Cleavage and polyadenylation: Ending the message expands gene regulation

Abstract

Cleavage and polyadenylation (pA) is a fundamental step that is required for the maturation of primary protein encoding transcripts into functional mRNAs that can be exported from the nucleus and translated in the cytoplasm. 3'end processing is dependent on the assembly of a multiprotein processing complex on the pA signals that reside in the pre-mRNAs. Most eukaryotic genes have multiple pA signals, resulting in alternative cleavage and polyadenylation (APA), a widespread phenomenon that is important to establish cell state and cell type specific transcriptomes. Here, we review how pA sites are recognized and comprehensively summarize how APA is regulated and creates mRNA isoform profiles that are characteristic for cell types, tissues, cellular states and disease.

Keywords: 3′end processing; alternative cleavage and polyadenylation; gene expression.

Figure 1.

The cis-elements that define pA sites. The cleavage and polyadenylation machinery relies on key cis elements to mediate 3′end processing. Canonical cis elements include the A[A/U]UAAA hexamer and its variants which lie ∼21 nucleotides upstream of the cleavage site (CS) and a downstream less well defined GU/U-rich element. Additional auxiliary elements may be positioned upstream and/or downstream of the cleavage site and are often U, GU and or G-rich.

Figure 2.

The core factors of the cleavage and polyadenylation complex. There are more than 80 proteins associated with the cleavage and polyadenylation machinery but fewer than 20 factors are considered to build the core of the processing complex. The major components are made up of multi-subunit factors including the cleavage and polyadenylation specificity factor CPSF (WDR33, hFip1, CPSF160, CPSF100, CPSF70, CPSF30); the cleavage stimulatory factor CstF (CstF77, CstF64, CstF50), the CFI (CFIm65, CFIm25) and CFII (∼15 subunits). The core factors involved in cleavage and polyadenylation, and the cis elements to which they bind are outlined here. Details of the individual factors are given in the text.

Figure 3.

Coding region APA (CR-APA) and UTR APA. Depending on the location of the different pA sites, APA events can be classed into 2 major groups. CR-APA is the result of differential usage of pA sites that are located within the body of the gene and alternative usage produces APA mRNA isoforms that differ in their coding potential. UTR-APA summarizes events where the different pA sites are located downstream of the stop codon and alternative usage modulates 3′UTR length but does not change the coding potential. pA sites can be found in the intron and in the UTR of a gene. Intronic pA sites (pAⁱ) are often cryptic poly A sites (pA^c) that need to be actively repressed to enable gene expression. pA sites in the 3′UTR are generally separated into proximal (pA^p) or distal (pA^d) sites. Usage of the proximal sites generates mRNA isoforms that have a so called constitutive 3′UTR (cUTR) and isoforms that are generated by usage of the distal site contain both the constitutive and alternative 3′UTR (aUTR) regions. The respective resulting APA mRNA isoforms are indicated, dotted lines refer to the removal on introns (i) and fusion of exons (E) and the 5′ splice sites and 3′ slice sites are indicated by the green and purple triangles respectively. The terminal exon is indicated by “^tE” and “^7meG” refers to the 5′ cap.

Figure 4.

Distinguishing between ACTIVE and PASSIVE APA. The APA profile can either be modified at the point of cleavage (Active), or at the post transcriptional level (Passive). In active APA, factors that inhibit or enhance one pA site over another produce APA isoforms that can avoid a particular regulatory pathway. On the other hand, in passive APA, the availability of factors such as RBPs (dark red circle) and miRNAs (navy) in the cytoplasm alter the APA profile by specifically downregulating a particular isoform. For example, as depicted here, miRNAs can target the aUTR which can recruit the RNA induced silencing complex (RISC) result in degradation by exoribonucleases (red “PacMan”). Different RBPs that bind to the aUTR can either stabilize or degrade the isoform. In this case although the whole cell APA profile is the same, the nuclear APA profile is different, highlighting the importance of assessing changes in the cytoplasm compared with the nucleus to distinguish Active and Passive APA. This gives a better resolution of the causes that enforce specific APA changes in different environments.

Figure 5.

Factors that regulate APA at the point of cleavage. Numerous RNA-binding proteins, and environmental stresses have been associated with modulating active APA at the point of cleavage. Factors are grouped into enhancing (green) or repressing (red) effects on a particular site and factors that are between a green and red bracket can either enhance or repress a site depending on the circumstances. For more details, see Tables 2, 3 and 4 or download the interactive slide. Red lines indicate inhibitory effects on pA sites and green lines indicate enhancing effects of factors on particular pA sites. Black and gray dots with arrows indicate the position of the different types of pA sites: (pAⁱ) = intronic pA site; (pA^c) = cryptic pA site; pA^p, proximal pA site; pA^d = distal pA site. The gene structure is detailed by specifying introns as blue double lines (i) and exons as black double lines (E) and the 5′ splice sites and 3′ splice sites are indicated by yellow and purple triangles respectively. The terminal intron is symbolised by ^tE.

Figure 6.

Consequences of APA: APA-isoform dependent decay rates and protein output. The 3′UTR length changes arising from APA can have implications on mRNA localization and transcript stability, which can impact on protein output and also determine the final destination of the encoded protein. This figure depicts the case where a short 3′UTR evades miRNA target sites in the aUTR, making it a more stable transcript, enabling increased protein output (protein symbolised by gray globules; ribosomes symbolised by mustard colored structures). The longer isoform shown here is bound by an RBP (dark green) in the nucleus, which prevents its export into the cytoplasm. The transcripts that are exported can be targeted for degradation by miRNA binding to the aUTR. The aUTR of the longer isoform can also bound by an RBP (dark red circle) in the cytoplasm which alters the localization of the transcript, for example in close proximity to the Endoplasmic Reticulum, for protein synthesis. Therefore, the UTR is important in mediating nuclear export, transcript stability, translatability and mRNA localization and the modulation of this is achieved by changing the expression of RBPs and miRNAs.