Loss of eIF4E Phosphorylation Engenders Depression-like Behaviors via Selective mRNA Translation

Abstract

The MAPK/ERK (mitogen-activated protein kinases/extracellular signal-regulated kinase) pathway is a cardinal regulator of synaptic plasticity, learning, and memory in the hippocampus. One of major endpoints of this signaling cascade is the 5' mRNA cap binding protein eIF4E (eukaryotic Initiation Factor 4E), which is phosphorylated on Ser 209 by MNK (MAPK-interacting protein kinases) and controls mRNA translation. The precise role of phospho-eIF4E in the brain is yet to be determined. Herein, we demonstrate that ablation of eIF4E phosphorylation in male mice (4Eki mice) does not impair long-term spatial or contextual fear memory, or the late phase of LTP. Using unbiased translational profiling in mouse brain, we show that phospho-eIF4E differentially regulates the translation of a subset of mRNAs linked to inflammation, the extracellular matrix, pituitary hormones, and the serotonin pathway. Consequently, 4Eki male mice display exaggerated inflammatory responses and reduced levels of serotonin, concomitant with depression and anxiety-like behaviors. Remarkably, eIF4E phosphorylation is required for the chronic antidepressant action of the selective serotonin reuptake inhibitor fluoxetine. Finally, we propose a novel phospho-eIF4E-dependent translational control mechanism in the brain, via the GAIT complex (gamma IFN activated inhibitor of translation). In summary, our work proposes a novel translational control mechanism involved in the regulation of inflammation and depression, which could be exploited to design novel therapeutics.SIGNIFICANCE STATEMENT We demonstrate that downstream of the MAPK (mitogen-activated protein kinase) pathway, eukaryotic Initiation Factor 4E (eIF4E) Ser209 phosphorylation is not required for classical forms of hippocampal LTP and memory. We reveal a novel role for eIF4E phosphorylation in inflammatory responses and depression-like behaviors. eIF4E phosphorylation is required for the chronic action of antidepressants, such as fluoxetine in mice. These phenotypes are accompanied by selective translation of extracellular matrix, pituitary hormones, and serotonin pathway genes, in eIF4E phospho-mutant mice. We also describe a previously unidentified translational control mechanism in the brain, whereby eIF4E phosphorylation is required for inhibiting the translation of gamma IFN activated inhibitor of translation element-containing mRNAs. These findings can be used to design novel therapeutics for depression.

Keywords: depression; eIF4E; inflammation; phospho-eIF4E; translation.

Figure 1.

Intact spatial learning and memory, contextual fear memory, and L-LTP in 4Eki mice. A, Representative confocal images of immunofluorescent staining of WT dorsal hippocampi with antibodies against total and phospho-Ser209 eIF4E. Scale bar, 100 μm. B, MWM task. Left, Graphic depiction of experimental design; latency (seconds) to find hidden platform during experimental days. Right, Platform crossings and quadrant occupancy during probe test (WT, n = 7; 4Eki, n = 8). ***p < 0.001 (repeated-measures ANOVA with Tukey's post hoc). C, CFC in 4Eki mice. Percentage freezing 24 h after initial shock (WT, n = 8; 4Eki, n = 8). D, CA1 L-LTP recordings in 4Eki mice. Normalized fEPSP slope over time (min) for 240 min. E, Summary quantification of percentage potentiation for L-LTP. (Student's t test).

Figure 2.

Ribosome profiling reveals preferential translation of a subset of mRNAs in the forebrain of 4Eki mice. A, Experimental design to assess genome-wide translational efficiency of mRNAs using ribosome profiling in whole brain tissue from WT and 4Eki mice. B, log₂ TE Plot showing translationally upregulated (red), downregulated (blue), and control (gray) mRNAs in 4Eki versus WT libraries (p < 0.05 and 0.75 ≥ TE ratio ≤ 1.5). Gray represents unchanged mRNAs; n = 2 for footprints and mRNA. C, UTR analysis using RegRNA in downregulated (651; blue), upregulated (52; red), and control (325; gray) mRNAs in 4Eki, compared with WT. Percentage of genes containing one or more of the depicted RNA sequence elements in 5′- or 3′-UTR. ^#Categories in downregulated or upregulated mRNAs, which are underrepresented compared with control mRNAs. (Fig. 2-1A, Fig. 2-1B, Fig. 2-1C,D). D, Length and GC content analysis of differentially translated mRNAs. Length (bp) or percentage of GC content is displayed for 5′-UTR (left) or 3′-UTR (right). ^#p < 0.05 difference from all other categories (one-way ANOVA with Tukey's post hoc); all other multiple comparisons between groups are not significant. E, Gene ontology analysis of 651 downregulated genes; plots for biological process, molecular function, and cellular component with number of genes in each category. p values next to each category are shown. F, KEGG pathway analysis for downregulated genes. G, Major genes downregulated in ribosome profiling organized in two categories: pituitary hormone genes and ECM genes with p value and FDR. H, Gene ontology analysis of 52 upregulated genes; plots for cellular component with number of genes in each category and p values and KEGG pathway analysis. Major genes upregulated in ribosome profiling organized in two categories: serotonin and ribosomal proteins; p value and FDR are shown for downregulated and upregulated genes.

Figure 3.

Exaggerated inflammatory responses and reduced serotonin levels in 4Eki brain. A, Quantitative ELISA for 6 mouse inflammatory cytokines in WT and 4Eki forebrains (n = 10 for each genotype). Blue represents Th1 cytokines. Gray represents Th2 cytokines. B, Left, Serotonin pathway genes upregulated in 4Eki brain (red). Quantitative ELISA for serotonin (5-HT) in WT and 4Eki forebrains (n = 20 for each genotype). Normalized concentration (pg/mg) is shown for all experiments. C, Quantitative ELISA for Iba-1, a marker of activated microglia (n = 10 for each genotype). A, C, ***p < 0.001 (one-way ANOVA with Bonferroni's post hoc). **p < 0.01 (one-way ANOVA with Bonferroni's post hoc). B, ***p < 0.001 (Student's t test).

Figure 4.

Depression and anxiety-like behaviors in 4Eki mice. A, Immobility time (seconds) as an indicator of depression-like behaviors in FST (left) and TST (middle) in WT(n = 14) and 4Eki (n = 18) mice. NSF (right), latency to start feeding in a novel environment, as a proxy for depression/anxiety-mediated hypophagia in WT and 4Eki (n = 18 each) mice. B, Open field exploration test (OF) in WT (n = 10) and 4Eki (n = 12) mice, as a measure of anxiety. Time (seconds) spent in the center square, in proximity of corners or walls, or outside the center square and total distance traveled. A, B, *p < 0.05 (Student's t test). **p < 0.01 (Student's t test). ***p < 0.001 (Student's t test). C, EPM in WT (n = 8) and 4Eki (n = 8) mice, as a measure of anxiety. Time (seconds) spent in the open or closed arms of the elevated maze. C, ***p < 0.001 (one-way ANOVA with Bonferroni's post hoc).

Figure 5.

Chronic fluoxetine intraperitoneal treatment does not reverse depression-like behaviors in 4Eki mice. A, Outline of chronic fluoxetine regimen. Intraperitoneal injection of 10 mg/kg/d for 21 d reduces immobility time (seconds) in WT (n = 12) but not 4Eki (n = 12) mice. B, FST. C, TST. **p < 0.01, ***p < 0.001 (one-way ANOVA with Bonferroni's post-hoc).

Figure 6.

Altered cap binding of GAIT complex protein rpL13a and eIF4A1 in 4Eki brains. A, Cap-column (m⁷GDP) pulldown from forebrain lysates (WT and 4Eki; n = 4 per genotype or n = 8 for Eprs, Gapdh). Left, Diagram of the closed loop model of translation depicting binding of repressive 3′-UTR GAIT elements to 5′-UTR cap-bound eIF4F complex, via rpL13a and below a depiction of a cap-column agarose bead. Representative immunoblot images from cap-bound and input lysates probed with antibodies against the indicated proteins (eIF4E, eIF4G, eIF4A1 rpL13a, Eprs, and Gapdh; β-actin is the loading control). B, Quantification of protein expression from input (5%) and cap-bound lysates. Protein expression (arbitrary units) normalized to input protein expression for cap-bound lysates and to β-actin for input lysates. *p < 0.05 (Student's t test).

Figure 7.

Depletion of eIF4E phosphorylation engenders inflammatory and depression-like phenotypes via selective translational control of a subset of mRNAs. A, Ablation of the single phosphorylation site on eIF4E (Ser209→Ala) does not affect global protein synthesis, but rather the translation of a subset of mRNAs harboring GAIT elements, which engenders a depression-like phenotype in 4Eki mice. 4Eki mice also display increased expression of inflammatory cytokines, which could be linked to disinhibition of GAIT translational repression and possibly to depression-like phenotypes. Altered cap binding of the helicase eIF4A1 and/or of the GAIT complex protein rpL13a could be the mechanism underlying altered translation initiation following depletion of Ser209 eIF4E phosphorylation. B, The SSRI fluoxetine requires eIF4E phosphorylation to exert its antidepressant action. C, Phosphorylation of eIF4E promotes anti-inflammatory and antidepressant pathways.