The mRNA decay pathway is necessary for adherence to plastic and wild-type FLO11 expression. (A) The Σ1278b strain (MATα ura3-52 his3:hisG leu2:hisG) was used for all experiments. The wild-type strain and the flo11Δ mutant were a generous gift from Todd Reynolds and are described in . Deletion mutants in the mRNA decay pathway components were constructed by standard methods. Growth of the indicated strains was assessed on 2% agar YPD plates, using 10× serial dilutions (starting from OD₆₀₀ = 0.5). Plates were photographed after 3 days of growth at 30°. (B) For testing adherence to polystyrene, overnight cultures were grown in 2% glucose synthetic media and then diluted to OD₆₀₀ ≈ 1 into 0.2% glucose media before inoculation into 96-well polystyrene plates. The cells were allowed to adhere for 90 min (at 30°, 75 rpm), followed by washes in PBS. Fresh 0.2% glucose media was then dispensed into each of the wells, and the adherent cells further incubated for 6 and 24 hr. Quantification was performed by crystal violet staining and reading OD₅₉₅ after washing and destaining in 95% ethanol. Shown are the averages of three biological repeats assayed in quadruplicate and the standard error. Independent deletion clones of the mutants (two to three) were tested and gave analogous results. For the ccr4Δ, pop2Δ, and dhh1Δ strains, the P-values were <0.001 at all time points. For the puf5Δ strain, statistically significant differences were observed at 90 min (P = 0.017) and at 24 hr (P < 0.001). At 6 hr, a slight reduction in adherence was observed, but this was not statistically significant (P = 0.38). (C) Mat formation was assessed on 0.3% agar YPD plates at room temperature as described previously (). The mutants were assayed against the wild type and the flo11Δ negative control on several separate occasions, and a composite of the performed experiments is shown. Two to three independently constructed deletion clones for each of the mutants were tested and gave similar results, except for ccr4Δ, for which independent clones displayed defective mat formation, although clone 1 was more affected than clone 2. Four other ccr4Δ clones were tested, of which two behaved as clone 1, one as clone 2, and one had an intermediate phenotype between clones 1 and 2 (Figure S1). The variability of ccr4Δ clones 1 and 2 was not observed in quantitative assays of attachment to plastic or FLO11 expression, with both strains displaying comparable defects (see B and D). (D) The expression levels of FLO11 were determined by quantitative real time PCR (qPCR) from cells grown in YPD at 30°. RNA was prepared by the hot phenol method. Reverse transcription was performed, and the resulting cDNA was used as the template in qPCRs essentially as described (). The qPCR data were analyzed with the LinReg PCR program (; ). ACT1 levels were used for normalization. All values were expressed relative to the average of the wild type. Shown are averages of at least three independent cultures and the standard error. The P-value was <0.001 for all mutants.

The mRNA decay pathway negatively regulates the expression of the FLO11 repressors NRG1 and NRG2. (A) qPCRs were performed as described in . Shown are averages of at least three independent biological replicates and the standard error. P-value was <0.001 for all mutants. (B) mRNA poly(A) analysis was performed using a modified version of the poly(A) test (PAT) assay (; ; ; ). This is a reverse transcription/PCR assay where the size of the PCR products reflects the size of the mRNA poly(A) tail. The shortest tail detected in this assay is ∼12 bp [shown by the TVN sample in B, for which a primer binding to the 3′ UTR–poly(A) junction is used]. The samples were spiked before reverse transcription with human HeLa RNA, and human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was assayed as a control, showing that the assay works equally well between samples for detecting the length of the poly(A) tail. Gene-specific PCR products were analyzed by 2% high-resolution agarose gel, prestained with SYBR safe, and imaged against a 100-bp ladder using an LAS 3000 imager and multiguage software (Fujifilm). Whether the tails are preferentially shorter or longer was determined by quantifying the signals corresponding to the longer or shorter forms in the wild-type samples (the dotted boxes in the second set of samples indicate how the quantification was performed). A ratio of long/short poly(A)-tailed forms of <1 is typical of a short-tailed transcript (). (C) The 3′ UTR regions of NRG1 and NRG2 (200 bp downstream of the STOP codon) were searched for putative PUF-binding sites. The NRG1 3′ UTR contains a sequence corresponding to the PUF5-1 motif identified by bioinformatic searches, which includes a core PUF consensus sequence UGUR starting 87 bp after the stop codon (). The 3′ UTR of NRG2 did not contain a consensus PUF-binding site. (D) The experiments testing adherence to polystyrene were performed as in . The yeast strain overexpressing NRG1 and NRG2 (bottom, red bars) was constructed by placing the genes under the control of the strong constitutive promoter TEF1 at the endogenous genomic locus using PCR and homologous recombination and with the plasmids pYM-N18 and pYM-N19 as PCR templates (). qPCRs were used to confirm elevated expression levels, which were 7.2-fold (±0.28 SE) for NRG1 and 2.8-fold (±0.38 SE) for NRG2. Deletion of NRG1 and NRG2 caused increased adherence in the wild-type strain, consistent with a previous report (), but the effect was more pronounced in the ccr4Δ background (statistical significance at all points was with P ≤ 0.05). For example, at 90 min, the fold difference between the wild type and the nrg1Δ nrg2Δ mutant was 1.23-fold, whereas the difference between the ccr4Δ strain and the triple mutant was 3.65-fold. (E) qPCR experiments were performed as described in . The cells were grown in conditions used to assay adherence to polystyrene, and FLO11 expression was assayed after 2 hr in 0.2% glucose synthetic complete media. Deletion of the repressors NRG1 and NRG2 in the ccr4Δ strain rescued FLO11 levels by 9.4-fold (P < 0.001), while, in the wild type, background deletion of the two repressors caused a 2.4-fold upregulation of FLO11 (P = 0.005).

Model for the control of FLO11 expression by the mRNA decay pathway. The mRNA decay pathway keeps the levels of the NRG1 mRNA low by targeted Ccr4-dependent deadenylation and decay. Puf5 provides specificity by binding to the NRG1 3′ UTR and recruiting the Ccr4-NOT-Dhh1 complex. Dhh1 acts on the mRNA 5′ cap as an activator of decapping and translational inhibitor. The length of the mRNA poly(A) tail positively correlates with translation, and thus targeted deadenylation by Ccr4 not only causes decay, but likely also limits translation of the NRG1 mRNA. In the mutants inactivated for the components of the mRNA decay pathway, the mRNA and protein levels for NRG1 are higher, causing increased repression of FLO11. A similar mechanism might be operating on the NRG2 transcript.