Fig 2: NMDA induces changes in nuclear and cytosolic levels of m6A demethylase FTO. (a) Representative images showing FTO and MAP2 staining in DIV15 cultured cortical neurons treated with 20 µm NMDA for 1, 5 and 20 min. (b) Quantification of cyto-dendritic FTO levels (normalized to MAP2) in DIV15 cultured cortical neurons treated with 20 µm NMDA for 1, 5 and 20 min. Data represents mean + /- SEM, n > 24 neurons for all groups, from at least 3 independent neuronal cultures, Kruskal–Wallis test followed by Dunn’s multiple comparison test. (c) Representative images showing FTO and MAP2 staining in DIV15 cultured cortical neurons treated with 20 µm NMDA for 1, 5 and 20 min. (d) Quantification of mean intensity of nuclear FTO levels in DIV15 cultured cortical neurons treated with 20 µm NMDA for 1, 5 and 20 min. Data represents mean + /- SEM, n > 24 neurons for all groups, from at least 3 independent neuronal cultures, Kruskal- Wallis test followed by Dunn’s multiple comparison test. (e) FTO immunoblot and control GAPDH blots for the total FTO protein levels treated with NMDA (40 µM) for 1, 5 and 20 min. (f) Quantification of FTO levels and plotted values are normalized to GAPDH. Data represent mean + /- SEM, N = 4 independent experiments, One way ANOVA with Tukey’s multiple comparison test. (g) m6A immunoblot for basal and NMDA 5 min condition in presence of FTO inhibitor Meclofenamic acid (MA). (h) Quantification of m6A signal comparing Basal and MA treatment for 2 h, N = 3, paired T-test. (i) Quantification of m6A signal in DIV15 cultured cortical neurons treated with 20 µm NMDA for 5 min with and without NMDA, N = 3, paired T-test. (j) Quantification of m6A signal in DIV15 cultured cortical neurons treated with 20 µm NMDA for 5 min and comparing Basal, N = 3, paired T-test. (k) Representation of temporal profiles of eEF2, total m6A, and total FTO levels.

Fig 3: Roles of rpS6 kinases and PP2A in ghrelin-induced activation of eEF2 and protein synthesis. a The schematic of HEK293GHS-R1a-EGFP+ cells treatment. b Changes in eEF2 and eEF2K phosphorylation induced by ghrelin after inhibition of either p90 RSK (10 µM BRD7389) or p70 S6K (1 µM PF4708671). c Effects of ghrelin on p-eEF2 (Thr56) and p-eEF2K (Ser366 and Ser500) levels in cells upon inhibition of p90 RSK and p70 S6K (double-treatment with BRD7389/PF4708671) or PP2A (OKA, 50 nM). In (b, c), cells were pre-treated with kinase or phosphatase inhibitors for 2 h prior to ghrelin addition (100 nM, 30 min). Changes in p-mTOR and p-ERK levels were assessed to confirm the GHS-R1a activation. d Changes in SNAP and GLuc production in cells treated with ghrelin upon inhibition of p70 S6K (1 µM PF4708671). Cells were pre-treated for 2 h and then treated with ghrelin (100 nM, 30 min) in the presence of PF4708671; an arrow shows the residual amounts of SNAP in cells lysed immediately after their treatment with SNAP block. e Comparative analysis of eEF2 response to ghrelin in resting and OKA-treated cells [extracted from (c)]. The bottom panel shows that the pre-treatment with OKA causes an increase in phosphorylation of ERK, a common target of PP2A. Red rectangles (b, c) highlight effects of BRD7389 on total levels of mTOR, eEF2 and eEF2K proteins. Red asterisks indicate significant difference in the p-eEF2 (Thr56) response to ghrelin in mock-treated vs BRD- and PF-treated cells (b), or in mock-treated vs BRD/PF- and OKA-treated cells (c), red hash signs—significant difference in p-eEF2 response to ghrelin between BRD/PF- and OKA-treated cells. Black asterisks show the same for p-eEF2K (Ser366) levels. Columns/horizontal bars and gradient circles show mean values and individual data points, respectively. Statistical details: b effects of p90 RSK or p70 S6K inhibition on the ghrelin-induced changes in eEF2 and eEF2K phosphorylation; (a) p-eEF2 (2,6) = 29.442, p < 0.001, one-way ANOVA, p < 0.001 for mock vs BRD, p = 0.001 for mock vs PF, Dunnett post hoc test vs mock treatment (N = 3); (b) p-eEF2K (2,6) = 10.13, p = 0.012, one-way ANOVA for the effect of treatments (BRD and PF), p = 0.024 for mock vs BRD, p = 0.01 for mock vs PF, Dunnett post hoc test vs mock treatment (N = 3). c Effects of rpS6 kinase and PP2A inhibition on the ghrelin-induced changes in eEF2 and eEF2K phosphorylation; (a) p-eEF2 (2,6) = 61.086, p < 0.001, one-way ANOVA for the effect of treatments (BRD/PF and OKA), p < 0.001 for mock vs BRD/PF, p = 0.043 for mock vs OKA, p < 0.001 for BRD/PF vs OKA, Tukey post hoc test (N = 3); (b) p-eEF2K (2,6) = 8.378, p = 0.018, one-way ANOVA for the effect of treatments (BRD/PF and OKA), p = 0.127 for mock vs BRD/PF, p = 0.015 for mock vs OKA, p = 0.263 for BRD/PF vs OKA, Tukey post hoc test (N = 3). d Effect of p70 S6K inhibition on the changes in SNAP (N = 3) and GLuc (N = 4) translation caused by ghrelin treatment, independent samples T test; SNAP: Alexa Fluor 594 t(4) = – 0.7822, p = 0.478; GLuc: t(6) = – 0.676, p = 0.524

Fig 4: Effects of GHS-R1α activation by ghrelin on eEF2 phosphorylation. a Proposed contribution of elongation into ribosome turnover and protein production de novo. Decreased eEF2 phosphorylation is expected to increase elongation rate, ribosome availability and protein production levels. b, c Western blotting analysis of time- and concentration-dependent changes in eEF2 (Thr56) phosphorylation induced in HEK293GHS-R1α-EGFP+ cells by ghrelin treatment; b 100 nM ghrelin for 15–60 min and c 4–100 nM ghrelin for 30 min. d Analysis of eEF2 phosphorylation in parental HEK293 cells lacking GHS-R1α. e Analysis of GHS-R1α protein expression in mouse embryonic hypothalamus cell lines E-N38 and E-N41; in E-N41, GHS-R1α protein levels are ~ 60% higher. f Ghrelin concentration-dependent decrease in p-eEF2 (Thr56) levels in E-N41 cells (30 min treatment). g Changes in the extracellular GLuc signals triggered by ghrelin treatment in E-N41 cells (200 nM, 30 min). Cells were pre-incubated for 14–16 h in serum-free DMEM supplemented with NEAA. Data are shown as individual data points (gradient circles), mean values (horizontal bars) and mean ± SD. Asterisks indicate significant difference between non-treated (0 time point or 0 ghrelin concentration) and ghrelin-treated cells. Statistical details: b effect of ghrelin treatment time: p-eEF2 (4,20) = 124.011, p < 0.001, one-way ANOVA, p < 0.001 for 15, 30, 45 and 60 min, Dunnett post hoc test vs mock treatment (N = 6). c Effect of ghrelin concentration: p-eEF2 (4,20) = 45.25, p < 0.001, one-way ANOVA, p < 0.001 for 3 nM, 11 nM, 33 nM and 100 nM, Dunnett post hoc test vs mock treatment (N = 6). f effect of ghrelin concentration: p-eEF2 (5,12) = 27.815, p < 0.001, one-way ANOVA, p < 0.001 for 167 nM and 500 nM ghrelin, Dunnett post hoc test vs mock treatment (N = 3). g effect of ghrelin treatment, N = 4, independent samples T test: GLuc t(6) = – 1.234, p = 0.181

Fig 5: (A) Luciferase production following co-expression of the indicated MSCV expression vectors in HEK293T cells with 3×BBox mRNA. n = 3, ±SD. (B) Western blot documenting expression levels of ?N fusions in transfected HEK293T cells. (C) Immunoprecipitations performed from cell extracts expressing the indicated FLAG-tagged ?N-4E fusion proteins. Following SDS-PAGE, Western blots were performed with antibodies targeting eIF4GI. (D) Luciferase production following co-expression of the indicated MSCV-eIF4A expression vectors in HEK293T cells with 3×BBox mRNA. n = 3, ±SD. (E) Western blot documenting expression levels of ?N fusions in transfected HEK293T cells. (F) Western blot documenting subcellular localization (C, cytoplasmic; N, nuclear) of the indicated ?N fusions in transfected HEK293T cells. eEF2 and hnRNPA1 are used as loading control for cytoplasmic and nuclear fractionation, respectively.