Transcription in the nucleus and mRNA decay in the cytoplasm are coupled processes

Abstract

Maintaining appropriate mRNAs levels is vital for any living cell. mRNA synthesis in the nucleus by RNA polymerase II core enzyme (Pol II) and mRNA decay by cytoplasmic machineries determine these levels. Yet, little is known about possible cross-talk between these processes. The yeast Rpb4/7 is a nucleo-cytoplasmic shuttling heterodimer that interacts with Pol II and with mRNAs and is required for mRNA decay in the cytoplasm. Here we show that interaction of Rpb4/7 with mRNAs and eventual decay of these mRNAs in the cytoplasm depends on association of Rpb4/7 with Pol II in the nucleus. We propose that, following its interaction with Pol II, Rpb4/7 functions in transcription, interacts with the transcript cotranscriptionally and travels with it to the cytoplasm to stimulate mRNA decay. Hence, by recruiting Rpb4/7, Pol II governs not only transcription but also mRNA decay.

Figure 1.

rpb6^Q100R cells exhibit defective mRNA decay. (A) Cells growing optimally at 30°C were shifted rapidly to 42°C to block transcription naturally (Lotan et al. 2005, 2007). Post-shift decay kinetics were determined by Northern analysis, using probes indicated on the left (see the Materials and Methods). T_1/2 were determined as in B. The ratio between the T_1/2 of a given mRNA in the mutant and its T_1/2 in the wild type (T_1/2 [rpb6]/T_1/2 [WT]) is indicated on the right. (B) Half-lives (T_1/2) were determined by plotting mRNA levels as a function of time post-transcription block. A graphic illustration of RPL29 mRNA decay kinetics is shown as an example. To obtain this graph, the intensity of each band, determined by PhosphorImager technology, was normalized to that of SCR1 (a Pol III transcript). The normalized band intensity at time 0 (before transcription arrest) was defined as 100% and the intensities at the other time points were calculated relative to time 0. Results were plotted as a function of time. The graph represents an average of three independent assays. Error bars indicate the standard deviation from the mean values. Half-lives were obtained from these graphs and are depicted on the right. Variations in the calculated half-lives were <15%. (C) Transcription was blocked as in A. RNA samples analyzed by the PAGE-Northern technique (Sachs and Davis 1989), using the probes indicated on the right. The positions of fully adenylated (A_n) and deadenylated (A_0–10) RNAs are shown on the left. Lane “Δ(A)_n” shows the position of fully deadenylated RNA (see Lotan et al. 2007). The asterisk (*) indicates the time point when deadenylation is estimated to be complete. The proportional decrease in radioactivity between this time point and the following one was assessed using PhosphorImager technology (normalized to SCR1) and is depicted underneath the respective lanes to estimate the stability of the deadenylated RNA. Quantitative illustration of the deadenylation kinetics is shown in Supplemental Figure S3. (D) rpb6^Q100R cells degrade MFA2pG mRNA abnormally slowly and accumulate abnormally high decay intermediate of MFA2pG mRNA. Wild-type and rpb6^Q100R cells expressing Tet-Off-MFA2pG were grown at 30°C to mid-logarithmic phase before transcription was blocked by adding doxycycline (2 μg/mL) (Hilleren and Parker 2003). Following drug addition, cultures continued to be shaken at 30°C and samples were taken at the indicated time points. RNA samples analyzed by the PAGE-Northern technique (Sachs and Davis 1989), hybridized with an MFA2pG-specific probe (see the Materials and Methods). The position of MFA2pG full-length mRNA (FL) is indicated on the left. “Frag.” indicates a degradation intermediate that accumulates due to a poly(G) tract that blocks 5′-to-3′ exonuclease activity by Xrn1p (Vreken and Raue 1992; Decker and Parker 1993). The ratios between this fragment and MFA2pG, the full-length one, at 0 time points (at steady state) were determined by PhosphorImager technology (normalized to SCR1 RNA) and are indicated at the bottom as percentages. Half-lives were calculated as in B; in wild type it was 12 min and in the mutant, 25 min. The positions of the size marker bands are shown on the right. (E) rpb6^Q100R cells are hypersensitive to deletion of XRN1. Strains, indicated on the left, carrying pRPB4/7/URA3/2μ (overexpressing Rpb4/7) were spotted, in 10-fold dilutions, on 5-FOA-containing plate (the drug kills cells expressing URA3). In parallel, cells were spotted on a synthetic medium lacking uracil that selects for cells carrying the plasmid (designated SC–uracil). Plates were incubated for 3 d at 30°C.

Figure 2.

Overexpression of RPB4/7 in rpb6^Q100R cells suppresses partially the defect in mRNA decay. mRNA decay was monitored as in Figure 1B. (Top panels) Graphic representation of RPL29 mRNA decay kinetics in the indicated isogenic strains was generated as in Figure 1B. (Bottom panels) Autoradiograms showing the decay kinetics of several mRNAs (indicated on the left). The presence (+) or absence (−) of the high-copy plasmid overexpressing RPB4/7 is indicated above the autoradiograms. Half-lives (T_1/2) were determined as in Figure 1B. (A) Overexpression of RPB4/7 does not change mRNA decay kinetics in wild-type cells. The ratios between the T_1/2 in wild-type cells expressing normal levels of Rpb4/7 and the T_1/2 in wild-type cells overexpressing Rpb4/7 (“T_1/2 ratio”) are indicated on the right. Variations were <20%. (B) Overexpression of RPB4/7 accelerates mRNA decay in rpb6^Q100R cells. T_1/2 are indicated on the right. Variations were <20%.

Figure 3.

Efficient association of Rpb4/7 with mRNAs is dependent on Rpb6p. Equal amounts of cell extracts from wild-type or from rpb6^Q100R cells carrying or lacking TAP-tagged RPB4, as indicated, were subjected to RNA immunoprecipitation (RIP) as detailed in the Materials and Methods. Prior to the RIP procedure, 5% of each extract was removed to determine the levels of the indicated mRNAs in the input material. cDNA was synthesized using either random hexamer (A) or oligo(dT) (B–D) as the primer, in the presence (+) or absence (−) of reverse transcriptase (RT). cDNA levels were determined by qPCR (see the Materials and Methods). The level of each cDNA produced from the IP material was normalized to that produced from the input RNA, diluted 1:1500-fold. Each bar represents average of three experiments; each one was done in triplicate. Error bars indicate the standard deviation. (A) Relative cDNA levels obtained using the random primers. Each bar represents the ratio of cDNA level of RPS22B mRNA (normalized to that of the input) divided by that of SCR1 cDNA (normalized to that of the input). (B) Levels of PCR-amplified RPS22B cDNA, obtained using oligo(dT), determined by Southern analysis. PCR was carried out by the real time machine. At late-exponsntial phase (cycle 31), the sample harvested and electrophoresed in 2% gel followed by Southern analysis using RPS22B probe. (C,D) Levels of the indicated cDNAs obtained using the oligo(dT) primer.

Figure 4.

Cells carrying rpb1^{C67S; C70S}, but not another ts allele of RPB1, exhibit defective mRNA decay. (A) Decay kinetics in various mutant cells carrying transcriptionally defective RPB1 allele. Decay kinetics was determined as in Figure 1A. Strains, indicated on top of the autoradiograms are described in Donaldson and Friesen (2000). Half-lives were determined as in Figure 1C. The ratios between the T_1/2 values in the mutant cells and those in the wild-type cells (“T_1/2 ratio”) are indicated on the right. (B) rpb1^{C67S; C70S} cells exhibit defective mRNA deadenylation and subsequent decay. Deadenylation kinetics was determined by PAGE-Northern as in Figure 1C. Half-lives were determined using Northern blot hybridization (not shown), as in A. Stability of the deadenylated RPL29 RNA was estimated as described in Figure 1C.

Figure 5.

Pol II controls the two major mRNA decays in the cytoplasm via Rpb4/7 that serves as a mediator. (A) Maintaining proper levels of mRNA: schematic representation of the roles played by Pol II and Rpb4/7 in a wild-type cell. Only a small portion of the nuclear Rpb4/7 is recruited to Pol II and is involved in transcription initiation (Choder 2004), elongation (Verma-Gaur et al. 2008), and termination (Runner et al. 2008). At some stage during transcription, Rpb4/7 interacts with the transcript. This conditional interaction is dependent upon its proper interaction with Pol II (Fig. 3). Following transcription, the Rpb4/7–RNA complex is exported out of the nucleus by Rpb4p-mediated manner (this feature is apparent only during stress) (Farago et al. 2003). Consistently, Rpb4/7 export is dependent on transcription (Selitrennik et al. 2006). At an undefined stage, Rpb4/7 helps recruiting Pat1p to the mRNA, by virtue of its capacity to interact with both Pat1p and the mRNA (Lotan et al. 2005, 2007). Rpb4/7 then stimulates shortening of the poly(A) tail by an unknown mechanism (Figs. 1C, 4B; Supplemental Fig. S3; Lotan et al. 2005, 2007). Following this stage, Rpb4/7 is involved in stimulating both major pathways of mRNA degradation (Lotan et al. 2005, 2007). The role of Rpb4/7 in the 5′-to-3′ pathway involves its interaction with Pat1p (Lotan et al. 2005, 2007) and probably stabilizing Lsm1–7 complex interaction with the mRNP (Lotan et al. 2005). The Rpb4/7–mRNP complex enters a P body (Lotan et al. 2005, 2007), where mRNA degradation is executed (association with P bodies is apparent mainly during starvation). Rpb4/7 also plays a role in the 3′-to-5′ degradation pathway. This role seems to be important as viability of mutant cells carrying some rpb7 alleles, unlike that of wild-type cells, is dependent on the 5′-to-3′ pathway (Lotan et al. 2007). (B) Effect of defect in Pol II capacity to recruit Rpb4/7. Since the interaction of Rpb4/7 with mRNAs is dependent on its binding with Pol II (Fig. 3), there is little interaction in the mutant cells. Consequently, every Rpb4/7-mediated stage is adversely affected.