Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 1.
Published in final edited form as: Virology. 2016 Sep 4;498:181–191. doi: 10.1016/j.virol.2016.08.021

Evidence That a Threshold of Serine/Arginine-Rich (SR) Proteins Recruits CFIm to Promote Rous Sarcoma Virus mRNA 3′ End Formation

Stephen W Hudson 1, Lisa M McNally 1, Mark T McNally 1,*
PMCID: PMC5045808  NIHMSID: NIHMS814843  PMID: 27596537

Abstract

The weak polyadenylation site (PAS) of Rous sarcoma virus (RSV) is activated by the juxtaposition of SR protein binding sites within the spatially separate negative regulator of splicing (NRS) element and the env RNA splicing enhancer (Env enhancer), which are far upstream of the PAS. Juxtaposition occurs by formation of the NRS – 3′ ss splicing regulatory complex and is thought to provide a threshold of SR proteins that facilitate long-range stimulation of the PAS. We provide evidence for the threshold model by showing that greater than three synthetic SR protein binding sites are needed to substitute for the Env enhancer, that either the NRS or Env enhancer alone promotes polyadenylation when the distance to the PAS is decreased, and that SR protein binding sites promote binding of the polyadenylation factor cleavage factor I (CFIm) to the weak PAS. These observations may be relevant for cellular PASs.

Keywords: Rous sarcoma virus, SR proteins, RNA splicing, polyadenylation, 3′ end processing, CFIm

Graphical Abstract

graphic file with name nihms814843u1.jpg

Introduction

Similar to host cell mRNAs, production of mRNAs from retroviral proviruses is initiated by RNA polymerase II and involves additional host RNA processing machineries to carry out capping at the 5′ end, RNA splicing, and 3′ end formation (or polyadenylation) to generate a poly(A) tail. There is considerable evidence that these processes are coupled for mammalian messages to ensure that all steps proceed efficiently, and defects in one process can be detrimental to another (Maniatis and Reed, 2002; Niwa and Berget, 1991; Rigo and Martinson, 2008). While most host mRNAs are subject to alternative splicing, which can include intron retention, most of the introns are removed in the final product (Lee and Rio, 2015). This contrasts with retroviruses where splicing is required to generate sub-genomic mRNAs (in the simplest cases, pol and env mRNAs), but splicing is regulated since most of the RNA remains full length to act as mRNA for gag-pol and serve as genome for progeny virions (Goff, 2007; McNally, 2008). This predominant lack of splicing of retroviral mRNA poses a conundrum in coupling of RNA processing events, and 3′ end formation in particular, since RNA splicing promotes polyadenylation (Cooke and Alwine, 2002; Kyburz et al., 2006; Millevoi et al., 2006; Niwa et al., 1990; Vagner et al., 2000b). Despite the fact that majority of Rous sarcoma virus (RSV) mRNA is unspliced and the PAS is weak (as evidenced by its poor use in vitro (Maciolek and McNally, 2008)), polyadenylation is reasonably efficient in infected cells (~85% polyadenylated), which indicates the existence of a novel mechanism(s) to ensure proper polyadenylation.

A number of mechanisms account for accumulation of substantial levels (~75%) of unspliced mRNA in RSV, including a novel RNA processing control element termed the negative regulator of splicing (or NRS) that also coordinates polyadenylation (McNally, 2008). The two spliced RNAs are generated from a single 5′ splice site (ss) that can be paired with the alternative env 3′ or src 3′ splice sites, each of which is suboptimal and contributes to splicing control (Katz and Skalka, 1990; Zhang and Stoltzfus, 1995). In contrast to these splice site-associated sequences, the NRS lies between the splice sites at ~400 nt downstream of the 5′ ss and ~4,000 nt from the env 3′ ss (see Fig. 1A)(Arrigo and Beemon, 1988; McNally et al., 1991; Stoltzfus and Fogarty, 1989). The NRS has been extensively studied and is thought to function as a pseudo 5′ ss that non-productively interacts with the 3′ ss, which prevents splicing by sequestering the 3′ splice sites from the authentic 5′ ss (Cook and McNally, 1998; Giles and Beemon, 2005; Hibbert et al., 1999; McNally and McNally, 1998, 1999). The central requirements for splicing inhibition by the NRS are strong upstream binding sites for SR protein family splicing factors (Cook and McNally, 1998; McNally and McNally, 1996, 1998) that recruit U1 snRNP to the downstream, non-functional pseudo 5′ ss where U1 snRNP initiates non-productive interactions with the 3′ splice sites (Giles and Beemon, 2005; Hibbert et al., 1999; McNally and McNally, 1999). The existence of the stalled NRS – 3′ss splicing complex suggested that one or more components of this complex might couple to polyadenylation in the absence of bona fide splicing.

Fig. 1.

Fig. 1

Open in a new tab

Polyadenylation efficiency increases with the number of synthetic SR protein binding sites. A. Schematic representation of proviral expression constructs (not to scale). Terminal boxes represent the long terminal repeats. Shown are the 5′ss and env and src 3′ splice sites, NRS (dark green box), Env enhancer (Enh, light green box), the gag, pol, env, and src genes, and the viral polyadenylation site (PAS); the downstream CAT gene and SV40 PAS are not shown. Deletions are represented as gaps in the boxes. ΔEnh lacks nt 5107 to 5186; ΔEnh-nX SRSF1 have inserted 1, 3, 5, or 7 high affinity SRSF1 binding sites; ΔEnh-3xMut has three mutated SRSF1 binding sites. At the bottom is a schematic of the RPA probe and protected products. B. CEFs were transfected with the indicated proviral clones and total RNA isolated 48 hr later was subjected to RPA. Products were resolved by electrophoresis on an 8M urea-6% polyacrylamide gel and analyzed with a phosphorimager. The probe and protected products representing read-through (RT) and correctly polyadenylated RNA (polyA), and unspliced (unspl) and spliced (spl) transcripts, are indicated on the right. M, size standards of the indicated sizes. C. Quantitation of percent transcripts processed at the RSV PAS from at least three independent experiments. Percent RSV PAS use was calculated as the units of polyA/polyA + RT, and for splicing spl/spl + unspl, and units were corrected for U content in the bands. Error bars represent +/− standard error of the mean.

Early work identified two far upstream elements that were required for optimal polyadenylation (Miller and Stoltzfus, 1992) that were later mapped to the NRS and the ASLV exonic splicing enhancer (referred to here as the Env splicing enhancer) (Fogel et al., 2002; Hudson and McNally, 2011; O’Sullivan et al., 2002). These observations suggested a role for SR proteins in polyadenylation. The upstream portion of the NRS binds several SR proteins and can act as a potent RNA splicing enhancer in reporter constructs (Fogel et al., 2002; McNally and McNally, 1996, 1998). The Env enhancer located just downstream of the env 3′ss and was first described as an element required for optimal use of the env 3′ss, and sequences within it are similar to cellular splicing enhancers (Fu et al., 1991; Xu et al., 1993). Subsequently, a role for the Env enhancer in splicing of RSV minigenes was shown, and it predominantly binds SRSF5 (Bouck et al., 1998; Staknis and Reed, 1994).

A number of observations point to SR protein splicing factors as the key proteins involved in promoting the use of the RSV PAS; i) the RSV PAS is poorly used in vitro but is stimulated by the portion of the NRS that binds SR proteins, or by synthetic SR protein binding sites (Maciolek and McNally, 2007, 2008), ii) synthetic SR protein binding sites support optimal polyadenylation in proviral constructs, but only when repositioned closer to the PAS via the NRS – 3′ss complex or genetically through deletions (Maciolek and McNally, 2007), iii) in addition to the NRS, SR protein binding sites within the Env splicing enhancer are required, and the two need to be juxtaposed through NRS – 3′ss complex formation or through deletions (Hudson and McNally, 2011), and iv) the Env splicing enhancer stimulates the viral PAS in vitro (Hudson and McNally, 2011). Collectively, the evidence supports a model in which polyadenylation is stimulated by SR proteins through a long range, splicing-independent coupling mechanism (the NRS – 3′ss complex) that repositions two spatially separate SR binding platforms closer to the PAS. Here, we provide further evidence for the SR protein threshold model of long-range PAS stimulation, and provide evidence for the recruitment of the CFIm polyadenylation factor by SR proteins as a potential mechanism.

Results

Polyadenylation efficiency increases with the number of synthetic SR protein-binding sites

It was previously shown that the NRS and Env enhancer, each of which bind SR proteins but are spatially separated, are required for proper polyadenylation in RSV, but the two elements must be juxtaposed through formation of the NRS complex (Hudson and McNally, 2011). Neither the NRS nor the Env enhancer is able to stimulate polyadenylation in the absence of the other (Hudson and McNally, 2011; Maciolek and McNally, 2007). Likewise, three high-affinity binding sites (called 3xASF) for the SR protein splicing factor SRSF1 could substitute for the NRS, but not in the absence of the Env enhancer (Hudson and McNally, 2011). These observations suggest that stimulation of the RSV PAS is sensitive to the number of upstream SR protein-binding sites and that a threshold level of SR proteins is required for proper polyadenylation. The stimulation seen with the 3xASF was weaker than with the NRS (Hudson and McNally, 2011), which binds several SR proteins and has strong splicing enhancer activity (McNally and McNally, 1996, 1998). This finding is consistent with the observation that stimulation of the RSV PAS is sensitive to the number of upstream SR protein sites. Therefore, we predicted that polyadenylation efficiency should be proportional to the number of SR protein-binding sites at the Env enhancer.

To test the idea that a threshold of SR proteins is required to stimulate RSV polyadenylation, the Env enhancer was substituted with increasing numbers (1, 3, 5 and 7) of synthetic SR protein-binding sites for the SR protein SRSF1 (Fig. 1A). One construct served as a negative control and contained three SRSF1 sites with the T to A changes (3xMut), which no longer binds SRSF1 and has no enhancer activity (Tacke and Manley, 1995). Plasmids were transfected into chicken embryo fibroblasts (CEFs), and use of the viral PAS was determined by RNase protection using a probe that simultaneously monitors viral splicing and polyadenylation (Fig. 1A)(Fogel et al., 2002). As shown in Figure 1B and quantitated in Figure 1C, approximately 90% of wild-type RNA utilized the viral PAS whereas deletion of the Env enhancer (ΔEnh) decreased correct polyadenylation to ~63% (lanes 4 and 5), as expected. A single (1x) or three high affinity sites (3x) failed to improve polyadenylation beyond what is seen in ΔEnh (lanes 6 and 7), but both 5x and 7x constructs showed polyadenylation levels that were similar to WT (lanes 8 and 9). The mutated control sequence was unable to enhance polyadenylation (lane 10).

Because the Env enhancer also has a regulatory function in promoting splicing to the env 3′ss, and splicing and polyadenylation are known to be coupled, the total splicing levels for each construct was also quantitated (Fig. 1C). It is important to note that the RPA probe only monitors splicing from the 5′ss and therefore total splicing, and not splicing to individual downstream 3′ splice sites. All constructs showed total splicing levels similar to WT (lanes 4–10). This is because any reduction in splicing to the env 3′ss is compensated by a shift to use of the src 3′ ss (Berberich and Stoltzfus, 1991). Since global splicing was not affected by the Env enhancer deletion, the differences in polyadenylation cannot be attributed to effects at the level of coupled splicing and polyadenylation. These data support the hypothesis that the number of SR protein binding sites is important to achieve a threshold of SR proteins to support proper polyadenylation; the threshold is achieved by juxtaposition of the NRS and Env enhancer.

The NRS/Env enhancer stimulate polyadenylation more efficiently when repositioned to within ~1,000 nt of the viral polyA site

If juxtaposition of SR protein binding sites contributes to accumulation of a threshold of SR proteins that is required to promote 3′ end formation at the distant viral PAS, then decreasing the distance between the NRS or Env enhancer and the PAS is predicted to eliminate the requirement for both elements. To test this prediction, deletions in the env and src regions (ΔAM and ΔAT, see below) were designed to reposition the NRS and Env enhancer closer to the PAS in an otherwise wild-type context or in a context where the NRS and Env enhancer were genetically juxtaposed by deleting gag/pol (ΔGP) sequences. The ΔGP deletion places the SR protein binding portion of the NRS and Env enhancer ~231 nt apart and also removed the env 3′ ss and part of the Env enhancer (Fig. 2A). We found that combining the ΔGP and ΔAM/ΔAT deletions, which removes both 3′ splice sites, resulted in poor RNA expression. The decreased expression may be due to inhibition of polyadenylation by the lone 5′ss or instability associated with the lack of a 3′ss (Gunderson et al., 1998; Phillips et al., 2004; Vagner et al., 2000a). To circumvent the expression defect, an alternative ΔGP construct, GP ΔGPII (Fig. S1), was created by making a gag/pol deletion that preserved the env 3′ss. The NRS and Env enhancer SR protein binding regions are ~403 nt apart and this construct directed expression of normal levels of correctly polyadenylated mRNA in transfected CEFs (Fig. S1) and deletion of the NRS resulted a polyadenylation defect (lane 8), as expected (Hudson and McNally, 2011; Maciolek and McNally, 2007).

Fig. 2.

Fig. 2

Open in a new tab

The NRS and Env enhancer stimulate polyadenylation more efficiently when repositioned to within 1,000 nts of the viral PAS. A. Schematic representation of proviral expression constructs (not to scale) as described in Fig. 1. ΔNRS lacks nt 706 to1006; ΔEnh lacks nt 5107 to 5186; ΔGPII lacks nt 1026 to 4995; ΔAM lacks nt 5442 to 7901; ΔAT lacks nt 5442 to 8297. B. Quantitation of percent transcripts processed at the RSV PAS from at least three independent experiments. CEFs were transfected with the indicated proviral clones and total RNA isolated 48 hr later was subjected to RPA. Products were resolved by electrophoresis on an 8M urea-6% polyacrylamide gel and analyzed with a phosphorimager. Percent RSV PAS use was calculated as the units of polyA/polyA + RT, and units were corrected for U content in the bands. Error bars represent standard error of the mean. Where indicated (*), p<0.01. NS, not significant.

To bring the NRS and the Env enhancer closer to the viral PAS, deletions were made in the env/src region that positioned the NRS and Env enhancer closer to the PAS. The first deletion, ΔAM, moved the Env enhancer (and NRS complex) to within ~1,400 nt of the PAS in the wild-type context (Fig. 2A) and caused a small but statistically significant decrease in polyadenylation efficiency compared to wild type, but the decrease was not as severe as ΔNRS or ΔEnh (Fig. 2B). The ΔAM decrease was also not corrected when the NRS and Env enhancer were fused by the ΔGPII deletion. The ΔAT deletion placed the Env enhancer ~1,000 nts from the PAS site and in the ΔGPII context, the NRS and Env enhancer are ~1,400 and 1,000 nts, respectively, from the PAS. As with ΔAM, a small decrease in polyadenylation was observed with the ΔAT deletion. The decreases with ΔAM and ΔAT suggest that there may be and additional, weak element(s) in the src region that influence 3′ end formation. When the NRS and Env enhancer were genetically fused in ΔGPIIΔAT, polyadenylation efficiency reverted to wild type (Fig. 2B). These results suggest that the src 3′ss and other putative elements are dispensable for proper polyadenylation in the ΔGPIIΔAT context, and that the NRS/Env enhancer are more potent when moved to within 1,000 nt of the PAS, as opposed to 1,400 nt. The latter point is an indication that the distance between the SR protein binding sites and the PAS is important.

The Env enhancer alone stimulates polyadenylation when repositioned to within ~1,000 nt of the viral polyA site

We next tested whether the Env enhancer alone could support optimal polyadenylation when positioned close to the PAS. Constructs containing the Env enhancer alone at either 1,400 nts (ΔAM) or 1,000 nts (ΔAT) from the PAS (Fig. 3A) were transfected into CEFs, and use of the viral PAS was determined by RPA. Upon deletion of the NRS (ΔNRSΔGPIIΔAM), leaving the enhancer alone at 1,400 nts, there was a significant decrease in PAS use compared to the construct harboring the both NRS and the Env enhancer (ΔGPIIΔAM, Fig. 3B). This demonstrates a requirement for the NRS and that the Env enhancer alone is incapable of recruiting SR proteins to a level that is sufficient to stimulate polyadenylation from 1,400 nt. However, there was no decrease in PAS use when the NRS was deleted from ΔGPIIΔAT, which shows that the Env enhancer alone can stimulate the viral PAS from 1,000 nts (compare ΔGPIIΔAT to ΔNRSΔGPIIΔAT. This observation is consistent with the idea that the threshold for SR binding sites is lower and fewer SR proteins are required to stimulate the viral PAS when the distance is decreased from 1,400 nts to 1,000 nts.

Fig. 3.

Fig. 3

Open in a new tab

The Env enhancer alone stimulates polyadenylation when repositioned to within ~1,000nt of the viral PAS. A. Schematic representation of proviral expression constructs (not to scale) as described in Fig. 1. Deletions are represented as gaps in the boxes. ΔAM lacks nt 5442 to 7901; ΔGPII ΔAM also lacks nt 1026 to 4995; ΔNRSΔGPIIΔAM further lacks nt 543 to 1026; ΔAT lacks nt 5442 to 8297; ΔGPII ΔAT also lacks nt 1026 to 4995; ΔNRSΔGPIIΔAT further lacks nt 543 to 1026. B. Quantitation of percent transcripts processed at the RSV PAS from at least three independent experiments. CEFs were transfected with the indicated proviral clones and total RNA isolated 48 hr later was subjected to RPA. Products were resolved by electrophoresis on an 8M urea-6% polyacrylamide gel and analyzed with a phosphorimager. Percent RSV PAS use was calculated as the units of polyA/polyA + RT, and units were corrected for U content in the bands. Error bars represent standard error of the mean. Where indicated (*), p<0.01.

The NRS alone stimulates polyadenylation when repositioned to within ~2,000 nt of the viral polyA site

Having examined the ability of the Env enhancer to function alone in polyadenylation control, we asked if the NRS alone can stimulate polyadenylation from either ~2,000 nts (ΔAM) or ~1,600 nts (ΔAT) from the PAS (Fig. 4A). Constructs containing the NRS but not the Env enhancer were transfected into CEFs, and use of the viral PAS was determined by RPA. In contrast to the decrease in PAS use with the Env enhancer alone in Fig. 3 (ΔNRSΔGPIIΔAM), there was no difference in polyadenylation efficiency when the Env enhancer was removed from ΔGPIIΔAM, which positioned the NRS alone ~2,000 nts from the PAS (Fig. 4B, ΔEnhΔGPIIΔAM); this suggests that the NRS alone is sufficient to promote polyadenylation at this distance. However, a small but significant decrease in polyadenylation was observed for the NRS alone at 1,600 nt from the PAS. This was unexpected and may reflect inhibition by the positioning the NRS pseudo 5′ ss closer to the PAS. Still, polyadenylation with this construct was equivalent to the NRS alone at ~2,000 nt from the PAS.

Fig. 4.

Fig. 4

Open in a new tab

The NRS alone stimulated polyadenylation when repositioned within 1,000 nts of the viral PAS. A. Schematic representation of proviral expression constructs (not to scale) as described in Fig. 1. Deletions are represented as gaps in the boxes. ΔAM lacks nt 5442 to 7901; ΔGPII ΔAM also lacks nt 1026 to 4995; ΔEnhΔGPIIΔAM further lacks nt 5107 to 5186; ΔAT lacks nt 5442 to 8297; ΔGPII ΔAT also lacks nt 1031 to 4995; ΔEnhΔGPIIΔAT further lacks nt 5107 to 5186. B. Quantitation of percent transcripts processed at the RSV PAS from at least three independent experiments. CEFs were transfected with the indicated proviral clones and total RNA isolated 48 hr later was subjected to RPA. Products were resolved by electrophoresis on an 8M urea-6% polyacrylamide gel and analyzed with a phosphorimager. Percent RSV PAS use was calculated as the units of polyA/polyA + RT, and units were corrected for U content in the bands. Error bars represent standard error of the mean. Where indicated (*), p<0.01.

Collectively, the data suggest that the requirement for both elements can be relieved if the distance to the PAS is decreased, which is consistent with a model where both elements are required to achieve a threshold of SR proteins to promote long-distance activation of the viral PAS.

RNA substrates that polyadenylate efficiently in vitro show increased cross-linking to an ~25kDa protein (CFIm25)

Previous work demonstrated that the NRS activates the RSV PAS in vitro and that SR proteins are sufficient for the effect (Maciolek and McNally, 2007; Wilusz and Beemon, 2006). Another RS-domain containing splicing factor, U2AF, is known to couple splicing and polyadenylation through an interaction of the U2AF arginine/serine-rich (RS) domain with a similar domain in the polyadenylation factor CFIm (Millevoi et al., 2006). Because SR proteins are also known to interact with CFIm (Dettwiler et al., 2004), we speculated that SR proteins activate the weak RSV PAS by recruiting CFIm through protein:protein interactions mediated by the RS domain. One prediction of this model is that the efficiency of CFIm binding to the RSV PAS substrates would correlate with the presence of upstream SR protein binding sites. To test this, a series of 32[P]-labeled in vitro polyadenylation substrates that harbored synthetic SR protein binding sites was assessed for in vitro polyadenylation and CFIm binding in HeLa nuclear extract. As shown in Fig. 5, a control substrate derived from the strong adenovirus L3 PAS was efficiently polyadenylated in vitro in a time-dependent manner whereas the RSV PAS was poorly used, as expected (lanes 1–6)(Maciolek and McNally, 2008). RSV substrates harboring highly specific, SELEX-derived binding sites for SRSF1 {Tacke, 1995 #678} and SRSF7 {Cavaloc, 1999 #103} showed improved polyadenylation whereas random sites (RAN) had little effect (lanes 7–15), consistent with previous work (Maciolek and McNally, 2008).

Fig 5.

Fig 5

Open in a new tab

Synthetic SR protein binding sites stimulate polyadenylation and CFIm25 crosslinking to the RSV PAS in vitro. A. In vitro polyadenylation assays. Above the panels are representations of the substrates; the larger and smaller boxes indicate regions upstream and downstream of the PAS. The three boxes upstream of the RSV substrate indicate 3x SRSF1 or SRSF7 protein binding sites, or random (RAN) sites. The indicated [α-32P]-labeled substrates were incubated in HeLa nuclear extract for the indicated times and resolved on an 8M urea-6% polyacrylamide gel and imaged with a phosphorimager. The upward smear is polyadenylated RNA. M, size markers. B. Quantitation of data in A as percent polyadenylated RNA (poly(A)/(substrate +poly(A)). Error bars represent standard error of the mean. C. CFIm25 crosslinking to RSV PAS. RNA substrates uniformly labeled with [α-32P] were incubated in HeLa nuclear extract, irradiated with 256nm light (or not, for controls), RNase treated, and samples were incubated with beads pre-bound with anti-NUDIX21 antibody specific for CFIm25 (α-25) or with secondary antibody only (α-mouse). After precipitation and washing, samples were resolved on a 12% SDS- polyacrylamide gel followed by phosphorimaging. XL, the L3 substrate subjected to cross-linking but no immunoprecipitation. Size standards are shown at the right. The arrow points to the CFIm25 band.

The substrates were then subjected to UV cross-linking/immunoprecipitation with an antibody to the 25 kDa subunit of CFIm (Fig. 5). Controls that received no UV irradiation or no specific antibody revealed only background bands (lanes 7 to 12). The efficiently polyadenylated adenovirus L3 site substrate produced a clear band consistent with the 25 kDa subunit of CFIm that was largely absent from the RSV substrate (lanes 2 and 3). The CFIm band was increased to a level similar to L3 for the RSV substrate harboring the SRSF1 sites and SRSF7 sites (lanes 4 and 5). The similar cross-link intensity to the more efficiently used L3 substrate is perhaps not unexpected since UV cross-linking is not a quantitative assay, and the L3 substrate has strong upstream and downstream elements that might render it less dependent on CFIm than the RSV PAS, which has poor upstream and downstream elements {Maciolek, 2008 #1046}. The control sample containing a random insert that was poorly polyadenylated unexpectedly produced a 25-kDa band (lane 6). It is possible that the random sequence fortuitously binds CFIm in nonfunctional manner. Still, the increase in CFIm25 cross-linking in the presence of SRSF1 and SRSF7 sites vs. RSV alone is consistent with SR proteins activating polyadenylation by recruiting CFIm.

CFIm recruitment by SR proteins may not be direct

The above data do not distinguish between direct and indirect recruitment of CFIm by SR proteins. To determine if SRSF1 could directly promote CFIm binding to polyadenylation substrates, myc-tagged SRSF1 purified from HEK293 cells (Fig. S2) and baculovirus-purified CFIm (25 kDa and 68 kDa dimer) were used in UV cross-linking assays. Salt conditions were optimized (20 mM KCl) such that non-specific binding was minimized (Fig. S3). Cross-linking of CFIm alone to the L3 positive control RNA showed two bands corresponding to CFIm68 and CFIm25 (Fig. 6, lane 13), consistent with previous reports (Wahle and Keller, 1996). Substrate RNA with 3x high-affinity binding sites for SRSF1 was incubated with purified CFIm at 1.0 and 2.5 pmol, and SRSF1 from 0.01 to 1.0 pmol. In the absence of protein or UV irradiation, two background bands were observed which likely represent incompletely degraded RNA (lanes 1 and 2). Because of their size, the background bands complicate interpretation of CFIm25 cross-linking. However, the 68-kDa band was observed when 2.5 pmol CFIm alone was used (lane 4). When CFIm and SRSF1 were incubated together with the 3xSRSF1-RSV polyA substrate, the dose-responsive appearance of SRSF1 cross-linking was observed (lanes 5–12), but there was no significant change in the intensity of CFIm68 band. These results suggest that SR proteins indirectly stimulate CFIm binding and polyadenylation or alternatively, SR proteins target a different CFIm large subunit isoform.

Fig. 6.

Fig. 6

Open in a new tab

Purified SRSF1 does not stimulate CFIm crosslinking to the RSV PAS. Top, schematic of the 3xSRSF1-RSV substrate used for cross-linking. RNA substrate uniformly labeled with [α-32P] was incubated with purified SRSF1-myc-His and CFIm25/68 protein at the concentrations indicated. Samples were irradiated with 256nm light, RNase treated, and samples were resolved on a 12% SDS- polyacrylamide gel followed by phosphorimaging.

Discussion

Polyadenylation of viral mRNAs is typically inefficient for retroviruses, and this holds for RSV where ~85% of transcripts have correct 3′ ends and 15% represent read-through transcripts. RSV has evolved a novel mechanism to promote polyadenylation that requires the juxtaposition of two spatially separate SR protein-binding platforms (the NRS and Env splicing enhancer) to activate the far distant PAS. These observations led to the model that juxtaposition of SR protein binding sites leads to a threshold of SR proteins that collectively promote long distance PAS stimulation (Hudson and McNally, 2011). In this report, we provide further evidence for the threshold model by showing that a minimum number of five synthetic SR protein binding sites substitute for the Env enhancer, and that the NRS or Env enhancer alone support polyadenylation when their distance to the PAS is decreased. We also established a correlation between SR protein stimulation of the viral PAS in vitro and binding of the CFIm polyadenylation factor, which points to SR protein recruitment of CFIm as a possible mechanism of SR protein-mediated stimulation of polyadenylation.

Polyadenylation of mRNAs involves distinct cleavage and poly(A) addition steps that are coordinated by conserved and auxiliary cis elements and the proteins that bind them (Lutz and Moreira, 2011; Millevoi and Vagner, 2010; Shi et al., 2009; Shi and Manley, 2015). While not all PASs have all elements, the core sequences include the AAUAAA poly(A) signal that resides ~15–30 nts upstream of the cleavage site, a downstream element (DSE) ~40 nts downstream that is U or GU rich, and an upstream U rich element (USE). The poly(A) signal and DSE are synergistically recognized by cleavage and polyadenylation specificity factors (CPSF) and cleavage stimulation factor (CstF), and the USE is often recognized by cleavage factor I (CFIm) (Brown and Gilmartin, 2003; Hu et al., 2005). The complexes then recruit other components of the machinery, including poly(A) polymerase (PAP) that adds the poly(A) tails subsequent to cleavage. Polyadenylation is also recognized as a regulated step in gene expression for transcripts that harbor more than one PAS, but the mechanisms for regulation are unknown in many instances (Di Giammartino et al., 2011; Elkon et al., 2013; Lutz and Moreira, 2011). Understanding polyadenylation control in RSV may inform on potential mechanisms for cellular regulation.

The RSV PAS has a consensus AAUAAA but lacks the DSE (and does not bind CstF) and has a poor USE, which in part explains its poor use in vitro (Maciolek and McNally, 2008). Some PASs are independent of the DSE/CstF and are substituted with G-rich DSEs and bind hnRNP H (Arhin et al., 2002; Decorsiere et al., 2011; Yao et al., 2012). However, there are no appropriately positioned hnRNP H sites near the PAS and there is no evidence to date of a role for hnRNP in RSV polyadenylation in viral constructs (Maciolek and McNally, 2007). Alternative mechanisms are required to overcome the deficiencies in conventional cis elements and activate the PAS, which include the NRS, the Env enhancer, and SR proteins.

RSV uses a novel form of coupling between the splicing and polyadenylation apparatus to promote 3′ end formation. For cellular mRNA, polyadenylation occurs through coupling of the RNA splicing (e.g., U2AF and U2 snRNP) and polyadenylation (CFIm, CPSF, and PAP) machineries across the 3′ terminal exons (Kyburz et al., 2006; Millevoi et al., 2006; Niwa and Berget, 1991; Niwa et al., 1990; Rigo and Martinson, 2008; Vagner et al., 2000b), but this is relatively short-distance coupling as 3′ exons are typically ~500 nt or less (Ji et al., 2009; Zhang, 1998). In RSV, most of the RNA is not spliced, and the 3′ splice sites are weak and are well beyond the standard coupling distance (> 2,220 nt from the PAS), so conventional coupling may not be possible. Instead, SR proteins have been enlisted as a solution for longer-range coupling (~4,100 nts) through juxtaposition of the NRS and Env enhancer binding sites (via NRS – 3′ss complex formation) in the absence of active splicing (Hudson and McNally, 2011; Maciolek and McNally, 2007). This model is consistent with observations that the number and nature of SR proteins, and their distance to an intron, is important for splicing enhancer activity (Graveley et al., 1998), which appears to hold true for polyadenylation as well.

Promotion of RSV polyadenylation is complex. We showed previously that three high-affinity binding sites for SRSF1 could partially substitute for the NRS (Hudson and McNally, 2011), but a minimum of five sites was required at the Env enhancer site (Fig. 1). This could reflect a requirement for a different SR protein-binding site in this region, as the Env enhancer binds SRSF5 whereas we used SRSF1 sites. SRSF1 has a shorter RS domain than SRSF5, which might make it less potent in long-range interactions with the polyadenylation machinery, in analogy to the splicing machinery (Graveley et al., 1998). We also showed that the NRS/Env enhancer functioned better when positioned ~1,000 nt from the PAS vs. 1,400 nt, which is consistent with distance playing a role in the effectiveness of SR protein stimulation of polyadenylation. The prediction that the NRS or Env enhancer alone would support polyadenylation when moved closer to the PAS held true (Figs. 3 and 4). Juxtaposition of the NRS and Env enhancer improved polyadenylation efficiency at 1,400 nt but curiously, PAS use decreased slightly when the Env enhancer was deleted (Fig. 4). This may be a result of repositioning the NRS pseudo 5′ ss closer to the PAS, since it is known that a 5′ splice site negatively impacts polyadenylation (Gunderson et al., 1998).

Like U2AF, SR proteins are RNA binding proteins that have an arginine-serine rich protein:protein interaction domain (the RS domain)(Long and Caceres, 2009) that can interact with a similar domain in the polyadenylation factor, CFIm, which contributes to CPSF binding (Dettwiler et al., 2004; Millevoi et al., 2006). CFIm is a dimer of CFI25 and one of two larger subunits, CFI59 or CFI68, and each can bind RNA (Yang et al., 2011; Yang et al., 2010). The larger subunits have an RS-like domain that can interact with SR proteins (Dettwiler et al., 2004; Ruegsegger et al., 1996). These parallels led to our demonstration that SR proteins can promote CFIm binding to RSV polyadenylation substrates in nuclear extracts (Fig 5). However, the interaction does not seem to be direct since there was no effect of SR proteins on CFIm binding using pure proteins. One possibility is that SRSF1 works through CFI59, which we did not test. This possibility is based on work showing that the RS domain of U2AF has a preference for CFI59 (Millevoi et al., 2006) and that the CFI68 RS domain did not interact with SRSF1 (Dettwiler et al., 2004).

While there is strong evidence of roles for SR proteins in promoting RSV polyadenylation, the small decreases in polyadenylation seen with the src deletions (ΔAM and ΔAT ) suggest that additional minor elements that promote polyadenylation may reside in the src region, but their deletion can be overcome by close proximity of the NRS/Env enhancer to the PAS. The SELEX binding sites for SR proteins that induced CFIm cross-linking in nuclear extracts (Fig. 5) are highly specific, but the lack of direct recruitment of CFIm to RSV substrates using purified proteins (Fig 6) leaves open the possibility that other factors recruit CFIm. Collectively, however, our data support a model (Fig. 7) whereby juxtaposition of the NRS and Env enhancer form a super SR protein-binding platform that promotes long-distance binding of CFIm to the RSV PAS. This mechanism may apply to cellular 3′ exons that exceed the standard coupling distance of ~ 300 to 500 nts.

Fig. 7.

Fig. 7

Open in a new tab

Model of SR protein activation of RSV polyadenylation. Top, schematic representation of RSV mRNA (not to scale) as described in Fig. 1. SR proteins are depicted as orange circles and the NRS and Env enhancer (Enh) are indicated. Bottom, juxtaposition of the NRS and the Env enhancer by the formation of the non-productive splicing complex (blue circle) loops out the intron and brings the SR binding sites together, allowing for a threshold level of SR proteins that stimulate polyadenylation at the distal PAS by directly or indirectly recruiting CFIm (red circles).

Materials and methods

Plasmid Constructs

Proviral constructs were generated in pJTM14 (Miller and Stoltzfus, 1992), which contains the RSV Prague A strain provirus upstream of a chloramphenicol acetyltransferase gene (CAT) and the early polyadenylation signal from the SV40 virus. Coordinates for RSV are as described by Schwartz et al. (Schwartz et al., 1983). pJTM14-ΔEnh (pSWH37) was made by inserting an a KpnI-AgeI fragment generated by overlap PCR that lacked the Env enhancer into the same sites of pSWH36 (Hudson and McNally, 2011). A plasmid (pSWH40) in which the Env enhancer was replaced with an EagI/Sal I linker was generated by inserting a KpnI – AgeI overlap PCR product that harbored EagI – SalI sites into the KpnI/AgeI sites of pSWH36. This plasmid was cut with EagI/SalI and the Env enhancer was replaced with annealed oligonucleotides harboring varying numbers of high affinity binding sites for SFSRF1 (1x, pSWH46; 3x, pSWH47; 3x mutant, pSWH48; 5x, pSWH49; 7x, pCVD97). pJTM14- ΔGPII (pSWH50) was made by digesting pSWH36 with FseI/KpnI and relegation after blunting the ends with Klenow. pJTM14-ΔGPIIΔAM (pSWH51) and pJTM14-ΔGPIIΔAT (pSWH52) were generated by digesting pJTM14-ΔGPII (pSWH50) with AgeI/MluI or AgeI/BstXI, respectively, and religating after Klenow treatment. pJTM14-ΔEnhΔGPIIΔAM (pSWH53) and pJTM14-ΔEnhΔGPIIΔAT (pSWH67) were made by digesting pSWH37 with FseI/KpnI, and AgeI/MluI or AgeI/BstXI, respectively, and relegation after blunting with Klenow. pSWH60 (pJTM14-ΔSK) was made by digesting pSWH37 with SacII/KpnI and relegation after blunting the ends. pJTM14-ΔSKΔAT (pSWH62) and pJTM14-ΔSKΔAM (pSWH66) were made by digesting pSWH60 with Age/BstXI and Age/MluI, respectively, and religating after Klenow treatment. To make pJTM14-ΔAT (pSWH68), pSWH36 was cut with AgeI/BstXI and religated after blunting the ends. pJTM14-ΔAM (pCVD76) and (pCVD77) were constructed by cutting pJTM14 and pBLF89 (Fogel et al., 2002) with AgeI/MluI, and religating after Klenow treatment. pMM371 was made by inserting the coding sequence of SFSR1 as an EcoRI/BamHI fragment into the same sites of pcDNA3.1/myc-his(−)A (Life Technologies); the SFSR1 cDNA was amplified from HeLa cell RNA. To make pMM372, a SRSF7 cDNA was generated from HeLa RNA and inserted as an XhoI/HindIII fragment into the same sites of pcDNA3.1/myc-his(−)A.

Cell culture, transfection, and RNA isolation

Chicken embryo fibroblasts (CEFs) were grown at 39°C in 5% CO2 in medium 199 (Invitrogen) supplemented with 2% tryptone phosphate, 1% heat-treated chick serum (Gibco), 1% bovine calf serum (HyClone), and 1x antibiotic-antimycotic (Invitrogen) in 15 cm dishes. The cells were transfected with 10 μg of RSV proviral plasmid DNA in the presence of 0.2 mg of DEAE-dextran/ml in serum-free medium 199 for 4 h, followed by a 10% dimethyl sulfoxide (DMSO) shock, and cells were incubated further in serum-containing medium for 48 h. At 48 hours post transfection total cellular RNA was isolated using RNeasy kits (Qiagen) according to the manufacturers instructions.

RNase protection assay

RNase protection assays (RPAs) were conducted as previously described (Fogel et al., 2002). Briefly, an [α-32P]UTP-labeled antisense probe was generated from p5′XHI (Stoltzfus and Fogarty, 1989), which spans nt 218 to 630 and simultaneously monitors splicing at the 5′ss and polyadenylation. The probe was hybridized overnight at 55°C with 2.5 to 5 μg of total RNA isolated from transfected CEFs and then digested with 10U of RNase T1 and 5 μg of RNase A/ml for 1 h at room temperature. Samples were then processed for resolution on an 8 M urea–4% polyacrylamide gel. Percent RSV PAS use was calculated as poly(A)/poly(A) + read-through, and the units were corrected for U content in the bands.

In vitro polyadenylation assay

In vitro polyadenylation reactions were conducted on gel-purified, [α-32P]UTP-labeled substrates as previously described (38). The pSP64L3 plasmid was a generous gift from Walter Keller (171) (172, 173). This polyadenylation substrate was chosen for their demonstrated ability to UV cross-link with CFIm (174) and was linearized with DraI. Template DNA for substrates ASF-RSV, 9G8-RSV and RAN-RSV were generated as PCR products from the vector p3Z RSV PvuI-PstI (NM29). Primer pairs are indicated in Table 4–1 (38). Briefly, 100,000 cpm of labeled substrate were incubated in 50% HeLa nuclear extract (Accurate Chemical & Scientific Corp.) for the times indicated in the figure legend and the products were resolved on an 8 M urea–6% polyacrylamide gel. Quantitation was done with a phosphorimager and the percent polyadenylation was calculated as the units of polyadenylated product divided by the units in the substrate plus the product. Units were corrected for U content in the bands.

UV cross-linking and immunoprecipitation

Substrates for UV cross-linking (p3Z-308, p3Z-308/15T, and p3Z-ASLV) were linearized with XbaI and generated 78-nt (308 and 308/15T) and 76-nt (ASLV) transcripts, including vector sequences. The substrates L3, L3-USE, RSV, ASF-RSV, 9G8-RSV, Ran-RSV were produced as described above. UV-cross-linking assays were conducted using 100,000 cpm of [α-32P]UTP-labeled RNA substrate and 15 μg of HeLa nuclear extract (Accurate Chemical & Scientific Corp.) under splicing conditions as previously described (Maciolek and McNally, 2008), but in the absence of creatine phosphate and ATP. For competitions, a 25- or 75-fold molar excess of unlabeled RNA was included in the reactions. Reactions were incubated for 20 min at 30°C and then irradiated with 254-nm light at a 5-cm distance at 4°C for 20 min. After RNase treatment, the samples were resolved by SDS–12% PAGE, and products were visualized using a phosphorimager. For immunoprecipitation reactions, UV cross-linking reactions followed the RNase treatment. The samples were incubated with a mouse antibody directed against CFIm (a generous gift from Dr. Gregory Gilmartin, University of Vermont). The samples were then incubated with Protein A beads pre-bound to a rabbit α mouse antibody. After washing, the samples were boiled in SDS solubilizing buffer and the samples were resolved by SDS–10% PAGE, and products were visualized using a phosphorimager.

For samples that were incubated with purified proteins, UV cross-linking was conducted using 15 fmol of [α-32P] UTP-labeled RNA substrate and the indicated concentration of purified SR proteins or CFIm25/68 (a generous gift from Dr. Gregory Gilmartin), under in vitro splicing assay conditions as described above. Reactions were incubated for 10 min at 30°C and then irradiated with 254-nm light at a 4-cm distance at 4°C for 30 min. After RNase treatment, the samples were resolved by SDS–10% PAGE and products were visualized using a phosphorimager.

Protein purification

Purification of recombinant SRSF1 and SRSF7 was conducted by Ni-NTA His-tag purification from 293T cells as described (Cazalla et al., 2005). Plasmids encoding recombinant SRSF1 (pcDNA-SF2) and SRSF7 (pcDNA-9G8) were transfected into 293T cells by Lipofectamine 2000 (Life Technologies). The cells were then scraped, sonicated, and resuspended in lysis buffer at 10 mM imidazole. The lysate was then combined with pre-equilibrated Ni-NTA beads and loaded on a column. The column was then washed with 20 mM imidazole and eluted at 130 mM imidazole. The elution fractions were then dialyzed against BC100 buffer. Protein purification was confirmed by SDS-PAGE and coomassie stain and western blot with the mouse primary antibody targeted against myc and a HRP conjugated secondary antibody directed against mouse.

Supplementary Material

1
2
3
4

Highlights.

  • At least five SRSF1 protein binding sites are required to replace the Env enhancer

  • The NRS and Env enhancer show position-dependent activation of polyadenylation

  • SR proteins promote binding of CFIm to polyadenylation substrates

  • A threshold of SR proteins are required to promote RSV polyadenylation

  • SR protein recruitment of CFIm may not be direct

Acknowledgments

We are grateful to members of the McNally lab and Dr. Amy Hudson for helpful comments on the manuscript, and to G. Gilmartin (University of Vermont) for recombinant CFIm protein and antibody.

This work was supported by Public Health Service Grant R01 CA78709 from the National Cancer Institute to MTM, and an award from the Medical College of Wisconsin Cancer Center.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Arhin GK, Boots M, Bagga PS, Milcarek C, Wilusz J. Downstream sequence elements with different affinities for the hnRNP H/H′ protein influence the processing efficiency of mammalian polyadenylation signals. Nucleic acids research. 2002;30:1842–1850. doi: 10.1093/nar/30.8.1842. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arrigo S, Beemon K. Regulation of Rous sarcoma virus RNA splicing and stability. Mol Cell Biol. 1988;8:4858–4867. doi: 10.1128/mcb.8.11.4858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Berberich SL, Stoltzfus CM. Mutations in the regions of the Rous sarcoma virus 3′ splice sites: implications for regulation of alternative splicing. Journal of virology. 1991;65:2640–2646. doi: 10.1128/jvi.65.5.2640-2646.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bouck J, Fu XD, Skalka AM, Katz RA. Role of the constitutive splicing factors U2AF65 and SAP49 in suboptimal RNA splicing of novel retroviral mutants. The Journal of biological chemistry. 1998;273:15169–15176. doi: 10.1074/jbc.273.24.15169. [DOI] [PubMed] [Google Scholar]
  5. Brown KM, Gilmartin GM. A mechanism for the regulation of pre-mRNA 3′ processing by human cleavage factor Im. Molecular cell. 2003;12:1467–1476. doi: 10.1016/s1097-2765(03)00453-2. [DOI] [PubMed] [Google Scholar]
  6. Cazalla D, Sanford JR, Caceres JF. A rapid and efficient protocol to purify biologically active recombinant proteins from mammalian cells. Protein Expr Purif. 2005;42:54–58. doi: 10.1016/j.pep.2005.03.035. [DOI] [PubMed] [Google Scholar]
  7. Cook CR, McNally MT. SR protein and snRNP requirements for assembly of the Rous sarcoma virus negative regulator of splicing complex in vitro. Virology. 1998;242:211–220. doi: 10.1006/viro.1997.8983. [DOI] [PubMed] [Google Scholar]
  8. Cooke C, Alwine JC. Characterization of specific protein-RNA complexes associated with the coupling of polyadenylation and last-intron removal. Molecular and cellular biology. 2002;22:4579–4586. doi: 10.1128/MCB.22.13.4579-4586.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Decorsiere A, Cayrel A, Vagner S, Millevoi S. Essential role for the interaction between hnRNP H/F and a G quadruplex in maintaining p53 pre-mRNA 3′-end processing and function during DNA damage. Genes & development. 2011;25:220–225. doi: 10.1101/gad.607011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dettwiler S, Aringhieri C, Cardinale S, Keller W, Barabino SM. Distinct sequence motifs within the 68-kDa subunit of cleavage factor Im mediate RNA binding, protein-protein interactions, and subcellular localization. The Journal of biological chemistry. 2004;279:35788–35797. doi: 10.1074/jbc.M403927200. [DOI] [PubMed] [Google Scholar]
  11. Di Giammartino DC, Nishida K, Manley JL. Mechanisms and consequences of alternative polyadenylation. Molecular cell. 2011;43:853–866. doi: 10.1016/j.molcel.2011.08.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Elkon R, Ugalde AP, Agami R. Alternative cleavage and polyadenylation: extent, regulation and function. Nature reviews Genetics. 2013;14:496–506. doi: 10.1038/nrg3482. [DOI] [PubMed] [Google Scholar]
  13. Fogel BL, McNally LM, McNally MT. Efficient polyadenylation of Rous sarcoma virus RNA requires the negative regulator of splicing element. Nucleic acids research. 2002;30:810–817. doi: 10.1093/nar/30.3.810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fu X-D, Katz RA, Skalka AM, Maniatis T. The role of branchpoint and 3′ exon sequences in the control of balanced splicing of avian retrovirus RNA. Genes Dev. 1991;5:211–220. doi: 10.1101/gad.5.2.211. [DOI] [PubMed] [Google Scholar]
  15. Giles KE, Beemon KL. Retroviral splicing suppressor sequesters a 3′ splice site in a 50S aberrant splicing complex. Molecular and cellular biology. 2005;25:4397–4405. doi: 10.1128/MCB.25.11.4397-4405.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Goff SP. Retroviridae: The Retroviruses and Their Replication. In: Knipe DM, Howley PM, Griffin DE, Lamb RA, Martin MA, Roizman B, Straus SE, editors. Fields Virology. 5. Lippincott Williams & Wilkins; Philadelphia, PA: 2007. pp. 1999–2069. [Google Scholar]
  17. Graveley BR, Hertel KJ, Maniatis T. A systematic analysis of the factors that determine the strength of pre- mRNA splicing enhancers. The EMBO journal. 1998;17:6747–6756. doi: 10.1093/emboj/17.22.6747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gunderson SI, Polycarpou-Schwarz M, Mattaj IW. U1 snRNP inhibits pre-mRNA polyadenylation through a direct interaction between U1 70K and poly(A) polymerase. Molecular cell. 1998;1:255–264. doi: 10.1016/s1097-2765(00)80026-x. [DOI] [PubMed] [Google Scholar]
  19. Hibbert CS, Gontarek RR, Beemon KL. The role of overlapping U1 and U11 5′ splice site sequences in a negative regulator of splicing. Rna. 1999;5:333–343. doi: 10.1017/s1355838299981347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hu J, Lutz CS, Wilusz J, Tian B. Bioinformatic identification of candidate cis-regulatory elements involved in human mRNA polyadenylation. Rna. 2005;11:1485–1493. doi: 10.1261/rna.2107305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hudson SW, McNally MT. Juxtaposition of two distant, serine-arginine-rich protein-binding elements is required for optimal polyadenylation in rous sarcoma virus. Journal of virology. 2011;85:11351–11360. doi: 10.1128/JVI.00721-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ji Z, Lee JY, Pan Z, Jiang B, Tian B. Progressive lengthening of 3′ untranslated regions of mRNAs by alternative polyadenylation during mouse embryonic development. Proceedings of the National Academy of Sciences of the United States of America. 2009;106:7028–7033. doi: 10.1073/pnas.0900028106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Katz RA, Skalka AM. Control of retroviral RNA splicing through maintenance of suboptimal processing signals. Molecular and cellular biology. 1990;10:696–704. doi: 10.1128/mcb.10.2.696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kyburz A, Friedlein A, Langen H, Keller W. Direct interactions between subunits of CPSF and the U2 snRNP contribute to the coupling of pre-mRNA 3′ end processing and splicing. Molecular cell. 2006;23:195–205. doi: 10.1016/j.molcel.2006.05.037. [DOI] [PubMed] [Google Scholar]
  25. Lee Y, Rio DC. Mechanisms and Regulation of Alternative Pre-mRNA Splicing. Annual review of biochemistry. 2015;84:291–323. doi: 10.1146/annurev-biochem-060614-034316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Long JC, Caceres JF. The SR protein family of splicing factors: master regulators of gene expression. Biochem J. 2009;417:15–27. doi: 10.1042/BJ20081501. [DOI] [PubMed] [Google Scholar]
  27. Lutz CS, Moreira A. Alternative mRNA polyadenylation in eukaryotes: an effective regulator of gene expression. WIREs RNA. 2011;2:23–31. doi: 10.1002/wrna.47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Maciolek NL, McNally MT. Serine/arginine-rich proteins contribute to negative regulator of splicing element-stimulated polyadenylation in rous sarcoma virus. Journal of virology. 2007;81:11208–11217. doi: 10.1128/JVI.00919-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Maciolek NL, McNally MT. Characterization of Rous sarcoma virus polyadenylation site use in vitro. Virology. 2008;374:468–476. doi: 10.1016/j.virol.2008.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Maniatis T, Reed R. An extensive network of coupling among gene expression machines. Nature. 2002;416:499–506. doi: 10.1038/416499a. [DOI] [PubMed] [Google Scholar]
  31. McNally LM, McNally MT. SR protein splicing factors interact with the Rous sarcoma virus negative regulator of splicing element. Journal of virology. 1996;70:1163–1172. doi: 10.1128/jvi.70.2.1163-1172.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McNally LM, McNally MT. An RNA splicing enhancer-like sequence is a component of a splicing inhibitor element from Rous sarcoma virus. Molecular and cellular biology. 1998;18:3103–3111. doi: 10.1128/mcb.18.6.3103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. McNally LM, McNally MT. U1 small nuclear ribonucleoprotein and splicing inhibition by the rous sarcoma virus negative regulator of splicing element. Journal of virology. 1999;73:2385–2393. doi: 10.1128/jvi.73.3.2385-2393.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. McNally MT. RNA processing control in avian retroviruses. Front Biosci. 2008;13:3869–3883. doi: 10.2741/2975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McNally MT, Gontarek RR, Beemon K. Characterization of Rous sarcoma virus intronic sequences that negatively regulate splicing. Virology. 1991;185:99–108. doi: 10.1016/0042-6822(91)90758-4. [DOI] [PubMed] [Google Scholar]
  36. Miller JT, Stoltzfus CM. Two distant upstream regions containing cis-acting signals regulating splicing facilitate 3′-end processing of avian sarcoma virus RNA. Journal of virology. 1992;66:4242–4251. doi: 10.1128/jvi.66.7.4242-4251.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Millevoi S, Loulergue C, Dettwiler S, Karaa SZ, Keller W, Antoniou M, Vagner S. An interaction between U2AF 65 and CF I(m) links the splicing and 3′ end processing machineries. The EMBO journal. 2006;25:4854–4864. doi: 10.1038/sj.emboj.7601331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Millevoi S, Vagner S. Molecular mechanisms of eukaryotic pre-mRNA 3′ end processing regulation. Nucleic acids research. 2010;38:2757–2774. doi: 10.1093/nar/gkp1176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Niwa M, Berget SM. Mutation of the AAUAAA polyadenylation signal depresses in vitro splicing of proximal but not distal introns. Genes and Development. 1991;5:2086–2095. doi: 10.1101/gad.5.11.2086. [DOI] [PubMed] [Google Scholar]
  40. Niwa M, Rose SD, Berget SM. In vitro polyadenylation is stimulated by the presence of an upstream intron. Genes and Development. 1990;4:1552–1559. doi: 10.1101/gad.4.9.1552. [DOI] [PubMed] [Google Scholar]
  41. O’Sullivan CT, Polony TS, Paca RE, Beemon KL. Rous sarcoma virus negative regulator of splicing selectively suppresses SRC mRNA splicing and promotes polyadenylation. Virology. 2002;302:405–412. doi: 10.1006/viro.2002.1616. [DOI] [PubMed] [Google Scholar]
  42. Phillips C, Pachikara N, Gunderson SI. U1A inhibits cleavage at the immunoglobulin M heavy-chain secretory poly(A) site by binding between the two downstream GU-rich regions. Molecular and cellular biology. 2004;24:6162–6171. doi: 10.1128/MCB.24.14.6162-6171.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rigo F, Martinson HG. Functional coupling of last-intron splicing and 3′-end processing to transcription in vitro: the poly(A) signal couples to splicing before committing to cleavage. Molecular and cellular biology. 2008;28:849–862. doi: 10.1128/MCB.01410-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ruegsegger U, Beyer K, Keller W. Purification and characterization of human cleavage factor Im involved in the 3′ end processing of messenger RNA precursors. The Journal of biological chemistry. 1996;271:6107–6113. doi: 10.1074/jbc.271.11.6107. [DOI] [PubMed] [Google Scholar]
  45. Schwartz DE, Tizard R, Gilbert W. Nucleotide sequence of Rous sarcoma virus. Cell. 1983;32:853–869. doi: 10.1016/0092-8674(83)90071-5. [DOI] [PubMed] [Google Scholar]
  46. Shi Y, Di Giammartino DC, Taylor D, Sarkeshik A, Rice WJ, Yates JR, 3rd, Frank J, Manley JL. Molecular architecture of the human pre-mRNA 3′ processing complex. Molecular cell. 2009;33:365–376. doi: 10.1016/j.molcel.2008.12.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shi Y, Manley JL. The end of the message: multiple protein-RNA interactions define the mRNA polyadenylation site. Genes & development. 2015;29:889–897. doi: 10.1101/gad.261974.115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Staknis D, Reed R. SR proteins promote the first specific recognition of Pre-mRNA and are present together with the U1 small nuclear ribonucleoprotein particle in a general splicing enhancer complex. Molecular and cellular biology. 1994;14:7670–7682. doi: 10.1128/mcb.14.11.7670. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Stoltzfus CM, Fogarty SJ. Multiple regions in the Rous sarcoma virus src gene intron act in cis to affect the accumulation of unspliced RNA. Journal of virology. 1989;63:1669–1676. doi: 10.1128/jvi.63.4.1669-1676.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Tacke R, Manley JL. The human splicing factors ASF/SF2 and SC35 possess distinct, functionally significant RNA binding specificities. EMBO Journal. 1995;14:3540–3551. doi: 10.1002/j.1460-2075.1995.tb07360.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Vagner S, Ruegsegger U, Gunderson SI, Keller W, Mattaj IW. Position-dependent inhibition of the cleavage step of pre-mRNA 3′-end processing by U1 snRNP. Rna. 2000a;6:178–188. doi: 10.1017/s1355838200991854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Vagner S, Vagner C, Mattaj IW. The carboxyl terminus of vertebrate poly(A) polymerase interacts with U2AF 65 to couple 3′-end processing and splicing. Genes and Development. 2000b;14:403–413. [PMC free article] [PubMed] [Google Scholar]
  53. Wahle E, Keller W. The biochemistry of polyadenylation. Trends Biochem Sci. 1996;21:247–250. [PubMed] [Google Scholar]
  54. Wilusz JE, Beemon KL. The negative regulator of splicing element of Rous sarcoma virus promotes polyadenylation. Journal of virology. 2006;80:9634–9640. doi: 10.1128/JVI.00845-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Xu R, Teng J, Cooper TA. The cardiac troponin T alternative exon contains a novel purine-rich positive splicing element. Molecular and cellular biology. 1993;13:3660–3674. doi: 10.1128/mcb.13.6.3660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Yang Q, Coseno M, Gilmartin GM, Doublie S. Crystal Structure of a Human Cleavage Factor CFI(m)25/CFI(m)68/RNA Complex Provides an Insight into Poly(A) Site Recognition and RNA Looping. Structure. 2011 doi: 10.1016/j.str.2010.12.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Yang Q, Gilmartin GM, Doublie S. Structural basis of UGUA recognition by the Nudix protein CFI(m)25 and implications for a regulatory role in mRNA 3′ processing. Proceedings of the National Academy of Sciences of the United States of America. 2010;107:10062–10067. doi: 10.1073/pnas.1000848107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yao C, Biesinger J, Wan J, Weng L, Xing Y, Xie X, Shi Y. Transcriptome-wide analyses of CstF64-RNA interactions in global regulation of mRNA alternative polyadenylation. Proceedings of the National Academy of Sciences of the United States of America. 2012;109:18773–18778. doi: 10.1073/pnas.1211101109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Zhang L, Stoltzfus CM. A suboptimal src 3′ splice site is necessary for efficient replication of Rous sarcoma virus. Virology. 1995;206:1099–1107. doi: 10.1006/viro.1995.1033. [DOI] [PubMed] [Google Scholar]
  60. Zhang MQ. Statistical features of human exons and their flanking regions. Human molecular genetics. 1998;7:919–932. doi: 10.1093/hmg/7.5.919. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS814843-supplement-1.PPTX (183.7KB, PPTX)
2
NIHMS814843-supplement-2.PPTX (87KB, PPTX)
3
NIHMS814843-supplement-3.PPTX (758KB, PPTX)
4
NIHMS814843-supplement-4.docx (141.3KB, docx)

RESOURCES