• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. May 13, 2008; 105(19): 6864–6869.
Published online May 5, 2008. doi:  10.1073/pnas.0800420105
PMCID: PMC2383978
From the Cover
Biochemistry

Cis- and trans-splicing of mRNAs mediated by tRNA sequences in eukaryotic cells

Abstract

The formation of chimeric mRNAs is a strategy used by human cells to increase the complexity of their proteome, as revealed by the ENCODE project. Here, we use Saccharomyces cerevisiae to show a way by which trans-spliced mRNAs can be generated. We demonstrate that a pretRNA inserted into a premRNA context directs the splicing reaction precisely to the sites of the tRNA intron. A suppressor pretRNA gene was inserted, in cis, into the sequence encoding the third cytoplasmic loop of the Ste2 or Ste3 G protein-coupled receptor. The hybrid RNAs are spliced at the specific pretRNA splicing sites, releasing both functional tRNAs that suppress nonsense mutations and translatable mRNAs that activate the signal transduction pathway. The RNA molecules extracted from yeast cells were amplified by RT-PCR, and their sequences were determined, confirming the identity of the splice junctions. We then constructed two fusions between the premRNA sequence (STE2 or STE3) and the 5′- or 3′-pretRNA half, so that the two hybrid RNAs can associate with each other, in trans, through their tRNA halves. Splicing occurs at the predicted pretRNA sites, producing a chimeric STE3-STE2 receptor mRNA. RNA trans-splicing mediated by tRNA sequences, therefore, is a mechanism capable of producing new kinds of RNAs, which could code for novel proteins.

Keywords: ENCODE, G protein-coupled receptors, genomics, tRNA endonuclease

In higher eukaryotes, most genes contain one or more introns, which are removed by the spliceosomal complex in a regulated manner. When multiple introns are present, alternative splicing provides a way to increase the number of mRNAs from each gene and hence to enhance protein diversity. In lower eukaryotes, like the yeast Saccharomyces cerevisiae, only ≈250 genes of >6,000 contain an intron, and for most of them, a single intron is present. Splice-site sequences in yeast introns are characterized by a strict consensus, and alternative splicing is rare (1, 2).

The formation of chimeric mRNAs constitutes another strategy that eukaryotic cells can use to increase proteome complexity. This phenomenon is much more widely spread than previously thought. Recent studies have shown that many of the human genes analyzed in the ENCODE project use exons lying outside their canonical boundaries (3, 4).

Chimeric mRNAs can be generated by several mechanisms. In one way, an exon of one gene is fused to an exon of a different gene in a reaction mediated by the spliceosome complex. This fusion can occur in cis, when the two exons derive from two adjacent cotranscribed genes (5, 6), or in trans, when the two exons come from different premRNA transcripts (7, 8). Another way for the generation of chimeric transcripts requires the use of a tRNA-splicing endonuclease. We have shown that an endonuclease of archaeal origin can carry out trans- and cis-splicing of mRNA in cultured mouse cells (9).

The expression of tRNA genes in most organisms requires the accurate removal of intervening sequences. In Bacteria, pretRNA splicing is autocatalytic, whereas in Archaea and Eukarya, these introns are removed by endonucleases. The archaeal tRNA splicing endonucleases recognize a structural element in the pretRNAs comprised of two 3-nt bulges separated by a 4-bp helix, the so-called bulge–helix–bulge (BHB) motif. The BHB motif is recognized and cleaved independently of the mature tRNA domain (10). In Eukarya, the splicing mechanism of pretRNAs is more complex. The scenario of tRNA splicing is normally dominated by the mature tRNA domain (1114). The three-dimensional structure of the mature domain is important in recognition by the eukaryal endonucleases, which bind to one or more recognition sites common to all pretRNAs and measure the distance to the equivalently positioned intron-exon junctions (15, 16). Several studies have shown that the intron itself contributes to specific splice site recognition and that important interaction elements also lie at the intron-exon boundaries (1719).

In a previous study, we have shown that long noncanonical pretRNA molecules, containing the BHB structure, are correctly cleaved and ligated in S. cerevisiae, producing functional tRNAs (20). Here, we addressed the question whether the yeast tRNA splicing system can specifically splice mRNA sequences. Because the yeast tRNA endonuclease, like its eukaryotic homologues, generally requires the presence of the mature tRNA domain to precisely recognize and cleave at the splice sites, we used premRNAs in which tRNA sequences were inserted. We demonstrate that splicing of hybrid premRNAs/pretRNAs indeed occurs at the sites predicted by the sequence and conformation of the pretRNA, producing functional mature mRNAs and tRNAs. Splicing of hybrid RNA molecules takes place both in a cis- and, more interestingly, also in a trans-state, i.e., when the pretRNA structure is split between complementary 5′- and 3′-pretRNA halves that are fused to different premRNA sequences. This way of producing trans-spliced chimeric RNAs may have important implications for the comprehension of the complexity of eukaryotic genomes.

Results

A PretRNA Sequence Can Be Removed from the Middle of a PremRNA.

We used the S. cerevisiae mating signal transduction pathway as a model system. The advantages of this system are that it is well known genetically and biochemically, and that the signals elicited by pheromones are amplified hundreds of times, so that a suitable reporter gene product located at the end of the pathway could be detected even in the presence of a weak signal.

We inserted a tRNA gene sequence, containing or lacking the BHB motif, into the sequence encoding the third cytoplasmic loop of the Ste2 or Ste3 receptor (which bind, respectively, the pheromones α- or a-factor). It should be mentioned that the STE2 and STE3 genes, like most S. cerevisiae genes, are devoid of introns. The tRNA inserted into the receptor premRNA was SUP4 tRNATyr, a well known suppressor of nonsense ochre (UAA) mutants (21). To ensure continuity of the hybrid transcript, the tRNA 5′- and 3′-ends were joined at the level of the acceptor stem. In addition, an artificial CCA sequence was added at the tRNA 3′-end. A mature tRNA can be obtained from this precursor if the tRNA splicing endonuclease cleaves at the pretRNA splicing sites, the cut ends are ligated, and a cleavage by RNase P occurs, leaving a mature 5′-end and a 3′-terminal CCA in the tRNA (20, 22). Very recently, a similar gene organization has been found in a red alga (23).

The hybrid premRNAs/pretRNAs transcribed by the gene constructs are shown in Fig. 1. In one type of construct, the SUP4 pretRNA carries the BHB feature (Fig. 1 Upper Left), whereas in another one, the pretRNA is identical to wild-type SUP4 pretRNA, which lacks the BHB (Fig. 1 Lower Left). Correct cleavage by the tRNA splicing endonuclease, followed by RNA ligation and RNase P cut, should release from both kinds of hybrid precursors mature SUP4 tRNA and receptor mRNAs with an additional stem-loop sequence (Fig. 1 Right).

Fig. 1.
Hybrid premRNAs/pretRNAs. The hybrid precursors consist of the receptor premRNA (STE2 or STE3) and the SUP4 pretRNATyr, containing the BHB structure (Upper Left) or without it (Lower Left). The first nucleotide at the 5′-side of the bottom stem ...

The tRNA Spliced from the Hybrid PremRNA/PretRNA Is Functional in Suppression.

The mature SUP4 tRNATyr produced by correct splicing of the hybrid precursors shown in Fig. 1 was analyzed by its ability to suppress nonsense mutations in appropriate genes in S. cerevisiae. The STE2 or STE3 genes, containing in their middle the pretRNA sequence (with or without the BHB), were introduced into a yeast strain (PJ17–1A) bearing ochre nonsense mutations in three different loci, met4-1o, lys2-1o, and ade2-1o. These mutations require increasing levels of SUP4 tRNATyr to be suppressed, i.e., met4-1o is suppressed by low levels, lys2-1o is suppressed by intermediate levels, and the ade2-1o mutation requires high levels of expression of the SUP4 gene (24). This difference is due to the functionality of the mutated protein when tyrosine is inserted as a replacement for the wild-type amino acid. The use of a multiple ochre mutated strain permits an evaluation of the levels of expression of the SUP4 gene and also avoids the potential deceit of misinterpreting revertants of the nonsense mutation as true suppression events. The yeast strain PJ17–1A carrying a plasmid containing the gene for the hybrid premRNA/pretRNA was plated onto selective media without methionine, lysine, or adenine and compared with cells containing the vector alone. In Fig. 2A, we show that both the STE2 and STE3 genes containing the pretRNA insert produce enough mature SUP4 tRNATyr to suppress the met4-1o mutation (the easiest to be suppressed). A slight suppression was obtained for the lys2-1o mutation, and there was almost no suppression of the ade2-1o mutation.

Fig. 2.
In vivo suppression in yeast strains with plasmids encoding hybrid premRNAs/pretRNAs. (A) The PJ17-1A strain (met4-1o, lys2-1o, ade2-1o), carrying the following hybrid precursors containing the NcoI site (unless otherwise stated), with or without the ...

To increase the amount of mature SUP4 tRNATyr, we made use of the observation (25) that the tap1-1 mutation in the RAT1/TAP1 gene, which encodes a 5′-3′ exoribonuclease (26, 27), increases the expression of a defective tRNA gene. When we analyzed the tap1-1 mutant containing the constructs bearing the hybrid premRNA/pretRNA, a very good level of suppression was obtained for the lys2-1o and even for the ade2-1o mutation (Fig. 2B).

Another way to determine suppression of the nonsense ade2-1o mutation is to grow the transformed cells on rich media. On these plates, ade2 mutant colonies appear red, because of the accumulation of a pigment, whereas the congenic wild-type ADE2+ strain is white. Depending on the level of suppression, yeast colonies carrying a suppressor tRNA will be white (high suppression level), pink (intermediate level), or red (no suppression at all) (20, 25). Fig. 2C shows a high level of suppression in cells carrying a gene encoding the hybrid premRNA/pretRNA.

The mRNA Spliced from the Hybrid PremRNA/PretRNA Is Translated.

Upon correct cleavage by the tRNA splicing endonuclease and ligation, the hybrid premRNAs/pretRNAs should generate, in addition to mature SUP4 tRNA, mRNAs containing a short stem–loop insertion that, after translation, should produce functional receptors active in signal transduction. We constructed the hybrid precursors such that only spliced RNA molecules would provide translatable mRNAs. If no splicing occurs, the precursors cannot be translated because of the presence of two stop codons in the inserted tRNA sequence.

The hybrid genes were introduced into a yeast strain (DDS2) whose main features consist of deletions of the endogenous STE2 and STE3 genes; deletion of FAR1 (allowing continuous growth even during mating pathway activation; ref. 28); and a mutation in the SST2 gene, which encodes a regulatory protein that, when mutated, renders the strain supersensitive to pathway activation (29, 30). Additionally, this strain carries a lacZ reporter gene placed under the control of the FUS1 promoter (which is activated by the signal transduction pathway when pheromones bind to their receptors). β-Gal activity is therefore a measure of the induction level of the signal transduction pathway.

In Fig. 3, we show that both for the hybrid STE2 premRNA/pretRNA (Fig. 3A) and STE3 premRNA/pretRNA (Fig. 3B), there is strong induction of FUS1-lacZ expression. Because the spliced mRNA carries an insertion of a stem–loop, an additional amino acid sequence is present in the protein. This sequence varies depending on whether the pretRNA contains or lacks the BHB structure. The hybrid precursor used in the experiment shown in Fig. 3A, column 1, carries a BHB motif and the amino acid sequence inserted in the third cytoplasmic loop as a result of splicing is LGARGKDILSHL. The level of β-Gal activity is ≈40% of that obtained with full induction, i.e., in the presence of a wild-type Ste2 receptor (Fig. 3A, column 5). Full induction was obtained also with a STE2 construct containing the same stem–loop as a within-frame insert (data not shown): this last result is an indication there are not appreciable effects on the translation of these mRNAs because of the presence of a stem–loop. Two other hybrid constructs, with a different nucleotide sequence, one containing the BHB motif (encoding the amino acid sequence HGRGKDILYH as a result of splicing) and the second one lacking the BHB motif (giving the amino acid sequence HGEDHYEFPHH), were not functional when made with the STE2 gene (Fig. 3A, columns 2 and 3) but gave a high level of induction when used with the STE3 gene (Fig. 3B, columns 1 and 2). A possible explanation of this difference is that the inserted amino acid sequence affects the activity of the Ste2 and Ste3 receptor differently, although we cannot exclude that there is also a variation in the splicing efficiency of the hybrid precursors.

Fig. 3.
Induction of the mating signal transduction pathway. β-Gal activity was assayed in permeabilized cells of strain DDS2. Data are in Miller units and represent averages of three samples; error bars correspond to 1 SD. (A) STE2 gene constructs: ...

Splicing of Hybrid PremRNA/PretRNA Takes Place Precisely at the tRNA Splicing Sites.

We conclude from the experiments shown in Fig. 3 that a hybrid premRNA/pretRNA can produce a functional mRNA that is correctly translated. These experiments, however, do not indicate the exact positions where the splicing takes place. To determine the splicing sites, we analyzed the RNA transcripts by RT-PCR. Total RNA was extracted from yeast cells containing the various plasmids encoding the hybrid precursors. The cDNA was generated by reverse transcription by using suitable primers, and the cDNA was then amplified by PCR by using specific primers corresponding to sequences upstream and downstream from the tRNA sequence inserted into the premRNA. Fig. 4 A and B show, respectively, the RT-PCR analysis for the STE2 and STE3 constructs. In Fig. 4A, lane 1, a major band is shown that has the size expected for the unspliced precursor and, below it, a minor band whose size corresponds to the spliced product. To increase the visibility of the putative spliced product, we exploited the fact that the SUP4 tRNA gene bears a SmaI site in its middle. We digested the unspliced cDNA band with SmaI and then reamplified it with the same pair of primers. Now, the putative spliced product was markedly increased (Fig. 4A, lanes 2 and 4). Similar results were obtained for the hybrid construct that does not contain the BHB sequence, but in this case, the band corresponding to the spliced product was seen only in the SmaI-digested reaction (Fig. 4A, lane 6).

Fig. 4.
RT-PCR analysis of the processing of hybrid premRNAs/pretRNAs. RNA was extracted from yeast cells containing different plasmids, reverse-transcribed, amplified, and electrophoresed on an agarose gel, together with size markers (M). STE2 constructs, containing ...

For the experiment shown in Fig. 4B, performed with the STE3 hybrid precursor, we improved the ability to detect the splicing products by using primers that were more specific for the spliced mRNA than for the unspliced one (the last two nucleotides of the primer, of 20, binding preferentially to the spliced product). In this way, we could easily detect the product corresponding to the spliced mRNA (Fig. 4B, lanes 2 and 4). We sequenced all of the bands purified from the agarose gel and confirmed that the unspliced and spliced products were indeed exactly as expected (see the chromatograms in Fig. 4).

Trans-Splicing Mediated by tRNA Sequences Can Produce Chimeric mRNAs.

We concluded from the experiments described so far that a pretRNA domain inserted into a premRNA context could direct the splicing reaction precisely at the sites typical of tRNA introns. This reaction takes place in a continuous RNA molecule where the pretRNA is placed in cis with respect to the premRNA. We reasoned it is plausible that the same reaction could occur even when the pretRNA domain is formed in trans from its two tRNA halves. Fig. 5A shows a scheme of this concept: two different premRNA sequences (1 and 2) are brought into association through their tRNA halves; splicing mediated by the tRNA domain produces a chimeric mRNA. In Fig. 5B, we show the sequences of the two components of the trans-splicing reaction. In this case, one element is the 5′-half of the STE3 gene joined to the 3′-half of the SUP4 tRNA gene. The second element is the 3′-half of the STE2 gene joined to the 5′-half of the SUP4 tRNA gene. The two tRNA halves are able to form a BHB structure and reconstitute a mature tRNA domain. A splicing reaction occurring at the predicted tRNA sites would generate mRNA encoding a Ste3-Ste2 chimera. The two DNA components were placed on different plasmids (which is analogous to having them in different chromosomes). The RNA extracted from yeast cells expressing both plasmids was reverse-transcribed, and PCR was performed by using two primers specific, respectively, for the STE3 and the STE2 genes. The reaction product was electrophoresed on an agarose gel (Fig. 5C), and a DNA band, with a size corresponding to the expected spliced product, was purified and sequenced. The chromatogram in Fig. 5D shows a sequence that is indeed a chimera between STE3 and STE2; moreover, the joint is precisely the one predicted by the position of the splicing sites of the pretRNA domain. This result is thus a clear demonstration that a previously undescribed kind of trans-splicing reaction, mediated by tRNA sequences, can produce chimeric mRNAs.

Fig. 5.
Chimeric mRNAs can be generated by trans-splicing of pretRNA sequences joined to two different premRNAs. (A) Scheme of the trans-splicing reaction. Splicing sites are those determined by the tRNA domain formed by the two complementary tRNA halves joined ...

The chimeric mRNA would produce, if translated, a chimeric receptor protein, which is very unlikely to be functional. We then performed the same trans-splicing experiment as in Fig. 5 using both halves of the STE3 mRNA joined to the two halves of the SUP4 tRNA. In this case, trans-splicing of the two hybrid RNA molecules reconstituted a full-length STE3 mRNA that, after translation, was active in signal transduction [supporting information (SI) Fig. S1]. This result shows that the trans-spliced mRNA is functional.

Discussion

In this work, we demonstrated that trans-spliced RNAs could be produced in eukaryotic cells by means of tRNA sequences adjacent to them, therefore potentially increasing the diversity of the transcriptome and proteome. We have first shown that premRNAs, in which pretRNA sequences have been inserted in cis, are spliced at the specific pretRNA splicing sites in yeast cells. The hybrid premRNA/pretRNA molecules release, upon splicing, a mature functional tRNA and a correctly translatable mRNA. In our case, we used the nonsense suppressor SUP4 tRNATyr and the mRNAs encoding the Ste2 or Ste3 G protein-coupled receptor, which activate the yeast mating signal transduction pathway. Subsequently, we constructed two fusions between the receptor mRNA sequence (STE2 or STE3) and the SUP4 5′- or 3′-tRNA half, so that the two RNA molecules can associate with each other, in trans, through the tRNA halves. In this case, too, splicing occurs at the pretRNA splicing sites, producing a chimeric STE3-STE2 receptor mRNA.

The enzyme responsible for the cleavage reaction is presumably the tRNA splicing endonuclease. In fact, the splice sites of the cis- or trans-spliced premRNA/pretRNA are precisely those one would expect for a cleavage performed by the tRNA splicing endonuclease. Many studies with purified tRNA endonucleases from different organisms, and the use of mutants, have given us a very detailed picture of the cleavage reaction, and we are able to predict the exact splice positions in the pretRNA on the basis of its sequence and conformation (11, 3133). Another enzyme, Ire1p, which has both kinase and endoribonuclease activities, cleaves HAC1 mRNA in yeast (34) and XBP1 mRNA in mammals (35), generating a 2′,3′-cyclic phosphate and a 5′-OH like the tRNA splicing endonuclease, but recognizes completely different RNA structural features, and therefore it is very unlikely that it could be involved in the findings shown here (36). The most likely candidate to ligate the cleaved premRNA sequences is the yeast tRNA ligase. This enzyme has already been shown to participate in the ligation of substrates other than tRNAs, such as HAC1 mRNA (37). In a previous work in mouse cells, we showed that both possible RNA recombinants were generated in the trans-splicing event initiated by the archaeal endoribonuclease (9).

We could not detect splicing in mRNA precursors containing the BHB structural motif alone, i.e., without the mature tRNA domain; however, it is possible that this reaction occurs at low efficiency, below detection level in our experimental conditions. In fact, we have already shown that, at least in vitro, eukaryotic tRNA splicing endonucleases can cleave RNA molecules containing a BHB structure (11, 19). Based on the vast body of data available on tRNA splicing, it is plausible to maintain that the major determinant for the splicing reaction would be the conformational feature of the RNA, rather than a particular sequence.

Trans-splicing of mRNAs mediated by tRNA sequences could take place in eukaryotic cells through the presence of numerous tRNA genes or pseudotRNA genes, which are usually interspersed in the genome. Their number ranges from several hundred, for bona fide genes, up to 22,000 for pseudogenes, as in mouse (38). PretRNA sequences (or parts of them) adjacent to premRNAs, if associating with each other, could in principle reconstitute a tRNA or a tRNA-like structure able to promote tRNA-mediated splicing of premRNAs. A support for this concept comes from Archaea, where several instances of functional tRNAs being created by trans-splicing of separate 5′- and 3′-halves have been identified (13, 39, 40).

Another feasible way to achieve trans-splicing of mRNAs by the tRNA endonuclease is to exploit the vast repertoire of repetitive sequences present in eukaryotic organisms. In the human genome, about half of the nucleotide sequence consists of repetitive elements (41). The short repeats (SINEs, short interspersed nuclear elements) are related to tRNA genes or other RNA Polymerase III-transcribed genes, and their number ranges from a few hundred to ≈500,000 for the MIR (mammalian interspersed repeats), which is a tRNA-derived family (42). Moreover, there are more than one million copies of the Alu elements, the most abundant family of repeats typical of humans and other primates, that by themselves comprise ≈10% of the whole genome (43). In the past, repetitive sequences were called “junk” DNA. Nowadays, however, considerable evidence points to a more complex picture, where repetitive elements can be recruited to reshape the genome and promote its evolution. Repetitive sequences thus constitute a large reservoir of potential regulatory elements, functioning, for instance, in alternative splicing, RNA editing, transcription and translation regulation (4345).

A repetitive sequence (or part of it) that is found inside the coding region of an mRNA, as shown in many cases (45), could be cleaved, in some instances, by the tRNA splicing endonuclease. Such a case would be equivalent to the cis splicing described in this study. Alternatively, two repetitive sequences, or parts of them, may be adjacent to two different genes: one repeat at the 5′ side of one gene and the other repeat at the 3′ side of a second gene. If the two repeats can associate with each other through complementary sequences, it is conceivable that trans-splicing mediated by the tRNA splicing endonuclease can occur. An extensive screening of human or other eukaryotic genome databases should reveal protein-encoding genes next to tRNA sequences or tRNA-like repeats capable of associating with each other to form RNA structures potentially cleavable by the tRNA splicing endonuclease. A targeted experimental analysis would then elucidate the actual presence of trans-spliced RNAs of this type and their splice sites.

The ENCODE project is committed to unraveling the complexity of human genome expression (46). Besides alternative splicing (47), spliceosomal trans-splicing (7, 48), start and stop of transcription at sites other than the canonical signals (3), we propose that trans-splicing of mRNAs mediated by tRNA-derived sequences is another possible way to produce novel proteins.

Materials and Methods

Strains and Plasmids.

S. cerevisiae strains were: PJ17-1A (MATa, trp1, ura3-1, ade2-1o, lys2-1o, met4-1o, can1-100o, gal10-1u, his5-2u, leu2-1u) (49); GDS4-16D, a derivative of PJ17-1A that carries the mutation tap1-1 in the TAP1/RAT1 gene (25, 26); SG2 (Mata, ste2Δ, ste3Δ::ADE2, ura3-1, leu2-3,112, trp1-1, ade2-1, his3-11,5); DDS2 (Matα, ste2Δ, ste3Δ::ADE2, far1Δ, FUS1-lacZ::TRP1, sst2, ura3-52, leu2-3, trp1, ade2-1, lys2-1, his3; derived from strain RM7, ref. 50). In strain DDS2, the STE2 and FAR1 genes were deleted by using the hit-and-run system as described in ref. 50. The sst2 mutant was isolated as a spontaneous blue colony on X-Gal plates; the SST2 gene was sequenced and found mutated at position 1650 (C→ G). This mutation creates a stop codon. Bacterial plasmid pUC19 and the yeast shuttle vectors (R&D Systems) PYX212 (URA3-2μ-TPI promoter), PYX222 (HIS3-2μ-TPI promoter), and PYX213 (URA3-2μ-GAL promoter) were used. Growth media were as described (20).

Hybrid Receptor/tRNA Gene Constructs.

The STE2 gene was provided by L. Hartwell (University of Washington, Seattle); the STE3 gene was obtained by PCR. The StyI site (CCTTGG) in STE2 (nucleotide position nos. 705–710) and the sequence TTCAGG in STE3 (nucleotide position nos. 594–599) were mutated by PCR-based techniques to the NcoI site (CCATGG). SUP4 tRNA gene sequences were introduced into, or joined to, the receptors' genes (at the StyI or NcoI sites) by means of one or more pairs of oligonucleotides, synthesized using the 392 DNA/RNA Synthesizer (Applied Biosystem). Annealing and ligation were according to standard procedures. The β-Gal assay to determine signal transduction activity of the receptors was performed as described (50).

RNA Extraction, RT-PCR, Agarose Gel Electrophoresis, and Sequencing.

Standard techniques were used. For experimental details, see SI Materials and Methods and Table S1.

Supplementary Material

Supporting Information:

Acknowledgments.

We thank T. Wagner for critical reading of the manuscript; P. Fruscoloni, R. Matteoni, G. D. Tocchini-Valentini, M. Zamboni, and the members of our laboratory for helpful discussions; D. De Simone for help in yeast strain construction; G. Di Franco for technical assistance; T. Passi for help with figures; and A. Ferrara and T. Cuccurullo for help with the manuscript. This work was supported by the Italian Ministry of Research Fund for Basic Research (G. Armenise–Harvard Foundation, Idee Progett. 2005 and SVIFASTA Grants) and European FP6 contracts (MUGEN, EURASNET, EUCOMM, and EUMODIC).

Footnotes

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

See Commentary on page 6793.

This article contains supporting information online at www.pnas.org/cgi/content/full/0800420105/DCSupplemental.

References

1. Pleiss JA, Whitworth GB, Bergkessel M, Guthrie C. Rapid, transcript-specific changes in splicing in response to environmental stress. Mol Cell. 2007;27:928–937. [PMC free article] [PubMed]
2. Yang L, Park J, Graveley BR. Splicing from the outside in. Mol Cell. 2007;27:861–862. [PMC free article] [PubMed]
3. Denoeud F, et al. Prominent use of distal 5′ transcription start sites and discovery of a large number of additional exons in ENCODE regions. Genome Res. 2007;17:746–759. [PMC free article] [PubMed]
4. Kapranov P, et al. Examples of the complex architecture of the human transcriptome revealed by RACE and high-density tiling arrays. Genome Res. 2005;15:987–997. [PMC free article] [PubMed]
5. Parra G, et al. Tandem chimerism as a means to increase protein complexity in the human genome. Genome Res. 2006;16:37–44. [PMC free article] [PubMed]
6. Akiva P, et al. Transcription-mediated gene fusion in the human genome. Genome Res. 2006;16:30–36. [PMC free article] [PubMed]
7. Yang Y, Walsh CE. Spliceosome-mediated RNA trans-splicing. Mol Ther. 2005;12:1006–1012. [PubMed]
8. Horiuchi T, Aigaki T. Alternative trans-splicing: a novel mode of pre-mRNA processing. Biol Cell. 2006;98:135–140. [PubMed]
9. Deidda G, Rossi N, Tocchini-Valentini GP. An archaeal endoribonuclease catalyzes cis- and trans-nonspliceosomal splicing in mouse cells. Nat Biotechnol. 2003;21:1499–1504. [PubMed]
10. Kleman-Leyer K, Armbruster DW, Daniels CJ. Properties of H. volcanii tRNA intron endonuclease reveal a relationship between the archaeal and eucaryal tRNA intron processing systems. Cell. 1997;89:839–847. [PubMed]
11. Fabbri S, et al. Conservation of substrate recognition mechanisms by tRNA splicing endonucleases. Science. 1998;280:284–286. [PubMed]
12. Tocchini-Valentini GD, Fruscoloni P, Tocchini-Valentini GP. Structure, function, and evolution of the tRNA endonucleases of Archaea: An example of subfunctionalization. Proc Natl Acad Sci USA. 2005;102:8933–8938. [PMC free article] [PubMed]
13. Tocchini-Valentini GD, Fruscoloni P, Tocchini-Valentini GP. Coevolution of tRNA intron motifs and tRNA endonuclease architecture in Archaea. Proc Natl Acad Sci USA. 2005;102:15418–15422. [PMC free article] [PubMed]
14. Tocchini-Valentini GD, Fruscoloni P, Tocchini-Valentini GP. The dawn of dominance by the mature domain in tRNA splicing. Proc Natl Acad Sci USA. 2007;104:12300–12305. [PMC free article] [PubMed]
15. Reyes VM, Abelson J. Substrate recognition and splice site determination in yeast tRNA splicing. Cell. 1988;55:719–730. [PubMed]
16. Mattoccia E, Baldi MI, Gandini-Attardi D, Ciafré S, Tocchini-Valentini GP. Site selection by the tRNA splicing endonuclease of Xenopus laevis. Cell. 1988;55:731–738. [PubMed]
17. Baldi MI, Mattoccia E, Bufardeci E, Fabbri S, Tocchini-Valentini GP. Participation of the intron in the reaction catalyzed by the Xenopus tRNA splicing endonuclease. Science. 1992;255:1404–1408. [PubMed]
18. Bufardeci E, Fabbri S, Baldi MI, Mattoccia E, Tocchini-Valentini GP. In vitro genetic analysis of the structural features of the pre-tRNA required for determination of the 3′ splice site in the intron excision reaction. EMBO J. 1993;12:4697–4704. [PMC free article] [PubMed]
19. Fruscoloni P, Baldi MI, Tocchini-Valentini GP. Cleavage of non-tRNA substrates by eukaryal tRNA splicing endonucleases. EMBO Rep. 2001;2:217–221. [PMC free article] [PubMed]
20. Di Segni G, Borghese L, Sebastiani S, Tocchini-Valentini GP. A pre-tRNA carrying intron features typical of Archaea is spliced in yeast. RNA. 2005;11:70–76. [PMC free article] [PubMed]
21. Kurjan J, Hall BD, Gillam S, Smith M. Mutations at the yeast SUP4 tRNATyr locus: DNA sequence changes in mutants lacking suppressor activity. Cell. 1980;20:701–709. [PubMed]
22. Carrara G, Calandra P, Fruscoloni P, Tocchini-Valentini GP. Two helices plus a linker: A small model substrate for eukaryotic RNase P. Proc Natl Acad Sci USA. 1995;92:2627–2631. [PMC free article] [PubMed]
23. Soma A, et al. Permuted tRNA genes expressed via a circular RNA intermediate in Cyanidioschyzon merolae. Science. 2007;318:450–453. [PubMed]
24. Hawthorne DC, Leupold U. Suppressors in yeast. Curr Top Microbiol Immunol. 1974;64:1–47. [PubMed]
25. Di Segni G, McConaughy BL, Shapiro RA, Aldrich TL, Hall BD. TAP1, a yeast gene that activates the expression of a tRNA gene with a defective internal promoter. Mol Cell Biol. 1993;13:3424–3433. [PMC free article] [PubMed]
26. Aldrich TL, et al. Structure of the yeast TAP1 protein: Dependence of transcription activation on the DNA context of the target gene. Mol Cell Biol. 1993;13:3434–3444. [PMC free article] [PubMed]
27. Amberg DC, Goldstein AL, Cole CN. Isolation and characterization of RAT1: An essential gene of Saccharomyces cerevisiae required for the efficient nucleocytoplasmic trafficking of mRNA. Genes Dev. 1992;6:1173–1189. [PubMed]
28. Elion EA. Pheromone response, mating and cell biology. Curr Opin Microbiol. 2000;3:573–581. [PubMed]
29. Dietzel C, Kurjan J. Pheromonal regulation and sequence of the Saccharomyces cerevisiae SST2 gene: a model for desensitization to pheromone. Mol Cell Biol. 1987;7:4169–4177. [PMC free article] [PubMed]
30. Dohlman HG, Thorner JW. Regulation of G protein-initiated signal transduction in yeast: Paradigms and principles. Annu Rev Biochem. 2001;70:703–754. [PubMed]
31. Trotta CR, et al. The yeast tRNA splicing endonuclease: a tetrameric enzyme with two active site subunits homologous to the archaeal tRNA endonucleases. Cell. 1997;89:849–858. [PubMed]
32. Di Nicola Negri E, et al. The eucaryal tRNA splicing endonuclease recognizes a tripartite set of RNA elements. Cell. 1997;89:859–866. [PubMed]
33. Li H, Trotta CR, Abelson J. Crystal structure and evolution of a transfer RNA splicing enzyme. Science. 1998;280:279–284. [PubMed]
34. Cox JS, Walter P. A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell. 1996;87:391–404. [PubMed]
35. Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell. 2001;107:881–891. [PubMed]
36. Gonzalez TN, Sidrauski C, Dorfler S, Walter P. Mechanism of non-spliceosomal mRNA splicing in the unfolded protein response pathway. EMBO J. 1999;18:3119–3132. [PMC free article] [PubMed]
37. Sidrauski C, Cox JS, Walter P. tRNA ligase is required for regulated mRNA splicing in the unfolded protein response. Cell. 1996;87:357–359. [PubMed]
38. Goodenbour JM, Pan T. Diversity of tRNA genes in eukaryotes. Nucleic Acids Res. 2006;34:6137–6146. [PMC free article] [PubMed]
39. Randau L, Munch R, Hohn MJ, Jahn D, Söll D. Nanoarchaeum equitans creates functional tRNAs from separate genes for their 5′- and 3′-halves. Nature. 2005;433:537–541. [PubMed]
40. Randau L, et al. The heteromeric Nanoarchaeum equitans splicing endonuclease cleaves noncanonical bulge-helix-bulge motifs of joined tRNA halves. Proc Natl Acad Sci USA. 2005;102:17934–17939. [PMC free article] [PubMed]
41. International Human Genome Sequencing Consortium. Initial sequencing and analysis of the human genome. Nature. 2001;409:860–921. [PubMed]
42. Smit AF, Riggs AD. MIRs are classic, tRNA-derived SINEs that amplified before the mammalian radiation. Nucleic Acids Res. 1995;23:98–102. [PMC free article] [PubMed]
43. Sela N, et al. Comparative analysis of transposed element insertion within human and mouse genomes reveals Alu's unique role in shaping the human transcriptome. Genome Biol. 2007;8:R127. [PMC free article] [PubMed]
44. Hasler J, Samuelsson T, Strub K. Alu RNAs in the human transcriptome. Cell Mol Life Sci. 2007;64:1793–1800. [PubMed]
45. Dagan T, Sorek R, Sharon E, Ast G, Graur D. AluGene: A database of Alu elements incorporated within protein-coding genes. Nucleic Acids Res. 2004;32:D489–D492. [PMC free article] [PubMed]
46. ENCODE Project Consortium. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature. 2007;447:799–816. [PMC free article] [PubMed]
47. Tress M, et al. The implications of alternative splicing in the ENCODE protein complement. Proc Natl Acad Sci USA. 2007;104:5495–5500. [PMC free article] [PubMed]
48. Fitzgerald C, et al. Mammalian transcription in support of hybrid mRNA and protein synthesis in testis and lung. J Biol Chem. 2006;281:38172–38180. [PMC free article] [PubMed]
49. James P, Hall BD. ret1–1, a yeast mutant affecting transcription termination by RNA Polymerase III. Genetics. 1990;125:293–303. [PMC free article] [PubMed]
50. Medici R, Bianchi E, Di Segni G, Tocchini-Valentini GP. Efficient signal transduction by a chimeric yeast-mammalian G protein α subunit Gpa1-Gsα covalently fused to the yeast receptor Ste2. EMBO J. 1997;16:7241–7249. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links