An accurate measure of the half-life of GFP-SRE+ 
mRNA can be found in Figure S2. doi:10.1371/journal.pone.0047121.gEap1p Functions in Vts1p-Mediated Transcript DecayFigure 2. Eap1p is required for the destabilization of GFP-YIR016W mRNA. mRNA expression was MedChemExpress PD168393 induced in the indicated strains and then shut-off with glucose and reporter mRNA levels were assayed at the times indicated after transcriptional shutoff by Northern blot. The results of at least three independent experiments were quantitated and normalized using the levels of SCR1 RNA and graphed with error bars representing standard deviation. Note that the half-life of GFP-YIR016W mRNA is significantly CASIN increased in both vts1D and eap1D cells as judged by a t test (P,0.01). doi:10.1371/journal.pone.0047121.gstrain this pool is no longer sequestered and as such is free to act on the GFP-SRE- mRNA, thereby accelerating its deadenylation and subsequent degradation. Consistent with this highly speculative model we see a similar stability for GFP-SRE+ and GFP-SREmRNAs in vts1Dcells in transcriptional pulse-chase experiments (Figure S2).Eap1p is not Required for Vts1p-mediated DeadenylationWe next set out to identify at which step in the mRNA decay process Eap1p functions. Given that Vts1p triggers mRNA decay via poly(A) tail shortening, mediated by the Ccr4p-Pop2p-Not deadenylase complex, we analyzed the poly(A) tail length of GFPSRE+ mRNA in eap1D cells to determine whether Eap1p was required for deadenylation of Vts1p 25837696 target mRNAs. We used an RNase H cleavage method to measure the poly(A) tail length of the GFP-SRE+ mRNA over an 8-min transcriptional pulse-chase time course. Total RNA samples from each time point were treated with RNase H in the presence of a specific antisense oligonucleotide that hybridized to the GFP open reading frame. After RNase H cleavage, the poly(A) tail of the resulting short 39-end fragment of the GFP reporter mRNA was measured by high resolution polyacrylamide electrophoresis and Northern blotting. To one sample specific oligonucleotide and oligo(dT) were added and, as such, treatment with RNase H yielded a 39-end fragment lacking a poly(A) tail, which served as a marker for the deadenylated 39end fragment. Comparison of the RNase H cleavage data from wild-type cells and eap1D cells (Figure 4) indicate that Eap1p is not required for deadenylation of the GFP-SRE+ mRNA. For example, wild-type cells at time zero showed two populations of GFP-SRE+ mRNA. One population had tail lengths from 30 to 50 nucleotides while the other had very short tails presumably representing transcriptsFigure 3. Eap1p is not required for the degradation of GFP-SREmRNA. GFP-SRE- gene transcription was induced in wild-type, vts1D and eap1D cells with galactose and then shut off with glucose and reporter mRNA levels were assayed at the times indicated after transcriptional shutoff by Northern blot. The results of at least two independent experiments were quantitated and normalized using the levels of SCR1 RNA and graphed with error bars representing standard deviation. doi:10.1371/journal.pone.0047121.gEap1p Functions in Vts1p-Mediated Transcript Decaythat were deadenylated during the 16-min induction with galactose indicating the transcript was being rapidly deadenylated. At time zero eap1D cells also showed these same two populations of GFP-SRE+ mRNA, again suggesting that this mRNA is rapidly deadenylated. In fact, eap1D cells show an increase in the fraction of deadenylated GFP-.An accurate measure of the half-life of GFP-SRE+ mRNA can be found in Figure S2. doi:10.1371/journal.pone.0047121.gEap1p Functions in Vts1p-Mediated Transcript DecayFigure 2. Eap1p is required for the destabilization of GFP-YIR016W mRNA. mRNA expression was induced in the indicated strains and then shut-off with glucose and reporter mRNA levels were assayed at the times indicated after transcriptional shutoff by Northern blot. The results of at least three independent experiments were quantitated and normalized using the levels of SCR1 RNA and graphed with error bars representing standard deviation. Note that the half-life of GFP-YIR016W mRNA is significantly increased in both vts1D and eap1D cells as judged by a t test (P,0.01). doi:10.1371/journal.pone.0047121.gstrain this pool is no longer sequestered and as such is free to act on the GFP-SRE- mRNA, thereby accelerating its deadenylation and subsequent degradation. Consistent with this highly speculative model we see a similar stability for GFP-SRE+ and GFP-SREmRNAs in vts1Dcells in transcriptional pulse-chase experiments (Figure S2).Eap1p is not Required for Vts1p-mediated DeadenylationWe next set out to identify at which step in the mRNA decay process Eap1p functions. Given that Vts1p triggers mRNA decay via poly(A) tail shortening, mediated by the Ccr4p-Pop2p-Not deadenylase complex, we analyzed the poly(A) tail length of GFPSRE+ mRNA in eap1D cells to determine whether Eap1p was required for deadenylation of Vts1p 25837696 target mRNAs. We used an RNase H cleavage method to measure the poly(A) tail length of the GFP-SRE+ mRNA over an 8-min transcriptional pulse-chase time course. Total RNA samples from each time point were treated with RNase H in the presence of a specific antisense oligonucleotide that hybridized to the GFP open reading frame. After RNase H cleavage, the poly(A) tail of 
the resulting short 39-end fragment of the GFP reporter mRNA was measured by high resolution polyacrylamide electrophoresis and Northern blotting. To one sample specific oligonucleotide and oligo(dT) were added and, as such, treatment with RNase H yielded a 39-end fragment lacking a poly(A) tail, which served as a marker for the deadenylated 39end fragment. Comparison of the RNase H cleavage data from wild-type cells and eap1D cells (Figure 4) indicate that Eap1p is not required for deadenylation of the GFP-SRE+ mRNA. For example, wild-type cells at time zero showed two populations of GFP-SRE+ mRNA. One population had tail lengths from 30 to 50 nucleotides while the other had very short tails presumably representing transcriptsFigure 3. Eap1p is not required for the degradation of GFP-SREmRNA. GFP-SRE- gene transcription was induced in wild-type, vts1D and eap1D cells with galactose and then shut off with glucose and reporter mRNA levels were assayed at the times indicated after transcriptional shutoff by Northern blot. The results of at least two independent experiments were quantitated and normalized using the levels of SCR1 RNA and graphed with error bars representing standard deviation. doi:10.1371/journal.pone.0047121.gEap1p Functions in Vts1p-Mediated Transcript Decaythat were deadenylated during the 16-min induction with galactose indicating the transcript was being rapidly deadenylated. At time zero eap1D cells also showed these same two populations of GFP-SRE+ mRNA, again suggesting that this mRNA is rapidly deadenylated. In fact, eap1D cells show an increase in the fraction of deadenylated GFP-.