Zc3h13/Flacc is required for adenosine methylation by bridging the mRNA-binding factor Rbm15/Spenito to the m6A machinery component Wtap/Fl(2)d

Abstract

N⁶-methyladenosine (m⁶A) is the most abundant mRNA modification in eukaryotes, playing crucial roles in multiple biological processes. m⁶A is catalyzed by the activity of methyltransferase-like 3 (Mettl3), which depends on additional proteins whose precise functions remain poorly understood. Here we identified Zc3h13 (zinc finger CCCH domain-containing protein 13)/Flacc [Fl(2)d-associated complex component] as a novel interactor of m⁶A methyltransferase complex components in Drosophila and mice. Like other components of this complex, Flacc controls m⁶A levels and is involved in sex determination in Drosophila We demonstrate that Flacc promotes m⁶A deposition by bridging Fl(2)d to the mRNA-binding factor Nito. Altogether, our work advances the molecular understanding of conservation and regulation of the m⁶A machinery.

Keywords: Flacc; RNA modifications; Zc3h13; m6A; methyltransferase complex; sex determination.

Figure 1.

Zc3h13/Flacc interacts with the m⁶A machinery. (A) TAP-LC-MS/MS of endogenously Flag-Avi-tagged Rbm15 mESCs. Parental cells were used as background control, and proteins were purified in the presence of 350 mM NaCl. Highlighted in the volcano plot are enriched proteins previously identified as Mettl3 interactors (red) as well as Zc3h13 (green). (B) Heat map comparing relative label-free quantification (LFQ) intensities of selected Mettl3-bound proteins across increasing NaCl concentrations. Statistical analysis was done with Perseus (see the Materials and Methods for details). MS raw data were deposited in ProteomeXchange. (C,D) Stable isotope labeling of amino acids in cell culture (SILAC) coupled to MS analysis using Nito-Myc as bait. Scatter plot of normalized forward versus inverted reverse experiments plotted on a log₂ scale. The threshold was set to a 1.5-fold enrichment (red dashed line). Proteins in the top right quadrant of C are enriched in both replicates. Gene ontology (GO) term analysis (Tyanova et al. 2016) for enriched proteins is shown in D. (E) Coimmunoprecipitation experiments were carried out with lysates prepared from S2R⁺ cells transfected with FlagMyc-Flacc and HA-Nito. In control lanes, S2R⁺ cells were transfected with FlagMyc alone and an identical HA-containing protein. Extracts were immunoprecipitated with Myc antibody and immunoblotted using Flag and HA antibodies. Two percent of input was loaded. The same experiment was repeated in the presence of RNaseT1. Nito and Flacc interact with each other in an RNA-independent manner. (F) Table representing orthologous proteins of the m⁶A–METTL complex (MAC) and the m⁶A-METTL-associated complex (MACOM) in mice and flies.

Figure 2.

Flacc/Zc3h13 regulates the m⁶A pathway. (A) LC-MS/MS quantification of m⁶A levels in mRNA extracts from wild-type mESCs, Mettl3 knockout and Mettl3 knockout plasmid rescue, and Zc3h13 knockout cells. The mean of two biological replicates and three independent measurements is shown. Errors bars indicate standard deviation (SD). (*) P < 0.01; (n.s.) not significant, Student's t-test. (B) University of California at Santa Cruz genome browser shots of m⁶A immunoprecipitation profiles of RNA isolated from Mettl3 knockout, Zc3h13 knockout, and wild-type cells and input samples for each genetic background at the Wtap-encoding locus. Scale is mapped reads in 100-base-pair (bp) bins normalized to mean library size. (C) Metaplot depicting reads from m⁶A immunoprecipitations at target genes (defined as genes overlapping or within 500 bp of MACS-identified peaks of m⁶A immunoprecipitation/input in wild-type cells) aligned to the coding sequence (“start” and “stop” refer to translation start and stop, respectively). (D) LC-MS/MS quantification of m⁶A levels in either control samples or mRNA extracts depleted for the indicated proteins in S2R⁺ cells. The bar chart shows the mean of three biological replicates and three independent measurements. Errors bars indicate SD. (*) P < 0.01, Student's t-test. Knockdown of the indicated proteins significantly reduces m⁶A levels. (E) Fold enrichment of m⁶A-regulated transcripts (Aldh-III and Dsp1) over input in Myc-Ythdc1 RIP after control or Flacc depletion. The bar chart shows the mean of three biological replicates. Errors bars indicate SD. (*) P < 0.01; (**) P < 0.001, Student's t-test. Loss of Flacc affects Ythdc1 binding. (F) Relative isoform quantification of m⁶A-regulated genes (Aldh-III, Hairless, and Dsp1) upon depletion of the indicated components. Flacc is required for m⁶A-dependent splicing events.

Figure 3.

Flacc regulates common transcripts with other components of the m⁶A complex. (A) Number of differentially expressed genes (5% false discovery rate [FDR]) upon knockdown of the indicated proteins (left) and common differentially expressed targets regulated by components of MACOM (right). (B) Scatter plot of the first two principal components of a principal component analysis (PCA) of the 500 most variable genes in all conditions. The biological replicates are indicated in the same color, with elliptical areas representing the SD of the two depicted components. (C) Gene length distribution for genes tested in the differential expression analysis and the differential expressed genes up-regulated or down-regulated in all conditions. The distributions were tested for difference using the Kolmogorov-Smirnov test. (D) Overlap between common up, down, or all differentially expressed genes and genes annotated to have m⁶A-modified transcripts (according to methylation individual nucleotide resolution cross-linking immunoprecipitation data from Kan et al. 2017). The significance of the overlap was tested using a hypergeometric test. (E) Fold change (log₂) expression of commonly misregulated genes. The heat map is clustered according to rows and columns. The color gradient was adjusted to display the 1% lowest/highest values within the most extreme color (lowest values as the darkest blue and highest values as the darkest red). (F) The GO term analysis of common up-regulated and down-regulated genes performed using the package ClusterProfiler. The top 10 GO terms are displayed.

Figure 4.

Flacc is required for sex determination via control of Sxl alternative splicing. (A,B) dome-GAL4-driven expression of shRNA or dsRNA in genital discs and first pair of leg discs against Nito or Flacc, respectively. (Top) Forelegs of a wild-type male fly and female flies depleted for Nito or Flacc show the appearance of male-specific sex comb bristles (red arrow). (Bottom) Depletion of Nito or Flacc results in transformations of female genitalia and loss of vaginal bristles (red arrowhead). (B) Quantification of female survival and transformations in escapers upon depletion of Nito or Flacc using the dome-GAL4 driver. (n) The number of analyzed flies with the expected number of escapers in brackets. Depletion of Nito and Flacc results in a high level of transformation in female genitalia and the appearance of male-specific sex combs on forelegs. (C) Semiquantitative RT–PCR analysis of Sxl isoforms in male and female heads from flies depleted for Fl(2)d, Nito, or Flacc, respectively, using the ElavGAL4 driver. Inclusion of male-specific exon L3 is observed in flies lacking m⁶A components. (D) The flacc locus (flacc^C) with a premature stop codon at amino acid Leu730. Sites of dsRNA fly lines KK110253 and GD35212 are shown below gene loci. (E) Viability of female flies from a cross of the indicated genotypes mated with Sxl^7BO males. The loss of one copy of flacc significantly reduces female survival in a genetic background where one copy of Sxl and da are absent. The same compromised survival is observed for other m⁶A components [Mettl3, Mettl14, Ythdc1, fl(2)d, vir, and nito]. Viability was calculated from the numbers of females compared with males, and statistical significance was determined by a χ² test (Graphpad Prism). (F) The viability of female flies with homozygous vir2F mutation can be rescued by the loss of a single copy of flacc and nito. Viability was calculated from the numbers of homozygous vir2F females compared with heterozygous balancer-carrying siblings, and statistical significance was determined by a χ² test (Graphpad Prism).

Figure 5.

Flacc bridges the methyltransferase complex to mRNA targets via binding to Nito. (A) Coimmunoprecipitation experiments were carried out with lysates prepared from S2R⁺ cells transfected with GFPMyc-Nito and Fl(2)d-HA. In control lanes, S2R⁺ cells were transfected with Myc alone and an identical HA-containing protein. Extracts were immunoprecipitated with Myc antibody and immunoblotted using Myc and HA antibodies. Two percent of input was loaded. The same experiment was repeated in Flacc knockdown conditions. Interaction between Nito and Fl(2)d is strongly reduced upon depletion of Flacc. (B) Fold enrichment of m⁶A-regulated transcripts (AldhIII, Hairless, and Dsp1) over input in Myc-Fl(2)d and Myc-Nito RIP upon depletion of Flacc or in control conditions. The bar chart shows the mean of three biological replicates. Errors bars indicate SD. (*) P < 0.01; (**) P < 0.001; (***) P < 0.0001; (n.s.) not significant, Student's t-test. Loss of Flacc strongly affects Fl(2)d binding and, to a milder extent, binding of Nito to m⁶A-regulated transcripts. (C,D) Coimmunoprecipitation experiments were carried out with lysates prepared from S2R⁺ cells transfected with either FlagMyc-Nito or Fl(2)d-HA. In control lanes, S2R⁺ cells were transfected with FlagMyc alone and an identical HA-containing protein. Extracts were immunoprecipitated with Flag antibody and immunoblotted using Myc and HA antibodies. Two percent of input was loaded. The same experiment was performed upon depletion of Flacc. Human ZC3H13 was transfected in an identical set of experiments. The interaction between Nito and Fl(2)d is strongly reduced upon loss of Flacc (lane 6) but can be rescued upon expression of human ZC3H13 protein (lane 8). Quantification of two replicates is shown in D.

Figure 6.

Zc3h13 stabilizes the interaction between RBM15 and WTAP. (A,B) Comparison of TAP-LC-MS/MS of endogenously Flag-Avi-tagged Rbm15 mESCs in either a wild-type or a Zc3h13 knockout background. Rbm15 and associated proteins were purified in the presence of 350 mM NaCl. (A) Volcano plot showing enriched proteins in wild-type cells (right) versus Zc3h13 knockout cells (left). (B) Table of spectral counts, unique peptides, and percentage coverage of TAP-LC-MS/MS data in A. (C) Split luciferase NanoBiT assay examining the interaction of mouse Rbm15 and Wtap. (Left) Scheme representing luciferase reconstitution upon transfection of large (LgBit) and small (SmBit) NanoLuc subunit fusions and the interaction of Rbm15 and WTAP. (Right) Comparison of Rbm15–Wtap NanoBiT NanoLuc signal in wild-type and Zc3h13 and Mettl3 knockout cells. The mean of three independent experiments, three transfections each, is shown. Errors bars indicate SD. (*) P = 0.026, calculated using two-tailed Student's t-test.

Figure 7.

Schematic representation of the role of the MACOM and the MAC. The MACOM can regulate gene expression in two ways: either on its own (MAC-independent functions) or by interacting with MAC components (m⁶A methylation). Flacc (Zc3h13) is a novel component of the MACOM that stabilizes the interaction between Fl(2)d and Nito (Wtap and Rbm15) proteins, thereby ensuring deposition of m⁶A to targeted transcripts.