COMMENTARY COMMENTARY Role of alternative polyadenylation in epigenetic silencing and antisilencing Liuyin Maa, Cheng Guob, and Qingshun Quinn Lia,b,1 oxidative responses (10–12). Genomewide aKey Laboratory of the Ministry of Education for Coastal and Wetland Ecosystems, College of analyses by deep sequencing have revealed the Environment and Ecology, Xiamen University, Xiamen, Fujian 361102, China; – b that 70 80% of human and plant genes are and Department of Biology, Miami University, Oxford, OH 45056 subject to APA (13–15). The role of such abundant APAs in cellular functions, how- ever, remains largely unknown. Epigenetic marks such as DNA methylation Eulgem (5), anti-silencing 1 (ASI1)by It was previously suggested that histone and histone modifications are widely in- Wangetal.(6),andincrease in bonsai modification might influence poly(A) choice volved in regulating different aspects of methylation 1 (IBM1) by Saze et al. (7). through correlation analysis (16). The signif- developmental and environmental responses What distinguishes the latter group of icance of the work on EDM2 and ASI1 is that (1). Meanwhile, DNA methylation and his- genes from ROS1 is that they seem to be they provide direct genetic evidence demon- tone modification are also used constitutively involved in posttranscriptional regulation strating an involvement of epigenetic marks to silence transposable elements and repeat through alternative polyadenylation (APA). in the selection of APA sites. EDM2, identi- elements (TREs) (2). Such TRE-mediated si- Polyadenylation is an essential mRNA fied as an enhancer of fungal disease resis- lencing should necessarily be limited to the processing step for almost all genes in tance (17), specifically recognizes intronic intended targets only and not spread to ad- eukaryotes (8, 9). The poly(A) site defines heterochromatin in its target genes presum- jacent genes and their regulatory elements. the end of the transcript and thus its 3′- ably through its composite plant homeodo- Higher eukaryotic organisms have evolved UTR. However, an alternate site of process- main (PHD) domain (4). This binding of antisilencing mechanisms to keep the bal- ing and poly(A) addition would create a EDM2 somehow blocks the poly(A) site ance between silencing and antisilencing different transcript by an inclusion or ex- choice of the transcript in the vicinity of that is required for precise gene expression clusion of certain RNA sequences such as the binding location. This is deduced from regulation. Earlier work revealed that re- the miRNA target site, a recognition site of thefactthatintheedm2 mutant, use of pressor of silencing 1 (ROS1)isonesuch an mRNA stability factor, or a translational intronic poly(A) sites [so called proximal antisilencing gene (3). Recently, a group of repressor. In recent years, APA has been poly(A) sites, in contrast to the full-length unique antisilencing genes has been dis- shown to change the fate of a cell or a de- distal poly(A) sites] results in the generation covered, including enhanced downy mil- velopmental process and to alter cellular of shortened transcripts that encode no or dew 2 (EDM2) reported both by Lei et al. responses to the environment, including nonfunctional proteins (4, 5). In other words, in PNAS (4) and earlier by Tsuchiya and tumorigenesis, flowering time control, and the normal function of EDM2 is to reduce the production of nonfunctional transcripts and to promote the generation of full-length mRNA ending at the distal poly(A) site. The net result is the avoidance of reduced gene expression caused by silencing due to TREs in the intron (hence antisilencing). The antisilencing phenomena of EDM2 were clearly demonstrated in the APA of a group of genes with large TRE-containing genes such as the fungal disease resistance gene resistance to Peronospora parasitica (RPP7)(5)andIBM1 (4). Interestingly, an- other gene (ASI1) was also found to have very similar function of generating APA of IBM1 transcripts (6). Adding an additional layer of complexity to this story is the fact that IBM1 itself is a histone H3K9 demethylase that pre- vents plant-specific CHG methylation (7). Fig. 1. Hypothetic model for antisilencing regulation of alternative polyadenylation. In WT Arabidopsis, EDM2 and ASI1 bind to the intronic heterochromatin region by recognizing enriched epigenetic marks. With the assistance of an Author contributions: L.M., C.G., and Q.Q.L. wrote the paper. unknown factor X, the three proteins form a complex that potently masks the poly(A) signals from being recognized The authors declare no conflict of interest. by the polyadenylation apparatus riding on the C-terminal domain of RNA polymerase II. Consequently, the distal poly(A) site is used. In the edm2 or asi1 mutant, absence of either EDM2 or ASI1 reduces the masking effect on the proximal See companion article on page 527. poly(A) site.The result is an extensive use of the proximal site, producing shortened transcripts encoding nonfunctional 1To whom correspondence should be addressed. E-mail: liq@ proteins, leading to a silencing effect of the target gene. miamioh.edu. www.pnas.org/cgi/doi/10.1073/pnas.1321025111 PNAS | January 7, 2014 | vol. 111 | no. 1 | 9–10 Downloaded by guest on September 27, 2021 Moreover, genomewide epigenetic modifica- of disease resistance conferred is tightly of intronic poly(A) sites in Arabidopsis and tion profiles of all three mutants (asi1, associated with its transcript level (5). rice (14, 15). Further analysis of these intronic edm2,andibm1)largelyoverlap,which Could the APA just be a byproduct of sites would broaden our view on the in- indicates that they work together through other modes of regulation? This possibil- terrelationship of RNA processing and a similar antisilencing pathway. In fact, it ity is very unlikely. There exist other epigenetic regulation. was suggested that both ASI1 and EDM2 examples showing that APA is a predom- The connection between gene silencing function mostly through IBM1 by modulat- inant way to attenuate the expression of and polyadenylation was demonstrated by ing APA of IBM1 transcripts (4, 5). Herr et al. (21) where mutations of poly(A) The question is how are these epigenomic The significance of the protein factors led to enhanced silencing of marks sensed by the polyadenylation ma- work on EDM2 and ASI1 some genes. This work is indicative that chinery so that a different poly(A) site is normal functions of these polyadenylation chosen? Polyadenylation is a cotranscriptional is that they provide factors prevent silencing of the tested genes, event where numerous factors affecting tran- direct genetic evidence an equivalent to an antisilencing effect. While scription would impact polyadenylation (18). demonstrating an there was not clear demonstration of the The polyadenylation machinery is composed of 20–25 different proteins in plants that rec- involvement of involvement of TREs, this is further evidence ognize a set of poly(A) signals on the pre- epigenetic marks in the that motivates investigations to elucidate mRNA (9). The use of one poly(A) site selection of APA sites. a potential direct engagement of poly(A) over the other is likely owing to the relative factors in antisilencing. Meanwhile, many interesting questions strength of the polyadenylation signal being genes. One involves the flower control locus remain. What are the specific functions of perceived by poly(A) machinery, as well as A (FCA) gene, a regulator of flower locus C the availability of the site. It is also possible (FLC) expression, and of flowering time in ASI1 and EDM2 during this process? If any, that associated protein(s) may recruit (or re- Arabidopsis (12). Another is the oxidative what are other factors involved? Is binding pel) the polyadenylation apparatus close to tolerant 6 (OXT6) gene in Arabidopsis that of ASI1 or EDM2 to intronic heterochroma- a particular site. It is equally possible that a encodes two proteins: a smaller one pro- tin region needed to influence poly(A) site protein could inhibit the function of the duced through use of an intronic poly(A) choice? Is RNA modification (e.g., methyla- complex. One possibility in the EDM2 or site and a larger one produced using a distal tion) or RNA secondary structure involved? ASI1 cases is that, while binding to epige- poly(A) site (11). Both FCA and OXT6 What is the role of the splicing of the large netic marks, these proteins also interact genes have large introns where the proxi- intron in all this? Addressing these questions with the poly(A) signals, thus blocking mal APA occurs. However, it seems that would ensure a fine-grained understand- the recognition of the proximal poly(A) these APA events use a different mechanism ing of posttranscriptional regulation of genes site.Indeed,ASI1wasfoundtohaveanRNA because these introns do not contain TREs in this ever-increasing complex world of recognition motif and a bromo-adjacent ho- (4). Recent work by Duc et al. indicated that the epigenome. mology domain involved in preventing the a spen family protein FPA might be involved genomewide CHG methylation (6). EDM2, in the selection of some intronic polyad- ACKNOWLEDGMENTS. Research in the authors’ labora- on the other hand, possesses an N6-adenine tory is supported by US National Science Foundation Grant enylation (20). Nonetheless, recent genome- IOS–0817829 (to Q.Q.L.) and US National Institutes of methyltransferase domain (4), providing a po- level profiling discovered a significant amount Health Grant 1R15GM94732-1 A1 (to Q.Q.L.). tential to methylate adenine residues in RNA. Methylated adenine was shown to affect 3′-end formation (19). Binding of EDM2 to 1 Chinnusamy V, Zhu J-K (2009) Epigenetic regulation of stress 11 Zhang J, et al. (2008) A polyadenylation factor subunit implicated in chromatin through its PHD domain could responses in plants. Curr Opin Plant Biol 12(2):133–139.