Differential genome-wide profiling of tandem 3' UTRs among human breast cancer and normal cells by high-throughput sequencing

Abstract

Tandem 3' UTRs produced by alternative polyadenylation (APA) play an important role in gene expression by impacting mRNA stability, translation, and translocation in cells. Several studies have investigated APA site switching in various physiological states; nevertheless, they only focused on either the genes with two known APA sites or several candidate genes. Here, we developed a strategy to study APA sites in a genome-wide fashion with second-generation sequencing technology which could not only identify new polyadenylation sites but also analyze the APA site switching of all genes, especially those with more than two APA sites. We used this strategy to explore the profiling of APA sites in two human breast cancer cell lines, MCF7 and MB231, and one cultured mammary epithelial cell line, MCF10A. More than half of the identified polyadenylation sites are not included in human poly(A) databases. While MCF7 showed shortening 3' UTRs, more genes in MB231 switched to distal poly(A) sites. Several gene ontology (GO) terms and pathways were enriched in the list of genes with switched APA sites, including cell cycle, apoptosis, and metabolism. These results suggest a more complex regulation of APA sites in cancer cells than previously thought. In short, our novel unbiased method can be a powerful approach to cost-effectively investigate the complex mechanism of 3' UTR switching in a genome-wide fashion among various physiological processes and diseases.

Figure 1.

SAPAS strategy. (A) Experiment outline. (B) Genomic locations of reads that were uniquely mapped to the nuclear genome after internal priming filtering. (C) Histogram of the number of reads for UCSC canonical genes.

Figure 2.

Poly(A) site characteristics. (A) Genomic locations of poly(A) sites. (B) Distribution of poly(A) site expression levels. (C) Examples of type A, B, and C poly(A) sites, which are strong peak, weak peak, and no peak sites, respectively.

Figure 3.

Histogram of the signal number of poly(A) sites. Motifs were searched in the upstream 50 nt of the cleavage cluster range combined with the range itself. (A) Canonical AATAAA and ATTAAA sequences and the other 10 variants were all considered. (B) Only the canonical signal AATAAA and ATTAAA were considered.

Figure 4.

APA site switching and gene expression levels of normal and cancer cell lines. Cancer tandem 3′ UTR length index (CTLI) is plotted against the logarithm of the expression level ratios between the cancer ([A] MCF7 and [B] MB231) and normal (MCF10A) cell lines. The x-axis denotes CTLI; a larger positive value indicates that longer tandem UTRs are prone to be used in the cancer samples. Genes with significant switching to longer (blue) or shorter (red) tandem UTRs in cancer samples (FDR = 0.01; see Methods) are colored. The y-axis denotes the logarithm of the expression level of genes from the cancer sample relative to the normal sample.