The Arabidopsis m6A-binding proteins ECT2 and ECT3 bind largely overlapping mRNA target sets and influence target mRNA abundance, not alternative polyadenylation

Gene regulation via N6-methyladenosine (m6A) in mRNA involves RNA-binding proteins that recognize m6A via a YT521-B homology (YTH) domain. The plant YTH domain proteins ECT2 and ECT3 act genetically redundantly in stimulating cell proliferation during organogenesis, but several fundamental questions regarding their mode of action remain unclear. Here, we use HyperTRIBE (targets of RNA-binding proteins identified by editing) to show that most ECT2 and ECT3 targets overlap, with only few examples of preferential targeting by either of the two proteins. HyperTRIBE in different mutant backgrounds also provides direct views of redundant and specific target interactions of the two proteins. We also show that contrary to conclusions of previous reports, ECT2 does not accumulate in the nucleus. Accordingly, inactivation of ECT2, ECT3 and their surrogate ECT4 does not change patterns of polyadenylation site choice in ECT2/3 target mRNAs, but does lead to lower steady state accumulation of target mRNAs. In addition, mRNA and microRNA expression profiles show indications of stress response activation in ect2/ect3/ect4 mutants, likely via indirect effects. Thus, previous suggestions of control of alternative polyadenylation by ECT2 are not supported by evidence, and ECT2 and ECT3 act largely redundantly to regulate target mRNA, including its abundance, in the cytoplasm.

tissues, and there was a clear correlation between the editing proportions of the common editing 135 sites, albeit with higher editing by ECT2-FLAG-ADAR overall (Figure 2A,B). Indeed, the pattern of 136 editing sites resulting from fusion of ADAR to ECT2 or ECT3 was similar for many targets, with a 137 few more sites typically detected in the ECT2-HT dataset (e.g. ATP-Q, Figure 2C left panel).

140
UBQ6, Figure 2C right panel), perhaps hinting to molecular explanations for the recently described Although the demonstration that ECT2 and ECT3 bind to strongly overlapping target sets is 153 consistent with largely redundant in vivo function, it does not constitute a direct proof. For example, 154 the proteins may bind to the same targets, but in different cells such they act de facto non-155 redundantly. We reasoned that HyperTRIBE might provide a means to observe directly whether 156 ECT2 and ECT3 act specifically or redundantly on shared targets, and whether one ECT protein  Although the tendency of ECT2/3 to show redundant target mRNA interactions was widespread, we 184 also looked for examples of specific interactions in the HyperTRIBE data in single and triple 185 mutants. A priori, we considered targets to be ECT2-specific if they were detected by ECT2-FLAG-

186
ADAR, but not ECT3-FLAG-ADAR, in single mutant backgrounds (strictly specific), or became 187 edited by ECT3-FLAG-ADAR only in the triple mutant background. The definition of ECT3-specific 188 targets followed analogous criteria. However, because ECT2 expression is much higher than ECT3 189 expression ( Figure 3-figure supplement 1), ECT2-specific targets identified in this way may simply 190 be below the detection limit of the less highly expressed ECT3-FLAG-ADAR transgene. Hence,

191
arguments for existence of bona fide specific targets must take detectability by ECT3-FLAG-ADAR 192 into account. Consistent with expectation from the different ECT2/ECT3 dosage, much larger 193 numbers of strictly ECT2-specific transcripts were identified compared to ECT3: 2,414 ECT2-194 specific and 93 ECT3-specific targets were identified in aerial tissues, while in roots, 1,738 were 195 ECT2-specific and 197 were ECT3-specific ( Figure 3D,E, Figure 3-figure supplement 2). In 196 addition, small sets of specific target mRNAs became targets of the other ECT protein upon 197 knockout of its genuine interacting protein (110 and 24 for ECT2-specific targets in aerial and root 198 tissues respectively, and 2 for ECT3-specific targets in roots) ( Figure 3D,E, Figure 3- figure   199 supplement 2). These sets constitute outstanding candidates for ECT2/3-specific mRNA targets.
Curiously, a few transcripts (21 in aerial tissues and 9 in roots) were edited by either ECT2 or ECT3 201 only in the triple mutant background ( Figure 3D,E, Figure 3-  supplement 2). We also analyzed the permissive target sets for groups of functionally related 222 genes, and found that ECT2/3 targets are enriched in housekeeping genes, many related to basic 223 metabolism and protein synthesis ( Figure 4C). These initial analyses provide well-defined common  to be isolated prior to transcriptome analyses to avoid confounding effects from cells that do not 231 express these m 6 A readers. We therefore used the fact that ect2-1/ECT2-mCherry exhibits root 232 growth rates similar to wild type while te234/ECT2 W464A -mCherry exhibits clearly reduced root            to be different in ECT2/3 targets than in non-targets ( Figure 5F). In fact, the most common PAS is 276 more likely to be unchanged in targets than in non-targets ( Figure      signal, probably due to lack of perpendicularity between the nuclear envelope and the optical 301 section in these areas. In such cases, the cytoplasm, nucleus and nuclear envelope may be 302 contained in the same region of the optical section and thus appear to be overlapping ( Figure

325
while the majority of differentially expressed non-targets were up-regulated compared to wild type 326 ( Figure 7D). Furthermore, ECT2/3 targets accounted for more than half of all significantly 327 downregulated genes, but only about 15% of upregulated genes ( Figure 7E). In contrast, highly 328 upregulated genes tended to be non-targets ( Figure 7B, right panel).

331
To test if these differentially regulated gene sets represented subsets of functionally related genes 332 within target and non-target groups, we analyzed their potential enrichment of GO terms. This Log 2 (TPM+1) ect2-1 ECT2-mCh Perm.

426
All data analyses were carried out using TAIR 10 as the reference genome and Araport11 as the 427 reference transcriptome. Unless otherwise stated, data analyses were performed in R

462
"Technical replicates" are understood to be independently conducted measurements using the 463 same technique on the same biological material (e.g. on one biological replicate as defined above).

464
Technical replicates were not carried out in this study, and the term "replicate" refers to biological 465 replicate as defined above.
Aerial Root             Uniquely mapped reads  40  30  20  10  0  Million reads  Aerial tissue   Roots  Aerial tissue   Roots   40  30  20  10  0  Million reads  40  30  20  10  0  L1  L2  L3  L4  L5  L1  L2  L3  L4 L5   Figure 3D,E).  Pseudotime  targets to have a different dominant PAC upon loss of ECT2/3/4 function depletion compared to non targets ( Figure 5E) could be due to differences in transcript abundance between the target and non-target groups. Looking at only the 2200 most highly expressed non-target genes, only 5.5% of these genes have a different dominant PAC in te234 ECT2 W464A -mCherry than ect2-1 ECT2-mCh samples (lower panel, dark shading refers to genes with different dominant PAC as in Figure 5E), significantly smaller than the percentage for all non-target genes (20.8%, Figure 5E) (p=3.2e-9, Fisher's exact test).    Figure 6B .