Two Related Proteins, Edc1p and Edc2p, Stimulate mRNA Decapping in Saccharomyces cerevisiae

Corresponding author: Roy Parker, Department of Molecular and Cellular Biology, University of Arizona, 1007 E. Lowell St., Tucson, AZ 85721., rrparker{at}u.arizona.edu (E-mail)

Communicating editor: P. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The major mRNA decay pathway in Saccharomyces cerevisiae occurs through deadenylation, decapping, and 5' to 3' degradation of the mRNA. Decapping is a critical control point in this decay pathway. Two proteins, Dcp1p and Dcp2p, are required for mRNA decapping in vivo and for the production of active decapping enzyme. To understand the relationship between Dcp1p and Dcp2p, a combination of both genetic and biochemical approaches were used. First, we demonstrated that when Dcp1p is biochemically separated from Dcp2p, Dcp1p was active for decapping. This observation confirmed that Dcp1p is the decapping enzyme and indicated that Dcp2p functions to allow the production of active Dcp1p. We also identified two related proteins that stimulate decapping, Edc1p and Edc2p (Enhancer of mRNA DeCapping). Overexpression of the EDC1 and EDC2 genes suppressed conditional alleles of dcp1 and dcp2, respectively. Moreover, when mRNA decapping was compromised, deletion of the EDC1 and/or EDC2 genes caused significant mRNA decay defects. The Edc1p also co-immunoprecipitated with Dcp1p and Dcp2p. These results indicated that Edc1p and Edc2p interact with the decapping proteins and function to enhance the decapping rate.

AN important control point in gene expression occurs at the level of mRNA stability (for reviews, see ROSS 1995 ; CAPONIGRO and PARKER 1996 ; JACOBSON and PELTZ 1996 ). Work in the yeast Saccharomyces cerevisiae has demonstrated that mRNAs are degraded through two general pathways. The main decay pathway in yeast occurs through deadenylation followed by Dcp1p-catalyzed removal of the 5' cap structure (BEELMAN et al. 1996 ), which exposes the body of the mRNA to Xrn1p-dependent 5' to 3' exonucleolytic degradation (DECKER and PARKER 1993 ; HSU and STEVENS 1993 ; MUHLRAD et al. 1994 , MUHLRAD et al. 1995 ). The second general mRNA decay pathway occurs via deadenylation of the polyadenylated mRNA, followed by 3' to 5' exonucleolytic digestion of the transcript body by the exosome complex (MUHLRAD et al. 1995 ; ANDERSON and PARKER 1998 ).

Multiple lines of evidence indicate that these general mRNA decay pathways are likely conserved in other eukaryotes. For example, deadenylated, decapped, full-length mRNA has been detected from murine liver cells (COUTTET et al. 1997 ). Messenger RNA decay intermediates shortened from their 5' ends have also been identified in both plant and animal cells (LIM and MAQUAT 1992 ; HIGGS and COLBERT 1994 ; GERA and BAKER 1998 ). Decay of mRNA in a 3' to 5' direction has also been shown to occur in plants (HIGGS and COLBERT 1994 ). Last, the yeast proteins that are involved in the decay of mRNA, such as Dcp1p, Dcp2p, Xrn1p, Lsm1p-Lsm7p, Ski2p, and components of the exosome, are conserved in higher eukaryotes (DANGEL et al. 1995 ; LEE et al. 1995 ; BASHKIROV et al. 1997 ; MITCHEL et al. 1997 ; SALGADO-GARRIDO et al. 1999 ; DUNCKLEY and PARKER 1999 ).

Decapping is a critical step in the 5' to 3' mRNA decay pathway since it both precedes and permits the degradation of the mRNA body. Also, individual mRNAs are decapped at different rates, indicating that mRNA decapping is a controlled process that contributes to differential mRNA decay rates (MUHLRAD et al. 1994 , MUHLRAD et al. 1995 ). Decapping is also a key step in the degradation of aberrant mRNAs through the mRNA surveillance pathway (MUHLRAD and PARKER 1994 ; HAGAN et al. 1995 ; BEELMAN et al. 1996 ; DUNCKLEY and PARKER 1999 ). The product of the DCP1 gene has been shown to be required for mRNA decapping in vivo and sufficient for decapping activity in vitro, suggesting that DCP1 encodes a mRNA decapping enzyme (BEELMAN et al. 1996 ; LAGRANDEUR and PARKER 1998 ). An additional protein, Dcp2p, is also required for mRNA decapping (DUNCKLEY and PARKER 1999 ). Dcp2p purified from wild-type yeast copurifies with Dcp1p as well as with mRNA decapping activity (DUNCKLEY and PARKER 1999 ). If purified from a dcp1 mutant, Dcp2p no longer copurifies with decapping activity, indicating that the presence of functional Dcp1p is required for decapping activity. Additionally, Dcp1p purified from a dcp2 mutant is enzymatically inactive for mRNA decapping in vitro (DUNCKLEY and PARKER 1999 ). These results suggest two possibilities. First, Dcp2p could be required for the production of active Dcp1p decapping enzyme. Second, Dcp1p and Dcp2p could be essential members of a multi-subunit decapping enzyme.

Determining the relationship between Dcp1p and Dcp2p, as well as identifying additional factors that control the activity of Dcp1p and Dcp2p, will be important for understanding mRNA decapping. For this reason, we have performed a combination of biochemical and genetic experiments addressing the relevance of the interaction between Dcp1p and Dcp2p to the production of mRNA decapping activity. In addition, we identify two Enhancer of mRNA DeCapping proteins, Edc1p and Edc2p, that stimulate mRNA decapping in vivo.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plasmids and strains:
All strains used in this study are listed in Table 1. Strain yRP1501 was constructed in three steps. First, a portion of genomic DNA 3' of the DCP2 gene was PCR amplified from plasmid pRP930 and cloned into the SmaI/ClaI sites of a pBluescriptII KS+ vector that also contained the URA3 gene in the BamHI site, generating plasmid pRP988. Next, the dcp2-7 allele was excised from plasmid pRP994 using SacI/SpeI and cloned into the corresponding sites of pRP988 to generate plasmid pRP989. pRP989 was then cut with ApaI/SacI to release the dcp2-7 allele, URA3 gene, and 3' region of DCP2. This DNA fragment was transformed into yRP1358 and all fast-growing URA+ colonies were then screened for loss of the TRP1 marker. Integration of the dcp2-7 allele at the dcp2 locus was confirmed by Southern analyses.

Strain yRP1503 was constructed by homologous recombination in strain yRP1514. The HIS3 gene was PCR amplified from genomic DNA using primers oRP827 and oRP828, containing regions of DNA complementary to the sequence immediately 5' of the EDC1 initiation codon and 3' of the stop codon. This PCR product was then transformed into yRP1514. Deletion of the EDC1 gene was confirmed by Southern analysis. Deletion of EDC2 was accomplished by PCR of a NEO deletion cassette from the commercially available edc2 mutant (Research Genetics, Birmingham, AL) using oRP990 and oRP991, followed by transformation of the PCR product into yRP841. Deletion of EDC2 was confirmed by Southern analysis. Strain yRP1504 was then later obtained from dissection of a diploid heterozygous for the edc1 and edc2 mutations.

To make plasmid pRP984, the EDC1 gene was amplified from genomic DNA using oRP987 and oRP988. This PCR product was then cloned into yEP351, a 2µ plasmid, using the BamHI and SacI sites introduced on the oligonucleotides. Plasmid pRP985 was made similarly through amplification of EDC2 using oRP931 and oRP932. Plasmid pRP982 was constructed by amplification of the GPD promoter, HIS-DCP1 gene, and PGK1 terminator from plasmid pRP785 using oligonucleotides oRP910 and oRP916. This PCR product was then cotransformed into a dcp1 strain (yRP1070) with NcoI/NdeI-digested pRP66. Plasmids generated from in vivo recombination that complemented the dcp1 growth defect were rescued to Escherichia coli using a yeast DNA isolation system (Stratagene, La Jolla, CA) and sequenced. The pRP983 plasmid was constructed by amplification of the GPD promoter, FLAG-DCP2 gene, and PGK1 terminator from plasmid pRP936 using oligonucleotides oRP899 and oRP915. The resulting PCR product was then cotransformed into a dcp2 strain (yRP1346) with EcoRI-digested pRS317. Plasmids that complemented the dcp2 growth defect were rescued to E. coli and sequenced. To make the FLAG-EDC1 plasmid (pRP986), the 5' untranslated region (UTR) and promoter region of EDC1 was first PCR amplified using oligonucleotides oRP959 and oRP960, which contained the reverse complement of the FLAG nucleotide sequence. Next, the EDC1 coding region and 3' UTR were PCR amplified using oRP962 and oRP961, which contains the FLAG nucleotide sequence at its 5' end. These two PCR products were cotransformed into the edc1 strain (yRP1503) with EcoRI-digested pRS317. Plasmids generated from in vivo recombination were rescued from the resulting transformants and sequenced. The V5-DCP2 plasmid (pRP999) was constructed similarly to the FLAG-EDC1 plasmid. The 5' UTR of DCP2 was PCR amplified using oRP985 and oRP968, which contained the reverse complement of the V5 nucleotide sequence. The DCP2 coding region and 3' UTR were amplified using oRP984 and oRP967, which contained the V5 nucleotide sequence. These PCR products were cotransformed into yRP1346 with EcoRI-digested pRS317 and complementing plasmids were rescued to E. coli and sequenced. The insert from pRP999 was then subcloned into pRS416 using ApaI/SacI, generating pRP987.

Genetic screening procedures:
EDC1 was identified using a screening procedure for suppressors of the dcp1-2 ski8 mutant as previously described (DUNCKLEY and PARKER 1999 ). To screen for high-copy suppressors of the dcp2-7 ski3 mutant, strain yRP1502 was transformed with a yeast 2µ genomic DNA library (ATCC 37323). Transformants were selected at 24°. Positive transformants were then replica plated and screened for growth at 24° and 37°. Plasmids were isolated from clones that supported growth at 37°, retransformed into yRP1502, and retested for growth at the restrictive temperature. The inserts from plasmids that conferred growth at high temperature were then sequenced.

RNA procedures:
RNA samples were prepared and isolated as previously described (CAPONIGRO et al. 1993 ). Half-lives were determined by quantitation of blots using a Molecular Dynamics (Sunnyvale, CA) PhosphorImager. Loading corrections for quantitation were determined by hybridization to the 7S RNA, a stable RNA polymerase III transcript.

Determination of dcp1-2 and dcp2-7 suppression:
To determine suppression of dcp1-2 and dcp2-7, strains yRP840, yRP1341, yRP1341 [pRP984], yRP1356, yRP1356 [pRP984], yRP1501 [MFA2pG/LYS2], yRP1501 [pRP985] [MFA2pG/LYS2], yRP1358 [MFA2pG/LYS2], and yRP1358 [pRP985] [MFA2pG/LYS2] were grown at 24° to midlog phase in selective media containing 2% galactose. For dcp1-2 strains, cultures were then shifted to 33° for 1 hr. For dcp2-7 strains, cultures were shifted to 37° for 1.5 hr. Following the temperature shift, cells were harvested and RNA was prepared as previously described (CAPONIGRO et al. 1993 ).

In vitro separation of Dcp1p and Dcp2p:
Four liters of strain yRP1500 containing pRP982 and pRP983 were grown at 30° in selective media to an OD₆₀₀ of 1.0. Cells were lysed and the FLAG-Dcp2p was purified as previously described (DUNCKLEY and PARKER 1999 ), with one exception. The immunopellet was washed with Tris magnesium sodium (TMN) buffer containing 2.5 M NaCl, rather than with TMN containing 750 mM NaCl. All protein samples were dialyzed against 4 liters of TMN-150 (10 mM Tris-HCl, pH 7.7, 150 mM NaCl, 5 mM MgCl₂).

The FLAG-Dcp2p and HIS-Dcp2p were analyzed by standard SDS-PAGE on a 10% gel (LAEMMLI 1970 ). FLAG-Dcp2p and HIS-Dcp1p were detected by Western analyses using -FLAG antibody (Kodak, Rochester, NY) and polyclonal antibodies to full-length Dcp1p, respectively. Protein size markers were purchased from GIBCO BRL (Gaithersburg, MD). The in vitro decapping assays were performed as previously described (BEELMAN et al. 1996 ; LAGRANDEUR and PARKER 1998 ).

Immunoprecipitation of FLAG-Edc1p:
Two liters of strain yRP1512 containing pRP986 and pRP987 was grown in selective media at 24° to an OD₆₀₀ of 0.8. Cells were harvested and lysed as previously described (DUNCKLEY and PARKER 1999 ). Immunoprecipitation of FLAG-Edc1p was performed as previously described for FLAG-Dcp2p (DUNCKLEY and PARKER 1999 ) with one exception. Binding of the FLAG-Dsp1p was performed in TMN-25 buffer containing 25 mM NaCl. The immunopellet was then washed either with TMN-50, containing 50 mM NaCl, or with TMN-150, containing 150 mM NaCl, prior to elution of FLAG-Dsp1p. The RNase treatment was performed as previously described (THARUN et al. 2000 ). The V5-Dcp2p fusion protein was detected using commercially available -V5 antibody (Invitrogen, San Diego) following the manufacturer's protocol.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Separation of Dcp1p from Dcp2p yields active decapping enzyme:
On the basis of the available data, there are two possible functions for Dcp2p in mRNA decapping. First, Dcp2p could be required for the activation of the Dcp1p decapping enzyme and would be dispensable for decapping once Dcp1p was activated. Alternatively, Dcp1p and Dcp2p could be essential members of a multi-subunit decapping enzyme, in which case Dcp2p would always be required for decapping activity. To distinguish between these two possibilities, we biochemically separated a fraction of Dcp1p from Dcp2p and assayed for mRNA decapping activity of the different protein fractions in vitro to determine which protein, or proteins, are required for decapping.

To do this experiment, we purified a FLAG-tagged Dcp2 protein from a yeast strain that overexpressed both HIS-tagged Dcp1p and FLAG-Dcp2p (see MATERIALS AND METHODS). As has been observed previously, purification of FLAG-Dcp2p yields both the full-length protein and a shorter form of the protein (Fig 1A; DUNCKLEY and PARKER 1999 ). Importantly, FLAG-Dcp2p is detectable by Western analysis only in the eluate (Fig 1A), even after very long exposures (data not shown). In contrast, Dcp1p is readily detectable in the unbound, high-salt wash, and eluate fractions (Fig 1B). By this paradigm, we have separated the Dcp1p present in the cell into three pools: a pool that is not bound by Dcp2p and is present in the unbound fraction, a pool that is bound to Dcp2p but is released in the high-salt wash, and a pool that is tightly bound to Dcp2p and elutes only when Dcp2p is released from the column by peptide elution.

View larger version (29K): In this window In a new window Download PPT slide   Figure 1. Dcp2p activates Dcp1p. (A) A Western blot of the FLAG-DCP2 preparation using anti-FLAG antibody is shown. The lanes shown are: Unbound, supernatant remaining following binding to -FLAG immunoaffinity matrix; eluate, fraction eluted from the -FLAG matrix; wash, 2.5 M NaCl wash fraction performed prior to elution of FLAG-Dcp2p. Numbers on the left indicate molecular weight standards in kilodaltons. (B) Western blot using antibodies specific for the Dcp1 protein. Lanes shown are the same as those for A. (C) Decapping assays using the same fractions as in A and B are shown, with the addition of a negative control consisting of substrate RNA with no added protein. The products produced at the end of a 20-min time course are shown. One-ninth of the unbound sample was used relative to the wash fraction so that comparable amounts of Dcp1p were present in the two reactions. The spots beneath the 7mGDP decapping product result from the activity of other nucleases normally present in the unbound and wash fractions (T. DUNCKLEY and R. PARKER, unpublished observations).

Assaying the protein fractions for mRNA decapping activity revealed several important observations. First, in vitro decapping activity is present in the unbound and wash fractions, which contain Dcp1p but no FLAG-Dcp2p (Fig 1C). Further, these two fractions contain comparable amounts of decapping activity relative to the amount of Dcp1p that was used in the reactions. In the figure shown, approximately 30 ng of Dcp1p was included in each decapping reaction based on comparisons of the amount of Dcp1p in each fraction to purified Dcp1p. This observation demonstrated that, once produced in an active form, Dcp1p no longer required Dcp2p for decapping activity. This result strongly argued that the essential function of Dcp2p in decapping is to activate the Dcp1p decapping enzyme. Furthermore, this result confirms previous observations suggesting that Dcp1p alone is sufficient for mRNA decapping (LAGRANDEUR and PARKER 1998 ).

A second important observation was that, following a 2.5 M NaCl wash, Dcp2p remains associated with a large fraction of Dcp1p, indicating that Dcp2p interacts tightly with a significant amount of Dcp1p in vivo. Moreover, the Dcp1p that is present in this fraction is inactive for decapping in vitro (Fig 1). This is a surprising result and suggests several intriguing possibilities. First, Dcp1p may exist in vivo in an active and inactive form. In this model, Dcp2p would associate more tightly with the inactive form than with the active form of Dcp1p. The affinity of Dcp2p for Dcp1p would decrease following Dcp2p-dependent activation of Dcp1p. An alternative, but related, possibility is that Dcp1p is inactive if bound by Dcp2p. This model requires that Dcp2p function both as an activator and as an inhibitor of Dcp1p (see DISCUSSION).

Identification of EDC1 and EDC2:
To further characterize the relationship between Dcp1p and Dcp2p we used a genetic screen to identify proteins that affect their function. In this case, we took advantage of the observation that loss-of-function mutations in either dcp1 or dcp2 are synthetically lethal with ski2, ski3, ski4-1, ski7, or ski8 mutations, which prevent efficient 3' to 5' mRNA decay (ANDERSON and PARKER 1998 ; VAN HOOF et al. 2000 ). Two observations suggest this synthetic lethality results from lack of sufficient mRNA degradation. First, mRNAs are extremely stable in a dcp1-2 ski8 strain (ANDERSON and PARKER 1998 ). Second, all suppressors of this lethality isolated to date restore mRNA decay to some extent (DUNCKLEY and PARKER 1999 and unpublished observations). We have used this synthetic lethality, in combination with a temperature-sensitive allele of the DCP1 gene (dcp1-2), to isolate high-copy suppressors that restore growth at high temperature to a dcp1-2 ski8 strain (DUNCKLEY and PARKER 1999 ). This genetic screen identified DCP2 and EDC1 [open reading frame (ORF) no. YGL222C] as high-copy suppressors of the synthetic lethality of dcp1-2 ski8 mutants (Fig 2A; DUNCKLEY and PARKER 1999 ). Characterization of the role of Edc1p in mRNA decay is presented below.

View larger version (106K): In this window In a new window Download PPT slide   Figure 2. EDC1 and EDC2 overexpression rescues the growth defect of dcp1-2ski8 mutants and of dcp2-7ski3 mutants, respectively. (A) Strain yRP1345, containing either a 2µ plasmid (yEP351), EDC1 on a 2µ plasmid (pRP984), or EDC2 on a 2µ plasmid (pRP985), was grown at 24° for 4 days on minimal media lacking leucine and then replica plated and grown for an additional 2 days at either 24° or 33°. (B) Strains yRP840 and yRP1502, containing either a 2µ plasmid (yEP351), EDC1 on a 2µ plasmid (pRP984), or EDC2 on a 2µ plasmid (pRP985), were grown at 24° for 4 days on minimal media lacking leucine and then replica plated and grown for an additional 2 days at either 24° or 37°.

To identify proteins that influence Dcp2p activity, we made the synthetic lethality of dcp2 ski3 double mutants conditional using a temperature-sensitive mutation in dcp2, termed dcp2-7. This allele contains three point mutations in the N terminus of the protein (N60D, I68V, and D142V) and allows nearly normal mRNA decay at 24° but is essentially a null at 37° (T. DUNCKLEY and R. PARKER, unpublished results). We then screened three genome equivalents of a yeast 2µ genomic DNA library for genes that, when overexpressed, would suppress the conditional growth defect of dcp2-7 ski3 mutants at 37° (see MATERIALS AND METHODS). The EDC2 gene (ORF no. YER035W) was identified as an overexpression suppressor of the conditional synthetic lethality of dcp2-7 ski3 mutants (Fig 2B).

EDC1 and EDC2 encode related proteins with distinct functions:
Analysis of the EDC1 and EDC2 coding regions revealed several interesting features. First, both proteins are relatively small (Edc1p = 19.1 kD, Edc2p = 16.1 kD) and basic (Edc1p pI = 11.01, Edc2p pI = 10.05). Most strikingly, a BLAST search revealed that Edc1p and Edc2p are related (Fig 3). The most significant homology is clustered in two regions, one at the C termini of the two proteins and the other toward the N termini. At the present time, no other proteins in the databases show these conserved motifs.

View larger version (17K): In this window In a new window Download PPT slide   Figure 3. Edc1p and Edc2p are related proteins. Shown is an alignment of Edc1p and Edc2p. Darker shaded boxes indicate identity and lighter shaded boxes indicate similarity. The two proteins are 26% identical and 42% similar to each other. The alignment was generated using the ClustalW alignment program with the default parameters. Accession numbers for the sequences are: EDC1, CAA96938.1; EDC2, AAB64570.1.

The conservation of these two regions suggested that Edc1p and Edc2p might share a similar function. For this reason, we tested if overexpression of Edc1p would suppress the conditional lethality of dcp2-7 ski3 mutants and if overexpression of Edc2p would suppress the lethality of the dcp1-2 ski8 mutant. Interestingly, overexpression of Edc1p suppressed the growth defect of the dcp1-2 ski8 mutant as seen previously, but not that of the dcp2-7 ski3 mutant (Fig 2, compare A and B). In contrast, overexpression of Edc2p suppressed the growth defect of the dcp2-7 ski3 mutant but not that of the dcp1-2 ski8 mutant (Fig 2, compare A and B). However, a screen for genes that would suppress the dcp1-2 mutation if highly overexpressed from the GAL promoter yielded the EDC2 gene as a suppressor (data not shown). This suggests that high levels of Edc2p can suppress the dcp1-2 mutation. Combined, these results suggest that, although Edc1p and Edc2p are homologous proteins, they might have subtly different functions with respect to their interactions with the decapping machinery (see below).

Edc1p and Edc2p overexpression suppresses mRNA decapping defects:
In principle, the restoration of growth to the conditional dcp ski strains could be accompanied by a partial restoration of either 5' to 3' mRNA decay or 3' to 5' mRNA decay. To determine whether EDC1 and EDC2 overexpression restored mRNA decay through either degradation pathway, we made use of a galactose-inducible MFA2 mRNA that contains a poly(G) insertion in its 3' UTR, termed MFA2pG (DECKER and PARKER 1993 ). The poly(G) tract forms a very stable RNA secondary structure that blocks both 5' to 3' and 3' to 5' RNA degrading exonucleases, thereby trapping intermediates in mRNA decay (VREKEN and RAUE 1992 ; DECKER and PARKER 1993 ; ANDERSON and PARKER 1998 ). Moreover, the specific degradation intermediates that are trapped can be used to infer the directionality of mRNA decay (for discussion see HE and PARKER 1999 ). Therefore, we introduced the EDC1 gene on a high-copy-number plasmid into either dcp1-2 or ski8 strains that also contained the MFA2pG-inducible reporter mRNA and determined the mRNA decay phenotype of the resulting strains. It should be noted that we examined the effects of Edc1p or Edc2p in strains deficient in only 5' to 3' decay or 3' to 5' decay to simplify the analysis. Similarly, the EDC2 gene on a high-copy-number plasmid was introduced into either dcp2-7 or ski3 strains with the MFA2pG plasmid, and the mRNA decay phenotypes of these strains were also determined (see MATERIALS AND METHODS).

As shown in Fig 4A, overexpression of Edc1p partially rescued the mRNA decay defect of the dcp1-2 mutation at the restrictive temperature. This was observed in the dcp1-2 mutant that was overexpressing Edc1p as an increased production of 5' to 3' mRNA degradation product relative to the dcp1-2 mutant alone. In contrast, EDC1 overexpression does not restore mRNA decay to a dcp1 mutant, indicating that EDC1 overexpression is not a bypass suppressor of dcp1 mutations (Fig 4A).

View larger version (61K): In this window In a new window Download PPT slide   Figure 4. Overexpression of EDC1 and EDC2 genes suppresses mRNA decapping defects. The distribution of full-length MFA2pG relative to 5' to 3' degradation product is shown. The amount of smaller 5' to 3' degradation product is a measure of decapping activity in the strains shown. The experiment was performed as described in MATERIALS AND METHODS. Hybridizations were performed with an oligonucleotide specific for the MFA2pG mRNA (oRP140, CAPONIGRO and PARKER 1996 ).

Overexpression of EDC2 in the dcp2-7 mutant at restrictive temperature revealed that EDC2 partially restores decapping in the presence of the dcp2-7 mutation (Fig 4B). Overexpression of EDC2 does not bypass the dcp2-7 mutation as there is no effect of EDC2 overexpression on the decapping defect seen in the dcp2 mutant (Fig 4B). Finally, neither EDC1 nor EDC2 overexpression had a detectable effect on the 3' to 5' decay defect of the respective ski mutation (data not shown), suggesting that neither EDC1 nor EDC2 overexpression significantly affects 3' to 5' mRNA decay. These results suggest that overexpression of EDC1p suppresses dcp1-2 whereas overexpression of Edc2p is more specific for suppression of dcp2-7. However, there is some partial redundancy regarding the roles of Edc1p and Edc2p in decapping (see below).

Edc1p and Edc2p are not rate limiting for mRNA decay:
The suppression of conditional decapping defects by Edc1p and Edc2p overexpression suggests that these proteins are involved in mRNA decapping. To determine the role of Edc1p and Edc2p in mRNA decay, we first analyzed the mRNA decay phenotypes of strains that are deleted for these two homologous genes (see MATERIALS AND METHODS). The edc1, edc2, and edc1edc2 double deletion strains do not result in growth defects at any temperature tested from 18° to 37°. Furthermore, analysis of mRNA degradation in these strains revealed no defect in the decay of either the unstable MFA2pG reporter mRNA or the stable PGK1pG mRNA (Fig 5A and data not shown). This suggests that in otherwise wild-type cells Edc1p and Edc2p are not rate limiting for mRNA decapping on these substrates.

View larger version (27K): In this window In a new window Download PPT slide   Figure 5. EDC1 and EDC2 are required for efficient mRNA decapping. The decay of the MFA2pG mRNA in the indicated strains is shown. Strains were grown at 24° to midlog phase in YEP medium containing 2% galactose. Cells were harvested and resuspended in YEP medium. Glucose was added to a final concentration of 4% to repress transcription. Time points represent minutes after transcriptional repression. Hybridizations were performed as for Fig 4.

The edc1 and edc2 slow mRNA decapping in strains compromised for decapping activity:
The observation that Edc1p and Edc2p overexpression can enhance decapping, in combination with the observation that the edc1 edc2 double mutant showed normal mRNA turnover, suggested that these proteins may be involved in aspects of mRNA decapping that are not normally rate limiting. One possibility is that the EDC1 and EDC2 proteins are members of a larger protein complex involved in mRNA decapping (see below). In this case loss of the individual Edc proteins may not significantly affect the assembly or function of the larger complex. This type of situation, where individual members of a larger complex of proteins are dispensable for the normal function of the assembled complex, has been observed previously. For example, certain individual members of the pre-mRNA splicing machinery are dispensable for the splicing of pre-mRNA under otherwise wild-type conditions (RUTZ and SERAPHIN 2000 ). However, under conditions where splicing is inefficient, as is the case for specific mutations in the consensus splice site sequences (JACQUIER et al. 1985 ), loss of these individual proteins can be seen to be required for efficient splicing of the intron and, hence, for the efficient function of the splicing machinery. This illustrates a useful principle for identifying important functions of individual members of large complex assemblies.

Given this logic, we examined the effects of the edc1 and edc2 mutations in strains that were partially compromised for mRNA decapping. Both the conditional dcp1-2 and dcp2-7 alleles are very slightly defective for mRNA decapping in vivo at the permissive temperature (THARUN and PARKER 1999 ; T. DUNCKLEY and R. PARKER, unpublished results). Therefore, a reasonable expectation is that a mutation that further compromises the decapping rate will be evidenced more clearly in these strain backgrounds than in an otherwise wild-type cell. Therefore, we introduced the edc1 and edc2 into either the dcp1-2 or dcp2-7 mutant and assayed the mRNA decay defects of the resulting strains at 24°, which corresponds to the permissive temperature for both dcp1-2 and dcp2-7 alleles in an otherwise wild-type background.

An analysis of the mRNA decay phenotypes of the edc1 and edc2 in the dcp1-2 background revealed four important observations. First, the edc1 in the dcp1-2 background slowed the decay rate of the MFA2pG mRNA greater than twofold (Fig 5B). Similar results were seen with the stable PGK1pG mRNA (data not shown). This result indicates that, when Dcp1p is partially defective, Edc1p functions to stimulate the mRNA decapping rate. Second, the edc2 did not significantly affect the decay of the MFA2pG mRNA in the dcp1-2 mutant. This result is consistent with the suppression data described above where Edc1p on a 2µ plasmid would suppress the conditional lethality of the dcp1-2ski8 strain, but not the conditional lethality of the dcp2-7 ski3 strain (Fig 2). A third important observation was that, in the dcp1-2 background, the combined edc1 edc2 double deletion slowed the decay rate of the MFA2pG mRNA greater than did the edc1 mutant alone. In fact, loss of both Edc1p and Edc2p in the dcp1-2 mutant results in mRNA half-lives comparable to those observed in a dcp1 (BEELMAN et al. 1996 ), suggesting that mRNA decapping is essentially completely eliminated in this strain. Consistent with a specific defect in decapping, none of these strains exhibited altered deadenylation rates compared to the wild-type strain (data not shown). Last, a comparison of the MFA2pG half-life in the dcp1-2 edc1 edc2 strains (14') to that in the dcp1-2 edc1 (11') strain also suggests that, if Edc1p is absent, Edc2p is able to partially substitute for Edc1p (Fig 5B). Consistent with this interpretation, the Edc2p can partially suppress both the growth defect of dcp1-2 ski2 mutants and the mRNA decay defect of the dcp1-2 mutation if very highly overexpressed from the GAL promoter (data not shown). Combined, the above results suggest distinct but partially redundant functions for Edc1p and Edc2p in the stimulation of decapping.

Analyses of the mRNA decay phenotypes of the edc1 and edc2 in the dcp2-7 background also yielded several interesting observations. First, the dcp2-7 edc2 mutant showed a longer half-life for both the MFA2pG (11', Fig 5C) and PGK1pG mRNA (data not shown) than did the dcp2-7 edc1 (MFA2pG t = 8'). This result was in contrast to the case with the dcp1-2 allele where the edc1 had the larger effect (Fig 5B). The edc1edc2 double deletion in combination with dcp2-7 had an increase in half-life similar to the value obtained from the dcp2-7 edc2 strain. These results are consistent with the view that the dcp2-7 allele is more sensitive to the function of the Edc2p than to the function of the Edc1p. These results indicated that, while not normally rate limiting under conditions where mRNA decapping is maximally efficient on the MFA2 and PGK1 mRNAs, the EDC1 and EDC2 proteins can function as enhancers of the mRNA decapping machinery (see DISCUSSION).

Edc1p associates with the mRNA decapping machinery in vivo:
The proteins encoded by the EDC1 and EDC2 genes could affect mRNA decapping in multiple ways. One simple interpretation of the above genetic data is that these proteins interact with Dcp1p and Dcp2p and form a larger protein complex that carries out mRNA decapping. To test this possibility, we utilized co-immunoprecipitation experiments to determine if the Edc1p interacts with either Dcp1p or Dcp2p in vivo. For this experiment, we constructed a version of Edc1p with a flag epitope at its N terminus and confirmed that this tagged version of the protein was functional (data not shown). We were unable to construct a functional epitope-tagged version of Edc2p so we have restricted our analysis to the Edc1p to date.

Results of the co-immunoprecipitation demonstrated that both Dcp1p and a functional epitope-tagged version of Dcp2p associate with Edc1p (Fig 6). In this experiment we purified the Flag-Edc1p over a flag affinity column. On the basis of gel analysis and silver staining of the various fractions, we determined that the vast majority of the cellular protein is in the unbound and first wash fractions, with only a small percentage of the total protein present in the eluted fraction (data not shown). Interestingly, essentially all of the Dcp1p and Dcp2p in the cell interacted with Edc1p under low-salt conditions. The interactions between Edc1p and Dcp1p or Dcp2p were reduced, but not eliminated, by washing the immunopellet with 150 mM NaCl prior to elution of the FLAG-Edc1p (Fig 6). These results argue that Edc1p physically interacts with Dcp1p and Dcp2p.

View larger version (26K): In this window In a new window Download PPT slide   Figure 6. Edc1p interacts with Dcp1p and Dcp2p. Shown is an -FLAG immunoprecipitation from an edc1 strain (yRP1512 containing the V5-DCP2 plasmid, pRP987) and from a FLAG-Edc1p-expressing strain (yRP1512 containing pRP987 and the FLAG-EDC1 plasmid, pRP986). The experiment was performed as described in MATERIALS AND METHODS. For the last panel, RNase treatment was performed on the cell extract prior to binding to the -FLAG matrix. Specific protein bands are indicated on the left and the antibodies used to detect them are indicated on the right of the Western blots. U, unbound fraction; W, wash fraction at the indicated salt concentrations; E, eluate.

Because the observed interaction was reduced under relatively low-salt conditions, we hypothesized that the interaction may involve RNA. To test this possibility, we performed the immunoprecipitation experiment using cell extracts that were pretreated with RNase prior to binding to the -FLAG immunoaffinity matrix. The RNase treatment reduced, but did not eliminate, the observed interaction. Following treatment with RNase, there was no detectable Dcp1p or Dcp2p in the unbound fraction, indicating that the bulk of cellular Dcp1p and Dcp2p remain associated with FLAG-Edc1p in the absence of RNA (Fig 6). Interestingly, after RNase treatment, the interaction between Edc1p, Dcp1p, and Dcp2p became more labile. A large amount of the Dcp1p and most of the V5-Dcp2p fusion protein no longer interact with Edc1p at 50 mM NaCl instead of at 150 mM NaCl, as was seen in the presence of RNA. This result suggests that the observed interaction between Edc1p and Dcp1p/Dcp2p was not RNA dependent, but rather was strengthened by the presence of RNA. The identity of the RNA involved is currently unknown, but the mRNA itself represents a reasonable possibility. These results also imply that Edc1p, and likely Edc2p as well, may be members of a larger protein complex that assembles either prior to or during decapping (see DISCUSSION).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Dcp2p functions to activate Dcp1p:
Two important observations indicate that Dcp2p's critical function in mRNA decapping is to activate the Dcp1p decapping enzyme. First, previous findings show that Dcp1p is inactive for mRNA decapping in vivo and in vitro if produced in a dcp2 strain (DUNCKLEY and PARKER 1999 ). Second, Dcp1p is a functional mRNA decapping enzyme if produced in the presence of Dcp2p and subsequently separated from Dcp2p (Fig 1). These results suggest that, once produced, Dcp1p must undergo a Dcp2p-dependent activation event that makes the protein competent for mRNA decapping. The specific mechanism of Dcp1p activation is unknown. However, the Dcp2 protein contains a functional MutT motif that is required for its mRNA decapping functions (DUNCKLEY and PARKER 1999 ). The MutT motif is found in a large family of pyrophosphatases (for review of MutT proteins see KOONIN 1993 ; BESSMAN et al. 1996 ). This suggests that Dcp2p likely functions in some manner as a pyrophosphatase to activate Dcp1p.

Dcp2p associates with active and inactive Dcp1p with differing affinity:
Dcp2p associates with a large fraction of Dcp1p that is not removed in a 2.5 M salt wash, suggesting a very tight interaction between these two proteins (Fig 1A and Fig B). Furthermore, this fraction of Dcp1p is catalytically inactive in vitro (Fig 1C). These observations suggest two possible interpretations. In one model, Dcp2p associates with Dcp1p, inhibiting its activity. Hydrolysis of Dcp2p's substrate, mediated by the protein's MutT motif, would release Dcp1p, thereby allowing the Dcp1p enzyme to function. This makes the prediction that Dcp1p will be functional for decapping provided that the protein is not bound by Dcp2p. The fact that Dcp1p in the dcp2 mutant is completely inactive in vivo and in vitro (DUNCKLEY and PARKER 1999 ) makes this model unlikely. A more likely possibility is that cleavage of Dcp2p's substrate results in either a specific covalent or noncovalent modification of Dcp1p that converts inactive Dcp1p to active Dcp1p. This model could also explain why the Dcp1p that remains bound to Dcp2p is inactive (Fig 1). The simplest explanation is that Dcp2p associates tightly with an inactive form of Dcp1p and that cleavage of Dcp2p's substrate converts the inactive Dcp1p to active Dcp1p, which may have a decreased affinity for Dcp2p. As an extension of this model, it is possible that Dcp2p may function to inactivate Dcp1p through essentially the reverse of the activation reaction. Following cleavage of the cap structure, this would provide a simple mechanism for inactivating Dcp1p and readying the protein for another round of decapping.

Edc1p and Edc2p stimulate the mRNA decapping machinery:
Several observations show that Edc1p and Edc2p enhance the function of the mRNA decapping machinery. First, EDC1 and EDC2 overexpression suppresses loss-of-function mutations in the Dcp1p decapping enzyme and Dcp2p, respectively (Fig 2A and Fig B, and Fig 4A and Fig B). Second, the edc1 and edc2 deletions greatly exaggerate the partial mRNA decapping defects seen at the permissive temperature of conditional dcp1 and dcp2 mutants (Fig 5B and Fig C). Third, Edc1p co-immunoprecipitated from cell extracts with the Dcp1p and Dcp2p (Fig 6). To date, we have been unable to construct a functional epitope-tagged version of the Edc2p, but the prediction is that this protein will also associate with the Dcp1p and Dcp2p. Taken together, we interpret these observations to argue that Edc1p and Edc2p interact with the Dcp1p/Dcp2p decapping machinery to stimulate the mRNA decapping rate.

An alternative formal possibility is that Edc1p and Edc2p function to stabilize aberrant proteins much like chaperones and that their interaction with the decapping proteins is revealing a broader role in protein metabolism. However, two observations make this possibility unlikely. First, Edc1p interacts with wild-type Dcp1p. Second, high levels of Edc2 (from the Gal promoter) can increase decapping in wild-type strains as assessed by the ratio of full-length mRNA to the 5' to 3' decay product. These observations demonstrate that the Edc proteins can interact with wild-type Dcp1p and Dcp2p and therefore their function is not likely to be limited to stabilizing mutant proteins.

One of the intriguing properties of Edc1p and Edc2p is their apparent specificity for Dcp1p and Dcp2p, respectively. For example, the EDC1 gene is a high-copy suppressor of dcp1-2 but not dcp2-7, whereas the EDC2 gene is a high-copy suppressor of dcp2-7 but not dcp1-2 (Fig 2). Similarly, the edc1 reveals a stronger decapping defect when combined with the dcp1-2 allele than when combined with the dcp2-7 allele, whereas the edc2 reveals a decapping defect when combined with the dcp2-7 allele but not with the dcp1-2 allele (Fig 5). One possibility is that the specificity of Edc1p for Dcp1p, and of Edc2p for Dcp2p, results from the Edc proteins best stabilizing the decapping protein they interact with directly. Alternatively, the specificity could result from the dcp1-2 and dcp2-7 lesions affecting distinct substeps in the overall process of decapping. This possibility is based on the finding that mRNA decapping requires multiple steps and interactions between multiple proteins (DUNCKLEY and PARKER 1999 ; BOUVERET et al. 2000 ; SCHWARTZ and PARKER 2000 ; THARUN et al. 2000 ). The dcp1-2 mutation affects the catalytic activity of the enzyme (THARUN and PARKER 1999 ). The specific step in decapping that is affected in the dcp2-7 lesion is unknown, but it is likely to be prior to Dcp1p cleavage. In this instance, Edc1p and Edc2p would show differential effects because each protein would stimulate a different substep of decapping. Different substeps that could be affected would include the assembly of a decapping complex on the mRNA, or potentially the activation of Dcp1p by Dcp2p (see below).

How might Edc1p and Edc2p function in decapping?
On the basis of our current understanding of the mechanism of mRNA decapping, Edc1p and Edc2p could function to enhance decapping in one, or more, of three possible manners. First, these proteins could function to promote the disassembly of the cytoplasmic cap binding complex. This possibility is based on the observations that the cytoplasmic cap binding complex is an inhibitor of decapping and needs to dissociate before decapping can occur (SCHWARTZ and PARKER 1999 , SCHWARTZ and PARKER 2000 ). Currently, there is no evidence to favor this mechanism of function for the Edc proteins. Two more likely models are that the Edc proteins either facilitate the assembly of a decapping complex or function to enhance the Dcp2p-dependent activation of Dcp1p.

Evidence that the EDC proteins are part of a larger assembly of proteins that performs the decapping reaction comes from the observation that Edc1p co-immunoprecipitates with Dcp1p and Dcp2p (Fig 6). Additional evidence for the assembly of a large protein complex to promote decapping is based on the analyses of other proteins functioning in decapping. In addition to Edc1p and Edc2p, decapping of mRNA is now known to involve the function of a set of interacting proteins, including Dcp1p, Dcp2p, Pat1p/Mrt1p, and the cytoplasmic Like-SM complex of proteins consisting of Lsm1p-Lsm7p (BEELMAN et al. 1996 ; HATFIELD et al. 1996 ; BOECK et al. 1998 ; DUNCKLEY and PARKER 1999 ; BOUVERET et al. 2000 ; THARUN et al. 2000 ). Four observations suggest that these proteins form two distinct complexes that associate together with the mRNA before decapping. First, Dcp1p and Dcp2p copurify (DUNCKLEY and PARKER 1999 ; Fig 1). Second, the Lsm complex, consisting of Lsm1p-Lsm7p, copurifies with Pat1/Mrt1p and with the Xrn1 exoribonuclease (BOUVERET et al. 2000 ). Third, both Dcp1p and Lsm proteins can be shown to co-immunoprecipitate mRNA (THARUN et al. 2000 ). Finally, Dcp1p co-immunoprecipitates with the Lsm complex, but in an RNase-sensitive manner. Our data that Edc1p co-immunoprecipitated with the majority of the Dcp1p and Dcp2p in the cell suggest that Edc1p and Edc2p may be part of this larger assembly.

The interpretation that Edc1p and Edc2p are part of a larger assembly of proteins promoting decapping provides an explanation for their genetic phenotypes. Prior observations have shown that some individual members of large complexes are dispensable for the function of the complex under optimal conditions. However, the function of these individual components can be revealed when the process in which the complex functions is inefficient. The interpretation that Edc1p and Edc2p are modulatory components of a larger decapping complex is consistent with this precedent. Additional evidence that decapping involves a number of redundant protein interactions has come from a mutational analysis of the Dcp1p where several mutations have been identified that individually do not affect decapping but, when combined, give rise to a strong defect in decapping (THARUN and PARKER 1999 ). Since these mutations do not affect the enzymatic function of purified Dcp1p, this suggests that the Dcp1p makes multiple redundant physical interactions that promote decapping. An interesting possibility for future work is to determine if these dcp1 point mutations disrupt the interaction of the Dcp1p with Edc1p or, potentially, Edc2p.

An important goal will be to determine the timing and location of the interaction between Edc1p and Dcp1p/Dcp2p, as well as other proteins involved in decapping. We have shown that the interaction between Edc1p and Dcp1p/Dcp2p is largely eliminated at low-salt concentrations following RNase treatment (Fig 6). This suggests that the presence of RNA stabilizes the interaction between Edc1p and the decapping machinery. Although the identity of the RNA is unknown, a likely possibility is the mRNA itself. The Edc1p and likely Edc2p, because they interact with Dcp1p and Dcp2p, would be in a complex known to associate in some manner with the mRNA. Because Edc1p interacts with most Dcp1p and Dcp2p in the cell, Edc1p would likely interact with Dcp1p and Dcp2p both prior to and after their recruitment to the mRNA. Edc1p and Edc2p may function either to promote the assembly of the decapping complex on the mRNA or, once on the mRNA, Edc1p and Edc2p may stimulate a transition that results in the Dcp2p-mediated activation of Dcp1p. Using the mRNA to stimulate the activation of Dcp1p would provide an elegant mechanism for the localization of decapping activity to mRNAs that have been targeted for degradation.

	ACKNOWLEDGMENTS

We thank all members of the Parker laboratory for thoughtful and helpful discussions of this work. This work was supported by the Howard Hughes Medical Institute and a National Institutes of Health grant to R.P. (GM45443). T.D. was supported by funds from a Cancer Biology Training Grant.

Manuscript received June 23, 2000; Accepted for publication September 13, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANDERSON, J. S. and R. PARKER, 1998  The 3' to 5' degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3' to 5' exonucleases of the exosome complex. EMBO J. 17:1497-1506[Medline].

BASHKIROV, V. I., H. SCHERTHAN, J. A. SOLINGER, J. M. BUERSTEDD, and W. D. HEYER, 1997  A mouse cytoplasmic exoribonuclease (mXRN1p) with preference for G4 tetraplex substrates. J. Cell Biol. 136:761-773[Abstract/Free Full Text].

BEELMAN, C., A. STEVENS, G. CAPONIGRO, T. LAGRANDEUR, and L. HATFIELD et al., 1996  An essential component of the decapping enzyme required for normal rates of mRNA turnover. Nature 283:642-646.

BESSMAN, M. J., D. N. FRICK, and S. F. O'HANDLEY, 1996  The MutT proteins or "Nudix" hydrolases, a family of versatile, widely distributed, "housecleaning" enzymes. J. Biol. Chem. 271:25059-25062[Free Full Text].

BOECK, R., B. LAPEYRE, C. E. BROWN, and A. B. SACHS, 1998  Capped mRNA degradation intermediates accumulate in the yeast spb8-2 mutant. Mol. Cell. Biol. 18:5062-5072[Abstract/Free Full Text].

BOUVERET, E., G. RIGAUT, A. SHEVCHENKO, M. WILM, and B. SERAPHIN, 2000  A Sm-like protein complex that participates in mRNA degradation. EMBO J. 19:1661-1671[Medline].

CAPONIGRO, G. and R. PARKER, 1996  Mechanisms and control of mRNA turnover in Saccharomyces cerevisiae.. Microbiol. Rev. 60:233-249[Free Full Text].

CAPONIGRO, G., D. MUHLRAD, and R. PARKER, 1993  A small segment of the MAT1 transcript promotes mRNA decay in Saccharomyces cerevisiae: a stimulatory role for rare codons. Mol. Cell. Biol. 13:5141-5148[Abstract/Free Full Text].

COUTTET, P., M. FROMONT-RACINE, D. STEEL, R. PICTET, and T. GRANGE, 1997  Messenger RNA deadenylation precedes decapping in mammalian cells. Proc. Natl. Acad. Sci. USA. 94:5628-5633[Abstract/Free Full Text].

DANGEL, A. W., L. SHEN, A. R. MENDOZA, L. C. WU, and C. Y. YU, 1995  Human helicase SKI2W in the HLA class III region exhibits striking structural similarities to the yeast antiviral gene SKI2 and to the human gene KIAA0052: emergence of a new gene family. Nucleic Acids Res. 23:2120-2126[Abstract/Free Full Text].

DECKER, C. J. and R. PARKER, 1993  A turnover pathway for both stable and unstable mRNAs in yeast: evidence for a requirement for deadenylation. Genes Dev. 7:1632-1643[Abstract/Free Full Text].

DUNCKLEY, T. and R. PARKER, 1999  The DCP2 protein is required for mRNA decapping in Saccharomyces cerevisiae and contains a functional MutT motif. EMBO J. 18:5411-5422[Medline].

GERA, J. F. and E. J. BAKER, 1998  Deadenylation-dependent and -independent decay pathways for 1-tubulin mRNA in Chlamydomonas reinhardtii.. Mol. Cell. Biol. 18:1498-1505[Abstract/Free Full Text].

HAGAN, K. W., M. J. RUIZ-ECHEVARRIA, Y. QUAN, and S. W. PELTZ, 1995  Characterization of cis-acting sequences and decay intermediates involved in nonsense-mediated mRNA turnover. Mol. Cell. Biol. 15:809-823[Abstract].

HATFIELD, L., C. A. BEELMAN, A. STEVENS, and R. PARKER, 1996  Mutations in trans-acting factors affecting mRNA decapping in Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:5830-5838[Abstract].

HE, W. and R. PARKER, 1999  Analysis of mRNA decay pathways in S. cerevisiae.. Methods 17:3-10[Medline].

HIGGS, D. C. and J. T. COLBERT, 1994  Oat phytochrome A mRNA degradation appears to occur via two distinct pathways. Plant Cell 6:1007-1019[Abstract].

HSU, C. L. and A. STEVENS, 1993  Yeast cells lacking 5' to 3' exoribonuclease I contain mRNA species that are poly(A) deficient and partially lack the 5' cap structure. Mol. Cell. Biol. 13:4826-4835[Abstract/Free Full Text].

JACOBSON, A. and S. W. PELTZ, 1996  The interrelationships of the pathways of mRNA decay and translation in eukaryotic cells. Annu. Rev. Biochem. 65:693-740[Medline].

JACQUIER, A., J. R. RODRIGUEZ, and M. ROSBASH, 1985  A quantitative analysis of the effects of 5' junction and TACTAAC box mutants and mutant combinations on yeast mRNA splicing. Cell 43:423-430[Medline].

KOONIN, E. V., 1993  A highly conserved sequence motif defining the family of MutT-related proteins from eubacteria, eukaryotes and viruses. Nucleic Acids Res. 21:4847[Free Full Text].

LAEMMLI, U. K., 1970  Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685[Medline].

LAGRANDEUR, T. E. and R. PARKER, 1998  Isolation and characterization of Dcp1p, the yeast mRNA decapping enzyme. EMBO J. 17:1487-1496[Medline].

LEE, S. G., I. LEE, S. H. PARK, C. KANG, and K. SONG, 1995  Identification and characterization of a human cDNA homologous to yeast SKI2.. Genomics 10:660-666.

LIM, S. K. and L. E. MAQUAT, 1992  Human ß-globin mRNAs that harbor a nonsense codon are degraded in murine erythroid tissues to intermediates lacking regions of exon I or exons I and II that have a cap-like structure at the 5' termini. EMBO J. 11:3271-3278[Medline].

MITCHEL, P., E. PETFALSKI, A. SHEVCHENKO, M. MANN, and D. TOLLERVEY, 1997  The exosome: a conserved eukaryotic RNA processing complex containing multiple 3' 5' exoribonuclease activities. Cell 91:457-466[Medline].

MUHLRAD, D. and R. PARKER, 1994  Premature translational termination triggers mRNA decapping. Nature 370:578-581[Medline].

MUHLRAD, D., C. DECKER, and R. PARKER, 1994  Deadenylation of the unstable mRNA encoded by the yeast MFA2 gene leads to decapping followed by 5'-3' digestion of the transcript. Genes Dev. 8:855-866[Abstract/Free Full Text].

MUHLRAD, D., C. DECKER, and R. PARKER, 1995  Turnover mechanisms of the stable yeast PGK1 mRNA. Mol. Cell. Biol. 15:2145-2156[Abstract].

ROSS, J., 1995  mRNA stability in mammalian cells. Microbiol. Rev. 59:423-450[Abstract/Free Full Text].

RUTZ, B. and B. SERAPHIN, 2000  A dual role for BBP/ScSF1 in nuclear pre-mRNA retention and splicing. EMBO J. 19:1873-1886[Medline].

SALGADO-GARRIDO, J., E. BRAGADO-NILSSON, S. KANDELS-LEWIS, and B. SERAPHIN, 1999  Sm and Sm-like proteins assemble in two related complexes of deep evolutionary origin. EMBO J. 18:3451-3462[Medline].

SCHWARTZ, D. C. and R. PARKER, 1999  Mutations in translation initiation factors lead to increased rates of deadenylation and decapping of mRNAs in Saccharomyces cerevisiae.. Mol. Cell. Biol. 19:5247-5256[Abstract/Free Full Text].

SCHWARTZ, D. C. and R. PARKER, 2000  mRNA decapping in yeast requires dissociation of the cap binding protein, eukaryotic translation initiation factor 4E. Mol. Cell. Biol. 20:7933-7942[Abstract/Free Full Text].

THARUN, S. and R. PARKER, 1999  Analysis of mutations in the yeast mRNA decapping enzyme. Genetics 151:1273-1285[Abstract/Free Full Text].

THARUN, S., W. HE, A. E. MAYES, P. LENNERTZ, J. D. BEGGS, and R. PARKER, 2000  Yeast Sm-like proteins function in mRNA decapping and decay. Nature 404:515-518[Medline].

VAN HOOF, A., R. R. STAPLES, R. E. BAKER, and R. PARKER, 2000  Function of the Ski4p (Csl4p) and Ski7p proteins in 3' to 5' degradation of mRNA. Mol. Cell. Biol. 20:8230-8243[Abstract/Free Full Text].

VREKEN, P. and H. A. RAUE, 1992  The rate limiting step in yeast PGK1 mRNA degradation is an endonucleolytic cleavage in the 3' terminal part of the coding region. Mol. Cell. Biol. 12:2986-2996[Abstract/Free Full Text].

This article has been cited by other articles:

				A. Chowdhury and S. Tharun lsm1 mutations impairing the ability of the Lsm1p-7p-Pat1p complex to preferentially bind to oligoadenylated RNA affect mRNA decay in vivo RNA, October 1, 2008; 14(10): 2149 - 2158. [Abstract] [Full Text] [PDF]

				M. J. Taylor and B. A. Peculis Evolutionary conservation supports ancient origin for Nudt16, a nuclear-localized, RNA-binding, RNA-decapping enzyme Nucleic Acids Res., October 1, 2008; 36(18): 6021 - 6034. [Abstract] [Full Text] [PDF]

				A. Georgiev, M. Sjostrom, and A. Wieslander Binding specificities of the GYF domains from two Saccharomyces cerevisiae Paralogs Protein Eng. Des. Sel., September 4, 2007; (2007) gzm041v1. [Abstract] [Full Text] [PDF]

A. Chowdhury, J. Mukhopadhyay, and S. Tharun The decapping activator Lsm1p-7p-Pat1p complex has the intrinsic ability to distinguish between oligoadenylated and polyadenylated RNAs RNA, July 1, 2007; 13(7): 998 - 1016. [Abstract] [Full Text] [PDF]