Positive and Negative Regulation of Poly(A) Nuclease

Abstract

With rare exceptions, mRNAs whose synthesis originates within nuclei contain a 3Ј poly(A) tail. Poly(A) tracts are not encoded within genes but are added to nascent pre-mRNAs in a processing reaction that involves site-specific cleavage and subsequent polyadenylation (11,16,40,42). In Saccharomyces cerevisiae, newly synthesized poly(A) tails of different transcripts are relatively homogeneous, with their final lengths determined by the combined actions of poly(A) polymerase holoenzyme (Pap1p and Fip1p), poly(A)-binding protein (Pab1p), poly(A) nuclease (PAN), and the Pab1p-associated factor, Pbp1p (10,24,43).Poly(A) tracts are generally bound by Pab1p, a highly conserved protein with four RNA recognition motifs (RRMs) connected to a carboxy-terminal helical domain via a prolineand methionine-rich segment (25,31). Association of Pab1p with poly(A) requires a minimal binding site of 12 adenosines, and multiple molecules can bind via RRMs 1 and 2 to the same poly(A) tract, spaced approximately 25 nucleotides apart (1, 2, 31, 33). In yeast, the relatively abundant 70-kDa poly(A)-binding protein is encoded by the PAB1 gene. Mutations in PAB1 cause a significant increase in the average steady-state poly(A) tail length of total cellular mRNA (32), and these effects have been attributed to two apparently nuclear functions of Pab1p: the regulation of a switch between processive and distributive activities in poly(A) polymerase (43) and the stimulation of PAN activity (7, 9, 21).Yeast poly(A) tails are initially synthesized to default lengths of 70 to 90 A's and then trimmed to mRNA-specific lengths by PAN. Analyses of three different mRNAs indicate that such trimmed tails have lengths ranging from 55 to 71 A's (8). PAN, a Pab1p-dependent 3Ј to 5Ј poly(A) exoribonuclease, requires magnesium, releases AMP as a product, and is regulated by cis-acting mRNA sequences (21). Purified PAN contains two proteins which are essential for nuclease activity: Pan2p is a 127-kDa protein with homology to the RNase T family of 3Ј35Ј exoribonucleases, while Pan3p is a 76-kDa protein which apparently acts as a positive activator of PAN activity (8). Deletion of PAN2 and/or PAN3 eliminates poly(A) nuclease activity but does not hinder cell growth. Physical interaction between Pan2p and Pan3p has been inferred from coimmunoprecipitation and two-hybrid analyses of the full-length proteins (9).Pbp1p (Pab1p-binding protein 1) specifically interacts with a 74-amino-acid segment encompassing the proline-and methionine-rich domain of the Pab1p C terminus. PBP1 is not essential for viability, but its disruption can suppress the lethality associated with a PAB1 deletion (23). In the absence of Pbp1p, 3Ј termini of pre-mRNAs are properly cleaved but receive poly(A) tails that are, on average, 15 to 30 nucleotides shorter than normal (23,25). In vitro polyadenylation reactions using extracts from wild-type and pbp1⌬ strains demonstrated that the mutant extracts initially produced full-length tails equivalent to their wild-type counterparts but subseque...

An Integrative Approach to Gain Insights into the Cellular Function of Human Ataxin-2

“…While ATX2 binds the PABC domain itself, the major interaction site of Pbp1 with Pab1 appears to be a region preceding the PABC domain of Pab1. 28 The observed different binding modes of ATX2 and Pbp1 to PABC point to common evolutionary changes of the protein binding interface and are consistent with the fact that ATX2 contains the PABC-interacting motif PAM2, but not Pbp1. In agreement with that, we could not observe an interaction of human ATX2 and the corresponding region of the yeast PABP homolog Pab1 in the yeast-2-hybrid system (data not shown).…”

“…44 We also included the recently described interactions of Pbp1 with Mkt1, Pan2, Fir1, Ufd1, Dig1, and the as yet uncharacterized protein Pbp4 in the network. [27][28][29] We observed that the putative nuclease Mkt1 and the poly(A)-nuclease subunit Pan2 not only interact directly with Pbp1, but are also linked to Pbp1 indirectly via other interaction partners. This may point to larger protein complexes containing Pbp1 and Mkt1 or Pan3 and other proteins.…”

“…29 Furthermore, the poly(A)-nuclease subunit Pan2 binds to Pbp1, 27 and the poly(A)-polymerase binding proteins Fir1 and Ufd1 interact with Pbp1 and have also been implicated in processing the 3 0 -end of mRNA. 28 Since a functional relationship of ATX2 and Pbp1 has been corroborated by our yeast suppression studies, it is worthwhile to reason that ATX2 might also bind nucleic acids and RNA-modifying proteins that are human equivalents or even homologs of the corresponding yeast Pbp1-interacting proteins. We confirmed the interaction of ATX2 with the PABC domain of PABP by yeast-2-hybrid studies and co-immunoprecipitation experiments using mammalian cells.…”

“…27 In addition, Pbp1 negatively regulates the activity of the poly(A)-nuclease (PAN) by interacting with the subunit Pan2 of the PAN complex Pan2-Pan3. 28 Interestingly, another recent study showed that loss of Pbp1 expression causes reduced expression of the HO endonuclease at the posttranscriptional level, and that a complex of Pbp1 and the putative nuclease Mkt1 regulates the translation of the HO endonuclease gene via the 3 0 untranslated region (UTR) of HO mRNA. 29 In this report, we will shed new light on related functions of ATX2 and its yeast homolog Pbp1, and provide data suggesting that ATX2 may function in RNA metabolism.…”

Crossing the borders: Poly(A)-binding proteins working on both sides of the fence

“…25 Further evidence supporting a functional role of Pab1 in mRNA biogenesis is the accumulation of polyadenylated mRNAs in the nucleus of pan2 and pan3 deletion mutants, 25 which code for two subunits of a deadenylase complex (PAN) recruited by Pab1. 30 Because the PAN complex is thought to participate in the last step of mRNA biogenesis, 31 these results suggest that Pab1 is involved in the final step of 3' end processing before mRNA export.…”