INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

Premature termination codons (PTCs) decrease the accumulated levels of most known mRNAs in which they reside (reviewed in reference 28). In the yeast Saccharomyces cerevisiae, the UPF genes are involved in the degradation of PTC-containing RNAs in the cytoplasm (review in reference 37). Similarly, the SMG proteins of Caenorhabditis elegans are required for the rapid decay of PTC-containing unc-54 myosin heavy-chain mRNAs (reviewed in reference 41).

PTC-mediated effects on nuclear RNA processing have been shown to be associated with a number of human genetic diseases (reviewed in references 28 and 29); however, how PTCs mediate such effects in the nucleus is still unclear. In mammalian cells, PTCs have been shown to decrease the levels of nucleus-associated mRNAs by a posttranscriptional mechanism, often attributed to mRNA decay (3, 6, 9, 28, 45). Experiments done with PTCs in the human TPI gene suggest that nonsense-mediated decay (NMD) occurs on a fully spliced nucleus-associated mRNA molecule, possibly during mRNA export to the cytoplasm (4, 5, 9, 28).

There is also increasing evidence that PTCs may affect mammalian RNA processing events other than decay. For example, several instances of PTC-associated exon skipping have been described (15, 19, 28, 33). Perhaps the best-studied example is skipping of the 66-nucleotide (nt) exon 51 of fibrillin FBN1 RNA, which was detected when nonsense but not missense mutations were present within this exon, which was independent of protein synthesis, and for which normal splicing was restored when the nonsense codon was shifted out of frame with the initiation codon (14, 23). Retention of introns upstream of PTC-containing exons has also been reported for P4-generated RNAs of the parvovirus minute virus of mice (MVM) (32) and for mouse immunoglobulin kappa light-chain (Ig)RNA (26). PTCs have also been reported, in one instance, to inhibit splicing of Ig RNA in vitro in a manner independent of protein synthesis (1). These observations have led to the suggestion, still controversial, that PTCs may affect splice site choice (28), implying that the template for nonsense-codon mediated events may, at least in some cases, be a partially spliced or unspliced mRNA, and further suggesting that recognition of PTCs within pre-mRNAs may occur in the nucleus before or concomitant with splicing (47).

The parvovirus MVM is organized into two overlapping transcription units (Fig. 1A) (2, 10, 39). Transcripts R1 and R2 are generated from a promoter (P4) at map unit 4 and encode the viral nonstructural proteins NS1 and NS2, respectively, while the R3 transcripts are generated from a promoter (P38) at map unit 38 and encode the viral capsid proteins (12, 39). Both NS1 and NS2 play essential roles in viral replication and cytotoxicity (13), and so maintenance of their relative steady-state levels, which is controlled at least partially by alternative splicing (40), is critical to the MVM life cycle. All MVM mRNAs generated during infection or following transfection are very stable (43). The alternative splicing of MVM pre-mRNAs is accomplished solely by the interactions between cellular factors and viral cis-acting signals (40).

View larger version (83K): [in this window] [in a new window]   FIG. 1.   A PTC nt 2018 in the NS2-specific exon does not target R2 mRNA for nuclear degradation in vivo. (A) Genetic map of MVM. The three major transcript classes and protein-encoding ORFs are shown. The two promoters (P4 and P38) are indicated by arrows. The large intron, small intron, and NS2-specific exon are indicated. The nonconsensus donor (ncD) and the poor polypyrimidine tract [poor (Py)n] of the large intron are also shown. The position of the PTC mutation at nt 2018 within the NS2-specific exon is indicated; the mutant sequence is shown in Fig. 2A. The bottom diagram shows nucleotide locations, the two probes (A [nt 385 to 650] and B [nt 1854 to 2378]) used for RNase protection assays, and the two primers (a [nt 326 to 345]) and (b [nt 2557 to 2538]) used for RT-PCR as described in Materials and Methods. (B) RNase protection analysis, using probe B, of either total, nuclear, or cytoplasmic RNA generated by A9 cells infected with WT MVM virus, infected with 2018TAA virus, or mock infected, as designated above each lane. The identities of the protected bands are shown on the left and described in Materials and Methods. NUCLEAR (+DRB) and CYTOPLASMIC (+DRB), nuclear and cytoplasmic RNAs, respectively, collected 0.5 and 5 h after cells were treated with DRB (40 µg/ml), which was added 24 h after infection with WT MVM virus, infection with 2018TAA virus, or mock infection, as designated above each lane; TOTAL 0.5 h, total RNA isolated 0.5 h after cells were either treated with 40 µg of DRB per ml (+DRB) or left untreated (DRB), which demonstrated that DRB did indeed inhibit transcription: the absence of unspliced R1 (R1un) and unspliced R3 (R3un) in total RNA isolated 0.5 h after treatment indicated that minimal (if any) new P4 or P38-directed transcription occurred after exposure to DRB. All transcripts were spliced to mature mRNA within 0.5 h of such treatment, as shown earlier (32, 43). RNA from equal cell equivalents was protected for all samples show. The average accumulated ratios of R2 relative to R1, as well as the average accumulated ratios of total P4 product (R2+R1) versus total P38 product (R3), in each RNA, as determined by phosphorimager analysis of duplicate experiments, are summarized in Table 1. The relative ratios of R1 and R2 in total, nuclear, and cytoplasmic RNAs were the same for RNA isolated either immediately after addition of DRB (i.e., 0 h) or 0.5 h after addition (data not shown). Nuclear RNA contains unspliced R1 and R3, which is absent from cytoplasmic RNA (32, 43)). Nuclear fractions were preparations were determined to be >95% pure as explained in Materials and Methods.

There are two types of introns in MVM P4-generated transcripts (40) (Fig. 1A). An overlapping downstream small intron, which undergoes an unusual pattern of overlapping alternative splicing using two donors (D1 and D2) and two acceptors (A1 and A2) (20), is located between nt 2280 and 2399 and is common to both P4- and P38-generated transcripts. An upstream large intron, located between nt 514 and 1989, is additionally excised from a subset of P4-generated pre-mRNAs to generate R2 mRNA. This upstream intron utilizes a nonconsensus donor at nt 514 and has a weak polypyrimidine tract at its 3' splice site (48).

The NS2-specific exon is a 290-nt alternatively spliced exon which is translated in two open reading frames (ORFs). In singly spliced R1, this region utilizes ORF3 to encode NS1; in doubly spliced R2, this exon utilizes ORF2 to encode NS2 (Fig. 1A). Efficient inclusion of the NS2-specific exon as an internal exon in vivo, and consequent excision of the upstream large intron from P4-generated pre-mRNA to generate R2, requires an internally redundant, bipartite exon splicing enhancer (ESE) comprised of 5' and 3' elements within the NS2-specific exon (18).

We have previously shown that PTCs in either the first exon (NS1/NS2 common exon) or second exon (the NS2-specific exon) of R2 (Fig. 1A) caused a decrease in the accumulated levels of R2 relative to R1, although the total accumulated levels of R1 plus R2 remained the same (32). This decrease was the consequence of the artificially introduced translation termination signal acting in cis rather than in the absence of a functional viral gene product and was shown for a PTC in the NS2-specific exon at nt 2018 to be evident in nuclear RNA and independent of stability of total RNA (32). Here we have demonstrated that the PTC at nt 2018, which lies one nucleotide downstream of the 5' element of the bipartite ESE within the NS2-specific exon, interfered with the excision of the upstream large intron in an ORF-dependent manner. This effect was evident in both nuclear and cytoplasmic RNA, yet the nonsense mutation had no effect on either the nuclear or cytoplasmic stability of either R1, R2, or the total P4 product (R1+R2). Further, when combined with a deletion of the 3' element of the bipartite NS2-specific exon enhancer, the PTC at nt 2018 prevented efficient inclusion of the NS2-specific exon, also in an ORF-dependent manner. Our results are consistent with a model in which the PTC at nt 2018 can interfere with the ability of the bipartite ESE to strengthen interactions at the upstream intron polypyrimidine tract in an ORF-dependent manner and suggest that this PTC is recognized in the nucleus.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Mutant construction. Construction of pD1/2, pSX, pXP, pPH, pHS, pSX+HS pCSDSX+HS, p4TSX+HS, and p2018_TAA has been previously described (18, 32, 48).

View larger version (50K): [in this window] [in a new window]   FIG. 2.   A nonsense but not a missense mutation at nt 2018 interferes with excision of the upstream intron in an ORF-dependent manner. (A) Sequences of the large intron polypyrimidine tract and nt 2018 in WT MVM and mutants, as well as the presence (fs [frameshift]) or absence of single nucleotide frameshift deletions at nt 505 and 2011, are shown below the appropriate map positions (deviations from the WT sequence are underlined). Quantitations of R2/R1 obtained by RNase protection analysis using probe B are also indicated. All values are averages of at least three separate experiments. Standard deviations are indicated in parentheses. (B) RNase protection analysis of RNA, using probe B (Fig. 1A), of RNA generated in A9 cells following transfection with WT MVM, transfection with mutants (as described in text), or mock transfection, as designated above each lane. Identities of the protected bands for WT are shown on the left and explained in Materials and Methods. *, undigested probe B. (C) RNase protection analysis of RNA, using probe B (Fig. 1A), of RNA generated in A9 cells following transfection with WT MVM transfection with mutants (as described in the text), or mock transfection, as designated above each lane. Identities of the protected bands are shown on the left and explained in Materials and Methods. RNA generated by pfs(505) and pfs(505)2018TAA did not generate P38 products (R3 transcripts), since the 505 frameshift mutation disrupts the ORF of NS1 which is required for transactivation of the P38 promoter. *, undigested probe B.

Transfection and RNA isolation. Murine A92L cells, the normal tissue culture host for MVM(p), and baby hamster kidney (BHK) cells, were grown and transfected with wild-type (WT) and mutant MVM plasmids, using either DEAE-dextran (31, 32) or Lipofectamine-Plus reagent (as described in the protocol for the Gibco-BRL Lipofectamine-Plus reagent kit). RNA was typically isolated 48 h posttransfection, after lysis in guanidinium thiocyanate, by centrifugation through CsC1 exactly as previously described (43).

RNA analysis. (i) RNase protection assays. RNase protection assays were performed as previously described (43), using an [-³²P]UTP-labeled, SP6-generated antisense MVM RNA probe from MVM nt 385 to 652 (probe A) or 1854 to 2378 (probe B) (Fig. 1A). Probe A identifies all P4 products and distinguishes between R1 and those RNAs that use the nt 514 donor (R2+ES [exon-skipped product]) (Fig. 3B). Probe B extends from before the acceptor site of the large intron to within the small intron common to all MVM RNAs and distinguishes between P4 RNA species using the large intron acceptor and either of the alternative small intron donors, designated by the suffixes "M" for the major splice donor (D1) at nt 2280 and "m" for the minor splice donor (D2) at nt 2317 (Fig. 2B). Thus, probe B can distinguish between unspliced (suffix "un"), minor, and major forms of both R1 and R2, as well as R3; however, it cannot detect the ES product. For analysis of RNA produced after transfection with mutants within the region covered by probe B, RNase protection probes homologous to the mutants being analyzed were used. No attempt was made to standardize between lanes for equivalent amounts of specific RNA, which vary from sample to sample depending on transfection or infection efficiencies. RNase protection products were analyzed on a Betagen B scanning phosphorimage analyzer, and molar ratios of MVM RNA were determined by standardization to the number of uridines in each protected fragment.

View larger version (46K): [in this window] [in a new window]   FIG. 3.   A nonsense but not a missense mutation at nt 2018 interferes, in an ORF-dependent manner, with the function of an ESE within the NS2-specific exon. (A) The restriction sites within the NS2-specific exon (SmaI, XhoI, PstI, HincII, and SacI), which divide the exon into four regions (SX, XP, PH, and HS) and were used to generate the different exon deletion mutants (SX, for example, is a deletion between the SmaI and XhoI sites), are indicated. Sequences of the large intron donor, polypyrimidine tract, and nt 2018 in WT MVM and mutants, as well as the presence (fs [frameshift]) or absence of a single nucleotide frameshift deletion at nt 2011, are shown below the appropriate map positions (deviations from the WT sequence are underlined). , deletion. At the bottom is shown the position of nt 2018 with respect to the CA-rich element within the 5' SX enhancer element of the NS2-specific bipartite exon enhancer. (B) RNase protection analyses, using probes A and B (Fig. 1A), of RNA generated in A9 cells following transfection with WT MVM, transfection with mutants (as described in text), or mock transfection as designated above each lane. Identities of the protected bands in the left-hand panel are shown on the left. The larger species (R1) represents mRNA R1, while the smaller species (R2+ES) represents RNA that uses the large intron donor at nt 514. For the right-hand panel, identities of the bands generated by WT RNA and those generated by the mutants are shown on the left and right, respectively. The mutants were protected by versions of probe B homologous to the mutant region between nt 2011 and 2018 but nonhomologous in the HS region, and therefore the R1, R2, and R3 RNAs generated by these mutants protect fragments that are shorter than those generated by WT RNA. The identity of the bands designated * and ** in the left-hand panel are unknown; however, they are likely breakdown products of the probe since they are not reproducibly seen and occasionally appear in lanes of mock-infected RNA (data not shown). The band designated * in the right-hand panel represents undigested probe B. (C) The two panels represent RT-PCR analyses of RNA generated in A9 cells following transfection with WT MVM, transfection with mutants (as described in text), or mock transfection, as designated above each lane, with primers a and b (Fig. 1A) and performed as explained in Materials and Methods. Samples were run on a 6% acrylamide-urea gel. WT (RT) is a control reaction utilizing WT RNA but excluding reverse transcriptase; pD1/2, a mutation in which both the donors of the downstream small intron were deleted and which results in almost uniform skipping of the NS2-specific exon (18, 48), was used as a control for amplification of the ES product. An RNase protection analysis, using probe B, of RNA generated by WT was used as a marker (sizes of the marker bands are shown on the left in the left-hand panel and on the right in the right-hand panel) for the sizes of the RT-PCR amplified bands. WT RNA generated a 658-nt amplified R2 product. RNA generated by the mutants showed R2 products of sizes consistent with the sizes of the deletions in these mutants, as well as two kinds of amplified ES products, both of which were considered in the quantitations: a larger 368-nt product and a smaller 346-nt product which represent exon skipping to acceptors A1 and A2 of the small intron, respectively. The ES product has previously been sequenced across the splice junction to confirm its identity (49).

(ii) Quantitative RT-PCRs. First-strand cDNA synthesis used 5 µg of total RNA isolated after transfection and oligo(dT) priming by standard techniques (22). PCR detection of R2 and ES products was performed with primers (a and b [Fig. 1A]) described previously (49), with minor modifications (11). The forward primer (primer a) was 5' end labeled with [-P³²]ATP (as described in reference 30) and added to a 15-cycle PCR (94°C for 1 min, 55°C for 1 min, and 72°C for 1 min, followed by extension at 72°C for 5 min). PCR products were run on 6% acrylamide-urea gels and analyzed on a Betagen B scanning phosphorimage analyzer, to calculate the molar ratio of R2 versus ES product (see Fig. 3C and 4C) and obtain a direct percent R2/(R2+ES) value (see Table 4).

Nuclear and cytoplasmic RNA stability experiment. WT and mutant (2018_TAA) virus stocks were titered, propagated, and used to infect A92L murine fibroblasts as previously described (31). The 2018_TAA virus is a host range mutant virus that can be propagated in 324K cells (31, 32). At 24 h after infection, A92L cells were treated with 40 µg/l of 5,6-dichloro-1--D-ribofuranosyl benzimidazole (DRB) (Sigma) per ml, and nuclear and cytoplasmic fractionation of infected cellular RNA was done as described previously (32) at 0.5 and 5 h after the addition of DRB. DRB was used rather than actinomycin D since the latter has been shown to stabilize some RNA species (24). We have previously determined that this concentration of DRB effectively inhibits the production of MVM-specific RNA in infected murine A9 cells (32, 43). Briefly, nuclei were isolated after pelleting through sucrose twice, and the nuclear fractions were determined to be >95% pure, as monitored by the relative absence of mature rRNA and the presence of rRNA precursors in these preparations, by ethidium bromide staining after gel electrophoresis (data not shown) as described previously (32). The relative ratio of unspliced to spliced messages in the nuclear RNA preparations is also consistent with the purity of the nuclear fractions (data not shown). The absence of unspliced R1 and R3 in total RNA isolated 0.5 h after treatment with DRB demonstrated that minimal P4 or P38 transcription occurred after exposure to the drug. RNase protection assays using probe B were done as described above.

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

A PTC at nt 2018 in the MVM NS2-specific exon results in an increased accumulated steady-state level of R1 relative to R2 but does not target R2 mRNA for nuclear or cytoplasmic degradation in vivo. We previously showed that virus bearing a PTC transversion mutation at nt 2018 within the MVM NS2-specific exon (2018_TAA [Fig. 1A]) generated approximately half of the nuclear levels of doubly spliced R2 mRNA relative to singly spliced R1 mRNA compared to the ratio generated during WT infection, although the total accumulated levels of P4-generated product remained the same (32) (Fig. 1B; Table 1). Now we demonstrate that in the presence of the transcriptional inhibitor DRB, 2018_TAA-generated RNAs were as stable as those generated by WT MVM; there was no detectable difference in the relative stability of either R1, PTC-containing R2, or the total P4 product, as measured relative to the P38-generated product R3 (Fig. 1B; Table 1). These results, together with evidence presented below, demonstrated that neither nuclear or cytoplasmic degradation of PTC-containing R2 nor an increase in the stability of R1 could account for reduced accumulated ratios of R2 relative to R1, suggesting that the decreased levels of R2 generated by 2018_TAA may have been due to PTC-mediated interference with excision of the upstream large intron from P4-generated pre-mRNA.

A nonsense mutation at nt 2018 in the NS2-specific exon interferes with excision of the upstream intron in an ORF-dependent manner. R2 encodes the viral protein NS2, which is normally translated in ORF2 within the NS2-specific exon (Fig. 1A). Deletion of nt 2011 in the NS2-specific exon (Fig. 2A) prior to the PTC at nt 2018 shifted the NS2-encoding ORF2 into the long ORF3 which is normally used for NS1. The PTC at nt 2018 was thus bypassed in R2 by this frameshift and the NS2 ORF was thus reopened until the authentic stop site for NS1 at nt 2278 (Fig. 1A). This frameshift [pfs(2011)2018_TAA (Fig. 2A)] resulted in the recovery to WT levels of accumulated R2 relative to R1 (Fig. 2A and B). While a missense transversion mutation at nt 2018 (p2018_CAA [Fig. 2A]) also resulted in a minor decrease in the accumulated levels of R2 relative to R1, a frameshift-causing deletion at nt 2011 [pfs(2011)2018_CAA (Fig. 2A)] was unable to suppress this small effect (Fig. 2A and B). Thus, the effect on excision of the upstream intron of the nonsense, but not the missense, transversion mutation at nt 2018 was dependent on the integrity of a previously open reading frame that had the potential, after splicing, to be in frame with the initiating AUG.

A nonsense mutation at nt 2018 in the NS2-specific exon interferes with the function of an ESE in an ORF-dependent manner. The PTC at nt 2018 resulted in decreased excision of the upstream large intron; however, nonsense mutations within internal exons in other genes have been, on occasion, shown to cause skipping of these exons. The NS2-specific exon contains a bipartite ESE that is required for efficient inclusion of this exon into mature spliced mRNA and consequently for efficient splicing of the upstream intron. This bipartite ESE consists of a 22-nt CA-rich element within the 5' region of the exon (called the SX region) and a 58-nt purine-rich element in the 3' region of the exon (called the HS region) (18). These elements have at least partially redundant functions; while deletion of either element alone (pSX or pHS [Fig. 3A and 3C, left panel; see Table 4] [18]) or extensive point mutagenesis of the SX enhancer element (18) permitted efficient inclusion of the NS2-specific exon, deletion of both elements together (pSX+HS [Fig. 3A and C, left panel; see Table 4] [18]) resulted in substantial skipping of this exon in P4-generated pre-mRNAs (18), whereby the large intron donor at nt 514 was joined to either of the small intron acceptors. (Skipping of the NS2-specific exon could not be directly assessed by RNase protection analysis in our system. To measure the relative accumulated levels of P4-generated R2 and potential NS2-specific ES products, we used a quantitative RT-PCR assay, the results of which were validated indirectly by quantitative RNase protection assays as described below and in detail in Materials and Methods.) Improvement of the upstream intron polypyrimidine tract (p4TSX+HS [Fig. 3A and C, left panel; see Table 4] [18]), but not improvement of the upstream intron donor (pCSDSX+HS [Fig. 3A and C, left panel; see Table 4] [18]) could suppress exon skipping and substantially restore inclusion of the NS2-specific exon to the mutant in which both elements of the enhancer had been deleted, which suggested a model in which the bipartite ESE functioned by strengthening interactions at the weak upstream intron polypyrimidine tract (18).

View larger version (56K): [in this window] [in a new window]   FIG. 4.   Exon skipping caused by p2018TAA+HS can be suppressed by a frameshift in the upstream exon as well as by improvements of the upstream intron polypyrimidine tract. (A) The four regions (SX, XP, PH, and HS) of the NS2-specific exon are shown. Sequences of the large intron donor, polypyrimidine tract, and nt 2018 in WT MVM and mutants, as well as the presence (fs [frameshift]) or absence of a single nucleotide frameshift deletion at nt 505, are shown below the appropriate map positions (deviations from the WT sequence are underlined). , deletion. (B) The two panels show RNase protection analyses of RNA, using probe B (Fig. 1A), of RNA generated in A9 cells following transfection with WT MVM, transfection with mutants (as described in the text), or mock transfection, as designated above each lane. Identities of the protected bands for WT are shown on the left, as explained in Materials and Methods, and those protected by the mutants are shown on the right of each panel. The mutants were protected by versions of probe B homologous to the mutant region of the upstream polypyrimidine tract and nt 2018 but nonhomologous in the HS region, and therefore the RNAs generated by the mutants protect fragments that are shorter than those generated by WT RNA. *, undigested probe B. (C) RT-PCR analysis of RNA generated by A9 cell transfections with WT MVM, transfection with mutants (as described in the text), or mock transfection, as designated above each lane, with primers a and b (Fig. 1A) as explained in Materials and Methods. Samples were run on a 6% acrylamide-urea gel. WT(RT) and pD1/2 controls and marker bands are as described for Fig. 3C. WT RNA generated a 658-nt amplified R2 product, while RNA generated by the mutants showed R2 products of sizes consistent with the sizes of the deletions in these mutants. Some mutants show two or three kinds of amplified R2 products, all of which were considered in the quantitation; the largest of these is the authentic R2 product, while the smaller R2 products apparently use cryptic donors within the NS2-specific exon. As explained in the legend to Fig. 3C, two kinds of amplified exon-skipped products using either A1 or A2 were observed. (D) RNase protection analysis with probe B (Fig. 1A) (left) and quantitative RT-PCR analysis (right) of RNA generated in BHK cells following transfection with WT MVM, transfection with mutants, or mock transfection, as designated above each lane. Identities of the bands for the RNase protection assays using probe B are shown on the left. RT-PCR samples were run on a 6% acrylamide-urea gel; pD1/2 and WR (RT) controls and markers for the RT-PCR are as described for Fig. 3C. In the RT-PCR analysis, RNA generated by WT shows 658-nt amplified R2 product, while RNA generated by the mutants show R2 products of sizes consistent with the sizes of the deletions in these mutants, as well as two kinds of amplified ES products, as explained in the legend to Fig. 3C. WT MVM generates more of the ES product in BHK cells than it does in murine A9 cells (Fig. 3C and 4C).

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

In this report we demonstrate that a PTC at nt 2018, which caused a decrease in the accumulated levels of R2 relative to R1, although the total accumulated levels of R1+R2 remained the same, had no effect on either the nuclear or cytoplasmic stability of either R2 mRNA, R1 mRNA, or the total P4 product. This suggests that neither nucleus-associated or cytoplasmic degradation of PTC-containing R2 nor an increase in the stability of R1 could account for the decrease in R2 relative to R1. This interpretation was supported by the following observations. First, minor improvements in the upstream intron polypyrimidine tract could overcome the 2018 PTC effect and return accumulation of R2 to WT levels. Since these improvements lie outside the final mRNA, their effect could not be attributed merely to restoration of mRNA stability. Second, the effect of the 2018 PTC on the accumulation of R2 could be suppressed by mutations in either NS2-encoding exon which shift this PTC out of the NS2 ORF. These observations are consistent with a model in which the 2018 PTC interfered with nuclear excision of the upstream intron from P4-generated pre-mRNA in an ORF-dependent manner.

The 2018 PTC interferes the function of an ESE within the NS2-specific exon that is required for efficient inclusion of this exon in an ORF-dependent manner. Naturally occurring PTCs that have been shown to result in exon skipping are present within candidate purine-rich ESE in the 3-hydroxyl-3-methylglutaryl coenzyme A, adenosine deaminase, FBN1, OAT, and dystrophin genes (15, 38, 42, 44). It may be that the skipped exons in these examples contain a single ESE element that is required for exon inclusion. Inclusion of the NS2-specific exon of MVM is governed by a bipartite ESE, which includes a CA-rich 5' element (SX) and a purine-rich 3' element (HS) that are redundant in function. The 2018 PTC transversion mutation, which when present alone resulted in retention of the upstream intron rather than exon skipping, is directly adjacent to the CA-rich motif within the 5' SX element. Combination of the 2018 PTC and the 3' HS enhancer element deletion (p2018_TAA+HS) resulted in a significant loss of NS2-specific exon inclusion. Consequently, the NS2-specific exon was skipped in P4-generated RNA, to levels similar to that seen in RNA generated by a mutant in which both the 5' SX element and the 3' HS element had been deleted (pSX+HS). This implied that that the 2018_TAA mutation was phenotypically similar to disabling of the SX enhancer element. Combination of HS and a missense transversion mutation at nt 2018 alone (p2018_CAA+HS) resulted in some exon skipping, as might be expected for a missense mutation in an ESE. Shifting the 2018 PTC out of the NS2 ORF in the context of the HS element deletion [pfs(2011)2018_TAA+HS] resulted in recovery of exon inclusion back to the levels seen for the missense p2018_CAA+HS; however, the effect of 2018_CAA+HS could not be suppressed by such a frameshift [pfs(2011)2018_CAA+HS]. These results suggested that while both the PTC and missense transversion mutations had direct, ORF-independent effects on the function of the 5' SX enhancer element, the PTC mutation had an additional effect which was ORF dependent.

A model to explain the effects of the 2018 PTC. (i) Nuclear detection of reading frame. 2018 PTC-mediated intron retention and exon skipping is nucleus associated and ORF dependent. These observations imply that PTCs in P4 pre-mRNAs can be recognized in the nucleus before or concomitant with exon definition and exon juxtaposition during splicing, as has been proposed for the DHFR gene (47). There is no direct evidence that such a reading frame recognition exists in the nucleus (28); however, a number of recent observations open the possibility that an appropriate apparatus may be in place. Ribosomal proteins and RNAs, elongation factor subunits eIF2A (8) and eIF4E (25), and aminoacyl-tRNAs (27) have recently been detected in the nucleus, and U5 snRNP, which binds to exon sequences adjacent to 5' and 3' splice sites and has a role in juxtaposing exons during splicing (34-36, 46), contains a 116 kDa protein which is both essential for splicing and closely related to the ribosomal translocase EF-2 (16). Finally, it has been reported in one instance that nonsense mutations can inhibit RNA splicing in an in vitro splicing-competent nuclear extract in a manner independent of protein synthesis (1), which is also consistent with the existence of such a function. We have also recently shown that the effect of the 2018 PTC was not dependent on the presence of an initiating AUG codon, suggesting that this effect was independent of cytoplasmic translation (17).

(ii) If there is ORF scanning in the nucleus, how might such a mechanism affect nuclear steps of RNA processing? Numerous models have been proposed to explain decreases in the nuclear abundance of PTC-containing RNAs, including effects on RNA stability and various steps of RNA processing (8, 28). For MVM P4-generated RNAs, although a PTC in the NS2-specific exon affects the relative accumulated levels of various spliced products, the PTC-containing RNAs in both the nuclear and cytoplasmic factions were very stable, suggesting that the 2018 PTC may have a direct effect on P4 pre-mRNA splicing. Since a significant component of the 2018 PTC effect was ORF dependent, and since this PTC interfered with the ability of the 5' element of the bipartite NS2-specific ESE to strengthen interactions at the weak upstream polypyrimidine tract, perhaps recognition of the 2018 PTC somehow prevented the exon definition machinery from interacting with the 5' exon enhancer element (SX). The net effect of this interaction would either be upstream intron retention (2018_TAA) or exon skipping (p2018_TAA+HS). We have previously shown that PTCs at nt 2159 and 2268, which do not fall within the bipartite ESE of the NS2-specific exon, have very small effects on upstream intron excision (32). Although ESEs are thought to act primarily early in splice site recognition (7), an effect at later times during the splicing process has not been ruled out.