Adenosine Methylation in Arabidopsis mRNA is Associated with the 3' End and Reduced Levels Cause Developmental Defects

Abstract

We previously showed that the N6-methyladenosine (m(6)A) mRNA methylase is essential during Arabidopsis thaliana embryonic development. We also demonstrated that this modification is present at varying levels in all mature tissues. However, the requirement for the m(6)A in the mature plant was not tested. Here we show that a 90% reduction in m(6)A levels during later growth stages gives rise to plants with altered growth patterns and reduced apical dominance. The flowers of these plants commonly show defects in their floral organ number, size, and identity. The global analysis of gene expression from reduced m(6)A plants show that a significant number of down-regulated genes are involved in transport, or targeted transport, and most of the up-regulated genes are involved in stress and stimulus response processes. An analysis of m(6)A distribution in fragmented mRNA suggests that the m(6)A is predominantly positioned toward the 3' end of transcripts in a region 100-150 bp before the poly(A) tail. In addition to the analysis of the phenotypic changes in the low methylation Arabidopsis plants we will review the latest advances in the field of mRNA internal methylation.

Keywords: IME4; METTL3; MT-A70; mRNA methylation; post-transcriptional.

Figure 1

Confirmation of the homozygous nature of the SALK_074069 insertion in plants containing the ABI3MTA complementation cassette. (A) Structure of the MTA locus showing the location of the SALK_0749069 insertion (a rearranged T-DNA insertion line with two left borders) and the oligonucleotides used in the analysis. (B) The presence of the SALK_074069 insertion was confirmed by the presence of a band of 1050 bp following PCR amplification with oligonucleotides a (Lba1) and c (A63UTRR) (lanes 1–4). The homozygosity of the SALK insertion was confirmed by the lack of the 1080 bp wild type band seen after amplification with oligonucleotides b (A6E4F) and c (A63UTRR) (lanes 6–9). Lanes 1 and 6 wild type, 2 and 7 Line 8, 3 and 8 Line 10, 4 and 9 Line 14.

Figure 2

Homozygous SALK_074069 plants complemented with the ABI3MTA transgene have reduced mRNA methylation and associated developmental defects. Samples were collected from three different plants from Line 8 transgenic plants. m6A measurements were carried out using the TLC method. Error bars represent SD of three replicates. Levels of m6A within mRNA is reduced by more than 90% in the leaves and flowers of the SALK_074069 ABI3MTA transgenic lines (A). Plants with reduced levels of m6A are more compact, have leaf crinkling, shorter inflorescence, and reduced apical dominance (B). Floral defects are common in the reduced m6A plants. Some stamens show partial conversion to petals [(i,ii), sepals and petals removed for clarity] and organ order is also sometimes affected (iii) (C).

Figure 3

Altered levels of trichome branching in 3-week-old seedlings. The third and fourth leaves from 3-weeks-old seedlings of wild type and homozygous SALK_074069 ABI3MTA plants were examined for changes in their trichome phenotypes. Light micrographs of trichomes from plants with low m6A levels and from wild type plants (A). The low m6A plants are characterized by a higher proportion of trichomes with four or more branches. Trichomes from the abaxial side were analyzed for the third and fourth leaves from four wild type and four low m6A plants, each from different transgenic lines (554 and 521 trichomes respectively). Sixty-three percentage of trichomes from low methylation plants had four or more branches compared to just 24% in wild type (B). Error bars represent SD of the replicates.

Figure 4

Venn diagram showing overlapping expression changes of genes between low methylation plants and a trichome specific set. For the overlap search all genes with significant change were used from the low methylation versus wild type dataset (1537 genes). The central region corresponds to genes with changed expression in both trichomes and low methylation plants. The gene list is in Table S5 in Supplementary Material.

Figure 5

N6-methyl adenosine is enriched in the 3′ end of transcripts. Fragmented high purity mRNA was fractionated into 3′, middle + 5′ and 5′ regions and the m⁶A content was measured for each fraction using the published TLC based method. The positions of m⁶A, A and C are indicated in the left hand panel. The spot corresponding to m⁶A is enriched on the TLC carried out on the 3′ end, compared to the 5′ and 5′ + middle samples (A). (The example is from mRNA fragmented to 180 nt). The graph shows the distribution of m⁶A for three different fragmentation experiments (B). When the fragment size is 90 nt or less, m⁶A enrichment associated with polyadenylated fragments is no longer seen.

Figure 6

Venn diagrams representing overlapping genes between the list of cDNAs with GGACU in the last 200 nt of the 3′ UTR and the differentially expressed genes in the low methylation plants. Ninety-nine Genes from the low methylation plants had at least one GGACU in the last 200 nt end of their 3′ UTR. Genes with no annotated 3′ UTR were not analyzed.

Figure A1

Measuring levels of MTA in ABI3 promoter driven transgenic plants. Reverse transcription was carried out using SuperScript II reverse transcription kit (Invitrogen). Real-time PCR was carried out using the MX3005P qPCR machine and the Brilliant SYBR Green qPCR master mix (Stratagene). MAXPro software was used for data analysis. Primers used were, MTA primers: 5′-GGAACCTTTGGAGTTGTTATG-3′ and 5′-CAAAGCTCCAAACATTCACG-3′, and the β-actin2 (normalizer gene) primers: 5′-GTACAACCGGTATTGTGCT-3′ and 5′-ATCAGTAAGGTCACGTCCA-3′. Total RNA was isolated from young buds of three wild type plants, and three MTA homozygous T-DNA insertion lines complemented with the ABI3 promoter driven construct. The error bars are representing SD of three replicates.

Figure A2

Quality control of fragmented mRNA. The poly(A) RNA used for the fragmentation experiments was purified using oligo(dT) at least three times, the quality of the mRNA was confirmed by RNA6000 LabChip (Agilent) (A). After fragmentation the average fragment size was confirmed by RNA6000 LabChip (Agilent) (B). A dot-blot was used for confirming that the fragments after fractionation were enriched in the predicted parts of the mRNA pool. Probes were made from the 5′, the middle and the 3′ regions of the ATPC mRNA and it was hybridized to membrane bound RNA pools of the different fractionated fragments (C).