Mettl1-dependent m7G tRNA modification is essential for maintaining spermatogenesis and fertility in Drosophila melanogaster

Abstract

Modification of guanosine to N⁷-methylguanosine (m⁷G) in the variable loop region of tRNA is catalyzed by the METTL1/WDR4 heterodimer and stabilizes target tRNA. Here, we reveal essential functions of Mettl1 in Drosophila fertility. Knockout of Mettl1 (Mettl1-KO) causes no major effect on the development of non-gonadal tissues, but abolishes the production of elongated spermatids and mature sperm, which is fully rescued by expression of a Mettl1-transgene, but not a catalytic-dead Mettl1 transgene. This demonstrates that Mettl1-dependent m⁷G is required for spermatogenesis. Mettl1-KO results in a loss of m⁷G modification on a subset of tRNAs and decreased tRNA abundance. Ribosome profiling shows that Mettl1-KO led to ribosomes stalling at codons decoded by tRNAs that were reduced in abundance. Mettl1-KO also significantly reduces the translation efficiency of genes involved in elongated spermatid formation and sperm stability. Germ cell-specific expression of Mettl1 rescues disrupted m⁷G tRNA modification and tRNA abundance in Mettl1-KO testes but not in non-gonadal tissues. Ribosome stalling is much less detectable in non-gonadal tissues than in Mettl1-KO testes. These findings reveal a developmental role for m⁷G tRNA modification and indicate that m⁷G modification-dependent tRNA abundance differs among tissues.

Fig. 1. Mettl1 is essential for maintaining fertility.

a Schematic diagram of the genomic structure of Mettl1 and a gRNA targeting Mettl1 exon 2 for CRISPR-mediated generation of Mettl1-KO lines. Mettl1 sequences of the WT (yw) and the Mettl1-KO mutants (Mettl1^KO1 and Mettl1^KO2). b Western blot showing Mettl1 and Wh in Drosophila ovaries. c Western blot showing Mettl1 in Drosophila testes. d Male fertility assay at 25 °C using WT (yw), WT with Mettl1 transgene (yw; [Mettl1]), Mettl1-KO (Mettl1^KO1) and Mettl1-rescue (Mettl1^KO1; [Mettl1]) males. e Female fertility assay at 25 °C using control (Mettl1^KO1/FM7, Mettl1^KO2/FM7), Mettl1-KO (Mettl1^KO1, Mettl1^KO2) and Mettl1-Rescue group (Mettl1^KO1; [Mettl1], Mettl1^KO2; [Mettl1]) females. FM7 is a balancer chromosome. f Male fertility assay at 25 °C using (yw; m + /z + ), and zygotic Mettl1-KO (Mettl1^KO1; m + /z–), maternal and zygotic Mettl1-KO (Mettl1^KO1; m–/z–) males. Each western blots were reproduced three times with similar results (b, c). In box plots of fertility assays, central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. The number (n = ) below the box plot indicates sample size. Upper and lower edges of boxes represent third and first quartiles, respectively (d, e). Two-sided P-values were calculated using Tukey’s HSD test (d, e). Source data of Fig. 1b–e are provided as a Source Data file.

Fig. 2. Loss of Mettl1 results in absence of mature sperm in the testis.

a Schematic diagram of Drosophila spermatogenesis. The genes indicated below are markers used to identify each stage of spermatogenesis. (GSC; germline stem cell, CySC; cyst stem cell) (b) Immunostaining of Mettl1-Flag in the testis. yw is a wild-type control. (Left) Overall view of testis. (Right) Zoom-in view of the white box in panel. c Transparent image of seminal vesicle (arrowheads) from control (yw) and Mettl1-KO (Mettl1^KO1) males. d Representative images of Boule localization in testes from WT (yw) and Mettl1-KO (Mettl1^KO1) adult males. Testes were stained for Boule (green). Dotted line surrounds the region of spermatid elongation defects. e Differential interference microscopy image of spermatid bundles (arrowheads) in control (yw) and Mettl1^KO1 testes. f Immunostaining of acetylated-Tub in WT (yw) and Mettl1-rescue (Mettl1^KO1; [Mettl1]) testes. Arrowheads show elongated spermatids. g Male fertility assay at 25 °C using WT (yw), WT with Mettl1 transgene driven by Nanos promoter (yw; [nosP-Mettl1]), Mettl1-KO (Mettl1^KO1) and Mettl1-KO with Nanos-driven Mettl1 (Mettl1^KO1; [nosP-Mettl1]). In box plots, central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. Two-sided P-values were calculated using Tukey’s HSD test. The number (n = ) below the box plot indicates the sample size. Each immunostaining were reproduced three times with similar results. Source data of Fig. 2g is provided as a Source Data file. Scale bars of b (left), c, d, e, and f = 100 μm, and b (right) = 50 μm.

Fig. 3. Catalytic activity of Drosophila Mettl1 is required for fertility.

a Volcano plot showing enrichment rates and significance levels of each protein as log2 fold change (anti-Flag M2/negative control) versus negative log 10 of the Fisher’s exact test two-sided P-value. The four red dots represent Mettl1 interactors, and the blue dot represents Mettl1 in HEPES-NP40 buffer. b (Top) Schematic diagram showing the positions of the amino acid residues important for methylation activity of Mettl1. Mettl1 harboring L157A and D160A mutations is termed Mettl1-Cd. Blue box indicates catalytic domain of Mettl1. (Bottom) In vitro m⁷G modification assay for testing Mettl1-Cd activity. The m⁷G containing tRNAs were visualized by ¹⁴C. Toluidine blue O staining was performed to visualize RNAs. MBP (negative control) or MBP-fused protein was incubated with the substrate tRNA TrpCCA and GST-Wh. c Male fertility assay for WT (yw), Mettl1-KO (Mettl1^KO1), and transgenes, [Mettl1] and [Mettl1-Cd], in Mettl1^KO1 background. In box plots, central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. The number (n = ) below the box plot indicate the sample size. Two-sided P-values were calculated using Tukey’s HSD test. Source data of Fig. 3b, c are provided as a Source Data file.

Fig. 4. Drosophila Mettl1 methylates tRNA in gonads.

a Schematic diagram of m⁷G site-specific reduction and cleavage. Cleaved 3′ fragments were detected by northern blotting. b Northern blot of chemically-treated total RNAs from wild-type control (yw) and Mettl1-KO (Mettl1^KO1) testes. The probe was designed at the 3′ end of TrpCCA. Arrowhead denotes the position of the cleaved 3′ fragment. c Northern blot of chemically-treated total RNAs from WT (FM7c) and Wh-KO (Wh⁵⁶) testes. The probe was designed around the 3′ end of tRNA TrpCCA. Arrowhead denotes the position of the cleaved 3′ fragment. d TRAC-seq scheme using Drosophila gonads. e Mettl1-dependent m⁷G modified tRNAs identified in testes by TRAC-seq. f Sequence motif of m⁷G modification sites identified in testes by TRAC-seq. The arrowhead corresponds to the m⁷G site. g Representative plot of the cleavage score (difference between cleavage score of WT and Mettl1-KO) of tRNA CysGCA (Cys-GCA-2) between WT and Mettl1-KO, which showed the highest score difference by TRAC-seq. Pink shading represents the variable loop region of tRNA CysGCA. h Quantitative comparison of cleavage scores (CleavageScore) of identified m⁷G-modified tRNAs (n = 36) between WT and Mettl1-KO testes. yw is treated as a wild-type control. Two-sided P-value was calculated by the Mann–Whitney U-test are indicated. Central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. In our TRAC-seq, 3 biological replicates were set for each sample. Each northern blot was reproduced for three times (Fig. 4b, c). Source data of Fig. 4b, c, g and h are provided as a Source Data file.

Fig. 5. Drosophila Mettl1 is required to maintain tRNA abundance in the testis.

a Volcano plot representing changes in tRNA abundance between Mettl1-KO (Mettl1^KO1) and wild-type (yw) testes. Each dot shows the change in abundance of one tRNA. The two-sided P-values were calculated by applying the Wald test with the Benjamini-Hochberg correction implemented in DEseq2 module of the tRAX pipeline. b Heatmap showing changes in m⁷G-modified tRNA abundance between Mettl1-KO (Mettl1^KO1) and WT testis. Drosophila tRNAs were classified into three groups according to the difference in abundance between Mettl1-KO (Mettl1^KO1) and wild-type (WT), and P-values indicate significant differences; group 1 (m⁷G-modified tRNAs and significantly decreased abundance, log2FC < 0, P < 0.05); group 2 (other m⁷G-modified tRNAs, P ≥ 0.05); group 3 (non-m⁷G tRNAs). The P-values indicated correspond to those of (a). c Quantitative comparison of changes in tRNA abundance between the three groups of tRNAs (Group1: n = 15, Group2: n = 8, Group3: n = 22) defined in this study. Two-sided P-values were calculated by the Mann–Whitney U-test are indicated. Box plots represent maximum, median and minimum values with outliers. Central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. In our tRNA expression analysis, 3 biological replicates were set for each sample. d Northern blot analysis of tRNA abundance in Mettl1-KO (Mettl1^KO1) and WT (yw) testes. 2S rRNA was used as a loading control. Each northern blot was reproduced for three times. Source data of Fig. 5 are provided as a Source Data file.

Fig. 6. Ribosome profiling in Mettl1^KO1 testes.

a Metagene plot for ribosome footprints around start codons. The 5′ ends of 29 nt reads are depicted. Pearson’s correlation coefficients (r), ribosome footprints (b), and RNA-Seq reads (c) on each ORF. The color scales for r are represented. d Ribosomal occupancy analysis across A-site codons in the Mettl1-KO testis. Source data is provided as a Source Data file. e Ribosome occupancy analysis for the codons decoded by the indicated tRNA groups (Group1: n = 15, Group2: n = 8, Group3: n = 22). Central bands, upper and lower edges of box plots represent median, first and third quartiles, respectively. Upper and lower whiskers represent maximum, and minimum values. f Metagene plot for ribosome occupancy around CCA codons. The positions of paused and collided ribosomes are highlighted by arrows. g Differences in ribosome occupancy changes on A-site codons in Mettl1-KO testes between all analyzed genes and genes that are essential for fertility (Supplementary Table S1).

Fig. 7. Analysis of TE in testes and comparative analysis with non-gonadal tissues.

a MA (M, log ratio; A, mean average) plot showing the fold change of translation efficiency by Mettl1-KO and the averaged read number. Each dot represents one transcript. Significantly changed transcripts (up-regulated, log₂ fold change ≥ 1 and FDR ≤ 0.05; down-regulated, log₂ fold change ≤ −1 and FDR ≤ 0.05) are highlighted. b Fraction of codons decoded by group 1 tRNAs in the indicated transcript group. Two-sided P-value was calculated using Pearson’s χ² test. c Gene ontology analysis of mRNAs with TE down-regulated genes in Mettl1-KO testes (P < 0.01). The color scales for significance and the size scales for the number of genes in each category are shown. One-sided P-values were calculated by Fisher’s exact test. d Comparison of Dj-GFP expression between WT (w; Dj-GFP) and Mettl1-KO (Mettl1^KO1; Dj-GFP) testes. Dotted line surrounds the outline of testis. Observation for each sample was reproduced for three times. Scale bars is 100 μm. e Northern blots of the chemically-treated total RNAs from testis and male body (testis removed) samples. The probe was designed at the 3′ end of TrpCCA. Arrowheads denote the position of the cleaved 3′ fragment. f Northern blot analysis of tRNA ProTGG abundance in WT (yw), Mettl1-KO (Mettl1^KO1) and Mettl1-KO with Nanos-driven Mettl1 (Mettl1^KO1; [nosP-Mettl1]) testes. 2S rRNA was used as a loading control. Each northern blot was reproduced for three times. Source data of Fig. 7a, e and f are provided as a Source Data file.