CMTr cap-adjacent 2′-O-ribose mRNA methyltransferases are required for reward learning and mRNA localization to synapses

Abstract

Cap-adjacent nucleotides of animal, protist and viral mRNAs can be O-methylated at the 2‘ position of the ribose (cOMe). The functions of cOMe in animals, however, remain largely unknown. Here we show that the two cap methyltransferases (CMTr1 and CMTr2) of Drosophila can methylate the ribose of the first nucleotide in mRNA. Double-mutant flies lack cOMe but are viable. Consistent with prominent neuronal expression, they have a reward learning defect that can be rescued by conditional expression in mushroom body neurons before training. Among CMTr targets are cell adhesion and signaling molecules. Many are relevant for learning, and are also targets of Fragile X Mental Retardation Protein (FMRP). Like FMRP, cOMe is required for localization of untranslated mRNAs to synapses and enhances binding of the cap binding complex in the nucleus. Hence, our study reveals a mechanism to co-transcriptionally prime mRNAs by cOMe for localized protein synthesis at synapses.

Introduction

Methylation of cap-adjacent or internal nucleotides in messenger RNA (mRNA) is a major post-transcriptional mechanism to regulate gene expression. Methylation of mRNA is particularly prominent in the brain, but the molecular function of methylated nucleotides and their biological roles are poorly understood^1,2,3,4,5.

Methylation of cap-adjacent nucleotides is an abundant modification of animal, protist, and viral mRNAs, that varies in different tissues and transcripts^{6,7,8,9,10,11,12,13,14,15,16,17}. The most common methylation of cap-adjacent nucleotides is O-methylation at the 2′ position of the ribose (cOMe). This modification is introduced co-transcriptionally by two dedicated cap methyltransferases (CMTr1 and CMTr2) after capping at the beginning of an mRNA to a characteristic 5′–5′ linked N7-methylated guanosine^18,19,20. Knock-out of CMTr1 leads to neurological defects in mice and is embryonic lethal, while Drosophila CMTr1 null is viable, but has minor defects in siRNA-mediated gene silencing. It has been postulated that CMTr1 methylates the first and CMTr2 the second nucleotide in humans^6,20, while in trypanosomes the three CMTrs methylate the first four nucleotides²¹, but the unequivocal determination of cOMe on other than the first position remains technically challenging^11,22,23.

In vertebrates, if the first nucleotide is adenosine it can also be methylated at the N6 position by PcifI, but the mechanism for cap adenosine N6-methylation is different from internal methylation of adenosine and requires the prior cOMe modification^{9,10,11,12,13,17,24,25,26}.

The main function of the cap is to protect mRNAs from degradation and to recruit translation initiation factors, but also to promote splicing and 3‘ end processing²⁷. The cap is initially bound in the nucleus by the cap-binding complex (CBC), consisting of CBP20 and CBP80. Upon export from the nucleus, CBC is replaced by eIF4E, which is predominantly cytoplasmic and rate-limiting for translation initiation^28,29. N7-methylation of the cap guanosine is critical for both CBC and eIF4E binding. The importance of cap-adjacent nucleotide methylation in animal gene expression, however, remains elusive but is known to be essential in trypanosomes and viruses including SARS-CoV-2 for propagation¹⁵.

Results

CMTrs act redundantly

To elucidate the biological function of cap-adjacent 2′-O-ribose methylation (cOMe) in animals we made null mutants of the CMTr1 (CG6379) and CMTr2 (adrift) genes in Drosophila, that are corresponding homologs of human CMTr1 and CMTr2 (Supplementary Fig. 1). We generated small intragenic deletions in each gene by imprecise excision of a P-element transposon to make CMTr1^13A and CMTr2^M32 mutant flies (Fig. 1a–c). Both of these genetic lesions remove the catalytic methyltransferase domain from the encoded CMTr1 and CMTr2 proteins. Perhaps surprising, these mutant flies are viable and fertile as single and double mutants, exhibiting a slightly reduced survival to adulthood after hatching from the egg (Fig. 1d), and reduced climbing activity in negative geotaxis assays (Fig. 1e). In addition, CMTr1 mutants, and to a greater extent CMTr2 mutants, have reduced numbers of synapses at neuromuscular junctions (NMJs) of third instar larvae (Fig. 1f).

To detect cOMe in purified mRNAs we replaced the cap guanosine with a ³²P-alphaGTP by first decapping mRNAs with yDcpS that leaves a di-phosphate at the first nucleotide, which is the substrate for vaccinia capping enzyme (Supplementary Fig. 2a). Digestion of such labeled mRNA with RNAse I, which is unable to cleave after 2′-O-ribose methylated nucleotides result in unmethylated ^m7GpppA di-nucleotide, and 2′-O-ribose methylated tri- and tetra nucleotides that can be analyzed on 20% denaturing acrylamide gels (Fig. 1g). In adult flies, about 80% of mRNAs carry cOMe on the first nucleotide, but we could not detect methylation of the second nucleotide as compared to a single nucleotide ladder and appropriate markers (Fig. 1g, h). In CMTr1, but not CMTr2 mutants, cOMe levels on the first nucleotide are strongly reduced, but still detectable indicating that CMTr1 is the main methyltransferase and that only CMTr1/2 double knock-out flies completely lack cOMe at the first nucleotide (Fig. 1g, h).

To specifically analyze methylation of the first nucleotide in polyA mRNA, we decapped polyA mRNA with the pyrophosphatase RppH and removed the first phosphate for labeling the first nucleotide by ³²P-gammaATP followed by digestion into individual nucleotides by nuclease P1 (Supplementary Fig. 2a) and separation on 2D thin-layer chromatography (TLC) (Fig. 1i). In S2 cells, we detected cOMe on adenosine (pAm) and cytosine (pCm, Fig. 1j, k), but in female flies predominantly pAm was present (Fig. 1k). By omitting decapping, residual rRNA in the polyA mRNA preparation was analyzed and this RNA does not show cOMe (Supplementary Fig. 2b). Gm runs at the same position as C, and thus can not be distinguished (Supplementary Fig. 2c) and Um runs at the same position as dT, which can be carried over as a contaminant of oligo dT purification (Fig. 1i)⁹. Although single mutants in CMTr1 or CMTr2 still had cOMe, the double mutants were devoid of cOMe further suggesting that these two enzymes have overlapping functions and are both able to methylate the 2′-O-ribose of the first transcribed nucleotide (Fig. 1l-n).

Our TLC analysis of the first nucleotide of mRNAs in Drosophila shows a strong preference for A from quantification of the four nucleotides in CMTr1/2 double knock-out flies (Fig. 1o). We validated the accuracy of the TLC data by analyzing CAGEseq from Drosophila. The CAGEseq data corroborated observations from the TLCs, demonstrating a strong preference for A as the first nucleotide in Drosophila mRNA (Fig. 1o) which is further consistent with the transcription initiator motif (Inr) sequence YYANWYY (Y: pyrimidine, N: any nucleotide and W: A or T) containing one A in the consensus sequence³⁰.

To further test that both Drosophila CMTrs can methylate the first nucleotide, we expressed Drosophila CMTr1 and CMTr2 in Drosophila S2 cells (Supplementary Fig. 2d). Both Drosophila CMTrs show equal activity in methylating the first nucleotide in vitro using a ³²P-GTP capped RNA substrate with the consensus start sequence AGU after digestion with RNase I as judged by comparison to a single nucleotide ladder and appropriate markers (Supplementary Fig. 2e). Likewise, when the first nucleotide from this RNA (lanes 8 and 9) is labeled, only pAm is detected after digestion with nuclease P1 on 2D TLCs, which confirms that the first nucleotide of the substrate is A (Supplementary Fig. 2f).