Genome-wide control of polyadenylation site choice by CPSF30 in Arabidopsis

Abstract

The Arabidopsis thaliana ortholog of the 30-kD subunit of the mammalian Cleavage and Polyadenylation Specificity Factor (CPSF30) has been implicated in the responses of plants to oxidative stress, suggesting a role for alternative polyadenylation. To better understand this, poly(A) site choice was studied in a mutant (oxt6) deficient in CPSF30 expression using a genome-scale approach. The results indicate that poly(A) site choice in a large majority of Arabidopsis genes is altered in the oxt6 mutant. A number of poly(A) sites were identified that are seen only in the wild type or oxt6 mutant. Interestingly, putative polyadenylation signals associated with sites that are seen only in the oxt6 mutant are decidedly different from the canonical plant polyadenylation signal, lacking the characteristic A-rich near-upstream element (where AAUAAA can be found); this suggests that CPSF30 functions in the handling of the near-upstream element. The sets of genes that possess sites seen only in the wild type or mutant were enriched for those involved in stress and defense responses, a result consistent with the properties of the oxt6 mutant. Taken together, these studies provide new insights into the mechanisms and consequences of CPSF30-mediated alternative polyadenylation.

Figure 1.

Strategy for Assessing Poly(A) Site Choice. (A) Illustration of the clustering approach used to group closely situated poly(A) sites and of hypothetical results that may be used to generate the metric values for further analysis. The hypothetical reference sequence is at the bottom, bounded by 5′ and 3′. This reference has two clusters of poly(A) sites that are defined by the existence of sequence tags that end at the indicated positions (vertical tics). One such cluster is expanded at the top. Clustering of poly(A) sites is constrained such that the maximum distance between distinct sites is set at 10 nucleotides and the maximum span of a single cluster set to be 24 nucleotides. The table beneath illustrates three cases for illustrative purposes. In the three cases, the fraction of all tags that map to one of the two clusters in the reference sequence is calculated. From this, the absolute values of the differences between the two data sets (here, the wild type [wt] and mutant oxt6) is calculated and summed and the result divided by two to yield the value for the metric. (B) Illustration of the two extreme hypothetical outcomes of the assay. For this, the set of metrics for a data set are divided into 20 steps of 0.05 [the values of the poly(A) metric] and the numbers of genes whose metrics fall into one of these 20 steps counted. The running sum (normalized so that the final value is 1.0) is then calculated and plotted as shown. The plots expected if two data sets are largely similar and largely dissimilar are shown. These curves represent the probable extremes, between which will fall the results obtained from actual data.

Figure 2.

Plots of the Pairwise Comparisons of the Three Wild-Type (wt1, wt2, and wt3) and Three oxt6 (oxt61, oxt62, and oxt63) Data Sets. (A) Results of the three wild type–wild type (wt) comparisons (wt1-wt2, wt1-wt3, and wt2-wt3) denote the three pairwise comparisons that were made. These datasets are derived from the sequencing samples as described in Methods. (B) Results of the three oxt6-oxt6 comparisons (oxt61-oxt62, oxt61-oxt63, and oxt62-oxt63) denote the three pairwise comparisons that were made. These data sets are derived from the sequencing samples described in Methods. (C) Plots of the averages (avg) of the three respective individual comparisons. [See online article for color version of this figure.]

Figure 3.

Plots of the Pairwise Comparisons of the Wild-Type Data Sets. The curves showing the averages of the wild type–wild type (wt) and oxt6-oxt6 comparisons from Supplemental Figure 3 online and Figure 2C are provided for comparison’s sake. (A) Plot of the average of the nine comparisons shown in Supplemental Figure 4 online. (B) Plot of the results of a comparison of mappings using the combined tag data of the three sequencing runs. For this, the pooled collection of wild-type tags (e.g., wt1 + wt2 + wt3) were mapped to the extended 3′-UTR database, as was the pooled collection of oxt6 tags. avg, average. [See online article for color version of this figure.]

Figure 4.

Gene-by-Gene Analysis of Poly(A) Site Choice in the Wild Type and oxt6 Mutant. (A) to (E) Two parameters were plotted. One was the difference, for each gene, of the average poly(A) metric (as described in Figure 1) obtained in different pairwise comparisons. The other was the log(10) of the P value derived from a two-tailed Student’s t test that tests the hypothesis that the means of the differences in the within-sample (e.g., wild type–wild type and oxt6-oxt6 comparisons) and between-sample (e.g., wild type–oxt6 comparisons) are the same. A flowchart is given in Supplemental Figure 6 online that elaborates on these calculations. The y axes in the plots are inverted, such that lower P values are plotted in increasing fashion. (A) to (D) show the results obtained with a small gene set, as described in the text. (E) shows the results obtained with the large gene set. (A) A plot of the differences between the wild type (wt)–oxt6 comparison (“comparison” in all panels of this figure) and the wild type–wild type comparisons. (B) A plot of the differences between the wild type–oxt6 comparison (“comparison” in all panels of this figure) and the oxt6-oxt6 comparisons (oxt6) as a function of the P value derived from the described Student’s t test. (C) A plot of the differences between the wild type–oxt6 comparison (“comparison” in all panels of this figure) and the means of the control comparisons (wild type–wild type and oxt6-oxt6). (D) A plot of the differences between the control comparisons (wild type–wild type and oxt6-oxt6). (E) A plot of the differences between the wild type–oxt6 comparison (“comparison” in all panels of this figure) and the means of the control comparisons (wild type–wild type and oxt6-oxt6). (F) Plot of the running sum of genes that possess increasing differences in the poly(A) metric; plots for the small (n = 196) and large (n = 1025) gene sets are shown.

Figure 5.

Position-by-position analysis of average base composition of the regions surrounding PACs. (A) PACs that fall within extended 3′-UTRs. (B) PACs that fall within introns. (C) PACs that fall in intergenic regions. In all cases, plots were generated as described previously (Loke et al., 2005). “common” - poly(A) sites seen in both the wild-type and mutant; “wt” - sites seen only in the wild-type; “oxt6” - sites seen only in the oxt6 mutant. The numbers in the parentheses are the number of sequences used for the plots. Blue traces - A; red traces - U; green traces - C; purple traces - G.