ALKBH1-Mediated tRNA Demethylation Regulates Translation

Abstract

tRNA is a central component of protein synthesis and the cell signaling network. One salient feature of tRNA is its heavily modified status, which can critically impact its function. Here, we show that mammalian ALKBH1 is a tRNA demethylase. It mediates the demethylation of N¹-methyladenosine (m¹A) in tRNAs. The ALKBH1-catalyzed demethylation of the target tRNAs results in attenuated translation initiation and decreased usage of tRNAs in protein synthesis. This process is dynamic and responds to glucose availability to affect translation. Our results uncover reversible methylation of tRNA as a new mechanism of post-transcriptional gene expression regulation.

Keywords: ALKBH1; N(1)-methyladenosine (m(1)A); codon usage; dynamic tRNA modification; tRNA demethylase; tRNA demethylation; tRNA methylation; translation elongation; translation initiation; translation regulation.

Figure 1. Human ALKBH1 Binds tRNA and Catalyzes Oxidative Demethylation of m¹A in tRNA

(A) TBE-Urea gel of ALKBH1-bound RNA species. The image shows RNA species immunoprecipitated by ALKBH1 (lane 1), eEF1α (lane 2), and IgG (lane 3) with purified total human tRNA (lane 4) used as a size marker for mature tRNAs; eEF1α was used as a known tRNA-binding protein control and IgG as a negative control. The majority of RNAs crosslinked to ALKBH1 are intact mature tRNAs. (B) Pie chart showing ALKBH1-bound tRNAs identified by CLIP-seq. (C) Pie chart showing tRNAs identified in the total RNA control. (D) Demethylation of m¹A in total tRNA isolated from HeLa cells by recombinant ALKBH1 in vitro. Ratios of m¹A versus the sum of all four bases are shown. Error bars represent mean ± s.d., n = 8 (four biological replicates × two technical replicates). (E) Proposed oxidative demethylation of m¹A at the 58 position in tRNA by ALKBH1. See also Figure S1.

Figure 2. ALKBH1-Mediated m¹A Demethylation of tRNAs inside Cells

(A) Quantification of m¹A/G ratio in total tRNA purified from HeLa cells by LC-MS/MS. The transient knockdown and overexpression of ALKBH1 in HeLa cells led to ~6% increase and ~16% decrease of the m¹A/G ratio in total tRNA, respectively. Stable overexpression of ALKBH1 in HeLa cells resulted in ~21% decrease of the m¹A/G ratio compared to control cells. (B) In the Alkbh1^−/− MEF cell line, the m¹A/G ratio is ~42% higher compared to control cells. Control samples are shown in light grey. G is the most abundant base in tRNAs and is used to more accurately quantify the m¹A level in tRNA. (C) Quantification of m¹A/G ratio in tRNA^Glu(CUC), tRNA^His(GUG), tRNA^Gln(CUG), tRNA^Lys(yUU), tRNA^Ala(hGC), tRNA^Val(mAC), tRNA^Gly(GCC), tRNA^Asn(GUU), and tRNA^Tyr(GUA) by LC-MS/MS. Knockout of Alkbh1 in MEF cells led to a noticeable increase of m¹A/G ratio in these tRNAs compared to controls. y = U,C; h = A,C,U; m = A,C. (D) Stable overexpression of ALKBH1 in HeLa cells led to a noticeable decrease of the m¹A/G ratio in these tRNAs compared to controls. p values were determined using two-tailed Student’s t-test for paired samples. Error bars represent mean ± s.d., n = 8 (four biological replicates × two technical replicates). *p < 0.05, **p < 0.01. See also Figure S2 and Table S1.

Figure 3. ALKBH1 Affects the Cellular Level of tRNA^iMet and Cell Proliferation

(A) The knockdown of ALKBH1 in HeLa cells increased the cellular level of tRNA^iMet compared to the control (5s rRNA was used as the loading control). One representative result was shown in the upper panel. Four biological replicates were performed with statistic errors calculated. The relative ratios of tRNA^iMet to 5s rRNA were compared in the ALKBH1 knockdown HeLa cells (red) and control samples (grey). (B) The knockdown of ALKBH1 promoted proliferation of HeLa cells. (C) The knockdown of ALKBH1 in HeLa cells increased the tRNA^iMet level in the initiating ribosome compared to the control. See also Figure S3.

Figure 4. The Methylated tRNA Speices are Preferrentially Used to Promote Translation

(A) Hypermethylated tRNA^Val(mAC), tRNA^His(GUG), and tRNA^Gly(GCC) preferentially associate with the polysomes. Quantification of the m¹A/G ratio in tRNA^Val(mAC), tRNA^His(GUG), and tRNA^Gly(GCC) extracted from the translationally inactive portions (< 80S) and the translationally active portions (> 80S) in ALKBH1 stable overexpression HeLa cells and the control cells. Error bars represent mean ± s.d., n = 8 (four biological replicates × two technical replicates). (B) The elongation factor-1 protein (eEF1α) was pulled down from HeLa cells using a specific anti-eEF1α antibody and then incubated with an excess amount of total tRNA purified from same cells. LC-MS/MS measurements showed enrichment of m¹A in the eEF1α-bound portion and depletion of m¹A in the flow through portion. Error bars represent mean ± s.d., n = 8 (four biological replicates × two technical replicates). (C) Quantification of protein synthesis in MEF cells by flow cytometry. HPG incorporation into MEF control cells (grey), Alkbh1^−/− MEF cells (red), HeLa control cells (yellow), and ALKBH1 stable overexpression HeLa cells (blue) were recorded 1 h after administration. Six biological replicates were performed; for ALKBH1 stable overexpression compared to the control, p = 0.012. For Alkbh1^−/− MEF vs. control, p = 0.003. P values were determined using two-tailed Student’s t-test for paired samples. *p < 0.05, **p < 0.01. Error bars represent mean ± s.d., n = 9 (three biological replicates × three technical replicates). See also Figure S4.

Figure 5. A Reporter Assay Confirming Translation Regulation by the ALKBH1-Meidated tRNA Demethylation

(A) Scheme of the reporter assay on the left: the RNA reporter vector encodes firefly luciferase (F-luc) as the primary reporter and Renilla luciferase (R-luc) on the same plasmid as the internal transfection control. The effect of tRNA^His(GUG) in protein translation was revealed by a reporter assay. 6×CAC(His)-coding sequences (recognized by tRNA^His(GUG)) were inserted after the PLK promoter region of F-luc as the positive reporter (noted as 6×CAC(His)); this insertion in the luciferase reporter led to a noticeable increase of protein synthesis in the Alkbh1^−/− MEF cells compared to the wild-type MEF cells. The control reporter without any insertion was used to normalize the translation differences between the two cells lines. Error bars represent mean ± s.d., n = 12 (three biological replicates × four technical replicates). (B) Scheme of the reporter assay on the left. The effect of tRNA^Gly(GCC) in protein translation was revealed by a reporter assay. 6×GGC(Gly)-coding sequences were inserted after the promoter region of F-luc as the positive reporter (noted as 6×GGC(Gly)); 6× glycine-coding sequences of GGA(Gly) were fused to the 5′end of F-luc as the negative control (noted as 6×GGA(Gly)); another control vector is devoid of any amino acid insertion (noted as control). The insertion of 6×GGC(Gly)-coding sequences in the luciferase reporter led to a ~1.8-fold increased protein translation in the Alkbh1^−/− MEF cells compared to the wild-type MEF cells; this effect was not observed with the negative control of 6×GGA(Gly) insertion. The control reporter without any insertion was used to normalize the translation differences between the two cells lines. Error bars represent mean ± s.d., n = 12 (three biological replicates × four technical replicates). See also Figure S5.

Figure 6. The ALKBH1-Mediated tRNA Demethylation in Response to Glucose Availability

(A) Western blot showing increased concentrations of glucose in the growth medium led to decreased ALKBH1 expression in HeLa cells; protein levels of the m¹A58-transferase Trmt6/Trmt61 heterodimer remained mostly unchanged. (B) LC-MS/MS quantification of the m¹A/G ratio in tRNA^His(GUG) and tRNA^Gly(GCC) in HeLa cells cultured with 5 and 25 mM of glucose. The increased concentration of glucose led to increased m¹A methylation in tRNA^His(GUG) and tRNA^Gly(GCC). Error bars represent mean ± s.d., n = 8 (four biological replicates × two technical replicates). (C) Glucose deprivation decreased the cellular level of tRNA^iMet, which could be reversed by ALKBH1 knockdown. (5s rRNA was used as the loading control). Glucose deprivation also noticeably decreased the level of tRNA^iMet in the initiating ribosome. The knockdown of ALKBH1 led to elevated tRNA^iMet in the initiating ribosome in comparison to the control in HeLa cells cultured with 5 mM and 25 mM glucose, respectively. 500 ng of total RNA extracted from the initiating ribosomes were applied in the northern blot analysis. One representative result was shown. (D) The levels of tRNA^iMet as shown in (C) were quantified by using Quantity One. (E) The 6×CAC(His)-reporter assay showing increased translation in HeLa cells with elevated glucose levels in the culture medium. This effect could be reversed by transient knockdown of ALKBH1. Error bars represent mean ± s.d., n = 12 (three biological replicates × four technical replicates). (F) Knockdown of ALKBH1 in HeLa cells growing at 25 mM glucose led to an increased translation of the 6×CAC(His)-reporter. The effect could be reversed by transient overexpression of the wild-type ALKBH1 but not a catalytically inactive ALKBH1 H228A/D231A mutant. P values were determined using two-sided Student’s t-test for paired samples. *p < 0.05, **p < 0.01. Error bars represent mean ± s.d., n = 12 (three biological replicates × four technical replicates). n.s. represents not significant. (G) A proposed model of ALKBH1-dependent regulation of translation. The m¹A58 methylated tRNAs are preferentially used in the translation-active pool; the ALKBH1-mediated tRNA m¹A demethylation, which responds to the glucose deprivation, tunes the association of specific tRNAs to polysomes and thus represses protein synthesis. See also Figure S6.