at Ser 65, and also blocks TSC2, leading to greater mTOR activity and increased phosphorylation of 4E-BP1. It is likely that ERK acts directly on 4E-BP1 and indirectly via TSC2/mTOR following ionizing radiation (IR) and stimulates protein synthesis via ATM-dependent ERK phosphorylation.^23,24 4E-BP1 phosphorylation makes it resistant to rapamycin, which has been correlated to the induction of PIM kinases.²⁵ PIM2 could work directly in the cytoplasm to phosphorylate 4E-BP1 at Ser 65 ²⁶ or regulate the activity of specific phosphatases.^27,28 ATM, the protein product of the ATM gene, phosphorylates 4E-BP1 at Ser 111 in vitro. Insulin treatment induces phosphorylation of 4E-BP1 at Ser 112 in vivo in an ATM-dependent manner.²⁹ Cyclin-dependent kinase, cdc2/CDK1, can phosphorylate 4E-BP1 at Thr 70 in HeLa cells.³⁰ Paclitaxel (PTX) induces the hyperphosphorylation of 4E-BP1 in the breast cancer cell line, MDA-MB-231, through a CDK1-dependent mechanism.³¹ LRRK2, leucine-rich repeat kinase 2, is one of the physiological kinases for 4E-BP. LRRK2 exerts an effect on 4E-BP1 primarily at the Thr 37/46 sites.³² Some studies have shown that mTOR kinase inhibitors block p4E-BP1 more effectively than rapalogs, suggesting that mTORC2 is a candidate for 4E-BP1 phosphorylation.⁴ But whether rapamycin-resistant and PIM-kinase-dependent 4E-BP1 phosphorylation is mediated by the mTORC2 complex ³³ remains unknown (Table 1).

Figure 1. mTOR-independent 4E-BP1 phosphorylation is associated with colorectal cancer (CRC) resistance to mTOR kinase inhibitors (mTKIs). TIEK: TOR- independent 4E-BP1 kinase.

Table 1. Possible kinases for mTOR-independent 4E-BP1 phosphorylation.

Kinases	Condition	Phosphorylation site	References
GSK3β	N/A	Thr 37, Thr 46 (possibly Ser 65 and/or Thr 70)	^19
p38MAPK	Viral infection, TNFα/CHX-mediated apoptosis	N/A	^20,21
	UVB	Thr 36, Thr 45, Ser 64, Thr 69	^22
ERK	Ionizing radiation	Ser 65 (in vitro)	^23,24
PIM2	N/A	Ser 65	^26
ATM	Insulin treatment	Ser 112	^29
CDC2/CDK1	Paclitaxel (PTX) treatment	Thr 70	^30,31
LRRK2	Physiological condition	Thr 37, Thr 46	^32

Possible mTOR-independent 4E-BP1 kinases and corresponding phosphorylation condition and site were listed. For a review see refs. 19–32.

These findings add to our understanding of the mechanisms by which cancer cells are resistant to mTOR inhibitor treatment. Some important questions remain regarding whether these potential kinases contribute to the intrinsic or acquired drug resistance, and why mTOR substrate 4E-BP1 is phosphorylated by certain kinases in some cancer cells but not in others?

Overexpression of 4E-BP1 in carcinomas

High level of phosphorylated 4E-BP1 was widely reported in human cancers, and has been associated with a worse outcome in several malignancies. However, the pattern of 4E-BP1 expression is not clearly understood. Because 4E-BP1 belongs to the family of translation repressor proteins, it is reasonable to assume that 4E-BP1 expression is lower in carcinomas as compared to normal tissue. However, overexpression of 4E-BP1 has been found in various carcinomas, which is counterintuitive to its known suppressive role in translation. Based on data from Oncomine (https://www.oncomine.com), most studies on brain, CNS, CRC, kidney and other cancer types have shown a higher level of 4E-BP1 in cancer tissue as compared to normal tissue (Fig. 2).

Figure 2. Overexpression of 4E-BP1 has been found in various carcinomas. Original data from https://www.oncomine.com was analyzed and plotted into a bar graph. Studies that showed higher 4E-BP1 level in cancer tissue vs. normal tissue are depicted in red and others in blue.

4E-BP1 overexpression is strongly associated with prostate cancer. Expression of 4E-BP1 was elevated in prostate intraepithelial neoplasia (PIN, the pre-cancerous lesion of prostate cancer), and was significantly higher in prostate cancer as compared to normal tissue. 4E-BP1 was highly overexpressed in 40% of cancers and at least some overexpression was observed in the other 60% of cancers in a tissue array. Nearly all cancer samples had over 3.3-fold higher IHC staining scores than normal lumenal cells.³⁴ 4E-BP1 was also overexpressed in several large advanced breast cancers.³⁵ High cytoplasmic 4E-BP1 protein levels predict less benefit from endocrine treatment.³⁶ 4E-BP1 overexpression also occurs in head, neck and some gastrointestinal cancers.^37-39

Possible mechanisms for 4E-BP1 overexpression in cancer

Amplification of the 4E-BP1 gene is one possible mechanism for its overexpression in tumors. The gene encoding 4E-BP1 is located at the chromosomal region 8p12, which is commonly amplified in breast cancer. Gene amplification and the resultant high expression of 4E-BP1 were shown to indicate a poor prognosis.^40,41 Another suggested mechanism is through transcription. High 4E-BP1 mRNA, independent of phosphorylation status, was associated with an adverse outcome in breast cancer, especially in ER-positive subgroup.³⁶ 4E-BP1 protein accumulates under different conditions by different transcription factors. In prostate cancer, amplification or induced expression of cMyc has been shown to promote its binding to the 4E-BP1 gene promoter and increase expression.⁴² ATF4 mediates induction of 4E-BP1 in pancreatic beta-cells under endoplasmic reticulum stress.⁴³ In pancreatic cancer cell lines, hypoxia-triggered induction of 4E-BP1 is dependent on HIF-1α and SMAD4.⁴⁴

Since hyperphosphorylation and overexpression of 4E-BP1 occurs simultaneously in human cancers, is there a possible correlation between 4E-BP1 phosphorylation and its expression level? Phosphorylation is the most common mechanism by which a stable protein can be targeted for ubiquitination and subsequent degradation. However, phosphorylation can also have an opposing effect of stabilizing a short-lived protein, especially those proteins that require multi-site phosphorylation.^45-48 Because more than one kinase is involved in the phosphorylation of multiple sites, dephosphorylation of a single site is not sufficient for ubiquitination and degradation. Thus, multi-site phosphorylation could set a higher threshold for susceptibility to degradation.^46,49 Therefore, hyperphosphorylation of 4E-BP1 at multiple sites may play an important role in its stabilization and overexpression.

Possible functions of overexpressed 4E-BP1 in cancer

Approximately 30% of the 4E-BP1 expressed in cells is located in the nucleus, where it regulates the availability of EIF4E for the cytoplasmic translational machinery, by retaining EIF4E in the nucleus.³⁶ Thus, what the remaining (70% or more) 4E-BP1 does in the cytosol remains unknown. A recent study revealed that cytoplasmic overexpressed 4E-BP1 orchestrates a hypoxia-activated switch from cap-dependent to cap-independent mRNA translation that promotes increased tumor angiogenesis and growth at the level of selective mRNA translation. Eukaryotic cells perform both cap-dependent and cap-independent (IRES-mediated) protein synthesis. Under normal growth conditions cap-dependent translation is dominant, while under stress such as low nutrients and hypoxia in advanced cancer, cap-dependent translation is restricted but some cellular mRNAs can utilize cap-independent (IRES-mediated) mechanism of initiation, including those that orchestrate angiogenesis (VEGF-A), hypoxia responses (transcription factor HIF1α),⁵⁰ and inhibition of apoptosis (Bcl2).⁵⁰ So, in advanced cancer, 4E-BP1 acts as a hypoxia-inducible switch, promoting cap-independent over cap-dependent translation, resulting in selective translation of IRES-containing mRNAs such as VEGF, HIF1α, and Bcl2 that promote tumor angiogenesis, survival,³⁵ and may be involved in the PI3K/mTOR inhibitor adaptive resistance in matrix-attached cancer cells.⁵¹ MYC expression is reported to abrogate sensitivity to rapamycin through increased expression of 4E-BP1 and reduced autophagy in prostate cancer.⁴² PC-1/PrLZ confers resistance to rapamycin in prostate cancer cells through increased 4E-BP1 stability.⁵² Thus, besides translation repression, 4E-BP1 may also act as a tumor promoter under certain cellular conditions, indicating that 4E-BP1 is a multi-functional protein. This hypothesis is currently being tested.

While these findings explain how certain advanced tumors keep progressing in an unfavorable environment, some important questions remain regarding whether 4E-BP1 overexpression is a cause for tumor development or a consequence of tumor overgrowth. Also, the involvement of any other factors dependent on the cellular context in the switch in 4E-BP1 functions needs to be investigated.

Conclusion

Several candidates are assumed to phosphorylate 4E-BP1, suggesting that the regulation of 4E-BP1 may be multifactorial, and thus highly dependent on the cellular context. Indeed, it is plausible to assume that cancer cells adapt by deregulating additional kinases, such as GSK3β, that substitute for mTOR in the phosphorylation and inactivation of some mTOR substrates, such as 4E-BP1. We believe that finding alternative kinases responsible for 4E-BP1 phosphorylation will be an important future task, especially to identify new targets for cancer therapy.

4E-BP1 overexpression is nearly universal in cancer, although the degree of overexpression varies, suggesting that dysregulation of translation may occur in cancer or may be responsible for cancer progression. Aberrant 4E-BP1 expression may contribute to the inability of cells to properly translate, or gaining the ability of selective mRNA translation. Transcriptional regulation and post-translational modification work together to modulate the level and function of 4E-BP1, making it a complex controller of protein synthesis and a powerful transfer station far beyond mTOR signaling pathway.

Abbreviations

4E-BP1	=	Eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1
mTOR	=	mechanistic target of rapamycin
S6K1	=	S6 kinase 1
PTX	=	Paclitaxel
PIN	=	prostate intraepithelial neoplasia
mTKIs	=	mTOR kinase inhibitors
TIEK	=	TOR- independent 4E-BP1 kinase

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Author contributions

Qin X, Jiang B and Zhang Y designed and wrote the review. Zhang Y took responsibility for revising the manuscript and final approval of the version to be published.

Funding

This work was supported by National Nature Science Foundation of China 81270035 and Shanghai Pujiang Program 15PJ1404900 (Zhang Y).