Nucleic Acids Research

Plasmids and RNAs

pLFMyc, which contains a 556 bp fragment of the 3[prime] untranslated region (3[prime]-UTR) of the human c-myc cDNA (positions 7217-7773), was constructed as previously described (9). Transcription of pLFMyc linearized with PvuII gave a 634 base RNA (mycNS). The construction of pSVL-GEM, which contains the 3[prime]-UTR of the simian virus 40 (SV40) late transcription unit (nucleotides 2533-2682), has been described previously (22). HindIII digestion generated template DNA that was used to produce the SV RNA.

Transcription templates for the generation of an exact copy of the U11 RNA were produced by PCR amplification from the plasmid pU11 (a pAR2463 derivative containing U11 DNA sequences inserted between the BamHI and StuI sites) using 5[prime]-ATTTAGGTGACACTATAGAATACACAAGGCTTCTGTCGT as the upstream primer and 5[prime]-AAGGGCGCCGGGAC as the downstream primer. Reaction products were used as a template to make the U11 RNA. The template to generate a U11 RNA variant that lacked the 3[prime]-terminal stem-loop structure (U11del RNA) was generated by cutting the PCR product with AluI.

Full-length U7 RNA was produced by hybrizing the following oligonucleotides and ligating them using T4 DNA ligase: 5[prime]-ATTTAGGTGACACTATAGAATACACCTGTGTTACAGCTCTTTTAGAA; 5[prime]-CAAATTCTAAAAGAGCTGTAACACAGGTGTATTCTATAGTGTCACCTAAAT; 5[prime]-TTTGTCTAGTAGGCTTTCTGGCTTTTCACCGGAAAGCCCCT; 5[prime]-AGGGGCTTTCCGGTGAAAAGCCAGAAAGCCTACTAGA. Ligated products were amplified by PCR using the primers ATTTAGGTGACACTATAGAATACACGTGTTACAGCTCTC and 5[prime]-AGGGGCTTTCCGG to increase the yield of appropriately ligated products. PCR products were either transcribed directly using SP6 RNA polymerase (U7 RNA) or cut with HpaII prior to transcription to produce a variant lacking the 3[prime]-terminal stem-loop structure (U7del RNA).

pT7VA1, obtained from Dr Michael Mathews (23), was cut with DraI and transcribed with T7 RNA polymerase to produce VA1 RNA. Transcription of pT7VA1 that was linearized with SalI produced a variant of VA1 RNA with a 64 base extension on its 3[prime]-end (VA1int).

pAAV contains a 232 bp PstI-XbaI fragment of adeno-associated virus inserted between the PstI-XbaI sites of pGEM3 as described previously (24). Transcription of EcoRI-linearized template yielded a 286 base RNA (AAV). The template to generate AAV-URE was prepared as described previously (24). Transcription of this template yielded a 283 base variant of AAV RNA that had its 3[prime]-end complementary to an internal region of the transcript (AAV-URE).

Alteration of 3[prime]-ends by the ligation/PCR approach

RNAs containing additional sequences on their 3[prime]-ends were generated from the starting plasmid or construct named in Table 1 by the ligation/PCR approach (9). First, pairs of oligonucleotides were generated that had the following general structure: HindIII linker-desired sequence-restriction site-primer binding site (PBS). Oligonucleotide pairs were hybridized to each other by incubating at 100°C for 2 min in 10 mM Tris-HCl (pH 8.0), 50 mM NaCl and 1 mM EDTA followed by slow cooling to room temperature. The hybridized oligonucleotides were then ligated to the desired DNA target (listed in Table 1) using T4 DNA ligase at 14°C overnight. Ligated products were amplified by PCR using an SP6 primer (5[prime]-CATACGATTTAGGTGACACTATAG) and the appropriate downstream primer (5[prime]-TACCTCGAGCACTC,5[prime]-TACCTCGAGCAGAC or 5[prime]-CCCATATGCATGGTA). Amplified products were digested with the appropriate restriction enzyme and used as the templates for in vitro transcription to produce the RNAs indicated in Table 1. A variant of SVHis RNA (SVintHis), that contains the histone stem-loop structure 15 bases away from the 3[prime]-end, was produced by transcribing the appropriate PCR product prior to restriction enzyme cleavage.

Table 1. Sense strands of oligonucleotide pairs and target DNA fragments used to construct the indicated RNAs by the ligation/PCR technique

RNA	Starting plasmid	Oligonucleotide
SVA60	pSVL-Gem	5[prime]-AGCT(A50)AAAAAAAAAATATTGAGGTGCTCGAGGT
SVHis	pSVL-Gem	5[prime]-AGCTCTACCAATAAGAGGCCCTTTTCAGGGCCCCTGGAGTGCTCGAGGTA
SVU5	pSVL-Gem	5[prime]-AGCTTTTTTAAAGAGTG CTCGAGGTA
MycU5	pLFMyc	5[prime]- AGCTTTTTTAAAGAGTG CTCGAGGTA
SVU30	pSVL-Gem	5[prime]-AGC(T25)TTTTTTAAAGAGTGC TCGAGGTA
SVstm	pSVL-Gem	5[prime]-AGCTCTACCAATAAGAGCGGCTATACAGGGCCCCTGGAGTGCTCGAGG TA

The underlined bases indicate the restriction enzyme site used to create templates for in vitro transcription.

3[prime]->5[prime] exonuclease assay

Nuclear salt wash extracts were prepared from HeLa spinner cells grown in Joklik’s modified minimal essential medium (JMEM) containing 10% horse serum as described previously (25). ³²P-labeled, gel-purified RNAs were added to the exonuclease assay system as described previously (9). Reaction mixtures were incubated at 30°C for the indicated times followed by the addition of 400 µl of HSCB buffer (400 mM NaCl, 50 mM Tris-HCl, pH 7.6, 0.1% SDS). RNAs were phenol extracted, concentrated with ethanol and analyzed on 5% acrylamide gels containing 7 M urea. Bands were visualized using autoradiography or a Phosphor-Imager. Radioactivity was monitored at every step to avoid spurious results and all assays were performed at least three times to ensure reproducible results. All quantitation was performed using a Molecular Dynamics PhosphorImager.

UV crosslinking analysis and immunoprecipitation

Crosslinking experiments were performed as described previously (26). Typically, gel-purified RNAs, ³²P-labeled to the same specific activity, were incubated at 30°C in the in vitro exonuclease assay for 10 min in the presence of 0.88 mM EDTA (to stabilize all RNA substrates to allow for accurate comparisons). The reaction mixture was then irradiated on ice for 10 min at 4°C. RNases A, T1 and T2 were added and the reaction mixture was incubated at 37°C for 15 min to generate crosslinked RNA binding proteins containing small radioactive RNA oligomers. The sample was then subject to electrophoresis on a 10% polyacrylamide gel containing SDS. ³²P-labeled proteins were visualized by autoradiography or using a Molecular Dynamics PhosphorImager. After UV crosslinking, some reactions were mixed with 300 µl RIPA buffer (150 mM NaCl, 10% NP-40, 0.5% sodium desoxycholate, 0.1% SDS and 50 mM Tris-HCl, pH 7.6) and pre-cleared by centrifugation. Antibodies were added to supernatants and the reaction mixtures were incubated at 4°C for 1 h. Aliquots of 50 µl of 100 µg/ml protein A-positive Staphylococcus aureus cells were added and samples were incubated for 10 min on ice. The mixture was washed five times with RIPA buffer, resuspended in 2× protein loading buffer and boiled for 5 min prior to electrophoresis on acrylamide-SDSgels. La protein-specific autoimmune antisera was obtained fromDr J. Keene (Duke University).

Immunodepletion of extracts

To deplete La protein from the HeLa nuclear extracts, 10 µl of La autoimmune antisera was added to 200 µl of HeLa nuclear extract and incubated on ice for 1 h. As a control, mock-depleted extracts were incubated in the absence of antisera in a similar fashion. Protein A-positive S.aureus that were pre-washed were added to extracts and the mixture was incubated on ice for 30 min. Staphylococcus aureus cells were removed by centrifugation at 2000 r.p.m. for 3 min. Supernatants were collected and used for exonuclease assays and UV crosslinking/immunoprecipitation analysis.

3[prime]-Terminal structures inhibit RNA degradation by a 3[prime]->5[prime] exonuclease

A 3[prime]-terminal stem-loop structure is important for histone mRNA stability in mammalian cells. In vivo, however, the stem-loop element cannot stabilize RNAs if it is not located at the precise 3[prime]-end of the mRNA (21). We tested if the in vivo positional requirement for RNA stability by the histone stem-loop structure was reflected in the action of a 3[prime]->5[prime] exonuclease in our in vitro system derived from HeLa cells (9). Using the ligation/PCR approach described in Materials and Methods, we added a histone stem-loop structure directly to the 3[prime]-end of the unstable SV RNA (SVHis) or positioned the structure 15 bases away from the 3[prime]-end (SVintHis). The in vitro synthesized ³²P-labeled SVHis and SVintHis RNAs were incubated in the 3[prime]->5[prime] exonuclease assay system for 30 min and reaction products were analyzed on a polyacrylamide gel containing 7 M urea. As seen in Figure 1, SVintHis RNA was degraded 4-fold faster than SVHis RNA by the exonuclease in the assay. We conclude that the histone stem-loop structure must be located directly at the 3[prime]-terminus in order to block the 3[prime]->5[prime] exonuclease present in HeLa extracts. These data, along with our previous observations (9), further associate the action of this 3[prime]->5[prime] exonuclease with in vivo observations of RNA turnover (21).

Figure 1. Placement of the histone stem-loop structure 15 bases internal to the 3[prime]-end inhibits its ability to protect RNAs from a 3[prime]->5[prime] exonuclease. Variants of the unstable SV RNA containing a 3[prime]-terminal histone stem-loop structure (SVHis) or a histone stem-loop structure located 15 bases from the 3[prime]-end (SVintHis) were incubated in the exonuclease assay system for the indicated times. Products of the reaction were analyzed on a 5% acrylamide gel containing 7 M urea.

A variety of small RNAs that lack 3[prime]-terminal modifications are extremely stable in mammalian cells. We next determined whether several of these small RNAs that are stable in vivo were refractory to the action of the 3[prime]->5[prime] exonuclease in vitro. Using available clones or synthetic oligomers and PCR, we obtained DNA templates to make exact copies of the small nuclear RNA U11 (27) and the small adenoviral cytoplasmic RNA VA1 (23) by in vitro transcription (Fig. 2). The susceptibility of U11 and VA1 RNAs to the 3[prime]->5[prime] exonuclease was then determined in comparison with the turnover of the unstable SV RNA or a stable variant that contains a poly(A) tail (SVA60) (Fig. 3A). Both U11 and VA1 RNAs were at least as stable as the poly(A)⁺ SVA60 RNA. We conclude, that small poly(A)^- RNAs that are naturally stable in vivo are also resistant to the 3[prime]->5[prime] exonuclease in the HeLa nuclear extracts.

Figure 2. Potential secondary structures of wild-type and mutant U11, U7 and VA1 RNAs. Secondary structures are based on computer-assisted RNA folding predictions, phylogenetic comparisons, nuclease sensitivity analyses and mutagenesis studies (41-43). U11del and U7del RNAs lack the 3[prime] stem-loop structure, while VA1int RNA contains 64 bases of additional sequence on its 3[prime]-terminus.

Figure 3. 3[prime]-Terminal sequences that form stable secondary structures protect RNAs from a 3[prime]->5[prime] exonuclease. (A) U11 and VA1 RNAs were incubated in the 3[prime]->5[prime] exonuclease assay for 5, 10, 15 and 30 min along with the unstable SV RNA and the stable SVA60 RNA. Reaction products were purified and analyzed on a 5% acrylamide gel containing 7 M urea. The percentage of RNA remaining at the times indicated was determined using a PhosphorImager. (B) U11 RNA or a variant that lacks the 3[prime] stem-loop structure (U11del) was incubated in the 3[prime]->5[prime] exonuclease assay system for 30 min. RNA reaction products were purified and analyzed on a 5% acrylamide gel containing 7 M urea. (C) VA1 RNA or a derivative containing a 3[prime] extension of 64 bases (VA1int) was incubated in the 3[prime]->5[prime] exonuclease assay for 30 min. Products of the reaction were analyzed on a 5% acrylamide gel containing 7 M urea. (D) U7 RNA or a variant that lacks the 3[prime] stem-loop structure (U7del) was incubated in the 3[prime]->5[prime] exonuclease assay for 30 min. Products of the reaction were analyzed on a 5% acrylamide gel containing 7 M urea.

We next tested if the U11, U7 and VA1 small RNAs were resistant to the exonuclease because of specific 3[prime]-terminal sequences. An examination of the proposed secondary structure of U11 RNA shows a large stem-loop at its 3[prime]-end (Fig. 2A). Similar 3[prime] stem-loop structures are present in a large variety of small stable RNAs (14). A variant of the U11 RNA was created (U11del) in which this 3[prime] stem-loop was deleted (Fig 2A). As seen in Figure 3B, the presence of the 3[prime] stem-loop in U11 RNA specifically inhibited the 3[prime]->5[prime] exonuclease. U7 RNA, which also contains a 3[prime] stem-loop structure (Fig. 2C), was also resistant to the 3[prime]->5[prime] exonuclease (Fig. 3D). A variant of U7 RNA that lacked a 3[prime]-terminal stem-loop structure (Fig. 2C), however, was rapidly degraded (Fig. 3D). VA1 RNA is a stable, highly structured pseudo-double-stranded cytoplasmic RNA (Fig. 2B). VA1 RNA is a member of the remaining group of small stable RNAs that lack a stem-loop structure at their 3[prime]-ends. Rather, the 3[prime]-ends of VA1 RNA (and other members of this group such as tRNAs, etc.) are base paired with other regions of the transcript. In order to test if these base paired structures involving the 3[prime]-end were important for stability of this class of small RNAs, we added a 3[prime] extension of 64 bases onto VA1 RNA as described in Materials and Methods (Fig. 2B, VA1int). While VA1 RNA was highly resistant to the 3[prime]->5[prime] exonuclease in our assays, VA1int RNA was rapidly degraded (Fig. 3C). We conclude that structures involving the 3[prime]-terminal sequences of U11 and VA1 RNAs are key determinants for protecting the transcript from the action of 3[prime]->5[prime] exonucleases.

Small stable RNAs have extensive sequence diversity at their 3[prime]-ends. It is likely, therefore, that structure rather than the specific nucleotide sequence protects these RNAs from degradation. To directly test this, a non-specific stem-loop structure whose sequence was chosen at random was placed at the 3[prime]-end of the normally unstable SV RNA. As seen in Figure 4A and B, the artificial stem-loop structure ([Delta]G -6.5 kcal/mol) at the 3[prime]-end of SV RNA (SVstm) inhibited the 3[prime]->5[prime] exonuclease as efficiently as a poly(A) tail or a histone stem-loop structure. In order to confirm these results, we used a 286 base RNA derived from the adeno-associated virus (AAV) or a variant of AAV RNA that contained a 20 base segment at its 3[prime]-end that could base pair with sequences in the body of the transcript (AAV-URE) (24). The formation of a structure involving the 3[prime]-end of the construct was confirmed using T1 digestion (24). As seen in Figure 4C, the AAV transcript containing the engineered base paired 3[prime]-end (lanes AAV-URE) was resistant to the 3[prime]->5[prime] exonuclease as compared with the wild-type AAV RNA. We conclude that many, if not all, 3[prime]-terminal RNA structures can inhibit the 3[prime]->5[prime] exonuclease activity in this assay.

Figure 4. 3[prime]-Terminal structure, rather than specific nucleotide sequences, protects RNAs from the 3[prime]->5[prime] exonuclease. (A) Sequence and hypothetical structure of the artificial stem-loop present at the 3[prime]-end of SVstm RNA. (B) Variants of the unstable SV RNA were prepared that contained known 3[prime]-terminal stability elements (SVA60 and SVHis) or a synthetic 3[prime]-terminal stem-loop (SVstm). RNAs were incubated in the 3[prime]->5[prime] exonuclease assay and aliquots were removed at the times indicated. Reaction products were analyzed on a 5% acrylamide gel containing 7 M urea and quantitated using a PhosphorImager. (C) An autoradiogram of the 30 min time point of the SV and SVstm RNAs for the experiment shown in (A). (D) An AAV RNA (that lacks detectable 3[prime]-terminal structure) and a variant that contains an engineered 3[prime]-terminal structure (AAV-URE) were incubated in the 3[prime]->5[prime] exonuclease assay for the times indicated. Reaction products were run on a 5% acrylamide gel containing 7 M urea and analyzed by autoradiography.

A poly(U) tract in association with poly(U) specific trans-acting factors inhibits RNA degradation bya 3[prime]->5[prime] exonuclease

Poly(U) tracts that are encoded by oligo(dT) termination signals are naturally found at the 3[prime]-end of RNA polymerase III precursor transcripts (28). Guide RNAs, which serve as templates for editing of mitochondrial transcripts in kinetoplastid protozoans, also receive poly(U) tails of ~20 bases (29). We tested whether a 3[prime]-terminal poly(U) tract could inhibit the degradation of the normally unstable SV and Myc RNAs by the 3[prime]->5[prime] exonuclease. Figure 5 shows that a five base poly(U) tract placed at the 3[prime]-end of either the SV (SVU5) or Myc (MycU5) RNA indeed protected the transcript from 3[prime]->5[prime] exonucleolytic decay.

Figure 5. A five base uridylate tract can efficiently protect RNAs from the 3[prime]->5[prime] exonuclease. (A) The unstable SV RNA and a variant containing a five base poly(U) tail (SVU5) were incubated in the 3[prime]->5[prime] exonuclease assay for 0 (Input) and 30 min. Reaction products were analyzed on a 5% acrylamide gel containing 7 M urea. (B) Myc RNA and a variant containing a five base poly(U) tail (MycU5) were incubated in the 3[prime]->5[prime] exonuclease assay for 0 (Input) and 30 min. Reaction products were analyzed on a 5% acrylamide gel containing 7 M urea.

Following synthesis, RNA polymerase III transcripts associate with the abundant nuclear 50 kDa La protein through their short poly(U) tracts (16). Recently, La protein has been suggested to stabilize a degradation intermediate of the histone mRNA in an in vitro system developed by Ross and colleagues (19). Perhaps La protein was responsible for poly(U)-mediated protection from the 3[prime]->5[prime] exonuclease in our assay. We first tested whether La protein was specifically binding to the poly(U) tract in our extracts. As shown in Figure 6A and B, La protein could be UV crosslinked to SVU5 RNA, but not to the SV RNA that lacked a terminal U tract. These data indicate that La protein specifically binds to RNAs containing poly(U) tracts in our assay. The band migrating faster than La protein in Figure 6 was non-specific, as it was present in control immunoprecipitations (data not shown).

Figure 6. La protein specifically interacts with RNAs that contain a short poly(A) tail in the in vitro system. (A) SVU5 RNA, a variant of SV RNA that contains a five base uridylate tail, was radiolabeled at U residues and incubated in the 3[prime]->5[prime] exonuclease assay system for 5 min. Protein-RNA complexes were analyzed by a UV crosslinking assay. Total crosslinked proteins (total lane) or crosslinked proteins immunoprecipitated using anti-La antisera (Ippt lane) were analyzed on a 10% acrylamide gel containing SDS. (B) SV and SVU5 RNAs were incubated in the 3[prime]->5[prime] exonuclease assay system in the presence of EDTA to stabilize all transcripts. UV crosslinking analysis was performed and proteins were immunoprecipitated using a specific La polyclonal antiserum. Precipitated proteins were analyzed on a 10% acrylamide gel containing SDS.

Next, we determined if La protein was required by the poly(U) tail to inhibit the 3[prime]->5[prime] exonuclease. Using La-specific autoimmune antisera (16), we depleted La protein from HeLa nuclear extracts as described in Materials and Methods. As seen in Figure 7A, no La protein was found crosslinked to the SVU5 RNA incubated in the La-depleted extract. This indicates that we have significantly depleted La protein in these assays. As seen in Figure 7B (right side), the unstable SV RNA was still degraded by the 3[prime]->5[prime] exonuclease in both the mock-depleted and La-depleted extracts. Both the mock- and La-depleted extracts, therefore, contained adequate levels of 3[prime]->5[prime] exonuclease activity. As shown in Figure 7B (left side), the poly(U) tract present on the SVU5 RNA still functioned efficiently as an inhibitor of the exonuclease in La-depleted extracts. Similar results were also obtained using a variant of SV RNA that contained a 30 base poly(A) tail (data not shown). We conclude, therefore, that the association of La protein with the poly(U) tail is not necessary for inhibition of the 3[prime]->5[prime] exonuclease in our assays.

We next determined if poly(U) binding protein(s) was important for poly(U) tail-mediated protection from the exonuclease. SVU30 RNA, which contains a 30 base 3[prime] poly(U) tail to increase the sensitivity of assays for poly(U) binding proteins, was fully protected from 3[prime]->5[prime] exonuclease degradation (Fig. 8A, lane 0). In the presence of 50 and 500 ng of poly(U) competitor, however, the 30 base poly(U) tail no longer protected SVU30 RNA from the exonuclease [Fig. 8A, poly(U) lanes]. The ability of cold poly(U) competitor RNA to destabilize SVU30 RNA was specific, as other homopolymer competitor RNAs did not influence poly(U) tail-mediated RNA stability [Fig. 8A, poly(G) lanes]. The poly(U) tail, therefore, requires trans-acting factors to protect RNAs from the 3[prime]->5[prime] exonuclease. Surprisingly, La protein remained bound to SVU30 RNA even in the presence of 500 ng of poly(U) competitor (Fig. 8B), a level of competitor at which the terminal U tract failed to protect the RNA from the exonuclease. This result confirms our conclusion that La protein was not required for the poly(U) tail to function as an RNA stability element in vitro.

The inability of the poly(U) tail to prevent degradation of the SVU30 transcript in the presence of poly(U) competitor (Fig. 8A) suggests that poly(U) binding proteins, rather than structure, protect the 3[prime]-end of the RNA from the exonuclease. We used UV crosslinking analysis to identify candidate poly(U) binding proteins that may be involved in this protection. As seen in Figure 8C, in the presence of 50 ng of poly(U) competitor (a concentration that destabilized the SVU30 RNA; Fig. 8A), two poly(U) binding proteins of ~30 and 40 kDa were competed from the poly(U) tail. The interaction of these two proteins with the poly(U) tail was specific, as they were not observed to crosslink to SV RNA derivatives that lack a poly(U) tract (Fig. 8D, lane SV). The 30 and 40 kDa species also specifically crosslinked to the SVU5 transcript that contains a five base uridylate tail (Fig. 8D). Two additional poly(U) binding proteins of ~60 and 70 kDa bound to SVU30 RNA and were competed at the 500 ng level of poly(U) competitor (Fig. 8C). The binding of these proteins, however, did not precisely correlate with the ability of the poly(U) tail to function as an RNA stability element. We conclude that a poly(U) tail inhibits the activity of a 3[prime]->5[prime] exonuclease by associating with poly(U) binding proteins. We have identified two poly(U) binding proteins of ~30 and 40 kDa that associate with poly(U) and may be important for protecting the transcript from 3[prime]->5[prime] exonuclease digestion. These proteins may play a role in vivo in the stability of RNAs that contain terminal uridylate tracts.

DISCUSSION

The data presented above suggest that all cellular RNAs have evolved a 3[prime]-terminal structure or protein-RNA complex to protect them from 3[prime]->5[prime] exonucleases. This observation suggests an important role for a 3[prime]->5[prime] exonuclease in RNA turnover in mammalian cells. A protein-dependent strategy for exonuclease protection affords more opportunities for regulation than does a protein-independent strategy. It is not surprising, therefore, that homopolymer tracts are used to protect mRNAs that require differential rates of turnover. Poly(U) tails, that are found on RNA polymerase III precursor transcripts, also protect RNAs in a protein-dependent fashion. La protein was not necessary to stabilize poly(U)⁺ RNAs in our system. Rather, the binding of proteins of 30 and 40 kDa was associated with protection from exonucleolytic decay. The identification of these two factors and their possible role in maintaining the stability of precursor RNAs in vivo awaits future experiments.

The results reported here do not suggest a role for La protein in protecting RNAs from the action of 3[prime]->5[prime] exonucleases. However, we cannot absolutely rule out such a role for La protein for the following reasons. First, a small amount of La protein that is below the limit of detectability in our crosslinking assays may remain in our immunodepleted extracts (Fig. 7). This small amount of La protein may still be sufficient to protect a substantial fraction of the ~50 fmol of input RNA containing a poly(U) tail. Furthermore, in the experiments described in Figure 8, La protein may be bound at the 5[prime]-side of the poly(U) tract, allowing enough free poly(U) for the exonuclease to assemble on the 3[prime]-end of the RNA (9). The best way to definitively address these possibilities is with a reconstituted system using purified components. Experiments are currently underway to develop such a system.

Figure 7. La protein is not essential for poly(U) tails to protect RNAs from the 3[prime]->5[prime] exonuclease. (A) Extracts were depleted of endogenous La protein using specific antisera or mock-depleted as described in Materials and Methods. SVU5 RNA was incubated in the La-depleted (-La lane) or mock-depleted (mock lane) extract for 5 min in the presence of EDTA. Reaction mixtures were irradiated with UV light, treated with RNase A and analyzed on a 10% acrylamide gel before (total lanes) or after immunoprecipitation with anti-La antiserum (Ippt. lanes). The position of the 50 kDa La protein is indicated by the arrow. (B) SVU5 or SV RNAs were incubated in the La-depleted (-La lanes) or mock-depleted (mock lanes) extracts for 0 (input) or 30 min in the 3[prime]->5[prime] exonuclease assay. RNA reaction products were analyzed on a 5% acrylamide gel containing 7 M urea.

A large protein complex called the exosome contains several 3[prime]->5[prime] exonucleases that are essential for mRNA turnover in yeast (10,11). The human gene RRP4 bears significant homology to a component of the yeast exosome and can functionally complement the rrp4-1 yeast mutant (10). Therefore, human cells are likely to contain similar enzymatic activities that play a role in mRNA stability and nucleolar RNA processing. The 3[prime]->5[prime] exonuclease activity observed in this study may be related to these activities and may play a role in cytoplasmic mRNA turnover.

Substrate RNAs containing internal RNA structures were efficiently degraded by the 3[prime]->5[prime] exonuclease as no detectable degradation intermediates were observed. Both the VA1int RNA (that is ~64% double-stranded) or the SVintHis RNA were unable to slow the 3[prime]->5[prime] exonuclease once it initiated at the 3[prime]-end of the RNA substrate. These data suggest that the 3[prime]->5[prime] exonuclease may possess RNA unwinding capabilities.

Several in vivo observations support our conclusion that 3[prime]-terminal structures in the absence of proteins can stabilize mammalian RNAs. First, replacing the terminal stem-loop element on histone mRNAs with a 3[prime] stem-loop structure found on the U1 RNA does not affect transcript accumulation in vivo (30). Second, elements in the 3[prime]-terminus of several snoRNAs, as well as a stem-loop structure in the stable Herpes simplex virus latency-associated transcript, are important for in vivo transcript accumulation (31,32). Third, a poly(G) tract, that slows down the action of exonucleases in yeast cells (33), also inhibited the initiation of the 3[prime]->5[prime] exonuclease in vitro when placed at the 3[prime]-terminus of a substrate RNA (data not shown). The protection by a terminal poly(G) tract did not require specific titratable trans-acting factors as the addition of poly(G) competitor failed to destabilize substrate RNAs in our assays (data not shown). Finally, the histone stem-loop structure used in the studies reported here does not contain the short flanking sequences that are required for the binding of the histone stem-loop binding protein (34). This suggests that the histone stem-loop structure in our assays is also not associated with proteins. Therefore, RNA alone can block the assembly of the 3[prime]->5[prime] exonuclease present in our extracts onto the 3[prime]-end of substrate RNAs. Trans-acting factors that bind to terminal structures may be required to regulate exonuclease assembly, as well as the rate of degradation, of histone mRNAs and perhaps other poly(A)^- RNAs.

Figure 8. Poly(U) binding proteins are important for uridylate tails to protect RNAs from the 3[prime]->5[prime] exonuclease. (A) A variant of SV RNA containing a 30 base poly(U) tail (SVU30) was incubated in the 3[prime]->5[prime] exonuclease assay in the presence of 0, 50 or 500 ng of poly(U) or poly(G) competitor RNA for 30 min. Reaction products were purified and analyzed on a 5% acrylamide gel containing 7 M urea. (B) SVU30 RNA was incubated in the 3[prime]->5[prime] exonuclease assay in the presence of 0 or 500 ng of poly(U) competitor RNA. EDTA was added to stabilize all RNAs to allow for valid comparison (9). Reaction mixtures were exposed to UV light for 5 min, treated with RNase A and crosslinked proteins were analyzed before (input lane) or after (Ippt. lane) immunoprecipitation with anti-La antiserum on a 10% acrylamide gel containing SDS. (C) SVU30 RNA was incubated in the 3[prime]->5[prime] exonuclease assay system in the presence of 0 or 500 ng of poly(U) competitor RNA for 5 min. EDTA was added to stabilize all RNAs to allow for valid comparisons. Reaction mixtures were exposed to UV light, treated with RNase A and total crosslinked proteins were analyzed on a 10% acrylamide gel containing SDS. (D) SV or SVU5 RNAs were incubated in the in vitro system for 5 min in the presence of EDTA to stabilize the transcripts. SVU5 RNA was also incubated in the presence of the indicated amount of poly(U) competitor RNA. Reaction mixtures were exposed to UV light, treated with RNase A and total crosslinked proteins were analyzed on a 10% acrylamide gel containing SDS.

Many prokaryotic RNAs contain terminal hairpins that function as RNA stability elements (41,42). Bacterial poly(A) polymerase adds a short poly(A) tract to the 3[prime]-end of these RNAs to target them for degradation by creating an assembly site for the 3[prime]->5[prime] exonucleases RNase II and PNPase (43). The refractory nature of terminal RNA structures to the 3[prime]->5[prime] exonuclease observed in this study is very reminiscent of the prokaryotic transcripts. Poly(A) or poly(U) tracts that are added post-transcriptionally to the 3[prime]-end of eukaryotic transcripts serve to protect RNAs from the action of exonucleases, but only in the presence of RNA binding proteins. In the absence of RNA binding proteins, the addition of these tracts creates an effective single-stranded assembly site for 3[prime]->5[prime] exonucleases. The evolution of the 3[prime]-terminal sequences of mRNAs, therefore, was probably not constrained by the need to provide accessibility for exonucleases.

The observation that 3[prime]-terminal structures serve as important RNA stability elements may be useful in streamlining the design of eukaryotic expression vectors for the therapeutic delivery of antisense RNAs and ribozymes. Recently, an improvement in the efficiency of targeting of antisense RNA to the fibrillarin gene was reported by placing the antisense sequence within the U1 RNA (35). Our data suggest that deleting most of the U1 RNA while maintaining the 3[prime]-terminal stem-loop structure may further refine these types of delivery systems. In addition, VA1 RNA has been used as a carrier for the cytoplasmic delivery of a ribozyme (36). Our data suggest that maintaining the 5[prime]- and 3[prime]-end base pairing of the carrier VA1 RNA may be sufficient for ribozyme or antisense transcript stabilization. Such refinements may help to minimize possible masking effects of surrounding RNA sequences or protein interactions, thereby enhancing the therapeutic activity of RNA compounds.

RNA polymerase III precursor RNAs contain short 3[prime]-terminal poly(U) tracts (28). Similar to poly(A) tails, poly(U) tracts can block the 3[prime]->5[prime] exonuclease activity present in HeLa nuclear extracts by associating with specific trans-acting factors (Figs 5-8). The 3[prime]->5[prime] exonuclease activity, therefore, can be blocked by two distinct 3[prime]-terminal RNA-protein interactions. These data suggest that any 3[prime]-terminal RNA-protein interaction may be capable of blocking the initiation of the exonuclease on RNA substrates. The binding of specific trans-acting factors to homopolymer tails, however, is likely to be important in regulating the rate at which polymerase III precursor RNAs or mRNAs are degraded.

VA1 RNA contains a 3[prime]-terminal uridylate tract in addition to extensive 3[prime]-end base pairing with the 5[prime]-end (Fig. 2). Since both 3[prime]-terminal poly(U) tracts and 3[prime]-terminal base pairing can block the initiation of the 3[prime]->5[prime] exonuclease, it is unclear why VA1 RNA would contain two 3[prime]-terminal stabilizing elements. Perhaps the terminal uridylate tract protects newly synthesized VA1 RNA from exonucleases prior to the transcript folding into its mature conformation.

ACKNOWLEDGEMENTS

This work was supported by grants from the Pew Foundation and the National Institutes of Health (GM56434) to J.W. L.P.F. was supported by Cancer Training Grant CA09665.

REFERENCES

1. Schoenberg,D.R. and Chernokalskaya,E. (1997) mRNA Metabolism and Post-Transcriptional Gene Regulation. Wiley-Liss, New York, NY, pp. 217-240.

2. Decker,C.J. (1998) Curr. Biol., 8, R238-R240. MEDLINE Abstract

3. Shatkin,A.J. (1976) Cell, 9, 645-653. MEDLINE Abstract

4. Murthy,K.G.K., Park,P. and Manley,J.L. (1991) Nucleic Acids Res., 19, 1685-2692.

5. Larimer,F.W., Hsu,C.L., Maupin,M.K. and Stevens,A. (1992) Gene, 120, 51-57. MEDLINE Abstract

6. LaGrandeur,T.E. and Parker,R. (1998) EMBO J., 17, 1487-1496. MEDLINE Abstract

7. Caponigro,G. and Parker,R. (1996) Microbiol. Rev., 60, 233-249. MEDLINE Abstract

8. Bernstein,P., Peltz,S.W. and Ross,J. (1989) Mol. Cell. Biol., 9, 659-670. MEDLINE Abstract

9. Ford,L.P., Bagga,P.S. and Wilusz,J. (1997) Mol. Cell. Biol., 17, 398-406. MEDLINE Abstract

10. Mitchell,P., Petfalski,E., Shevchenko,A., Mann,M. and Tollervey,D. (1997) Cell, 91, 457-466. MEDLINE Abstract

11. Anderson,J.S.J. and Parker,R. (1998) EMBO J., 17, 1497-1506.

12. Marzluff,W.F. and Pandey,N.B. (1998) Trends Biochem. Sci., 13, 49-52.

13. Wang,Z.F., Whitfield,M.L., Ingledue,T.C., Dominski,Z. and Marzluff,W.F. (1996) Genes Dev., 10, 3028-3040. MEDLINE Abstract

14. Lerner,M.R. and Steitz,J.A. (1981) Cell, 25, 298-300. MEDLINE Abstract

15. Stefano,J.E. (1984) Cell, 36, 145-154. MEDLINE Abstract

16. Chamber,J.C. and Keene,J.D. (1985) Proc. Natl Acad. Sci. USA, 82, 2115-2119. MEDLINE Abstract

17. Yoo,C.J. and Wolin,S.L. (1997) Cell, 89, 393-402. MEDLINE Abstract

18. Gottlieb,E. and Steitz,J.A. (1989) EMBO J., 8, 841-850. MEDLINE Abstract

19. McLaren,R.S., Caruccio,N. and Ross,J. (1997) Mol. Cell. Biol., 17, 3028-3036. MEDLINE Abstract

20. Preiser,P., Vasisht,V., Birk,A. and Levinger,L. (1993) J. Biol. Chem., 268, 11553-11557. MEDLINE Abstract

21. Williams,A.S., Ingledue,T.C., Kay,B.K. and Marzluff,W.F. (1994)Nucleic Acids Res., 22, 4660-4666. MEDLINE Abstract

22. Wilusz,J., Feig,D.I. and Shenk,T. (1988) Mol. Cell. Biol., 8, 4477-4483. MEDLINE Abstract

23. Mellits,K.H., Manche,T.L., Robertson,H.D. and Mathews,M.B. (1990) Nucleic Acids Res., 18, 5401-5406. MEDLINE Abstract

24. Chen,F. and Wilusz,J. (1998) Nucleic Acids Res., 26, 2891-2898. MEDLINE Abstract

25. Dignam,J.D., Lebovitz,R.M. and Roeder,R.G. (1983) Nucleic Acids Res., 11, 1475-1489. MEDLINE Abstract

26. Wilusz,J. and Shenk,T. (1988) Cell, 52, 221-228. MEDLINE Abstract

27. Kolossova,I. and Padgett,R.A. (1997) RNA, 3, 227-233. MEDLINE Abstract

28. Watson,J.B., Chandler,D.W. and Gralla,J.D. (1984) Nucleic Acids Res., 12, 5369-5384. MEDLINE Abstract

29. Benne,R. (1994) Eur. J. Biochem., 221, 9-23. MEDLINE Abstract

30. Gallie,D.R., Lewis,N.J. and Marzluff,W.F. (1996) Nucleic Acids Res., 24, 1954-1962. MEDLINE Abstract

31. Ganot,P., Caizergues-Ferrer,M. and Kiss,T. (1997) Genes Dev., 11, 941-956. MEDLINE Abstract

32. Krummenacher,C., Zabolotny,J.M. and Fraser,N.W. (1997) J. Virol., 71, 5849-5860. MEDLINE Abstract

33. Muhlrad,D., Decker,C.J. and Parker,R. (1994) Genes Dev., 8, 855-866. MEDLINE Abstract

34. Williams,A.S. and Marzluff,W.F. (1995) Nucleic Acids Res., 23, 654-662.41. McLaren,R.S., Newbury,S.F., Dance,G.S.C., Causton,H.C. and Higgins,C.F. (1991) J. Mol. Biol., 221, 81-95.42. Higgins,C.F., McLaren,R.S. and Newbury,S.F. (1988) Gene, 72, 3-14.43. O'Hara,E.B., Chekanova,J.A., Ingle,C.A., Kushner,Z.R., Peters,E. and Kushner,S.R. (1995) Proc. Natl Acad. Sci. USA, 92, 1807-1811. MEDLINE Abstract

35. Montgomery,R.A. and Dietz,H.C. (1997) Hum. Mol. Genet., 6, 519-525. MEDLINE Abstract

36. Prislei,S., Buonomo,S.B., Michienzi,A. and Bozzoni,I. (1997) RNA, 3, 677-687. MEDLINE Abstract

37. Mellits,K.H. and Mathews,M.B. (1988) EMBO J., 7, 2849-2859. MEDLINE Abstract

38. Pe'ery,T., Mellits,K.H. and Mathews,M.B. (1993) J. Virol., 67, 3534-3543. MEDLINE Abstract

39. Ma,Y. and Mathews,M.B. (1996) J. Virol., 70, 5083-5099. MEDLINE Abstract

40. Wassarman,K.M. and Steitz,J.A. (1992) Mol. Cell. Biol., 12, 1276-1285. MEDLINE Abstract

---

*To whom correspondence should be addressed. Tel: +1 973 972 5218; Fax: +1 973 972 3644; Email: wilusz@umdnj.edu

---

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 29 Jan 1999
Copyright©Oxford University Press, 1999.
CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				N. L. Garneau, K. J. Sokoloski, M. Opyrchal, C. P. Neff, C. J. Wilusz, and J. Wilusz The 3' Untranslated Region of Sindbis Virus Represses Deadenylation of Viral Transcripts in Mosquito and Mammalian Cells J. Virol., January 15, 2008; 82(2): 880 - 892. [Abstract] [Full Text] [PDF]

				C. J. Wilusz and J. Wilusz New ways to meet your (3') end oligouridylation as a step on the path to destruction Genes & Dev., January 1, 2008; 22(1): 1 - 7. [Full Text] [PDF]

				M.-G. Song and M. Kiledjian 3' Terminal oligo U-tract-mediated stimulation of decapping RNA, December 1, 2007; 13(12): 2356 - 2365. [Abstract] [Full Text] [PDF]

				O. S. Rissland, A. Mikulasova, and C. J. Norbury Efficient RNA Polyuridylation by Noncanonical Poly(A) Polymerases Mol. Cell. Biol., May 15, 2007; 27(10): 3612 - 3624. [Abstract] [Full Text] [PDF]

				E. L. Murray and D. R. Schoenberg A+U-Rich Instability Elements Differentially Activate 5'-3' and 3'-5' mRNA Decay Mol. Cell. Biol., April 15, 2007; 27(8): 2791 - 2799. [Abstract] [Full Text] [PDF]

				H. J. Boot and S. B. E. Pritz-Verschuren Modifications of the 3'-UTR stem-loop of infectious bursal disease virus are allowed without influencing replication or virulence Nucleic Acids Res., January 12, 2004; 32(1): 211 - 222. [Abstract] [Full Text] [PDF]

				K. Tanabe, N. Ito, T. Wakuri, F. Ozoe, M. Umeda, S. Katayama, K. Tanaka, H. Matsuda, and M. Kawamukai Sla1, a Schizosaccharomyces pombe Homolog of the Human La Protein, Induces Ectopic Meiosis when Its C Terminus Is Truncated Eukaryot. Cell, December 1, 2003; 2(6): 1274 - 1287. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (12) Request Permissions Commercial Re-use Guidelines for Open Access NAR Content Google Scholar Articles by Ford, L. P. Articles by Wilusz, J. Search for Related Content PubMed PubMed Citation Articles by Ford, L. P. Articles by Wilusz, J. Social Bookmarking          What's this?